Glosario Webislam

hgira [hira] emigracin de los musulmanes de Meca a Medina. El muhir (emigrante) es el que emula a Haar. El Profeta Muhammad llega a dicha ciudad el sexto da desde que parti de Meca. Este acontecimiento histrico marca el inicio del calendario islmico (hay quienes lo sitan el da 1 del mes de muharram, aunque el consenso no es total). Si as fuera, estaramos hablando del 28 de julio del 622 de la era cristiana, ao 4.382 del calendario judo. Esta emigracin sac a los musulmanes de la clandestinidad y les permiti emanciparse. Ha habido dos emigraciones en el Islam, durante la vida del Profeta, y sta fue la segunda. La primera fue a Abisinia, donde el rey cristiano de ese pas los acogi y dio asilo a los ms vulnerables, aquellos que no tenan algn pariente que los protegiera de los ataques (embargos, torturas, muertes) que sufran por haberse hecho musulmanes.

hanif [hanf] (plur. hunaf), abrahmico, unitario, monotesta. Es alguien que sigue la tradicin de Abraham (Ibrhm) en un entorno caracterizado por el ahl y la ausencia de Islam. Se usa para referirse al Islam y al musulmn en el sentido de que el Islam no es algo nuevo, sino la restauracin de lo mismo que trajo Abraham (Ibrhm). Actuar como hanf es trabajar a diario despojndose de todo lo superfluo, de todo lo que se convierte en una adherencia que nos ata a ciertos convencionalismos o prejuicios y nos aleja de Al-lh. La mayor dificultad reside en que el ser humano se aferra con vehemencia a cualquier concepto, idea o imaginacin sobre lo que cree que le ofrece cierta seguridad o bienestar. Su precariedad aflora en todo momento y su impotencia pretende substituir con algo que supone que le beneficia. Puede convertir en dolo cualquier cosa. Aquello por lo que se sacrifica y le impide ser l mismo en su dimensin ms ascendente, es un dolo. El Faran o el becerro de oro slo son ejemplos de lo que el hombre es capaz en su autoengao. Su inclinacin a fabricar dolos, hace que quede prisionero en un mundo ilusorio del que slo puede liberarse cuando deja de escapar de s mismo y ya no teme enfrentarse a la realidad. El vrtigo que produce la sinceridad y la humildad del que se desnuda ante Al-lh y se arranca la coraza de las mentiras es la va por la que se llegara a la plenitud humana, es decir ms all del velo

cegador-limitador del mundo. La fragilidad y sensibilidad del hanf le acerca a la fitra perdida. hayy [ha] peregrinacin a Meca, que debe hacer el musulmn una vez en la vida, si puede. Es uno de los cinco pilares en los que se fundamenta la prctica del Islam. El ha para el musulmn supone el viaje al lmite, al lmite de lo sagrado, al lmite de lo humano, al lmite de las fuerzas, al lmite de s mismo. Es fsica y psquicamente una experiencia de extincin. El ha es la peregrinacin mayor, y la menor es la umra. En lo material, el ha es revivir la migracin de Agar (Haar). Nuevamente nos encontramos con la importancia fundacional de lo femenino en el Islam. hadra [hadra] presencia. En realidad, uno de los modos o niveles de la Presencia divina. Invocacin colectiva acompaada, a veces, con la danza o movimientos rtmicos, que se llama a su vez, en un trmino tcnico, imra (dar vida). Tambin es cualquier percepcin de energa viva no siempre visible materialmente, pero que se hace notar, por el odo, por la piel, o manifestando a travs de cualquier cosa, de un objeto, del aire, de la luz, etc islam [islm] la abdicacin del ser humano ante Al-lh. Viene de la raz salama (estar sano y salvo). Los significados etimolgicos derivados de Islm hacen referencia no slo a la salud sino tambin a la paz. Textualmente, islm es la aplicacin de la paz de Al-lh, del salm. El Islm es la tradicin espiritual que vehicula desde una cadena proftica universal, el mensaje divino revelado a todos los pueblos de la humanidad de todos los tiempos. Es una cosmovisin unificadora que tiende a reunir la multiplicidad creada en su Centro, y, an expandindose infinitamente, no se fractura ni se separa de la Razn de su origen ni de la energa generatriz que lo mueve. Quien acepta (abraza) y practica el Islm es musulmana/musulmn y vive con la certeza de que No hay dioses, hay Al-lh, y Muhammad es uno de Sus Enviados. Esta declaracin o testimonio se llama shahda. Cuando alguien la pronuncia con su boca y la contiene en su corazn est en el camino del Islm. Los arkn(sing. rukn),los pilares del Islm, son: testimoniar el Islam (shahda), azal (sal), azaque (ak), peregrinacin a Meca y ayuno de ramadn. yahila [hilya] tiempo de la insensatez, de la inmadurez. Aunque se define as la poca preislmica, como tiempo de la ignorancia, conviene recordar que ahl no es ignorancia en su origen. El ahil no es el ignorante sino el desbocado que se deja llevar rpido por la ira (del verbo rabe enfadarse con facilidad) desconociendo la autntica naturaleza armnica de las cosas. No se opone a ilm, sino a hilm, es decir, a la sensatez del que no se encoleriza fcilmente, del paciente. A los rabes de antes de la Revelacin la Kaba en Meca les recordaba a Allh, pero perdidos y aislados en sus mundos rotos sin el lazo de la Unicidad que lo conjuga todo, caan con facilidad en reconocer el seoro de unos hombres sobre otros, de las fuerzas de la naturaleza, de seres invisibles..., dando preeminencia al poder, la fortuna o el xito, y se desviaban de la orientacin unitaria, dispersando su atencin. As es como el ahil (inmaduro) se convierte en mushrik (idlatra; Ver shirk). meca [makka al-mukarrama] ciudad de Arabia donde naci Muhammad. En el Corn se usa la palabra bakka primer nombre de Meca- para designar el lugar donde dam estableci la primera casa de culto al Uno para la humanidad. Siglos ms

tarde, cuando todava el valle de Meca era un inhspito desierto, fue all mismo donde Har (la segunda mujer de Abraham) se asent con su hijo Isml (de quien descienden los rabes). Ella es, en realidad, la fundadora de esta ciudad. La tradicin nos cuenta que Har, expulsada al desierto de la casa de Abraham, recorri el valle de Meca circunvalando las dos colinas de Saf y Marua buscando sustento para no morir con su hijo a cuestas. Hasta que encontr por una revelacin divina- el lugar donde brota el agua de amam (borboteo). Para cumplir con el ha,los peregrinos musulmanes repiten los gestos de Haar como ritual, segn nos ense a hacer Muhammad. Las azoras o surat del Corn reveladas en este perodo son, en general, visiones apocalpticas del Profeta. Algunos lo interpretan como predicciones, y aunque pueden considerarse como advertencias o incluso como amenazas, tambin es interesante tener en cuenta que este tipo de acontecimientos ocurren ya en el mismo instante en que se produce la Revelacin en Muhammad. shahada [shahda] testimonio. Reconocer el Islam como la va que se va a seguir. Con la shahda uno no dice en qu cree sino cmo experimenta el mundo. Lo que te lleva a la shahda es verificar que el mundo tiene un orden interno, una lgica, un sentido. Cuando pronuncias la shahda no aceptas un credo, sino que declaras pblicamente tu intencin de seguir un camino espiritual con seriedad y sin concesiones. salat [sal] tambin transcrita salat y mal traducida por oracin o rezo, es una ibda (acto ritual) trascendental y uno de los pilares en los que se fundamenta la prctica del islam (arkn al-islm). Se realiza en direccin a Meca. El estado de despertar del cuerpo, al realizar las abluciones, crea un vaciamiento que nos abre a Al-lh. Se puede realizar individualmente, pero se aconseja practicarla en grupo (ama). Jams puede decirse que se realice en soledad, porque cuando se hace el universo entero se nos une para alcanzar con nuestros gestos la Unidad ms absoluta (tauhd). Debemos esperar la confluencia del todo, y ello es simbolizado por instantes especiales, que son cinco: el amanecer, el medioda solar, la tarde, el ocaso, el principio de la noche. Aunque el Corn slo habla de tres momentos, por eso algunas escuelas agrupan algunas veces las cinco veces en tres, como ocurre tambin en situaciones especiales, como el viaje. Es recomendable hacerla justo en su momento o poco despus, sin embargo, cuando no sea posible o exista alguna dificultad, existen opciones alternativas. En el Corn se acompaa frecuentemente esta palabra con el trmino ak, como dos pares de complementarios que se precisan uno a otro. Existen dos radicales distintas para el verbo que da origen a la palabra sal, que terminan en ya o enwau. La primera significa: tocar, alcanzar, ir detrs, seguir al primero (el asno que sigue a la hembra), o deslomar (tener la parte del lomo abatida, como ocurre en el ruku de la sal,en que se doblega la espalda).El plural de sal (salaw) significa sinagoga. Decimos ir a la sal, lo cual, implica movimiento. En cada movimiento se expresa una actitud que provoca un estado. La otra raz verbal es yasl: arder, calentar, asar, calentarse a la lumbre, sufrir ardores del fuego, acercar al fuego para asar, acercar a alguien al fuego con el fin de que se caliente; incluso para decir que se queme en el fuego del infierno en rabe se utlilizara el verbo sal (con ya). Si nos basamos en esta raz para la interpretacin del sentido del trmino sal como prctica islmica, significara consumirse en la inmensidad. Ms que oracin se convertira en una disciplina de ir incrementando nuestro sometimiento a la Divinidad. Durante su sal, la musulmana y el musulmn, no usan sus propias palabras sino las de Al-lh. No se sumerge en Al-lh quien habla desde su yo.

wudu [wud] ablucin ritual, previa a la sal. Dice un hadz que el wud es la mitad de la sal. El cuerpo reacciona y cambia su anterior situacin, lo cual predispone fsicamente a un nuevo estado propicio para ir a la sal limpio-libre-vaco de adherencias. Dicen los ntimos de Al-lh que se hace el wud para que el cuerpo est presente en la sal. El wud produce la sobriedad necesaria contra la ebriedad que nos ocasiona el duni (el mundo de los apegos), pues tenemos una prohibicin cornica de ir a la sal con ningn tipo de ebriedad. jutba [jutba] unas palabras que se dicen a la comunidad en la sal del medioda de los viernes (umua). Hay quienes lo traducen como sermn u homilia, no teniendo la jutba que ceirse al iguales que los sermones y las homilas- a temas religiosos. Muy al contrario, la segunda parte de la jutba de los viernes suele estar dedicada a comentar cuestiones sociopolticas que interesan a la comunidad. zakat [ak] azaque; porcentaje sobre los bienes de un musulmn que debe entregar obligatoriamente a la comunidad para paliar el estado de necesidad de algunos de sus miembros. Contribuye a una re-distribucin de la riqueza. Consiste en desprenderse aquella parte de los bienes propios que en realidad no nos pertenece. Uno de los pilares en los que se fundamenta la prctica del Islam. En el Corn se cita este trmino a menudo junto con la palabra sal (introspeccin con la que se busca a Al-lh). La sal y la ak son complementarios. Si la sal es introversin, la ak es la extroversin y el contacto reunificador con las criatura kaaba [kaba] el Cubo, edificacin construida por Abraham (Ibrhm) y su hijo Isml (Isml). Es con la expulsin de Haar de la casa de Ibrhm, Haar adquiere la condicin de mujer libre y busca para ella y su hijo un lugar donde asentarse. Tras el descubrimiento del pozo de amam, Haar se asentar en lo que luego ser Meca. A la muerte de Sara, Ibrhm vuelve a vivir all con Haar y edificar all con su primognito un monumento muy especial: la kaba, calificado por el Corn -nada menos que- como la Casa de Al-lh (baitu l-lh). El nombre de kaba hace referencia a su formadecubo y alberga una misteriosa piedra negra (al-haar al-asuad), incrustada en su parte oriental. Con el tiempo, hasta la llegada del Islam, fue convertida en templo de dioses. Un rayo la destruy y tras su reconstruccin, fue precisamente Muhammad, todava antes de que le llegara la revelacin, quien volvi a colocar la piedra en su lugar. Los musulmanes cuando hacen la peregrinacin(ha) la circunvalan, dando siete vueltas a su alrededor (tauwaf). La kaba es la orientacin (qibla) de la comunidad islmica al realizar la sal, y simblicamente cumple la funcin de centro del universo. s. saum [saum] ayuno. Ayunar en Ramadn (ver) es uno de los pilares que constituyen los fundamentos de la prctica del Islam (arkn al-Islm). Desde que empieza la luna de este mes hasta que se termina (29 o 30 das) los musulmanes se abstienen durante el da hasta el ocaso, de comer, beber, tener relaciones sexuales y en general, de todo lo que proporciona alimento o placer al cuerpo o que entretiene nuestros sentidos, agudizando as otras capacidades. No estn obligados a ayunar las mujeres durante la menstruacin, los nios, los enfermos, los viajeros, las embarazadas y las que estn dando de mamar. Es un ejercicio de disciplina fsica que

acta como depurativo y mtodo de autoconocimiento. Tambin se puede practicar voluntariamente fuera de este mes y hay das recomendados especialmente para ello. yihad [ihd] literalmente, y en su sentido fundamental, significa esfuerzo para lograr el salm. De la raz -H-D (esforzarse, aplicarse, insistir, trabajar con celo y asiduidad, empujar, fatiga, lucha). El ihd espiritual o ihd mayor (al-ihd al-akbar) se hace hacia dentro (hacia uno mismo). El muhid ha aceptado el devenir, el cambio, como su forma de vida, como su norma, como su paz. Para lograrla hay que combatir con vigor todo aquello que la obstaculice. kun [kun] verbo de Al-lh para crear el mundo, y que significa: S (t)!. O, incluso mejor, hzte!, plsmate!, pues est relacionado etimolgicamente con el hecho de constituirse, forjarse. Respecto a ese verbo prodigioso que hace emerger un universo, kun, Al-lh se dirige a s mismo y no al cosmos -que todava no es- ni a la nada -que no puede or ni obedecer. El mundo surge como una explosin desde lo ntimo de Al-lh, y llega hasta la aparicin de la conciencia, que es conciencia de ser humano. Como sucede con un ojo que todo lo puede ver excepto a s mismo, nos recuerda el clebre hadz: Yo era un tesoro escondido y quise ser conocido, y cre el mundo, y fui conocido. El mundo es necesariopara Al-lh porque no es algo substancialmente diferente ni Le resulta ajeno. Escriba el shaij al-Alaw que el mundo era una esterilla y Al-lh el esparto de la esterilla; la esterilla es slo el modo de trenzarse del esparto. E Ibn Aba plantea para esto una metfora: Al-lh sera como el sustantivo y el mundo un adjetivo, y al igual que un adjetivo no puede existir sin algo que calificar, un sustantivo alude a una realidad que -para existir- tiene que ser de algn modo. ramadn [ramadn] noveno mes del calendario islmico, en el cual se practica el ayuno (ver tambin saum). Fue durante el mes de ramadn cuando ya cumplidos los cuarenta aos- Muhammad se retir a la cueva de Hir(como haba hecho otras veces)y una noche (conocida como lailat al-qadr) desciendi la primera revelacin del Corn. A partir de entonces empez su misin como mensajero de Al-lh isnad [isnd]genealoga de una informacin sobre el Profeta, la cadena de transmisin de un hadz. sirat al mustaqin [sirt al-mustaqm] hasta ahora se ha traducido camino recto. Sirt significa camino (tambin el puente fino que pasa por encima del Pantano de fuego), y mustaqm viene del verbo qma-yaqm [levantarse, enderezarse], lo cual supondra ms que el sentido horizontal de camino recto, una fuerte verticalidad (ascendente o descendiente). El sirt al-mustaqm es algo que ante todo une el cielo con la tierra. hadiz [hadz] dicho del Profeta. Literalmente, relato o noticia. Constituye el fundamento de la tradicin (sunna), despus del Corn. Se divide en dos categoras: 1) El que la tradicin ha denominado quds, que son esos dichos del Profeta de temtica especialmente trascendente. Una palabra divina no-cornica (porque en estos hadices se dice que es Al-lh el que habla). 2) El hadz sharf (noble), clasificado segn su grado de autenticidad. El hadz sharf consta de dos partes: el matn (el mismo texto del acontecimiento relatado y que, habitualmente pone en escena al Profeta y a sus compaeros); y el isnd (apoyo) que contiene la cadena de transmisin y asegura la autenticidad del texto. Entre las recopilaciones de hadices se encuentran: Los Msnnaf (clasificados por temas) en los que los ms respetados son los de Muhammad

ibn Isml al-Bujr, y Abd al-Husain Muslim ibn Ha. Los sunan (recopilaciones que tratan especficamente de los actos y dichos de Muhammad) entre los que se cita a Ab Dd as-Siistn y a Ab s Muhammad al-Tirmidz. Los musnd son los de Mlik ibn Anas, Ibn Hanbal. Posteriormente, numerosas recopilaciones sern extradas de las primeras Corn [qurn] la revelacin de Al-lh al profeta Muhammad.Deriva de la misma raz que iqra, que es la primera palabra que revel ibrl a Muhammad. Qaraa significa recitar, transmitir oralmente, decir, citar, estudiar, reunir diversas partes de un todo (como se ha hecho compilando en forma de libro), hacer la sntesis, tener dentro (la mujer a su feto), tener sus reglas terminadas. Otra derivacin de la misma radical es qurn (ciclo menstrual de una mujer, ciclo, periodo, tiempo, momento, medida, rima de los versos). El Corn es llamado tambin Discernimiento (furqn), Descenso (tanl), Recuerdo (dzikr), Libro (kitb). El Corn fue revelado por fragmentos durante un periodo de 23 aos y no como un libro completo en una sola vez. Ha sido compilado a partir de lo que nos han transmitido un gran nmero de compaeros del Profeta. Dos han sido las formas de dicha transmisin: oral y escrita. La transmisin oral est basada en la memorizacin. Muhammad la memoriza cuando le llega, la hace pblica inmediatamente y pide a sus compaeros y compaeras que la memorizasen. Luego se repiten los pasajes cornicos aprendidos y se escuchan unos a otros. As es como lo haca el Profeta y as lo practicaron sus compaeros y los seguidores de estos. La memorizacin del Corn se hizo en vida de Muhammad y se sigue haciendo hasta hoy. La transmisin escrita se realiz escribiendo sobre diversos materiales (trozos de piel, omoplatos, hojas de palmera) los pasajes cornicos, a medida que se iban transmitiendo y memorizando. Antes de morir, Muhammad dej claro el orden final de la revelacin (que no fue cronolgico) y ms tarde fueron compilados los captulos en un solo volumen. No hubo en vida del Profeta ninguna duda ni disputa en cuanto al contenido integral del Libro y la ltima revelacin ocurri nueve das antes de su muerte, pero dejaba tras de s todas las partes del Corn conservadas por escrito y memorizadas sura [sra] azora castellanizado. Las surat son las 114 partes (captulos) que conforman el Coran. Estas surat estan a su vez compuestas por aleyas (ayt en plur. vide ya). Las ltimas surat del Corn son ms cortas y pertenecen al primer perodo de Meca, mientras que las reveladas en Medina son ms extensas y estn al principio del libro. aya [ya] aleya, signo (trazo para indicar, seal). El Corn est compuesto de yat (pl.), a menudo traducido como versculos. Pero este significado lo limita a una mera funcin literaria, como si el Corn se tratara simplemente de un libro. Sera ms apropiado usar la palabra castellanizada aleya que deriva del trmino rabe. En el fenmeno cornico cabe interpretar que cada ya se manifiesta como una seal de la Revelacin; una ya [femenino sing.] es un signo celeste que Al-lh hace estallar como garanta de su palabra. La raz verbal significa: resguardarse,retirarse, dar hospitalidad, reunirse, tratar con sentimiento de piedad y ternura, recibir en tu casa. Una aleya es un lugar de y (reposo, residencia). Cada ya es un signo, una casa para el sentido. Cada aleya es el retorno a la morada de los significados. En el secreto del signo hay una seal. Cuando la ya se abre te cobija; por eso el Corn es proteccin. Otro trmino cornico de la misma raz es maw: lugar en el que uno puede protegerse y al que puede acudir. En el Corn, los signos de Al-lh no son nicamente las palabras del soporte del Libro. Todo en la Creacin es una ya. A

travs de la cual, Al-lh se revela: En los cielos y en la tierra hay aleyas. Los ciclos de la naturaleza y lo que se despliega alrededor y en el interior de cada criatura son signos cornicos del gran Libro, cuyo conocimiento conecta al ser humano con el Todo. Los mundosse componen de yt (pl.). Y hay yt en cada persona (nafs) y en los fq el mundo material. A partir de esto es comprensible el amplio desarrollo de las ciencias naturales en el islam de los primeros siglos tras la Hgira (ver hira). Ibn Rushd (Averroes) consideraba a Aristteles como mutahid, que es quien hace itihd sobre el Libro de la Naturaleza fiqh [fiqh] derecho islmico. En el Corn significa conocimiento profundo, pero con el tiempo cambi su sentido y pas a significar conocimiento normativo del Islam. El fiqh permite la prctica del Islam y su concrecin como realidad comunitaria. Se centra, no en cuestiones abstractas sino en la realidad de la convivencia necesaria. El fiqh analiza la exterioridad del Islam de acuerdo a la enseanza de Muhammad. Durante la vida de Muhammad no existi, naturalmente, el fiqh como ciencia independiente: la Revelacin proporcionaba a la comunidad la respuesta a cada una de sus cuestiones. No obstante, aparecen dos principios formulados por el Profeta que son la base para el desarrollo posterior de la comunidad cuando la Revelacin a Muhammad quedase completada: el itihd (sentido crtico) y la shr (consulta asamblearia). Ambos pilares generan en el Islam el nacimiento de muchas escuelas o corrientes de fiqh. El experto en fiqh (al-faqh) responde al cmo del Islam, mientras que el fikr responde al por qu. Son disciplinas independientes entre s pero tambin auxiliares del fiqh: el tasauwuf o aprendizaje del Islam a travs de una cadena inicitica, el tafsr o exgesis cornica, los usl o criterios para fundamentar una argumentacin, las diversas ciencias del hadz que investigan las fuentes e incluyen las grandes compilaciones de la sunna y de sus maestros, etc. sharia [shara] lit. va que conduce al abrevadero. El musulmn se protege con la ley que rige la vida diaria. La shara es el resultado de la permanente discusin que existe en el seno del Islam acerca de la adecuacin del Corn a los fines de la sociedad humana. El Tao de la antigua China, el Dharma budista, el Nomos griego, la shara islmica, la Halaha juda son diferentes conceptos que con pequeos matices, expresan un mismo principio fundamental: el orden, la armona, la ley primordial que gobierna el cosmos y la naturaleza deben extender su dominio a la sociedad humana. Esta ley universal se considera reflejo del Principio divino (Ver Corn 45:18). La shara no est relacionada solamente con la forma correcta de vivir, sino tambin con aquellos principios, leyes y ritmos, por los cuales las cosas son como son. Desde esta perspectiva, seguirla es andar por el nico camino que conduce a la fuente. De hecho, todo posee su propia shara. Por lo tanto la shara no es una ley para los seres humanos, sino para todo el cosmos. No hay que confundir esta Ley natural o sagrada, con las leyes que los hombres aplican concretamente en su contexto y tiempo, a pesar de que estas suelen estar forjadas bajo una tica o moral basada en principios fundamentales, que intentan coincidir o al menos mantener un cierto equilibrio con dicha ley natural. No se debe fragmentar la unin de los dos aspectos que forman la plenitud islmica, shara y haqqa (ver). adhn [adzn] la llamada para convocar a los musulmanes a la sal. Se hace desde las mezquitas, o desde cualquier lugar en donde se hace una sal pblica. Lo proclama en alto una persona, limitndose el resto a repetirlo en silencio. Quien anuncia el adzn est reproduciendo lo que hizo el Profeta, es decir, convocar a la humanidad al Islam. Para hacer la sal se necesita el permiso de Al-lh. Adzn viene de

idzn, que en cierto modo significa permiso (en el sentido de darse por enterado). Necesitas que algo fije tu caminar, que no seas guiado por el capricho. La denominacin de quien se encarga de decir el adzn tiene su propio trmino en castellano: almudano. La palabra rabe es muadzdzin (del verbo adzdzana: llevar algo a los odos). Vulgarmente, se le ha llamado tambin muecn, sin embargo se considera un galicismo. El primer almudano fue Bilal, compaero (sahb) de Muhammad. baya [baia] el pacto. Es el pacto inicitico por el cual el maestro tiende la mano al aspirante en la va que sigue como revivificador de la cadena proftica. La ceremonia de investidura a un rey en rabe se llama mubyaa. Es de la misma raz que ba yabu: vender (Venta se dice en rabe bai), porque una venta es un pacto. haram [harm] lit. aquello que excluye. De la raz H-R-M (harama, yahrimu) alejar, defender, excluir, provienen adjetivos como inviolable, reservado Los musulmanes hablamos del harm de Meca, que es el lugar donde se efecta el ritual de la peregrinacin. Protege la Kaba, que es el primer espacio sagrado construido para una Divinidad nica. Deriva de esta raz la palabra castellanizada harn, el lugar reservado para las mujeres y del que quedan excluidos los hombres. Harm no es un concepto moral, al menos de una moral abstracta y universal. Lo que es harm para unos est permitido para otros. El acceso a la Kaba es harm para el que no sea musulmn, pero est permitida al peregrino. El vino o el cerdo es harm para el musulmn, pero no para el dzimm cristiano de un pas islmico, a quien las autoridades de ese pas estn obligadas a garantizarle la proteccin de dichos bienes rasul [rasl] mensajero, alguien que tiene un mensaje que dar. A Muhammad se le llama El enviado de Al-lh: Raslul-lh. Su raz etimolgica da idea de movimientos amplios, airosos y rtmicos, caminar despacio, cabellos largos. Emisario que transmite una noticia, alguien encargado de una misin: mursalun. El trmino en femenino mursalt se refiere a aquellas realidades que se envan, como ngeles, vientos, etc. Se distingue entre dos tipos de profeca: la del rasl y la del nab (ver). El rasl es el portador de una risla (anuncio, carta). Es la misma risla de Al-lh enviada anteriormente, que durante el intervalo de tiempo entre los mensajes (fatra) las gentes han olvidado nabi [nab] profeta, anunciador, el que trae un naba (una noticia). Todo rasl es un nab, pero no todo nab es un rasl. El rasl, adems de tener algo que decir, tiene una gran misin que cumplir. sa es la diferenciacin tradicional, que llama rasl a todo fundador de una comunidad espiritual (judasmo, cristianismo, etc.). Ibn Arab de Murcia o Ibn Hazm de Crdoba comparten la creencia de la posible nubuwya de Mariam (Mara), ya que rene todas las condiciones: penetra en ella la Revelacin a travs de ibrl y concibe a Jess (s). De hecho, el Corn la nombra entre los Profetas. yinn [inn] (pl. unn) la leyenda traduce genios, pero esta acepcin nos llevara con demasiada facilidad a considerarlos seres de mitologa. Es atribuida al inn en el mundo islmico toda aquella energa est encubierta a nuestra comprensin que altera nuestro mundo y nos incomoda. Segn el Corn el inn es de la naturaleza del fuego, y por tanto destructiva, pero tiene la posibilidad de destruir para causar un bien si es un inn mumin (sometido a Al-lh) o para causar un dao si es un inn kfir (un shaitn: rebelde a Al-h). De la misma raz que inn son los trminos unna (escudo), ann

(feto: lo que est protegido en el interior de la madre), anna (Jardn: sombra protectora en el desierto), ann (corazn: sede de la intimidad, lugar que el ser humano protege de s mismo); y el verbo que lo rige es anna: envolver. No cabe la menor duda de que los inn fueron tenidos como seres protectores en la Arabia preislmica. El Corn mismo deja constancia de ello: Haba humanos varones que se refugiaban en los inn varones (72:6). Durante la hilya (etapa preislmica) los inn eran los seores de la tierra, y su tiempo propicio era la oscuridad, haciendo alusin al carcter envolvente de la noche y el reinado de los unn en ella. La Revelacin someti el poder omnmodo de estos seores de la tierra y los trasform en energa sutil y/o presencia fsica. En el idioma rabe, por tanto,el concepto que se esconde tras este trmino ha pasado de ser lo divino en s mismo (en poca preislmica) a categorizar (ya con la Revelacin) lo que va de la influencia sutil de lo anglico a lo que mueve a actuar al demonio de carne y hueso, aunque siempre bajo el dominio de Al-lh. Muestra una vez ms el Islam que no consiente la fractura entre los mundos, entre la materia y el espritu, conceptualizando como inn todo aquello que pertenece al mundo de energas que se escapan al hombre, sin determinarlo tampoco a nivel moral: un inn no es, as sin ms, una categorizacin positiva ni negativa. Dice Muhammad que cada uno de nosotros tiene un inn que lo acompaa y al que est atado; y cuando le preguntaron que si l tambin repondi: Incluso yo. Pero mi rabb me ha dado fuerzas con las que puedo someterlo, y ya slo me susurra el bien. ansar [ansr] viene de la raz de nasara (ver nasr): ayudar, auxiliar. Se usa en plural (sing: ansr) para referirse histricamente a los medinenses que acogieron en sus casas a los musulmanes huidos de Meca, tratndoles como a hermanos y compartiendo con ellos todo lo que tenan. qibla [qibla]
orientacin. Es el punto de referencia, el eje o punto de convergencia, hacia donde nos dirigimos los musulmanes para hacer la sal. Durante algn tiempo la qibla de los musulmanes fue Jerusaln y actualmente se sita en la Kaba(Meca).

umma [umma] comunidad que comparte un linaje espiritual. Tambin es familia, asamblea, matriz, pas. A partir de la poca de los nacionalismos se usan umma y qaum para traducir nacin.De su raz se extraen significados como: ir en cabeza, dar ejemplo, o ser madre masyid [masid] mezquita. Literalmente, lugar de sud (postracin). En un hadz quds (el hadz que se pone en boca de Al-lh) se nos dice: Cre la tierra como una mezquita, como un espacio sagrado-limpio. La mezquita, gracias a su simbologa y a su geometra, resume y recuerda el ritmo del cosmos, en una rplica humana, as como la ciudad lo es del universo y el jardn lo es del paraso. En cierta forma, el masid no es sino una reproduccin de la mezquita primordial, que es la naturaleza virgen. Toda la creacin se muestra como espacio teofnico, como se desprende del citado hadz. El mundo entero constituye ese espacio sagrado. Cada musulmana y musulmn se considera a s mismo responsable (ver jalfa) de ese espacio y de ese modo se constituye en un autntico templario, es decir, cuidador del templo de la creacin. El masid es una construccin materializada por el hombre que simboliza esa mezquita telrica. din [dn] trmino que traducido como religin resulta confuso. El dn es una actitud social, una forma de enfrentarse a lo Absoluto.

Es de la familia lxica de dain, deuda. El dn es el cumplimiento de una deuda. No de una deuda con Dios, sino con el mundo y con uno mismo. Dn es tambin la Revelacin, el conjunto de enseanzas y prescripciones con las que los musulmanes se caminan hacia la Unidad. Es de la raz de un verbo con sentidos diversos: endeudarse, prestar, imprimir, retribuir, someterse, juzgar.... rida [rid] satisfaccin, del verbo estar contento, estar satisfecho.Tambinradiatun: que colma o satisface.Es la alegra que encuentra el corazn ante los acontecimientos del destino; el abandono ante todo lo que dispone y ejecuta Al-lh en ti; la ausencia de desaprobacin en todo lo que te acontece y que proviene de Su Voluntad. Tambien puede decirse que ar-rid se puede entender como sinnimo de at-taslm (entrega), aunque es ms general en su acepcin, pues se ha dicho ar-rid se produce en el descenso del Destino, mientras que at-taslm se produce antes del descenso de aquel. La rid establece la relacin ntima con Al-lh (wilya): radi al-lhu anh wa rad anh (Al-lh est satisfecho con vosotros y vosotros con l). De la misma famila semntica ridwn. sunna [sunna]ejemplo de Muhammad, su modo de hacer las cosas (ver hadz). Tradicin, costumbre, hbito. A la Naturaleza en rabe se la llama sunnat Al-lh, el modo habitual de hacer las cosas Al-lh. sunna [sunna]ejemplo de Muhammad, su modo de hacer las cosas (ver hadz). Tradicin, costumbre, hbito. A la Naturaleza en rabe se la llama sunnat Al-lh, el modo habitual de hacer las cosas Al-lh sura [sra] azora castellanizado. Las surat son las 114 partes (captulos) que conforman el Coran. Estas surat estan a su vez compuestas por aleyas (ayt en plur. vide ya). Las ltimas surat del Corn son ms cortas y pertenecen al primer perodo de Meca, mientras que las reveladas en Medina son ms y estn al principio del libro mulk [mulk] el mundo de los seres humanos. El mulk y el mundo de los malika -llamado malakt- no son mundos diferentes, sino dos aspectos de lo mismo, es decir, del universo del poder, slo tenemos que darnos cuenta de que la raz de ambas palabras es la misma. Ambos mundos (mulk y malakt) responden a la triltera rabe M-L-K, que hace alusin a poder, reino, gobierno (de ah el trmino rabe malik, rey, en hebreo mlej). Si el mulk es el universo del poder aparente del ser humano, el malakt es el universo del poder anglico; pero ambos universos pertenecen a la expansin natural del ser humano Universal, que es el seor de los ngeles de la existencia. Fue por eso que Al-lh cuando cre a dam dijo a todos los malika y a todas las criaturas que ante l hicieran sud (postracin). tawwaf [tauwaf] girar alrededor (de la Kaba). Circunvalar. Es parte esencial del rito de la Peregrinacin mayor (ha) y menor (umra), y consiste en dar siete vueltas en torno a la casa construida por Abraham e Isml, en direccin contraria a las agujas del reloj. La energa de una multitud de personas alrededor de la Casa abrahmica buscando el Centro. Es como gravitar en el cosmos. Durante las siete vueltas, el peregrino se acerca, toca o besa la piedra incrustada en una de las esquinas del Cubo (Kaba). En el tauwaf el viajero vuelve hacia su s mismo, atrado como el

hierro por un imn hacia esa Kaba que simboliza el ombligo del mundo, el corazn del universo. Como los girasoles y los planetas, como el derviche. El musulmn tiene en el tauwaf la oportunidad de vivir fsicamente la necesidad y la tendencia natural de lo creado, mltiple, partido y separado, de encontrarse (o acercarse ms) a lo que ama realmente: la Unin.

Jerusalem (hebrisch /Jeruschalajim, arabisch /al-Quds (asch-Scharif), die Heilige, altgriechisch , lateinisch: Hierosolyma) ist mit 729.100 Einwohnern (2006) die Hauptstadt des Staates Israel. Jerusalem liegt in den Judischen Bergen zwischen dem Mittelmeer und dem Toten Meer. In der Stadt befindet sich der Sitz des Prsidenten ebenso wie der Legislative, der Judikative und der Exekutive. Jerusalem ist eine der ltesten Stdte der Welt: erste Erwhnungen gehen auf das 19. Jahrhundert v. Chr. zurck. Das Spannungsverhltnis zwischen Antike und Moderne ist in dieser Stadt mit ihrer multikulturellen und multiethnischen Bevlkerung besonders sichtbar. Die Altstadt ist von einer Mauer umgeben und hat vier Teile: das jdische, christliche, armenische und muslimische Viertel. Der politische Status der Stadt ist bis heute international umstritten. Dies betrifft insbesondere Ost-Jerusalem, welches bedeutende Heiligtmer der drei groen, monotheistischen Weltreligionen beherbergt.

Inhaltsverzeichnis
[Verbergen]

1 Aktueller rechtlicher Status Jerusalems 2 Geschichte

2.1 Geschichte bis zur Zerstrung durch die Rmer 2.2 Nach der Zerstrung des Tempels 2.3 Jngere Entwicklungen und derzeitiger Status 2.4 Einwohnerentwicklung 3 Religion o 3.1 Jerusalem als heilige Stadt 4 Kultur und Sehenswrdigkeiten o 4.1 Theater o 4.2 Museen o 4.3 Bauwerke 5 Wirtschaft und Infrastruktur o 5.1 Verkehr o 5.2 Bildung 6 Persnlichkeiten o 6.1 Shne und Tchter der Stadt 7 Quellen 8 Literatur 9 Weblinks
o o o o

10 Siehe auch

Aktueller rechtlicher Status Jerusalems

Die Klagemauer, im Hintergrund der Felsendom

Umgebung von Jerusalem; Karte von 1888

Jerusalem; Karte von 1888 Nach dem Ersten Weltkrieg wurde in Jerusalem der Sitz des Hohen Kommissars und der Mandatsverwaltung eingerichtet. Dies geschah auf Basis des Vlkerbundmandats fr Palstina. In den Jahren 194749 versuchten die Vereinten Nationen, Jerusalem ihrer Gebietshoheit zu unterstellen. Am 29. November 1947 nahm die UNVollversammlung den UN-Teilungsplan fr Palstina, die Resolution 181, an; es folgten die Resolutionen 194 vom 11. Dezember 1948 und 303 vom 9. Dezember 1949.[1] Der UN-Teilungsplan sah vor, Jerusalem zu einem corpus separatum zu machen, der unter einem besonderen internationalen Regime stehen und von den UN durch einen Treuhnderrat und einem Gouverneur verwaltet wrde. Die Stadt sollte Teil eines gemeinsamen Handelsraumes sein, Brgern beider Staaten sollte es erlaubt sein die Stadt zu betreten und dort zu wohnen. Der Plan war von dem Wunsch getragen, den Zugang zu den heiligen Sttten der drei Weltreligionen gleichberechtigt zu sichern. Die lokale Verwaltung lge bei den bestehenden autonomen Verwaltungseinheiten, die Stadt sollte demilitarisiert werden und neutral sein. Die Polizeieinheit der Stadt wrde aus auslndischen Truppen rekrutiert. Die Gesetzgebung sollte in den Hnden eines Rates liegen, der von den Stadtbewohnern nach den Regeln der Verhltniswahl gewhlt wrde. Gegen die Entscheidungen dieses Rates sofern sie den Status der Stadt betrfen behielten sich die UN ein Vetorecht vor. Der Teilungsplan wurde jedoch nie umgesetzt, sodass die Vereinten Nationen keine Souvernitt ber Jerusalem erlangten.

Stattdessen folgte die israelische Unabhngigkeitserklrung. In ihr wird Jerusalem nicht erwhnt, der Text spricht jedoch davon, dass Israel die heiligen Sttten aller Religionen beschtzen werde. Im Unabhngigkeitskrieg unmittelbar nach Grndung des Staates verlor Israel das jdische Viertel der Altstadt Jerusalems und den Osten der Stadt an die jordanische Arabische Legion. Die Stadt war deshalb zwischen 1948 und 1967 in das israelische Westjerusalem und das jordanische Ostjerusalem geteilt. 1948 erlie der israelische Verteidigungsminister eine Verordnung, dass im Westen der Stadt wie in jedem Teil Palstinas, den der Verteidigungsminister als von israelischen Truppen gehalten erklre, israelisches Gesetz gelte. Ende 1949 erklrte der Premierminister, David Ben Gurion, vor dem israelischen Parlament, der Knesset, Jerusalem zum untrennbaren Teil Israels und seiner ewigen Hauptstadt. Diese Position wurde vom Parlament besttigt. 1950 annektierte Abdallah ibn Husain I., der Knig von Jordanien, das von seinen Truppen eroberte Westjordanlandes und Ostjerusalem. Die Annexion wurde nur von Grobritannien und Pakistan anerkannt. Grobritannien betonte, dass sich seine Anerkennung der Annexion nicht auf Jerusalem beziehe. In den Jahren 19481952 wurde von den Vereinten Nationen mehrmals der Versuch unternommen, eine Lsung der Frage des Status Jerusalems herbeizufhren, dann aber ergebnislos eingestellt. Seit 1952 akzeptierte die internationale Staatengemeinschaft die de facto Anwendung israelischen Rechts in Westjerusalem. Die Forderung, die Stadt zu internationalisieren, war immer weniger mit der Realitt zu vereinbaren und wurde deshalb im Laufe der Zeit nicht mehr von den UN erhoben. Die israelische Position besagt, dass der Westen der Stadt ohne Souvern gewesen sei, als sich Grobritannien 1948 aus seinem vormaligen Mandatsgebiet zurckgezogen hatte, und Israel so in einem Akt der Selbstverteidigung gegen die angreifenden arabischen Armeen die rechtmige Souvernitt ber das Gebiet erhalten habe.

Jerusalem

Jerusalem nach 1967 Als im Juni 1967 der Sechstagekrieg ausbrach, kontaktierte Israel Jordanien ber die UN und die amerikanische Botschaft. Die israelische Regierung erklrte, Israel wrde Jordanien nicht angreifen, sofern Jordanien seinerseits davon absehe, Israel anzugreifen. Jordanien ignorierte dies, beschoss whrend des Krieges Westjerusalem und eroberte das neutrale Hauptquartier der UN, den ehemaligen Sitz des High Commissioners. Einige Tage spter wurde dieses Gebiet von der israelischen Armee zurckerobert und die jordanische Armee aus der gesamten Westbank einschlielich Jerusalems vertrieben; damit hatte Israel das gesamte ehemalige Mandatsgebiet erobert. Nach dem Ende des Krieges verabschiedete die Knesset das Law-and-AdministrationOrdinance-Gesetz, das es der Regierung erlaubte, das israelische Gesetz, Israels Jurisdiktion und Verwaltung auf alle Gebiete des ehemaligen Mandatsgebiets auszuweiten. Gleichzeitig wurde die Gemeindeverwaltungsordnung gendert, wodurch es mglich wurde, die Verwaltungsgrenzen Jerusalems auf den Osten der Stadt auszuweiten. Allerdings wurden bestimmte gesetzliche Arrangements zugunsten der arabischen Bewohner der Stadt beschlossen, die im Legal and Administrative Matters (Regulation) Law von 1970 festgeschrieben sind. Die arabischen Stadtbrger wurden auch nicht automatisch Israelis, es wurde ihnen jedoch ermglicht, recht unkompliziert die israelische Staatsbrgerschaft zu erwerben, wovon allerdings nur wenige Gebrauch machten. Der Auenminister Israels, Abba Eban, erklrte daraufhin in einem Brief vom

Juli 1967 an den UN-Generalsekretr, dass Israel Ostjerusalem nicht annektiert, sondern nur verwaltungstechnisch integriert habe. Trotzdem wurde dieser Schritt von UNEinrichtungen kritisiert. In der Resolution 242 des UN-Sicherheitsrates wird Jerusalem nicht explizit erwhnt. Die Position der israelischen Regierung ist, dass weder Jordanien und noch ein anderer Staat auer Israel jemals Souvernitt ber die Stadt erhalten habe. Jordanien habe Jerusalem 1948 in einem Akt der Aggression unter seine Kontrolle gebracht, wogegen Israel 1967 in Selbstverteidigung gehandelt habe und schon deshalb bessere Ansprche geltend machen knne. Annexionen sind nach dem Vlkerrecht allerdings auch nach Verteidigungskriegen verboten. Die israelische Position besagt, dass die Resolution 181 der Vollversammlung als vlkerrechtlich nicht bindendes Dokument keine Gltigkeit besitze und aufgrund der arabischen Ablehnung niemals relevant gewesen sei, weshalb auch der Status Jerusalems als corpus separatum obsolet worden sei. Darber hinaus gebe es weder einen vlkerrechtlichen Vertrag dahingehend, noch sei der Status Jerusalems als corpus separatum Vlkergewohnheitsrecht. Bezglich der heiligen Sttten wurde von der Knesset 1967 das Preservation of the Holy Places Law erlassen, das den freien Zugang zu diesen und deren Schutz vor Entweihung gewhrleistet. In Berufung auf dieses Gesetz verhindert die israelische Polizei, um die ffentliche Ordnung und die Sicherheit zu gewhrleisten, dass nationalreligise Juden wie die in der Nationalist Groups Association organisierten auf dem Tempelberg ffentliche Gottesdienste abhalten. Im Vertragswerk von Camp David wurde Jerusalem ausgeklammert. In den beigefgten Briefen an den Gastgeber von Camp David, den Prsidenten der USA, Jimmy Carter, erklrte Menachem Begin fr Israel, dass Jerusalem die unteilbare Hauptstadt Israels sei. Sadat erklrte, dass das arabische Jerusalem ein integraler Teil des Westjordanlands sei und unter arabischer Souvernitt stehen solle. Er sprach sich jedoch gleichzeitig dafr aus, bestimmte Funktionen der Stadt einem gemeinsamen Rat zu berantworten. In diesem Sinne solle die Stadt ungeteilt sein, schrieb Sadat. Im Jahre 1980 verabschiedete die Knesset ein Gesetz von Verfassungsrang, das ganz Jerusalem (also auch Ostjerusalem) zur ewigen und unteilbaren Hauptstadt Israels erklrte. Mit der Resolution 478 verurteilte der UN-Sicherheitsrat diese Annexion OstJerusalems und forderte als Strafmanahme alle Staaten, deren Botschaften ihren Sitz in Jerusalem hatten, dazu auf, diese aus Jerusalem abzuziehen. Zu diesem Zeitpunkt hatten von 45 Staaten 13 den Sitz ihrer Botschaften in Jerusalem: Bolivien, Chile, Kolumbien, Costa Rica, die Dominikanische Republik, Ecuador, El Salvador, Guatemala, Haiti, die Niederlande, Panama, Uruguay und Venezuela. Alle anderen Botschaften hatten ihren Sitz in Tel Aviv. Alle 13 betroffenen Staaten folgten der Resolution. 1982 verlegten zwei Staaten, Costa Rica und El Salvador, ihre Botschaften zurck nach Jerusalem, revidierten diese Entscheidung im Sptsommer 2006 jedoch wiederum und verlegten ihre Botschaften erneut zurck nach Tel Aviv. Es befinden sich Generalkonsulate von Griechenland, Grobritannien und den USA in Jerusalem. Der US-Kongress beschloss 1995, die US-Botschaft nach Jerusalem zu verlegen, da Israel wie alle Staaten das Recht habe, seine Hauptstadt selbst zu bestimmen. Aus Furcht vor den auenpolitischen Folgen wurde diese Willenserklrung jedoch bis heute (2006) nicht umgesetzt.

1988 gab Jordanien seinen Anspruch auf Souvernitt ber das Westjordanland und damit auch Ost-Jerusalems auf. Im selben Jahr rief die PLO den Staat Palstina aus und erklrte Jerusalem zu seiner Hauptstadt, was zu diesem Zeitpunkt obwohl diese Unabhngigkeitserklrung von vielen arabischen Staaten anerkannt wurde reine Fiktion war. Vlkerrechtlich mssen neben der Ausrufung eines Staates vier Voraussetzungen erfllt sein, um einen Staat entstehen zu lassen: Es muss ein Staatsgebiet sowie ein Staatsvolk geben, ber die es eine effektive Regierung und Kontrolle gibt. Auerdem muss der neue Staat die Fhigkeit besitzen, internationale Beziehungen einzugehen. Die PLO war zu diesem Zeitpunkt weit davon entfernt, effektive Kontrolle ber irgendeinen Teil der umstrittenen Gebiete auszuben. In der Declaration of Principles, die am 13. September 1993 zwischen Israel und der PLO unterzeichnet wurde, wird die palstinensische Selbstverwaltung, wie sie in zwei Formen fr das Westjordanland festgeschrieben wurde (Gebiete A und Gebiete B), fr keinen Teil Jerusalems bestimmt. Der Endstatus der Stadt soll im Zuge des OsloFriedensprozesses in einem endgltigen Vertrag bestimmt werden. Die Declaration of Principles erlaubt es allerdings den palstinensischen Brgern Jerusalems, nach einem Extra-Abkommen zwischen den beiden Seiten an den Wahlen zur Palstinensischen Autonomiebehrde teilzunehmen.

Geschichte
Geschichte bis zur Zerstrung durch die Rmer
Dieser Artikel oder Abschnitt bedarf einer berarbeitung. Nheres ist auf der Diskussionsseite angegeben. Hilf bitte mit, ihn zu verbessern, und entferne anschlieend diese Markierung.

Jerusalem im 1. Jahrhundert Die Existenz Jerusalems als kanaanischer Stadtstaat (akkadisch KURU-ru-a10-limKI) ist durch gyptische Quellen seit dem 18. Jahrhundert v. Chr. als Uruschalim belegt. Der Name Uruschalim/Jerusalem (nach der Bibel vor der Eroberung durch David Salem oder Jebus) bedeutet Stadt des Schalim oder Stadt des Friedens. Schalim war ein kanaanitischer Gott der Abenddmmerung. Schalom bedeutet im Hebrischen, Salam im Arabischen Frieden. In den Amarna-Briefen wird ein gewisser Abdi-Heba als Herrscher von Jerusalem erwhnt, der Probleme mit den Hapiru hatte. Dieser Name ist

hurritisch, dies muss jedoch nicht unbedingt etwas ber die ethnische Zugehrigkeit des Trgers aussagen. Gegen Ende des 2. Jahrtausends v. Chr gehrte die Stadt nach dem biblischen Bericht den von JHWH zur Vernichtung vorgesehenen Jebusitern. Der Bibel nach entriss Knig David Jerusalem um das Jahr 997 v. Chr. den Jebusitern. Er machte es zur Davidsstadt und zum politischen und religisen Mittelpunkt des Israelitenreiches. Fr die jdische Religion und Kultur ist Jerusalem seitdem Zentrum und Hauptstadt. Sein Sohn, Knig Salomo (um 969930), erbaute einen Palast und einen ersten Tempel fr Jahwe. Der Tempelberg ist heute dem islamischen Waqf unterstellt, Ausgrabungen sind dort nicht mglich. Der Waqf erstellte in den vergangenen Jahren eine neue Moschee in den so genannten Stllen Salomos, was wegen der mglichen unbemerkten Zerstrung von Resten der beiden jdischen Tempel auf israelische Ablehnung stie. Allerdings drfte bereits der Bau des herodianischen Tempels zu einer weitgehenden Beseitigung frherer Spuren gefhrt haben. Dessen nochmalige Zerstrung, die Errichtung eines rmischen Heiligtums und schlielich die islamischen Bauarbeiten drften allerdings wenig berreste frherer Zeiten brig gelassen haben. Knig David verlegte im Jahre 1003 v. Chr. die Hauptstadt seines Reiches von Hebron nach Jerusalem. Bis zur Herrschaft der Babylonier blieb die Stadt ca. 400 Jahre unter jdischer Herrschaft. Auch nach der Rckkehr aus dem babylonischen Exil diente wieder Jerusalem als Hauptstadt der Juden. Die Rmer und Byzantiner, sechshundert Jahre die Herrscher ber Palstina, machten Caesarea zur Hauptstadt. Nach Salomos Tod 926 v. Chr. und der Spaltung des Knigreichs in die Staaten Juda (Sden) und Israel (Norden) wurde Jerusalem die Hauptstadt des Sdreiches Juda. Knigin Atalja (845840) entweihte den Tempel und fhrte im Tempel den Baalskult ein. Unter Knig Ahas (741725) wurden vielleicht auch assyrische Gtter verehrt. Erst Hiskija (725697) reinigte den Tempel und sicherte die Stadt durch Mauern und einen Tunnel zur Wasserversorgung. Joschija machte 628 v. Chr. Jerusalem zur alleinigen legitimen israelitischen Kultsttte statt der frheren weiteren Heiligtmer auch auf Anhhen. Im Knigreich Israel wurde ein Heiligtum in Jerusalem, auerhalb der eigenen Grenzen, abgelehnt - abgesehen davon, dass der wirtschaftlich, militrisch und politisch lange Zeit fhrende Staat vor allem unter den Omriden Israel mit Zentrum in Samaria und nicht das Knigreich Juda war. Nebukadnezar II. eroberte Jerusalem zweimal (605 v. Chr. und 597 v. Chr.), fhrte die jdische Oberschicht in die Gefangenschaft und setzte Zedekia als Vasallenknig ein. Nach dessen Bruch mit den Babyloniern wurde Jerusalem 586 v. Chr. [2] von den Babyloniern eingenommen. Die Stadt und insbesondere der Tempel wurden zerstrt. Nach der Einnahme des babylonischen Reichs durch Kyros II. 538 v. Chr. zogen sich die Wiederaufbauarbeiten des Tempels in Jerusalem ber mehrere Jahrzehnte hin (siehe auch: Kyros-Edikt und Nehemia). Dabei trennte man sich von den als mit den Nachbarn blutvermischt und hretisch angesehenen Bewohnern des ehemaligen Knigreiches Israel, die sich hierauf ihr eigenes Heiligtum auf dem Garizim schufen. Unter rmischer Herrschaft wurde der von Herodes dem Groen ausgestattete zweite Tempel im Jahre 70 n. Chr. am Ende des Jdischen Krieges durch Titus zerstrt.

Nach der Zerstrung des Tempels

Hadrian verbot nach dem Bar-Kochba-Aufstand Juden den Zutritt zur Stadt und benannte sie in Aelia Capitolina um. Auf dem Tempelberg wurde ein rmischer Tempel zu Ehren von Jupiter erstellt. Die jdischen Bewohner emigrierten in die jdische Diaspora rund ums Mittelmeer, viele wanderten auch ins Perserreich aus. Nachdem Kaiserin Helena im Heiligen Land Grabungen veranlasst hatte, lieen sie und ihr Sohn Konstantin am Ort der vermuteten Kreuzauffindung die Grabeskirche erbauen.

Hartmann Schedel: Die Zerstrung von Jerusalem (1493)

Die lteste gedruckte Stadtansicht von Jerusalem von Hartmann Schedel, Nrnberg 1493 Nach einer kurzen Besetzung durch die Perser (614628) und ihre jdischen Verbndeten, in deren Verlauf es zu Massakern an Christen gekommen war,[3] wurde die Stadt nach dem Sieg des ostrmischen Kaisers Herakleios an Byzanz zurckgeben. Im Jahre 638 eroberten Araber im Zuge der islamischen Expansion die Stadt; sie wurde vom Patriarchen Sophronius (560638) freiwillig bergeben, nachdem keine Hoffnung auf Entsatz mehr bestanden hatte und die Araber dem Abzug der Christen, welche die Stadt verlassen wollten, zugestimmt hatten. Unter muslimischer Herrschaft gab es sowohl Phasen von Toleranz gegenber christlichen Pilgern und Bewohnern als auch Phasen einer explizit christenfeindlichen Haltung. So wurde Kaiser Karl der Groe (9. Jahrhundert) vom muslimischen Herrscher als Schirmherr der heiligen Sttten eingesetzt. Hingegen wurde im Jahr 1009 die Grabeskirche auf Befehl des FatimidenKalifen al-Hakim zerstrt, was auch den Ansto zum Kreuzzug gab. Im Verlauf des ersten Kreuzzuges eroberten die Kreuzritter unter Gottfried von Bouillon 1099 Jerusalem und richteten dort ein Blutbad unter der Bevlkerung an. Innerhalb von drei Tagen wurden bis zu 20.000 Bewohner gettet. Nach der christlichen Eroberung von Jerusalem grndeten die Kreuzritter das Knigreich Jerusalem. Im Jahre 1187 gelang es Saladin (arabisch Salah ad-Din Yusuf ibn Ayub), dem kurdischstmmigen Sultan von gypten, Jerusalem zu erobern. Er lie nach seinem Sieg ber die Kreuzfahrer das von ihnen errichtete goldene Kreuz auf der Kuppel des Felsendoms des Templum Domini - der Kirche der Kreuzfahrer - und die Marmorverkleidung des Felsens samt Altar entfernen. Ein kurzes Intermezzo bildete die Herrschaft von Kaiser Friedrich II. von 1229 bis 1244.

Im Jahre 1244 besetzten die Mamluken, eine trkische Sldnerschicht, welche nach dem Tode Saladins die Herrschaft in gypten bernommen hatten, die Stadt. Vom 13. bis zum frhen 16. Jahrhundert wechselten sich Ayyubiden, Mamluken, Tataren und Mongolen in der Herrschaft ab. Jerusalem hatte damals weniger als 10.000 Einwohner und keine politische Bedeutung. Unter muslimischer Herrschaft galten nur die Muslime als vollgltige Brger. Christen und Juden mussten sich durch ihre Kleidung kenntlich machen. Sie durften ihre Religion als Anhnger einer Buch-Religion zwar im Allgemeinen ausben, wurden aber rechtlich in fast allen Lebensbereichen diskriminiert und mussten eine Zusatzsteuer zahlen. Dennoch existierte in dieser Zeit immer ein christliches und ein jdisches Viertel in der Stadt und ein stndiger, wenn auch kleiner Strom von christlichen und jdischen Besuchern und Pilgern. Im Jahre 1516 wurden die Mamluken in Syrien von den osmanischen Trken besiegt. Sultan Selim I. (14651520) gewann gypten, Arabien und Syrien. Jerusalem wurde zum Verwaltungssitz eines osmanischen Sandschaks (Regierungsbezirk). Die ersten Jahrzehnte der trkischen Herrschaft brachten Jerusalem einen deutlichen Aufschwung. Nach 1535 lie Sultan Sleyman I. (14961566) die Befestigungen der Stadt in zum Teil vernderter Linie erneut errichten, so wie sie gegenwrtig zu sehen sind. Durch diese Mauern erhielt die Altstadt ihre heutige Struktur. Die viel zu groen neuen Mauern um den heiligen Symbolort sollten fr die neue Herrschaft ein Zeichen setzen. Jerusalem gewann in der Folgezeit viel an Bedeutung. Die osmanische Verwaltung war sich uneinig in ihrer Haltung gegenber den Juden und Christen und schwankte zwischen Gewaltherrschaft und Toleranz. Die verarmten Juden und Christen lebten berwiegend vom Pilgergewerbe. Der Besitz der Heiligtmer Jerusalems war wegen der damit verbundenen Almosen eine lebenswichtige Einnahmequelle. Dies erklrt auch die damals teilweise erbitterten, manchmal gewaltsamen Konflikte unter den christlichen Kirchen um einzelne Besitzrechte. Ab 1860 kamen durch die zionistische Alija (Einwanderung) immer mehr Juden in die Stadt, und es wurden erste Wohngebiete auerhalb der Stadtmauern gegrndet. Jerusalem blieb osmanische Stadt, bis sie am 9. Dezember 1917 von britischen Truppen unter Fhrung des Generals Edmund Allenby besetzt wurde, die im Ersten Weltkrieg gegen das Osmanische Reich kmpften.

Jngere Entwicklungen und derzeitiger Status

Ben-Yehuda-Strae in Jerusalem

Basar in der Altstadt

In der Altstadt von Jerusalem

Mahane Yehuda Market Schon seit Beginn der Konflikte zwischen Juden und Muslimen war Jerusalem bzw. der Status der Stadt ein zentraler Streitpunkt. Die Vertreter beider Religionsgruppen beanspruchen die Stadt oder zumindest Teile davon als Hauptstadt von Israel bzw. von Palstina. Einen Versuch, zu einer einvernehmlichen Lsung zu kommen, stellte der Teilungsvorschlag der Vereinten Nationen des Jahres 1947 dar, der vorsah, auf dem Gebiet des heutigen Israels einen vorwiegend jdischen und einen muslimischen Staat zu schaffen und Jerusalem unter internationale Verwaltung zu stellen. Whrend der Plan als UN-Resolution 181 von ber zwei Dritteln der berwiegend westlichen Staaten der UN-Vollversammlung angenommen wurde, betrachtete ihn die arabische Seite als einen

unzumutbaren Verzicht auf einen Teil des Dar al Islam. Die Araber lehnten den Plan ab und griffen einen Tag nach der Grndung Israels diesen Staat an. Nach dem Israelischen Unabhngigkeitskrieg bzw. Palstinakrieg hatten die Israelischen Streitkrften groe Gebiete des Landes erobert; Jerusalem wurde geteilt. Die westlichen Stadtteile fielen an Israel, Ostjerusalem kam mit dem Westjordanland (Juda und Samaria) unter jordanische Besatzung. Die jdische Bevlkerung Ostjerusalems wurde vertrieben, das jdische Viertel in der Altstadt zerstrt, und der Zugang zur Klagemauer als heiligster Ort des Judentums blieb Juden fortan versperrt. Israel erklrte am 4. Januar 1950 Jerusalem zur Hauptstadt. Dies wurde von der internationalen Gemeinschaft mit wenigen Ausnahmen nicht anerkannt. Im Sechstagekrieg (1967) eroberten israelische Fallschirmjger im Verlauf der Kampfhandlungen, die mit Angriffen des jordanischen Militrs auf die Stadt begannen, das bis dato jordanisch kontrollierte Ostjerusalem. Zur Schonung von Moscheen und Kirchen verzichtete die israelische Armee dabei auf den Einsatz schwerer Waffen und nahm dafr erhebliche Verluste in Kauf: Von insgesamt rund 800 israelischen Kriegstoten fielen 183 in Jerusalem. Erstmals seit der Staatsgrndung konnten Juden dadurch berhaupt wieder an der Klagemauer beten. Anders als die arabische Seite 1949 den Juden, verweigerte Israel aber den Moslems grundstzlich nicht den Zugang zu den heiligen Sttten, sondern unterstellte den Tempelberg einer autonomen moslemischen Verwaltung (Waqf). Beide Stadtteile und einige Umlandgemeinden wurden am 30. Juli 1980 durch das Jerusalemgesetz zusammengefasst und die Stadt zur untrennbaren Hauptstadt Israels erklrt. Dies ist aus palstinensischer Sicht eines der Haupthindernisse auf dem Weg zum Frieden. Der Sicherheitsrat der Vereinten Nationen erklrte die Annexion Ostjerusalems fr nichtig (UN-Resolution 478), sie wird daher von der Staatengemeinschaft ganz berwiegend nicht anerkannt, so dass sich alle diplomatischen Vertretungen, mit Ausnahme von Costa Rica und El Salvador, in Tel Aviv befinden. Jerusalem ist der Sitz der israelischen Regierung, des israelischen Prsidenten, des israelischen Parlamentes der Knesset, der 1918 gegrndeten Hebrischen Universitt und der Holocaustgedenksttte Jad waSchem. 1979 lebten bereits wieder 50.000 Juden in Ost-Jerusalem, 1993 waren es schon 160.000. Der Groteil der inzwischen rund 200.000 Juden in und um Jerusalem versteht sich anders als z. B. in Gaza oder Hebron nicht als nationalreligise Siedler, sondern ist einfach aufgrund staatlicher Frderung (insbesondere Steuervorteile und infrastrukturelle Investitionen) in diese Gebiete gezogen. Das bekannteste Beispiel fr den Siedlungsbau ist die 1975 gegrndete Trabantenstadt Ma'ale Adumim, die zur Zeit von 32.000 Menschen bewohnt wird. Nach Angaben der Christian Peacemaker Teams aus Chicago wurden zwischen 1967 und 1991 in Jerusalem 40.000 neue Husereinheiten fr jdische Israelis gebaut, aber nur 555 fr Palstinenser. Zudem wurden palstinensische Huser von Israel zerstrt, um Platz fr den Sicherheitszaun zu schaffen. Die Palstinenser stellen zwar rund 30 Prozent der Einwohner von ganz Jerusalem, bewohnen aber nur mehr 14 Prozent von dessen Flche. Siehe auch: Israelisch-palstinensischer Konflikt, Al-Aqsa-Intifada

Einwohnerentwicklung

Religiser Jude in der Ben-Jehuda-Strae

Altstadt von Jerusalem Die folgende bersicht zeigt die Einwohnerzahlen nach dem jeweiligen Gebietsstand. Jahr 1525 1538 1553 1562 1800 1838 1844 1876 1896 1905 1913 1917 Einwohner 4.700 7.900 12.384 12.650 8.750 11.000 15.510 25.030 45.430 60.000 75.200 53.410 Jahr 1922 1931 1946 1948 1967 1977 1980 1985 1990 1995 2000 2005 Einwohner 62.053 90.451 205.100 164.440 262.609 345.600 407.100 457.700 524.400 617.042 657.500 718.900

Religion

Rekonstruktion des herodianischen Tempels In Jerusalem findet sich eine groe Anzahl von Religionen und religisen Bewegungen. Mission ist in Israel und Jerusalem in jeglicher Form verboten. Die wichtigsten religisen Gruppierungen in der Stadt sind die zum jdischen Spektrum gehrenden Ultraorthodoxe nicht-zionistische Juden und Orthodoxe zionistische Juden. Zum muslimischen Spektrum gehren die Sunniten, Schiiten, Alawiten und Drusen. Zum christlichen Spektrum gehren die Griechisch-Orthodoxen, GriechischKatholischen, Rmisch-Katholischen, die Lutheraner, Protestanten, Anglikaner, Armenier, die Russisch-Orthodoxen, Georgisch-Orthodoxen, die thiopier , Syrischorthodoxen und die Altkatholiken. zur Aufteilung der Religionen siehe Religionen in Israel zu Kirchengebuden siehe Liste der Kirchengebude in Jerusalem

Jerusalem als heilige Stadt

Jerusalem ist von groer Bedeutung in der Bibel und wird insgesamt ber 800-mal erwhnt. Immer wieder steht die Stadt im Mittelpunkt der Heils- und Gerichtsankndigungen des biblischen Gottes, so vor allem bei den Propheten Daniel, Jeremia, Jesaja, Ezechiel, Sacharja und den Psalmen, aber auch in der Offenbarung. Beispiele: Ezechiel 5,5: So spricht Gott der Herr: Das ist Jerusalem, das ich mitten unter die Heiden gesetzt habe und unter die Lnder ringsumher! Joel 4,17: Und ihr sollts erfahren, dass ich, der Herr, euer Gott, zu Zion auf meinem heiligen Berge wohne. Psalm 137,5: Vergesse ich dich, Jerusalem, so verdorre meine Rechte.

Sowohl die Stadt Jerusalem als auch das Land und Volk Israels stellt die Bibel als Gottes Eigentum dar. Interessant ist diesbezglich die literarische Darstellung Jerusalems als ein Findelkind, das von Gott aufgezogen wird in Ezechiel 16, ebenso wie die Zusagen Gottes an die Stadt in den Psalmen, die wie Eheversprechen formuliert sind. Jerusalem wird von Christen, Juden und Muslimen als Heilige Stadt angesehen. Im Gegensatz zur Bibel und dem Tanach erwhnt aber der Koran Jerusalem kein einziges Mal namentlich. Den Christen ist die Altstadt von Jerusalem heilig, da sie der Ort der Leidensgeschichte, Kreuzigung und Auferstehung von Jesus Christus ist. Fr die Juden ist sie als Hauptstadt des ersten jdischen Knigreiches heilig, fr die Muslime, weil sie mit der

Himmelfahrt des Propheten Mohammed in Verbindung gebracht wird. Fr alle drei Religionen ist Jerusalem als Wirkungsort verschiedener Propheten bzw. Heiligen wie Abraham, Salomon, David, Zacharias und anderen bedeutend.

Kultur und Sehenswrdigkeiten

Theater
Das bekannteste Theater in Jerusalem ist das Stadttheater.

Museen
Die bedeutendsten Museen in Jerusalem sind das Israel Museum mit dem Schrein des Buches, das Rockefeller Museum fr Archologie, die Holocaustgedenksttte Jad waSchem, das Naturhistorische Museum und das Bazabel-Museum fr Volkskunst und Folklore.

Bauwerke

Historischer jdischer Friedhof von Jerusalem Die Altstadt von Jerusalem wurde 1981 von der UNESCO zum Weltkulturerbe der Menschheit erklrt. Sie ist seit dem Mittelalter in das armenische Viertel im Sdwesten, das christliche im Nordwesten, das jdische im Sdosten und das muslimische Viertel im Nordosten unterteilt und wird von einer aus dem 16. Jahrhundert stammenden, fast vollstndig erhaltenen Stadtmauer umgeben. Die Mauer der alten Stadt Davids umfasst mehrere Trme sowie ursprnglich sieben Tore, davon drei groe und vier kleine, und wurde 1889 durch ein achtes ergnzt. Im christlichen Teil der Altstadt befindet sich das Neue Tor, an der Grenze zum armenischen Teil das Jaffator und zum muslimischen Teil das Damaskustor. In den muslimischen Teil fhren das Herodestor, Goldene Tor (durch die Trken versiegelt) und das Stephanstor. Im jdischen Teil steht das Zionstor und das Dungtor; sdwestlich davon erhebt sich der Berg Zion mit dem mutmalichen Grab Knig Davids. stlich der Altstadt liegt der lberg mit dem Garten Getsemani. Wichtige christliche Sttten sind die auf den Grundmauern einer Basilika aus dem 4. Jahrhundert erbaute Grabeskirche und die Via Dolorosa. Die circa 400 Meter lange, von den Juden Westliche Mauer genannte Klagemauer ist ein Teil der Sttzmauer des Plateaus, auf dem der groe Tempel Herodes des Groen

stand. Wichtige muslimische Bauwerke auf dem Tempelberg sind heute der Felsendom und die Al-Aqsa-Moschee. Weitere bedeutende Bauwerke in der Altstadt oder ihrer unmittelbaren Umgebung sind der Cardo (Sulengang), die Dormitio-Kirche, die Erlserkirche, die vier sephardischen Synagogen und die Zitadelle. Im Norden, Westen und Sden der Altstadt breitet sich die Neustadt von Jerusalem aus, die sich seit der Mitte des 19. Jahrhunderts entwickelt hat. Sie erstreckt sich ber die umliegenden Hgel und weiter bis in das wstenhafte Umland der Stadt. Die modernen Wohn- und Geschftsgebude und die breiten Straen der Neustadt bilden einen starken Kontrast zu den rmlichen Behausungen und engen Gassen der Altstadt. In der Neustadt befinden sich die Knesset (das israelische Parlament), die Synagoge des Hadassa-Klinikums mit ihren Chagallfenstern und zahlreiche bedeutende staatliche Einrichtungen. Dazu gehren unter anderem das Finanzministerium, das Auenministerium, das Innenministerium und der Sitz des Premierministers.

Wirtschaft und Infrastruktur

Das wirtschaftliche Leben der Stadt Jerusalem basiert zum berwiegenden Teil auf ihrer religisen und kulturellen Bedeutung sowie auf ihrer Funktion als Verwaltungszentrum. Der Dienstleistungssektor ist dementsprechend gut ausgebaut. Viele Bewohner von Jerusalem sind in der staatlichen und stdtischen Verwaltung sowie im Bildungswesen beschftigt. Eine eher untergeordnete Rolle spielt demgegenber das produzierende Gewerbe. Die Industriebetriebe der Stadt stellen unter anderem Glas-, Metall- und Lederwaren, Druckerzeugnisse, Schuhe und Zigaretten her. Die Produktionsbetriebe sind vorwiegend in den ueren Bezirken von Jerusalem angesiedelt. Der Tourismus ist jedoch der mit Abstand bedeutendste Wirtschaftsfaktor, da die Altstadt ein bedeutendes Ziel fr Touristen ist.

Verkehr
Durch die Berglage liegt Jerusalem abseits der wichtigsten Verkehrsstrme Israels, die vor allem in der Kstenebene und dem dahinter liegenden Landstreifen flieen. Innerhalb der Stadt muss sich die Straenfhrung der hgeligen Landschaft anpassen. Die wichtigste Straenverbindung Jerusalems ist die Autobahn nach Tel Aviv, in die anderen Richtungen bestehen Landstraen. Besonders bemerkenswert ist die Strae an das Tote Meer, die auf ihrem Weg durch das Westjordanland 1200 Hhenmeter abfllt. Nrdlich der Stadt befindet sich ein kleiner Flughafen, der nur Ziel von innerisraelischen Flugverbindungen war, aber seit 2001 geschlossen ist. Der internationale Flughafen fr Jerusalem ist der Ben-Gurion-Flughafen Seit April 2005 verkehren nach siebenjhriger Unterbrechung wieder Zge der Israel Railways zwischen Jerusalem und Tel Aviv; die Gebirgsstrecke ber Bet Schemesch

wurde seit Juli 1998 saniert. Eine Hochgeschwindigkeitsstrecke zwischen Jerusalem und Tel Aviv soll 2009 fertig gestellt werden. Die Anbindung mit ffentlichen Verkehrsmitteln erfolgt in erster Linie ber Busse der staatlichen Busgesellschaft Egged, die von Freitagmittag bis Samstagabend (Schabbat) ihren Dienst einstellt. 2007 ist geplant, Jerusalems neues Straenbahnsystem zu erffnen. Eine der bekanntesten und zugleich wichtigen innerstdtischen Straen ist die Jaffastrae, die von der Altstadt in Richtung Tel Aviv fhrt. Sie ist eine wichtige Einkaufsstrae und war bereits mehrfach Schauplatz blutiger Attentate.

Jesusschauspiel in der Via Dolorosa

Bildung
Zu den bekannten Bildungseinrichtungen in der Stadt gehren die 1918 erffnete Hebrische Universitt von Jerusalem, die 1959 gegrndete Israelische Akademie, das Planetarium, das Zionistische Zentralarchiv, die Gulbenkian-Bibliothek und die Jdische National- und Universittsbibliothek. In der Stadt befinden sich zahlreiche religise Lehr- und Forschungsinstitute. Dazu gehren unter anderem die 1890 erffnete cole Biblique et cole Archologique Franaise, das 1927 gegrndete Ppstliche Bibelinstitut und das 1963 erffnete Institut der Jdischen Religion.

Persnlichkeiten
Shne und Tchter der Stadt
Knig Salomo, 3. Knig Israels Mili Avital, israelische Film- und Theaterschauspielerin David Gerstein, israelischer Knstler Amira Hass, jdische Journalistin Flavius Josephus, jdischer Feldherr und Geschichtsschreiber Eliahu Inbal, israelischer Dirigent Jigael Jadin, israelischer Archologe, Politiker und Militr Abraham B. Jehoshua, israelischer Schriftsteller

Kyrill von Jerusalem, Kirchenvater der Orthodoxie und Kirchenlehrer Isaak Luria, jdischer Kabbalist Natalie Portman, US-amerikanische Schauspielerin Rehabeam, erster Knig von Juda Tom Segev, Journalist und Historiker Ahmet Kutsi Tecer, trkischer Dichter und Politiker Theodor I., Papst Wilhelm von Tyrus, Erzbischof von Tyros und Geschichtsschreiber des Mittelalters Hadsch Mohammed Amin al-Husseini, islamischer Geistlicher und palstinensischer Nationalist

Quellen
1.
Resolutionen der Generalversammlung der Vereinten Nationen sind im Unterschied zu Resolutionen des Sicherheitsrats nicht verbindlich. 2. siehe z.B. Lexikon zur Bibel von Fritz Rienecker R.Brockhaus Verlag Wuppertal, Seite 683 3. es gab geschtzte 90.000 Tote, s. Elliott S. Horowitz, Reckless Rites, 2006

Literatur
Christoph Gerhard: Marco Polo Reisefhrer Jerusalem. Mairs Geographischer Verlag, Stuttgart 2001, ISBN 3-89525-928-4 Helmut Hubel, Tilman Seidensticker (Hrsg.): Jerusalem im Widerstreit politischer und religiser Interessen. Die Heilige Stadt" aus interdisziplinrer Sicht. Peter Lang Verlag, Frankfurt am Main 2004, ISBN 3-631-51057-8 Gerhard Konzelmann: Jerusalem. 4000 Jahre Kampf um eine heilige Stadt. Hoffmann und Campe, Hamburg 1984, ISBN 3-455-08660-8 Hans Kchler (ed.): The Legal Aspects of the Palestine Problem With Special Regard to the Question of Jerusalem. Wilhelm Braumller, Wien 1981 ISBN 3-7003-0278-9

Weblinks
Stadtverwaltung Jerusalem (engl.) Verzeichnis wissenschaftlicher Literatur zu Israel/Jerusalem/Palstina, dem Salomonischen

Tempel zu Jerusalem, Templern allgemein, Papsturkunden, Kreuzzgen sowie der Ordensregel der Templer

Siehe auch
Portal: Jerusalem bersicht zu Wikipedia-Inhalten zum Thema Jerusalem Commons: Jerusalem Bilder, Videos und Audiodateien

Liste der Kirchen in Jerusalem Flagge Jerusalems Wappen Jerusalems

Von http://de.wikipedia.org/wiki/Jerusalem Kategorien: Wikipedia:berarbeiten | Ort in Israel | Hauptstadt in Asien | Weltkulturerbe (Asien und Ozeanien) | Jerusalem | Ort in der Bibel | Archologischer Fundplatz in Israel | Antike israelische Stadt
Ansichten

Artikel Diskussion Quelltext betrachten Versionen/Autoren

Persnliche Werkzeuge

Anmelden

Navigation

Hauptseite ber Wikipedia Themenportale Von A bis Z Zuflliger Artikel

Mitmachen Suche

Hilfe Autorenportal Letzte nderungen Spenden

Artikel

Volltext

Werkzeuge

Links auf diese Seite nderungen an verlinkten Seiten Hochladen Spezialseiten Druckversion Permanentlink Seite zitieren

Andere Sprachen

Afrikaans Alemannisch Aragons Asturianu Azrbaycan () Brezhoneg Bosanski Catal Chavacano de Zamboanga esky Cymraeg Dansk English Esperanto Espaol Eesti Euskara Suomi Na Vosa Vakaviti Franais Gaeilge Gidhlig Galego Hak-k-fa

Hrvatski Magyar Bahasa Indonesia Ido slenska Italiano Basa Jawa Taqbaylit Latina Ladino Ltzebuergesch Limburgs Lguru Lingla Lietuvi Latvieu Bahasa Melayu Nederlands Norsk (nynorsk) Norsk (bokml) Novial Nouormand Occitan Polski Portugus Runa Simi Rumantsch Romn Sicilianu Srpskohrvatski / Simple English Slovenina Slovenina Gagana Samoa Shqip / Srpski Svenska Kiswahili Tagalog Tok Pisin Trke

Ting Vit Volapk Walon Yorb

Diese Seite wurde zuletzt am 11. Januar 2008 um 13:14 Uhr gendert. Ihr Inhalt steht unter der GNU-Lizenz fr freie Dokumentation. Wikipedia ist eine eingetragene Marke der Wikimedia Foundation Inc. Datenschutz ber Wikipedia Impressum

1. Recuerda que la mejor informacin sobre la pelcula que ests viendo te la dan tus emociones al verla. Prate a pensar qu te gusta y qu no te gusta de la pelcula. OJO: incluso en la peor pelcula puedes encontrar temas interesantes. Incluso en tu pelcula favorita puedes encontrar cosas que no te gustan. 2. Recuerda. Mira ms all de la superficie: lo importante son los temas tab:emociones, sexo, dinero, salud, poder, raza, estatus, adicciones, secretos 3. Recuerda que una pelcula cuenta la historia en imgenes. Nada es casual. 4. Busca un ttulo para la pelcula 5. Divdela en partes: presentacin, desarrollo, climax, conclusin, eplogo. Recuerda que estas partes deben relacionarse con el ttulo (que se refiere al tema ms importante en tu opinin de la pelcula) 6. Recuerda que cada secuencia de la pelcula es una pequea historia completa en s misma con sus correspondientes partes (presentacin, desarrollo, climax, conclusin, eplogo). 7. Establece cul es el conflicto. Establece cual es el objetivo (o los objetivos) de los personajes, qu objetivos son ms superficiales y qu objetivos son ms profundos 8. Establece cules son las sorpresas y como hacen que cambie lo que nos dijo la pelcula en la presentacin 9. MIRA la pelcula. Observa las imgenes y presta atencin a los detalles.

10. Cmo son fsicamente los personajes? Qu ropa llevan? Qu objetos tienen? Cmo se mueven y actan? 11. Cmo es la luz en la pelcula? De dnde viene? A quin o a qu ilumina? Cmo son las sombras en la pelcula? 12. Cmo son los colores? Los colores se oponen o no? Qu emociones expresan? 13. Cmo es la composicin?: hay muchos o pocos objetos en la pantalla, vemos claramente o no, vemos slo a una persona o a muchas? 14. Cmo estn ordenados los objetos? Estn en equilibrio o amontonados a un lado? 15. Recuerda: cerca= ms emocin. Primer plano=emocin mxima. 16. Qu profundidad tiene la imagen? Qu ves mejor, el fondo o la imagen central? 17. Qu personajes u objetos tienen ms importancia en la imagen? Por qu? 18. Dnde est la cmara? arriba, dominando a los personajes o por debajo de ellos, presentndolos como superiores a ella? adopta la cmara el punto de vista de un personaje en concreto? 19. Cmo es el montaje? Las tomas son largas y sin cortes o hay un cambio contnuo de imgenes? Se mueve la cmara o no? Cmo se mueve la cmara: suavemente, bruscamente, violentamente. tiembla? 20. ESCUCHA Cmo es el sonido? Claro, difcil de entender, con ruidos de fondo? Hay silencio? 21. Cmo es la msica? Qu nos dice la msica de las emociones de los personajes? Hay canciones? Tiene relacin la letra de las canciones con la historia que narra la pelcula? 22. Todos estos elementos, son coherentes o se contradicen entre s? Qu elementos crees que son ms importantes en la pelcula que ests analizando? 23. Recuerda que todos los detalles tcnicos representan opiniones del director y estn intentando manipularnos, hacernos sentir simpata ciertas personas o cosas y antipata por otras
Aqu trato de incluir todo el amplio espectro de materiales que podemos denominar en sentido amplio imagen en movimiento, desde las pelculas de ficcin a los noticieros o documentales, desde los vdeos educativos a otro tipo de reportajes especficamente histricos de televisin. Debemos distinguir por lo pronto imgenes primarias, aquellas que han sido rodadas o filmadas en el mismo momento en el que se producen los hechos, o imgenes secundarias, elaboraciones o reconstrucciones posteriores. Evidentemente, podemos aplicar el mismo criterio a la imagen fija, ya sea foto, cuadro, dibujo... Al analizar una pelcula se deben de dar los siguientes pasos: Anlisis de la pelcula: o Resumen breve del argumento o Momentos o etapas claves del film. tema central? personajes claves? temas que se tratan? o El autor (ideologa, trayectoria artstica), circunstancias, fecha y lugar donde se rod la pelcula. (Esta parte se har siempre que sea posible o sea significativa la informacin) Anlisis de los personajes de la pelcula: o Si se trata de una pelcula de ficcin, tendremos que sealar como se

corresponde el film con la realidad histrica y la poca que trata, si nos hallamos con un documental o reconstruccin histrica, habr que analizar cul es la relacin entre el film y la verdad histrica que la que nosotros participemos. o Sealar si los protagonistas son individuales o colectivos, qu grupos sociales son representados, etc. Anlisis de los hechos histricos recogidos en la pelcula: o Comentario de los antecedentes, hechos histricos y consecuencias de estos reflejados en la pelcula. o Valoracin crtica es significativa la informacin que aporta? est prejuiciada ideolgicamente? si es as en qu sentido?

Caractersticas Generales en estas pelculas:

Las situaciones: Son realmente extremas, sobre todo en el caso de la "delgada linea" o "Happiness". Nadie en hollywood se atrevera a hacer una pelcula con uno de los protagonistas fuera un pederasta y otro un obseso sexual. Los actores: Todos estn muy medidos, ninguno sobresale por encima del otro. En la pelcula indy se cuida mucho que el conjunto resulte homogneo, para que sea un todo. En el cine de estudio se cuida mas , por ejemplo, los efectos especiales, los protagonistas, etc... En el cine indy todos los actores tienen una historia detras, y muy bien no sabes a veces quien es el protagonista y quien el secundario. El presupuesto: Salvo la "Delgada linea", todas las demas son producciones de menos de 10 millones de dolares. EL presupuesto medio de una pelicula de estudio son 25 millones. Como se puede ver, la diferencia es notable. Los escenarios: Normalmente son peliculas intimistas, en las que los exteriores no asumen un papel destacado. "Celebrity" si usa los exteriores, pero como si fuesen interiores. Para Woody Allen , rodar en Manhattan es como rodar en el salon de su casa. La Finalidad: ninguna de estas peliculas pretende desbordar el "Box Office". Todas buscan llegar a un pblico amplio, pero aun bastante selecto. No son "episodios I", vamos. Ideologia: Normalmente , son peliculas progresistas, contra el sistema establecido, pero a veces, se esconde un cierto mensaje fascista si no se sabe distinguir a tiempo ( caso de American ...) Hay que verlas en v.o.: Como dan poco en taquilla, normalmente el doblaje que se hace es penoso. Hay que darle

gracias a Dios por la mente iluminada que no permiti doblar "Bullworth ".

Desnudos: Los actores que estan dentro del sistema Hollywood aqui no tienen ningun tipo de pudor, como es el caso de Mcgregor. Si G.Lucas viese "Velvet"... Msica: Uso de musica tambien independiente , o recuperacion de viejos hits.

Tabla de contenidos
[ocultar]

1 Definicin 2 Distincin terminolgica entre Drama y Teatro 3 El drama como gnero literario 4 El drama como gnero cinematogrfico 5 Referencias

Definicin [editar]
Segn la definicin griega clsica, drama es la forma de presentacin de acciones a travs de su representacin por actores y por medio del dilogo. Existe una confusa tendencia dramtica nicamente a todo lo que incluya elementos trgicos, especialmente cuando se da el llamado "final trgico". Por ello, una tendencia generalizada, por un lado, al drama como gnero literario (el texto), al teatro como de puesta en escena del propio texto dramtico; pues, segn esta posicin, el texto dramtico no es teatro hasta que se lleva a escena para ser visto por el pblico. En tanto gnero literario, el drama se dividi desde sus inicios en la Grecia Antigua (donde se considera que el teatro tiene su nacimiento occidental) en Tragedia y Comedia. Posteriormente, el drama se divide en gneros realistas y gneros simblicos; entre los primeros quedaron inscritas la tragedia y la comedia ya existentes y, entre los segundos, el melodrama, la farsa y la tragicomedia, reconocidos como tales desde el Renacimiento. Gnero cinematogrfico de temtica seria y penosa. Un importante subgnero es el melodrama, que se centra en las pasiones humanas y los conflictos individuales, en los que se destacan lo trgico y lo dramtico; entre los primeros quedaron inscritos la tragedia y la comedia ya existentes y, entre los segundos, el melodrama, la farsa y la tragicomedia, reconocidos como tales desde el Renacimiento.

Distincin terminolgica entre Drama y Teatro [editar]

Algunos tericos del siglo XX insisten en la diferenciacin categrica entre el Drama y el Teatro, siendo la primera la versin constituida en lo absoluto por elementos

lingsticos, formando parte entonces, de lo que se considera como gnero literario, siendo la particularidad de este la predominancia de la funcin apelativa del lenguaje, la ausencia de un mediador entre el mundo creado o realidad ficticia y el lector, su posibilidad virtual de ser Teatralizado, etc. El Teatro sera la concrecin del Drama que incluye el espectculo, la actuacin, la msica, etc. Es decir, elementos que no le son propios al drama como realidad lingstica acotada solamente al discurso. El anlisis de un Drama puede ser hecho desde la crtica literaria, en cambio el anlisis del teatro debe incluir factores como actuacin, evaluacin del espectculo, los msicos, la iluminacin, etc. Se puede aadir la precisin, desde la perspectiva de la etnoescenologa,campo interdisciplinario que estudia los Fenomenos y Comportamientos Humanos Espactaculares Organizados (PCHSO) del Teatro que es un subgrupo dentro del conjunto de formas espectaculares organizadas. Se pueden distinguir tres aspectos claves dentro del anlisis de una forma espectacular: la "espectacularidad" la "performatividad"; y el fenmeno de "relacin simbitica", o de "empata" que se construye en la relacin al publico (Pradier, 1996). La pertinencia de la evaluacin de estos aspectos del teatro como forma espectacular, esta en el hecho que toda forma espectacular responde a un contexto social. En este aspecto, la dramaturgia, y su manifestacin espectacular por medio del teatro, se organizan por medio de codigos que no son universales, pero particulares a un contexto histrico y cultural.

El drama como gnero literario [editar]

En tanto gnero literario, el drama se dividi desde sus inicios en la Grecia Antigua (donde se considera que el teatro tiene su nacimiento occidental) en Tragedia y Comedia. Posteriormente, el drama se divide en gneros realistas y gneros simblicos; entre los primeros quedaron inscritos la tragedia y la comedia ya existentes y, entre los segundos, el melodrama, la farsa y la tragicomedia, reconocidos como tales desde el Renacimiento. Obras de dramaturgos como Pedro Caldern de la Barca, dentro del Siglo de Oro Espaol, y Antn Chjov, a principios del siglo XIX y principios del XX en Rusia, abri la discusin de dos nuevos gneros dramticos: el auto sacramental y la pieza, que pasaron a formar parte, el primero, de los gneros simblicos y, el segundo, de los realistas. El siglo XX y uno de sus escritores ms emblemticos, el alemn Bertolt Brecht, aportaran un nuevo gnero dramtico en lo que dio en llamarse la obra didctica; que, si bien tuvo sus primeras expresiones en los autos sacramentales de los autores del Siglo de Oro Espaol, con Brecht encontr temas no slo religiosos, sin perder el ingrediente moral (moraleja) que le caracteriza. El gnero realistico de forma general es el gnero literario que se caracteriza porque el lenguaje quiere presentar hechos de forma real y utilizando un dilogo y para eso su

lenguaje utiliza una funcin apelativa. Este gnero posee diversos elementos internos con el acto, escena y cuadro, que son: Acto: La subdivisin ms importante de la obra dramtica, marcado por el cierre o cada del teln.

Escena: Divisin interna de acto, en donde actan los mismos personajes. Se cambia por la entrada o salida de un personaje.

Cuadro: Es la ambientacin fsica de la escenografa (lo que se quiere mostrar).

La funcin del texto dramtico es llevar este texto a la puesta en escena o la representacin, lo que marca la diferencia entre obra dramtica y obra teatral. Luego se puede marcar los dos subgneros ms importantes: -Comedia: su funcin es producir la reflexin en el espectador sobre un vicio humano, aspectos ridiculos y graciosos de la vida, por medio del elemento cmico. -Tragedia: su funcin es producir la purificacin de alma y espritu al ver un hecho sumamente desgraciado, porque eso causa una empata con el protagonista y este espectador al ver el hecho desgraciado tambin recuerda sus penas y dolores y libera eso con el llanto y la utilizacin del elemento trgico.

El drama como gnero cinematogrfico [editar]

Como gnero cinematogrfico el drama siempre plantea conflictos entre los personajes principales de la narracin flmica provocando una respuesta emotiva en el espectador, conmoviendo a ste, debido a que interpela a su sensibilidad. Los problemas de los personajes estn relacionados con la pasin o problemas interiores. La temtica de este tipo de pelculas es diversa, pero su eje principal es el amor. Cuando esto ltimo es bastante exagerado, al drama se le conoce por melodrama: En l, los personajes quedan limitados al carcter simplista de buenos o malos. La mayora de las pelculas promueven la moralidad como eje central y, por ello, tienen una dimensin didctica acusada. La presencia de la msica y la fotografa acentan la credibilidad de la obra. El drama es para llorar.

Referencias [editar]
Gnero cinematogrfico es el estilo narrativo de una pelcula que sirve para su clasificacin comercial. Los gneros cinematogrficos, como los gneros de otros campos artsticos, son clasificaciones formales originadas en la cultura clsica, los dos gneros mayores griegos: comedia y tragedia; uno de estilo ligero, tema superficial y final fliz, otro afectado, profundo y de triste desenlace. Estos gneros se fueron diversificando en el

teatro, gneros que los primeros largometrajes intentaron imitar. Sin embargo las posibilidades del cine lo desmarcaron completamente de los gneros tradicionales creando nuevos gneros caracterizados por la escasa complejidad de su regulacin.

Tabla de contenidos
[ocultar]

1 Evolucin 2 Clasificacin o 2.1 Estilo o tono o 2.2 Ambientacin o 2.3 Formato o 2.4 Audiencia 3 Lista de gneros

Evolucin [editar]
Es muy importante sealar que en las primeras producciones el gnero de las pelculas era encorsetado, con caractersticas muy delimitadas que ayudaban al espectador a comprender rpidamente la pelcula y a que el autor crease lo que quisiese y el espectador lo esperase. Sin embargo aproximadamente tras la Segunda Guerra Mundial los gneros empezaron a mezclarse o a adulterar su esencia creando grandes producciones y extraos especmenes.

Clasificacin [editar]
Los gneros cinematogrficos se clasifican segn los elementos comunes de las pelculas que abarquen, originalmente segn sus aspectos formales (ritmo, estilo o tono y, sobre todo, el sentimiento que busquen provocar en el espectador, ajuste). Alternativamente, los gneros cinematogrficos se definen por su ambientacin o por su formato. Los gneros siguientes son a menudo concretados para formar subgneros, y tambin pueden ser combinados para formar gneros hbridos.

Estilo o tono [editar]

Comedia en el cine este gnero slo recoge las pelculas con la intencin de provocar la risa. Drama principalmente se centra en el desarrollo del personaje.

Negro presenta los personajes principales con un estilo nihilista o existencialista. Accin generalmente implica una interaccin moral entre bueno y malo llevada a su fin por la violencia o la fuerza fsica. Aventura situaciones de peligro y riesgo, mezclado a menudo con mucha fantasa. Terror con la intencin de provocar el miedo en la audiencia. Cine de misterio la progresin de lo desconocido a lo sabido (conocido) por el descubrimiento y solucin de una serie de pistas. Cine romntico hace hincapi en los elementos amorosos y romnticos. Suspense (Espaa)/Suspenso (Amrica) con la intencin de provocar tensin a la audiencia.
o

Ambientacin [editar]
Policiaco la derrota del Mal en el reino de la actividad criminal. Histrico la accin ocurre en el pasado, a menudo de forma estereotipada. o Blico campos de batalla y posiciones que pertenecen a un tiempo de guerra. o del Oeste (o Western) ambientado desde el perodo colonial a la era moderna de los Estados Unidos de Amrica. Ficcin coloca a los personajes en una realidad alternativa, tpicamente en el futuro, en el espacio o en universos imaginarios. o Ciencia ficcin ficcin cientfica: futura, espacial... o Fantasa ficcin especulativa irreal. Deportivo acontecimientos deportivos y situaciones relacionadas con un deporte.

Formato [editar]
Animacin con fotogramas dibujados a mano o con ordenador que, pasados rpidamente, producen ilusin de movimiento.

Audiencia [editar]

Infantil pelculas para nios pequeos; a diferencia de las familiares, ningn esfuerzo especial es hecho para hacer la pelcula atractiva para otros pblicos. Familiar con la intencin de ser pelculas atractivas para gente de todos las edades y sobre todo para una audiencia joven. Adulta para ser vistas slo por una audiencia adulta, el contenido suele incluir violencia, temas inquietantes, palabrotas o sexo explcito.

Lista de gneros [editar]

Cine cristiano Cine de accin Cine de animacin Cine arte Cine de aventuras Cine blico Clase B Ciencia Ficcin Cine de autor Cine denuncia Cine pico Cine fantstico Cine negro Cine pornogrfico Cine BDSM Comedia (cine) Comedia romntica Cine costumbrista Documental Experimental Drama Melodrama Cine histrico Intriga y suspense Cine musical Cine policiaco Cine poltico Cine propagandstico Cine de terror Thriller Western Cine gore Snuff movies

Obtenido de "http://es.wikipedia.org/wiki/G%C3%A9nero_cinematogr%C3%A1fico" Categora: Gneros cinematogrficos

Etnoescenologia

0AD^ |UDS^ (sacred, or holy tradition), also called ad9 ilh , or rabbn (divine tradition), is a class of traditions which give words spoken by class of traditions which give words spoken by God God , as distinguished from ad9 nabaw (prophetical tradition) which gives the words of the Prophet. Although ad9 uds is said to contain God's words, it differs from the |ur"n which was revealed through the medium of Gabriel, is inimitable, is recited in the alt , and may not be touched or recited by the ceremonially unclean. 0ad9 uds does not necessarily come through Gabriel, but may have come through inspiration ( ilhm ), or in a dream. One statement, not generally accepted, says God revealed these traditions to the Prophet on the night of the Mi#r3 . The words are not God's exact words, but express their meaning. They may not be used in alt , and there is no harm if one touches them when ceremonially unclean. Disbelief in the |ur"n is infidelity, but this does not apply to ad9 uds . When quoting a ad9 uds one must not say simply, God said as when quoting the |ur"n, but either, God's messenger said in what he related from his Lord, or, God most high said in what God's messenger related from Him. Some of these traditions quite clearly have their source in the Bible. For example, what eye has not seen, nor ear heard, nor has entered into the heart of man (cf. Isaiah lxiv, 4; 1 Cor. ii, 9), and a tradition telling that on the Day of Resurrection God will say, O son of Adam, I was sick and you did not visit me, continuing on the lines of Matthew xxv, 41 ff. The ad9 uds do not form a separate group in the books of tradition, but some collections have been compiled from the six Sunn books and, more commonly, from others. The largest collection, al-Itft al-saniyya fi 'l-ad9 al-udsiyya, by Muammad al-Madan, or al-Madyan (d. 881/1476), publ. 0aydarbd 1323/1905, contains 858 traditions divided in three groups: (1) those beginning with la; (2) those beginning with yalu; (3) those given alphabetically, this last containing 603. The isnd is not given, but as the collection from which each tradition comes is mentioned, | [III:29a] those who desire can find its isnd there. A collection of 101 uds traditions entitled Mi9kt al-anwr by Muyi 'l-Dn Ibn al-#Arab (d. 638/1240) was published in Aleppo (1346/1927) along with a collection of 40 compiled by Mull #Al al-|ri" (d. 1014/1605). Ibn al-#Arab, who divides his collection into three parts, two of 40 traditions and one of 21, gives a full isnd in the first, sometimes in the second and usually in the third. #Al al-|ri" merely mentions the Companion reputed to have heard the tradition from the Prophet. Another collection, not published, is by #Abd al-Ra"f Muammad b. T3 al-Dn al-Munw (d. 1031/1621). It is divided into two parts (cf. 033 9alfa, ed. Flgel, i, 150 f.), the first with traditions beginning with la and the second arranged in alphabetical order. It would appear that al-Munw, whose smaller work has the same title as al-Madan's, was largely dependent on that work.

SABA" or the Sabaeans (Greek Sabaoi), the name of a folk who were bearers of a highly developed culture which flourished for over a millennium before Islam, together with three other folks, Ma#n, |ataban and 0a'ramawt [q.vv.]. The main Sabaean centre was at Maryab (later Mrib, see ma"rib) in Yemen with its fertile oasis on the western edge of the desert known to Arab geographers as -ayhad (modern Ramlat al-Sab#atayn). In early historical times there were also Sabaean settlements in the Wd A9ana above the great dam which waters the oasis of Mrib, in some smaller oases to the north, and in parts of the Wd 3awf or Wd Ma9b. All these locations are approximately 1,000 m above sea level. The montane plains lying west of Mrib and having an average level of 2,000 m above sea level were the home of other folks who spoke the same language as the Sabaeans proper, and seem to have formed some kind of federation under the hegemony of Saba". Towards the end of the first millennium B.C., these highland folks became politically dominant in the Sabaean federation. Our knowledge of the Sabaeans is derived principally from their own inscriptions. Modern scholarship was first made aware of these by Carsten Niebuhr, member of a Danish exploratory mission in the end of the 18th century. A sporadic number of inscriptions were published and studied during the earlier part of the 19th century, but it was Eduard Glaser's travels in the last decades of the century which produced a large number of copies (mostly squeezes) forming the real foundation of subsequent research. It must be admitted, however, that later 19th and early 20th century scholars indulged too freely in speculative deductions based on insufficient | [VIII:663b] evidence. A turning point came in 1950; from then onwards, an ever more rapid archaeological activity resulting in the discovery of new texts has overturned not a few conclusions too confidently advanced by earlier researchers. At the time of writing, the flow of new material is still in full course, and it has to be anticipated that some of the presently current hypotheses may in their turn prove to be invalid. Any account that can be written at the moment must be taken as still tentative.

1. Script and language.

The monumental inscriptions are drafted in a variety of South Semitic alphabet, the socalled musnad script [see musnad . 1]. The Sabaic language, with the languages of Ma#n, |ataban and 0a'ramawt, forms an independent branch of Semitic, having in common one distinctive feature that is found nowhere else in Semitic: the use of suffixed -n in the function of a definite article corresponding to the Arabic prefixed al-. Within this language group, Sabaic is distinguished from the other three by using h as prefix of the causative verb and as base of the 3rd person pronouns, where the others have a sibilant. On the southern borders of the Sabaean domain, the area between the Yisli pass and 9amr used Sabaic language, as did the non-Sabaean Radmn folk to the east thereof, in the Rad# area. By the end of the 3rd century A.D. the other three languages had fallen into disuse, at least for epigraphic purposes, and the inscriptions throughout Yemen, now under Himyarite domination (see below), are in a late form of Sabaic; there are indications that the language may by this time have become a prestigious learned language, not in everyday use (this is comparable with the case in

North Arabia, where the Nabataean inscriptions are in Aramaic, though the everyday language was probably Arabic).

The general consensus today is to assign the oldest substantial body of Sabaic inscriptions (apart from a handful of seemingly earlier examples) to the 8th century B.C. Yet it still remains not altogether easy to discount completely one point which led Pirenne to propose a dating a couple of centuries later. Inscriptions of this period have rigidly geometrical forms, subjected to strict canons of proportion, astonishingly like Greek inscriptions of the 6th-5th century, but wholly unlike any other Semitic script of that or any earlier dating. It is hard to envisage how this style can have evolved totally independently, with a time-lag of two centuries, in two adjacent cultures with ancient trade links between them. In the latter part of the first millennium B.C., the musnad script developed (as was the tendency in Greco-Roman inscriptions) more decorative embellishments, at first with the introduction of serifs at the ends of the strokes.

2. History.

For the pre-history of Saba", that is, before the beginning of the epigraphic record, there is no evidence available as yet. Silt deposits in the Mrib oasis point to intensive agricultural exploitation by artificial irrigation going back to at least the early second millennium B.C.; but what, if any, connection there may have been between these ancient agriculturalists and the Sabaeans as we know them, is wholly obscure. Trade links between South Arabia and Mesopotamia there must have been, judging by Akkadian references to South Arabian products such as frankincense and myrrh; but the first specific mention of Sabaeans in Akkadian sources is in the 8th century B.C., when the governor of Suu (approximately #$na on the middle Euphrates) and Mari intercepted and plundered a caravan of folk from Taym" and Saba" (whose home is far away), seemingly for making a detour to evade transit dues in Suu (A. | [VIII:664a] Cavigneaux and B.K. Ismail, Die Statthalter von Suu und Mari, in Baghdader Mitt., xxi [1990], 351). Two other texts, known to us for a long time and recording gifts made by Sabaeans to Assyrian rulers in ca. 715 and 685 B.C., have led some scholars to postulate a Sabaean group living close to Assyria, since gifts was interpreted as tribute. In fact, it is now clear that the gifts were such as a trade mission would normally bring (and still do today) in order to smooth their path.

In this archaic phase, the Sabaean rulers used regnal names chosen from a total list of only six, but accompanied optionally by a cognomen chosen from a list of four; the use of these styles was exclusive to the rulers, hence a reference to an individual by these styles was sufficient to indicate ruler status. However, the inscriptions drafted as from the ruler himself commonly added the title mkrb of Saba"; this term is now believed to have much the same signification as Arabic mu3ammi# unifier (which was applied to |uayy [q.v.]), possibly implying that he was head both of Sabaeans proper

and of non-Sabaean elements in the federation. A few early inscriptions do contain references to kings ("mlk) of Marib, as well as to kings of other small communities such as Haram, Na9, etc. It remains uncertain whether or not a mkrb of Saba" was simultaneously a king of Marib.

The archaic flowering of Sabaean culture lasted until some time after the middle of the first millennium B.C. The fact that through the fourth, third and second centuries B.C. the important frankincense trade was in the hands of the Ma#n folk suggests some falling-off in Sabaean ascendancy.

The second great flowering of Sabaean culture was in the first three centuries A.D., by which time a very different political picture had emerged. The various folks of the 2,000 m highland zone played a much more dominant role; and some of their leaders, who traditionally bore the title wl prince (in later Sabaic and in Arabic, ayl ), founded dynasties who ruled as king of Saba" or king of Saba" and 9(lord of) Raydan. The dual title has been presumed to be the origin of the remark in the late first century A.D. Greek document known as the Periplus that a single ruler named Charibael was king of two nations, Himyarites and Sabaeans, and had his residence at afr (near modern Yarim), of which the adjacent citadel is named Raydan (see imyar ). Throughout these three centuries there was confused situation, with Sabaeans and Himyarites sometimes at war with each other, sometimes united under a single monarch as had been the case under Charibael; occasionally there appear to have been two rulers reigning simultaneously in Mrib and afr and both claiming the dual title king of Saba" and 9uRaydan. Somewhat oddly, the indigenous inscriptions of this period never speak of a king of the Himyarites; on one occasion when a king of Saba" was at war with the Himyarites, he alludes to his antagonist simply as the Raydanite (much as a European might speak of the Hapsburg).

In addition to the Sabaeo-Himyarite conflicts, there were wars waged by varying alliances among Saba", |ataban, Radman (see above) and 0a'ramawt, and also involving Abyssinians (0aba9at [q.v.]) settled in the Red Sea coastal region. But by the beginning of the 4th century A.D., 9ammar Yur#i9, whom the Arab writers call the first Tubba# [see tubba#] had put an end to these conflicts by eliminating |ataban and 0a'ramawt, and for the first time uniting the whole of what is today Yemen, employing the title king of Saba" and 9uRaydan and 0a'ramawt and | [VIII:664b] the South (Ymnt); in the 5th century this title was further enlarged by the addition of and their Arabs (i.e. Bedouin) in the highland and the Tihma.

The 4th to 6th centuries A.D. are thus politically speaking a Himyarite period and do not properly belong to the history of Saba", despite the fact that what is called the Late Sabaic language continued to have great prestige value and to be employed for epigraphic purposes.

Mediaeval Arab writers have preserved for us from the Himyarite period a mass of oral traditions which contain much authentic material mingled with folklore motifs. But they knew practically nothing about the genuine history of Saba" before the 4th century A.D., though they do mention the names of one or two of the most prominent individuals of the first-third centuries, notably king Il9ara Ya'ib, whom they credit with the building of the famous palace of 9umdn [q.v.] in -an#".

3. Religion.

The religion of all four South Arabian folks down to the beginning of the 4th century A.D. was a polytheistic paganism. Though it is probable that this may have survived among the peasantry and in remoter parts of the kingdom through the 4th to the 6th centuries, the upper classes, who are the authors of our inscriptional material, went over to some form of monotheistic creed, a cult of the Merciful (Rmn-n), the Lord of Heaven, which could perhaps best be described as 0anafite [see anf ] since it is devoid of explicit marks of either Judaism or Christianity. At the same time, already from the end of the 4th century, a few explicitly Jewish texts attest an influential Jewish presence, and in the 6th century under Abraha [q.v.] Christianity prevailed.

For the period down to the early 4th century A.D., few would now agree with the excessive reductionism of D. Nielsen, who in the 1920s held that all the many deities in the pagan pantheon were nothing more than varying manifestations of an astral triad of sun, moon and Venus-star; yet it is certainly the case that three deities tend to receive more frequent mention than the rest. The first, in the sense that his cult is found among all four of the South Arabian folks and that in invocations of several deities his name normally comes first, is #A9tar, a male counterpart of north Semitic I9tar/#A9tore9/Astarte. He is often qualified by the epithet eastern and occasionally by the complementary one western, which tends to support the commonly accepted identification of him with the planet Venus, regarded as morning star and evening star.

But just as the Greek local patron deities such as Athene in Athens, Artemis in Ephesus, etc., figure more prominently than the remoter and universal Zeus, so in South Arabia the most commonly invoked deity was a national one, who incorporated the sense of national identity. For the Sabaeans this was "lmh (with an occasional variant spelling "lmhw). A probable analysis of this name is as a compound of the old Semitic word "l god and a derivative of the root hw meaning something like fertility (cf. Arabic ah flourish); the h is certainly a root letter, and not, as some mediaeval writers seem

to have imaginged, a t" marba, which in South Arabian is always spelt with t. The federal significance of this deity appears notably in the fact that at the shrine on 3abal Riym (Arab) the worshippers of the local folk-deity Ta"lab were instructed that they must not omit to make an annual pilgrimage to "lmh in Mrib.

Many European scholars still refer to this deity in a simplicistic way as the moon god, a notion stemming from the triadic hypothesis mentioned above; | [VIII:665a] yet Garbini has produced cogent arguments to show that the attributes of "lmh are rather those of a warrior-deity like Greek Herakles or a vegetation god like Dionysus. (The remarks made in the art. atabn on this topic as it affects the situation in |ataban went to press before Garbini's article had appeared; they now need modifying in the light of that article.)

Nevertheless, the moon certainly had much religious significance. A very common symbol engraved on altars and religious buildings shows a crescent embracing a disc. It is presumably this symbol that the Muslim writers had in mind when they say that the first act of the day for an ancient Yemeni king was to bow down to the images of the sun and moon. This is not to say, of course, that they were right in seeing the disc as representing the sun; some modern scholars have been inclined to think it represents the planet Venus.

The place occupied by the sun in the pantheon is not easy to assess. The Radman folk had as their national patron-deity s 2 ms 1 #lyt Lofty Sun, and elsewhere there are mentions simply of Sun without qualification. But it is dubious whether the majority belief is justified, that numerous references to a feminine deity described simply as She-who-is-possessed of ( 9t ) a certain quality, are necessarily to a solar goddess (too often, the interpretation proposed for the term describing the quality has been dictated by the preconception that it must be a quality of the sun).

Certain of the ancient religious practices have a special interests in that they have survived in some form or another until the present day. Worth mentioning are the communal pilgrimages (ziyrt) on prescribed days; a code of ritual purity (see Ryckmans); and a ritualised hunting of the ibex, thought of as connected with the divine blessing of rain (Ryckmans and Serjeant).

4. Saba" in Bible and |ur"n.

The visit of the Queen of Sheba to king Solomon, and the abundant accretions of legend around it [see bils ], have been too extensively discussed to need mention here, except for the remark that there is a possibility that such a visit might have been associated with a trade mission, like the missions to the Assyrian kings (see above). In the |ur"nic allusion (XXVII, 27 ff.) the name Saba" does not occur; she is simply the queen of the south. But Saba" does feature in a passage (XXXIV, 15-16) which is one of those where the fate of ancient peoples is mentioned as a warning against worldly pride. The prosperity of the Mrib oasis (situated on each side of the wadi bed, hence the garden of the left and the garden of the right) had been dependent on the maintenance of the great dam in good order; after the death of king Abraha the political fabric that had made repairs possible crumbled, the irrigation system was destroyed and the oasis was devastated.

5. Sabaeans in Africa.

Around the middle of the first millennium B.C., there were Sabaeans also in the Horn of Africa, in the area that later became the realm of Aksum (Eritrea). The evidence consists of only a scanty number of inscriptions, which, however, make it clear that we have to do with genuine Sabaeans, holding to the national cult of "lmh. They were mixed up with various non-Sabaean communities, and it is still much in dispute how one can envisage the actual demographic (and political) situation. There are five places in the Bible where the writer distinguishes 9eba () son of Yoan (who appears in the Arab genealogies as |an [q.v.]), i.e. | [VIII:665b] the Yemenite Sabaeans, from Seba () son of Ku9, implying an African habitat. This spelling differentiation, however, may be purely factitious; at all events the indigenous inscriptions make no such difference, and both Yemenite and African Sabaeans are there spelt in exactly the same way. |A0$N , according to the consensus of opinion among Muslim genealogists, historians, and geograph- | [IV:448a] ers, and in popular tradition, the ancestor of all the South-Arabian peoples the ancestor of all the South-Arabian peoples [see yaman ], whence he is sometimes known as father of all Yaman, the Yamans themselves being called ban |an, ab"il |an , or simply |an . He thus corresponds to #Adnn [q.v.], the common ancestor of the northern Arabs, though some authorities prefer to contrast him with one or other of #Adnn's descendants, e.g., his son, Ma#add (al-Dnawar, 281; al-abar, ii, 1056, 1084; al-Mas#d, al-Tanbh, 88), or his grandson, Nizr (al-Mas#d, Mur3, v, 223, vi, 42 f., 46, 143, 150; Ibn al-A9r, iv, 273). The normal genealogy given for |an is NSmArfa99a99la9#$bar|an, and he is also credited with a brother, Fla9. In this we may see an adaptation of the Yahwist tradition in the Table of the Nations (Genesis, X, 21-5; I Chronicles, I, 4, 17-9); Noah ShemArphaxadShelahEberPeleg + Joktan, and this is confirmed by the Arab insistence on the identity of |an with Joktan (Yaan), who was the ancestor of several peoples of patent South Arabian reference. Some genealogists do admittedly make Yaan a brother or son of |an (e.g., Ibn |utayba, 14; al-Mas#d, Mur3, i,

79 f.; al-abar, i, 217), but this is certainly a confusion on their part. The claim, however, that |an is an Arabicized form of Yaan (e.g., T3 al-#Ars , s.v.) is phonologically hazardous and need not be pressed. Hebrew Yon is rather to be related to the verb an to be small, weak, thus the younger (brother), and one might compare the occurrence in Old South Arabian of the term qn to denote collateral descent from a clan-head. |an, on the other hand, is now attested as a tribal name even before Islam. An inscription of the time of the second century Hamdnid ruler, #R m "WTR (Jamme 635/26-7) alludes to campaigns in the country to the north of Na3rn against a king of Kinda and |an (mlk/kdt/wqtn), and this may provide additional support for the identification of Ptolemy's Katantai (Geogr., book vi, chapter 7, 20, 23) with |an. Though virtually nothing is known of the rle of this people in pre-Islamic times, it seems reasonable to suppose that the apparent similarity of the name with Yaan led the Arab genealogists to make the identification in order to provide the South Arabian peoples with a respectable biblical ancestry, just as the Northern Arabs, under the influence of the Bible and the |ur"n, had been linked with Ishmael, son of Abraham, through the fictitious #Adnn. The tribal confederation of the |an is subdivided into two groups, the smaller of the 0imyar and the larger of the Kahln. The two were officially regarded as brothers and their descent from |an was established by the interpolation between biblical Joktan and his son Sheba of two further generations represented by Ya#rub and Ya93ub. The 0imyar, as progenitors of the great pre-Islamic kingdoms of South Arabia, were probably settled, while the Kahln were essentially nomads (cf. Landberg, Arabica, v, 116 ff.). Most, indeed, of the Southern tribes which had settled in North Arabia, Syria, and #Ir by the advent of Islam claimed descent from Kahln.

The hostility between the |an and the Ma#add seems to go back to pre-Islamic times and may find its origin in the opposition between the desert and the sown. This enmity was intensified by the repeated raids of the Yamans into the lands of the Ishmaelites as well as by the later antagonism between the Anr (Medinans) and the |uray9, which came to a head after the death of the Prophet and influenced the | [IV:448b] history of the first two centuries of Islam in the most baneful fashion. It was perhaps this feud that first linked the Yaman tribes on the one side and the Ishmaelites on the other into closer ethnological unities. One of its more innocuous results was the muf9ara , the struggle for rank and glory, which continually prevailed between the two rivals. The | an, in view of the splendour of the ancient South Arabian kingdoms, had the more right at first to feel the more distinguished. But the mission of Muammad and the primacy of the |uray9 brought the Ma#add a tremendous superiority. The Yamans endeavoured to counterbalance this by creating a South Arabian saga, which pictured their past greatness in the most splendid colours. They then made |an son of the prophet Hd [q.v.], whom they next partly identified with #$bar. (The present-day | an still regard Hd as |an's father, though they are unaware of his shrine in the Wd 0a'ramawt.) Finally they tried to connect themselves with the #Adnn genealogy, partly by making the ancient 3urhum [q.v.], the brothers-in-law of Ism#l, to be direct descendants of |an, but especially by giving |an a genealogy direct from Ism#l, who thus became father of all the Arabs. They may also be responsible for the theory that the |an, together with the so-called lost Arabs ( al-#Arab alb"ida), represent the genuine (primary) Arabs ( al-#Arab al-#riba or al-#arb" etc.),

while the Ma#add were Arabicized (secondary) Arabs ( al-#Arab al-muta#arriba ). Another theory makes the lost Arabs al-#Arab al-#riba, while the |an are al#Arab al-muta#arriba , and the Ma#add al-#Arab al-musta#riba (cf. Lane, Lexicon, s.v. al-#Arab ).

The |an at present are for the most part beduin and form a very large group of tribes covering the area between B9a in #Asr and 0awa in Central Arabia. Those that Doughty met in 0"il claimed descent from Hd and traced their ancestral home back to the mountainous country around al-r in #Asr. Perhaps because of their isolated habitat on the northern fringe of the great southern desert, strange tales were reported of their customs and way of life and they acquired a reputation, perhaps justified, for savagery and cruelty. They are said to be 0anbal Sunns and derive their livelihood from camel breeding. The |an of #Asr differ from their nomadic relatives by living well-conducted lives and prospering in trade and agriculture. They form a federation of six autonomous tribes in the region just east of Abh, living independently of one another and uniting only in times of crisis. Although al-Hamdn knew of the individual tribes, he was apparently unaware of their description as |an. It is not improbable, however, that they are survivals of the pre-Islamic tribe. Al-Muaddas, writing some fifty years after al-Hamdn, mentions a district of |an between Zabd and -an#" and alludes also to a clan, the $l |an, northwest of Na3rn, whom he describes as the oldest princes of Yaman. ADN$N , ancestor of the Northern Arabs ancestor of the Northern Arabs according to the genealogical system which received its final form in the work of Ibn al-Kalb, about 800 A.D. The name occurs twice in Nabatean inscriptions from N.W. Arabia (#Abd #Adnn, #Adnon; Jaussen et Savignac, Mission Archologique en Arabie, Paris 1909-14, nos. 38, 328) also in Thamudic (Lankester Harding/Littman, Some Thamudic Inscriptions , Leiden 1952) and was taken to South Arabia along the incense-route (Corpus Inscriptionum Semit., iv, no. 808). As already noted by al-3uma, abat(Hell), 5 (cf. also Ibn #Abd al-Barr, al-Inbh #al |ab"il al-Ruwh, Cairo 1350, 48), it does not occur in pre-Islamic poetry at all (Labd, x1i, 7 is spurious), and only very rarely in early Islamic literature. This means that the name does not owe its place in the system to the conflict of parties in the Umayyad period, like Nizr and Rab#a, but is of pre-Islamic origin, although it does not spring from bedouin tradition. It may come, like other rudimentary elements of the system, from the Meccan tradition.It is noteworthy that, owing to the revival of national feeling, the name #Adnn again became current in Turkey by the last quarter of the 19th century. This is explained by the fact that the Young Turkish movement represented in its earliest stage on Ottoman nationalism which included also the Arabic traditions. LA|AB (a.) nickname nickname, and at a later date under Islam and with a more specific use, honorific title

honorific title (pl. alb). For suggestions about its etymology, see L. Caetani and G. Gabrieli, Onomasticon arabicum. i. Fonte-introduzione, Rome 1915, 144-5; and for its place in the general schema of the composition of Islamic names, see ism . The laab seems in origin to have been a nickname or sobriquet of any tone, one which could express admiration, be purely descriptive and neutral in tenor or be insulting and derogatory. In the latter case, it was often termed nabaz, pl. anbz, by-form labaz; cf. alBay'w on sra XLIX, 11, In common usage, nabaz is particularised for an unpleasant laab. The grammarians, in their love for schematisation, divided alb into simple ( mufrad ) ones, comprising only one word, and compound (murakkab) ones, comprising two or more words, but this division has neither historical nor semantic significance, beyond the obvious point that as Islamic society developed and became more complex, honorific titles tended to become lengthy and grandiloquent. 1. The pre-Islamic and the earliest Islamic periods . Already in these times, we have many examples of Bedouin leaders, poets, orators, etc. with nicknames, e.g. $kil al-Murr, al-Abra, al-Mutalammis, #$"id al-Kalb, Muabbil al-Rih, etc., and from the #Abbsid period we have Dk al-3inn, -ar# al-9awn, etc. These names might relate to physical characteristic or defects, to striking lines in a poet's verses, to incidents in the holder's life, but often to events whose significance is now lost in the mists of time. Explanation of the more unusual and bizarre names drew the authors of literary biographical works into much fanciful | [V:619a] theorising; cf. A9n , ed. Bl, xviii, 209, ed. Beirut, xviii, 480-1, on the origins of the name Ta"abbaa 9arran, and for a useful list of such names, together with explanations drawn from the sources, A.-C. Barbier de Meynard, Surnoms et sobriquets dans la littrature arabe, in JA , Ser. 10, ix (Jan.-June 1907), 173-244, 365-428, x (July-Dec. 1907), 55-118, 193-273. The giving of nicknames was clearly widespread in early Arab society, where the number of personal names was limited and where there was, as in the Arab world today, keen observation of personal foibles, physical peculiarities, etc.; and from the Arabs, the practice spread to the Persian, Turkish and wider Islamic world. The personal connection of such nicknames is presumably behind the oftquoted anonymous line of poetry, It is rare that you see a man whose character is not revealed, if you consider the matter, in his laab Al-9a#lib, who devotes two out of the ten chapters of his La"if al-ma#rif to the alb of prominent literary and political figures, quotes this verse in reference to the philologist al-Mubarrad [q.v.], and makes the curious observation that the people of Ba9dd and N9pr were celebrated for their facility in coining appropriate nicknames; he gives numerous examples from each town, some quite grotesque, but unfortunately omitting to explainif he ever knewthe origins of the names (ed. I. Abyr and 0. K. al--ayraf, Cairo 1960, 46, 53-4, tr. C. E. Bosworth, The book of curious and entertaining information, Edinburgh 1968, 63, 66-7). Some nicknames were clearly given as nomina boni augurii, aimed either at attracting a favourable future for the recipient or else at averting harm. Thus arose the practice of giving antiphrastic and apparently unpleasant names and nicknames in order to avoid attracting the evil eye [see #ayn ] which might otherwise fix itself on something wholesome and perfect; whilst slaves and other persons of little importance might be given euphemistic and euphuistic names, such as those of flowers, jewels, perfumes, etc. Thus the caliph al-Mutawakkil called one of his slave concubines, the mother of al-

Mu#tazz, |aba the hideous one because she was famed for her beauty. Euphuism is seen in such names as Yt (lit. jacynth, ruby) and 33ak (< Turkish iek flower, the slave mother of the caliph al-Muktaf), and a good example of antiphrasis in the name of Kfr al-I99d [q.v.] ( kfr camphor being white and fragrant, whereas Kfr was a black eunuch, proverbially noisome and malodorous [see 9a. i. In the Central Islamic lands]). In these connections, there is an interesting passage in Ibn Durayd's K. al-I9ti , ed. Wstenfeld, 4-5, in which he quotes al-#Utb as having once been asked, How is it that the Arabs give their sons names which are considered unpleasant (musta9na#a), yet give their slaves names considered pleasant (mustasana)? Al-#Utb replied that the Arabs gave their sons names with their enemies in mind, whereas they gave their slave any names they wanted (i.e. without any ulterior motive). Ibn Durayd then goes on to give examples of names given to sons which would augur well against their enemies (taf"ul an #al a#d"ihim), expressing violence, harshness, bellicosity and endurance, e.g. 9lib, lim, Mubil, 9bit, etc. (see further, M. P. Kister, Call yourself by graceful names ..., in Lectures in memory of Professor Martin B. Plessner, Jerusalem 1970). We are here perhaps straying into | [V:619b] the realm of the ism proper rather than the laab , although the names given to slaves and slavegirls may be considered as alb replacing those persons' first given names. The fondness of the Arabs, and of |uray9 in particular, for giving satirical and opprobrious nicknames is condemned in |ur"n, XLIX, 11, where there is a warning against the practice of groups of both men and women mocking at each other, Do not scoff at each other or give each other derisory nicknames (wa-l talmiz anfusakum wal tanbaz bi 'l-alb). This sra is accounted late Medinan, from a year or two before Muammad's death; by Nldeke (Gesch. des Qorns, i, 221) and Montgomery Watt (Companion to the Qur'an, London 1962, 237); it may be that the Prophet was feeling particularly aggrieved by nicknames given to him in the past by his opponents and by the Hypocrites, like that of al-Abtar the childless one, even though the sabab al-tanzl was, according to the commentators, the women's insults to Muammad's wife -afiyya bint 0uyayy because of her origin from the Jewish tribe in Medina of al-Na'r (cf. SaleWherry, A compendious commentary on the Qurn, London 1896, iv, 70-1). Yet whilst retaining for some time to come its derogatory and insulting aspects, the laab tended to be transmogrified into higher roles within Islamic society. In one direction, it evolved into the nomde-plume of authors and artists, the ma9la or ta9allu [q.v.], above all in the Persian, Turkish and Indo-Muslim worlds, such as we find in the name of the Persian poet 9n [q.v.] or the pseudonym adopted by the -afawid 9h Ism#l I for his Azeri Turkish verses [see Ism#l I. 2. His poetry]. Most significantly for the development of Islamic culture, the laab developed from being a nickname of praise or admiration (as in such examples as Mul#ib al-Asinna he who played with lances for the pre-Islamic poet #$mir b. Mlik, famed for his prowess in battle) into becoming an honorific title, conferring status and prestige on its owner, since it frequently implied a specially close relationship to the sovereign or the divinity or else a reward for personal bravery or services to the state. As such, the laab was to have an extended life-span in the Islamic worldmore particularly, in those parts of lying between Egypt and India, though these honorifics were also to be found to a more limited extent in Muslim Spain and North Africa, see below, 3right up to the 19th and early 20th centuries, when floridity and hyperbole went out of Islamic titulature just as they did from literature and other aspects of culture.

It is not difficult to discern why such titles became eagerly sought after; the desire to stand out above the rest of mankind is a universal one. Indeed, al-|ala9and, in his -ub al-a#9 , v, 440, following the line of traditional Islamic aw"il literature writers, traces what he calls the alb al-mad wa-nu#t back to such figures as Abraham al9all the friend of God, Moses al-Kalm the one who spoke with God, Jonah 9u 'l-Nn the man in the fish, etc. Within a formalised, hierarchial society such as the Islamic one became in the #Abbsid period, modes of address, insignia or rank and office, dress, etc. all contributed to the fixing of a man's status in society and the state, and as such were prized as the visible and audible symbols of success in the temporal world. In this process of the spread of honorifics, there wereas for other aspects of early Islamic culturethe precedents of | [V:620a] the earlier Near Eastern civilisations, the Romano-Byzantine and the Persian ones, to provide an element of continuity with the past. The early Arabs knew from their contacts with the Byzantines on the Syro-Palestinian borders that a complicated system of honorific titles and modes of address existed for Byzantine civil and military dignitaries. Allies of the Greeks, such as chiefs of the 9assnids [see 9assn ] like al-0ri9 b. 3abala and his son al-Mun9ir, were honoured by having the honorary title of patricius bestowed on them; this term found its way into general Arabic usage as bir/bar [see bir]. In Achaemenid Persia, the characteristic expression of pomp and magnificence (megaloprepea) through grandiloquent titles was familiar to Herodotus (i, 139), who further mentions (viii, 85) that meritorious servants of the state were enrolled in the list of the king's benefactors (orosgghw). Under the Ssnids, the great men of state were likewise honoured by the award of a wide range of honorifics, e.g. the title buzurg framadr chief executive for the emperor's chief counsellor and minister; Mahi9t the greatest [of the ruler's servants]; Hazrmard [with the strength of] 1,000 men for military commanders; and various compounds including the sovereign's name, e.g. Zhye9n 9usraw eternal 9., Hormuzd Varz wild boar of H., and many with tahm strong like Tahm 9pr or Tahm Yazdigird. These honours were usually completed by the grant of robes of honour, the later Islamic 9il#a [q.v.] (see Nldeke-abar, Geschichte der Perser und Araber zur Zeit der Sasaniden, 8-9, 443 and n. 1; F. Justi, Iranisches Namenbuch, 318; A. Christensen, L'Iran sous les Sassanides 2, Copenhagen 1944, 113-14, 326, 40911). 2. The period of the caliphate . The wide dissemination of honorific titles in the Islamic world began at the top, with the adoption of regnal titles by the caliphs. Later historians, projecting backwards what had become common practice in their own time, attributed the use of such titles to the Umayyads and even to the Orthodox caliphs. Al-Mas#d, in his, K. al-Tanbh, 335, tr. 431-3, cites pro-Umayyad traditions to the effect that the Umayyads had honorifics of the theocratic type which had by two centuries later become familiar from #Abbsid practices, e.g. al-Nir li-a Allh for Mu#wiya, al-|hir bi-#awn Allh for Maslama b. #Abd al-Malik and al-Muta#azziz bi'llh for Ibrhm b. Wald b. #Abd al-Malik; but he rightly rejects these traditions as weak and uncorroborated by the historians and biographers (cf. E. Tyan, Institutions du droit public musulman. i. Le califat, Paris 1954, 486-7). Similarly, the epithets attributed to the Orthodox caliphs, such as al--idd he who testifies to the veracity of the Prophet's mission for Ab Bakr, al-Fr the just

for #Umar and 9u 'l-Nrayn possessor of the two lights for #U9mn, are probably of early growth (cf. al-9wrazm, Maft al-#ulm, ed. van Vloten, 105), but must nevertheless have been applied only after the deaths of their owners. Honorifics of a theocratic nature, expressing dependence on God, reliance on Him, or participation with Him in the work of ruling, are really an innovation of the early #Abbsid period. B. Lewis has stressed the importance of the atmosphere of messianic expectation around the time of the #Abbsid revolution, when the worldly Realpolitik of the Umayyads and their animus against the Prophet's kinsfolk, the #Alids, was replaced by what it was | [V:620b] hoped would be the reign of divinely-guided justice; and these feelings favoured the adoption of religious-based titles by al-Manr and his successors. The first #Abbsid caliph, Abu 'l-#Abbs al-Saff, had no formal regnal title, though his partisans may well have attributed to him various names with messianic implications. His successor Ab 3a#far formally adopted the laab of al-Manr bi'llh he who is helped by God to victory, and the chiliastic tone of early #Abbsid titulature is especially clearly seen in the alb of his successor al-Mahd bi'llh the divinely-guided one and then of al-Hd il 'l-a the one who directs towards the divine truth (cf. Lewis, The regnal titles of the first Abbasid caliphs, in Dr. Zakir Husain presentation volume, New Delhi 1968, 13-22; F. Omar, A note on the laqabs (i.e. epithet) of the early #Abbasid caliphs, in #Abbsiyyt, studies in the history of the early #Abbasids, Ba9dd 1976, 141-7). In subsequent #Abbsid usage, the apocalyptic element is less prominent, but the honorifics still express reliance upon and submission to the deity, or confidence in His guiding power, e.g. al-W9i bi'llh he who puts his trust in God and al-Mu# bi'llh he who shows himself obedient to God (cf. Tyan, op. cit., 483 ff.). It is notable that the #Abbsid honorifics up to the later 4th/10th century and the period of Byid control in Ba9dd are formed essentially with participles or adjectives of a passive or reflexive character, thereby emphasising the supremacy of God's controlling power and the vital securing of His favour for the business of ruling in conformity with the 9ar#a , the ideal, if not always the practice, of the #Abbsids. A. Abel has noted that the #Abbsids' great rivals in North Africa and Egypt, the Fimids, had, in order to press their superior claim, as they saw it, to the caliphate and immate, to adopt a more aggressive and active form of titulature (Le khalife, prsence sacre, in Stud. Isl. , vii [1957], 38 ff.). This new type of titulature reflected the role assigned in Ism#l cosmogony to the Imm s in the hierarchy of intelligences emanating from the godhead, and it actively associated the holders of these titles with God's direct working in the world. Thus we have al-|"im bi-#amr Allh he who takes charge of the execution of God's command for the Mahd #Ubayd Allh's successor Abu 'l-|sim Muammad; and from the time of al-Mu#izz's accession onwards (341/953), the honorifics of the Fimids are all active participles or adjectives emphasising the ruler's decisive part in the implementation of God's will in this world, e.g. al-0kim, al-$mir, al-fir, etc., usually with a complement like ... li-dn Allh or ... bi-amr Allh . Not surprisingly, the #Abbsids strove from the latter years of the 4th/10th century onwards to emulate these activist forms of titulature, in order to emphasise their own position as upholders of the Sunna; whence titles like al-|"im bi-#amr Allh appear in the middle years of the 5th/11th century and ones like al-Nir li-dn Allh and alhir bi-amr Allh in the early 7th/13th one. In all these cases, the laab might be adopted as a new regnal title on the caliph's accession; but increasingly, the wal 'l-#ahd

or heir to the throne was given a laab as soon as he was invested formally as heir to the reigning monarch, with his own kunya and/or honorifics thereafter appearing on the coinage side-by-side with those of the actual caliph. This laab usually remained with the heir once he acceded to the caliphate proper; but occasionally, the laab adopted by such an heir-apparent was exchanged | [V:621a] for a different, often more grandiloquent one on his obtaining the throne. Thus al-Mutawakkil's original laab , during the period when he was his brother's heir, was al-Muntair, but this was changed on the second day of his succession to the caliphate (232/847) by the chief ' Ab Du"d to al-Mutawakkil (al-Mas#d, Mur3 al-9ahab , vii, 189 = ed. Pellat, 2872; cf. Tyan, op. cit., 280, 486). If the caliphs could assume splendid and sonorous honorifics themselves, it lay also within their power to confer titles on their servants and supporters, in the first instance to their viziers and secretaries and to their military commanders. The process dates from the early 3rd/9th century, and at least until the following century it was practised comparatively sparingly by the #Abbsids. The Persian secretary al-Fa'l b. Sahl [q.v.], former protg of the Barmaks and eventually vizier to the caliph al-Ma"mn, exercised the functions both of wazr and amr , and by 196/811-12 he is found with the honorific 9u 'l-Riysatayn possessor of the two executive functions, sc. of civil administration and military leadership. Around the same time, the Persian general hir b. al-0usayn [q.v.], founder of the line of hirid governors in 9ursn and #Ir, received the laab of 9u 'l-Yamnayn possessor of two right hands, the ambidextrous (various explanations for this phrase are in fact given in the sources; it had already been used in the earliest Islamic times as a synonym for liberal, open-handed, e.g. by the great poetess al-9ans" [q.v.] when elegising her brother -a9r); al-Fa'l b. Sahl's cousin #Al b. Ab Sa#d received that of 9u 'l-|alamayn possessor of the two pens (either alluding to the two principal government departments of finance and the army, or else to the two modes of writing, sc. Arabic and Persian); whilst later in the century, in 269/885, the vizier Sa#d b. Mu9allad was granted by al-Muwaffa the title 9u 'l-Wizratayn possessor of the two vizierates (referring either to the two spheres of power, civil and military, or recognising Sa#d's role as servant both of alMuwaffa and the titular caliph al-Mu#tamid) (see D. Sourdel, Le vizirat #abbside, i, 201-3, 319-20, ii, 678, 681). This type of laab containing a dual expression was, however, one known before this time, as is attested by the name 9u 'l-Nrayn for the caliph #U9mn and 9u 'l9ahdatayn, applied to the Companion and partisan of #Al, 9uzayma b. 9bit alAnr (because the Prophet had promised him double the normal martyr's rewards, see Ibn 0a3ar, Iba, i, 425, and Ibn al-A9r, Usd al-9ba , ii, 114). The type was, indeed, to enjoy a long life in Islamic titulature, just as dual forms of the simple ism (Muammadayn, al-0asanayn) have been current in the Islamic world until the present day. The #Abbsids' successors in western Persia and #Ir, the Byids, continued the usage of the caliphs for their own ministers and secretaries; thus we find Rukn alDawla's great vizier Abu 'l-Fat b. al-#Amd with the laab of 9u 'l-Kifyatayn possessor of the two capabilities (i.e. of the sword and pen), though this title was in fact awarded directly by the caliph [see ibn al-#amd ]. In the following period of the Sal3s, titles in al-0a'ratayn, referring to the separate courts of the #Abbsid caliphs and the Sal3 sultans, occur, e.g. 9iat al-0a'ratayn confidant of the two courts, and Nim al-0a'ratayn support of the two courts, the title held by the Nab al-Nuab" #Al b. arrd al-Zaynab in Ba9dd during the | [V:621b] latter decades of the

5th/11th century ( Ibn 9allikn, ed. Isn #Abbs, Beirut 1968-72, iv, 454, tr. de Slane, iii, 151). See for all these types of dual titles, I. Goldziher, Ueber Dualtitel, in WZKM , xiii (1899), 321-9 = Gesammelte Schriften, Hildesheim 1967-73, iv, 195-203, and Caetani and Gabrieli, Onomasticon arabicum, i, 167-9. An interesting fact of usage has recently been highlighted by P. Balog. Pious invocations after names are familiar from the early use of the taliya after the name of the Prophet, the taslm after those of earlier prophets, the tar'iya after the names of the early caliphs, etc. From the latter Umayyad period (from the time of Yazd II b. #Abd al-Malik [101-5/720-4] onwards, in regard to the evidence of inscriptions on glass weights and measure stamps from Egypt) onwards, pious invocations like alaahu Allh may God set him in the way of righteousness!, abhu Allh may God grant him long life! and akramahu Allh may God honour him! are consistently appended to the names of caliphs and of high officials and governors. It seems that a particular formula like those just mentioned remained with an official or governor throughout his career, unless he were awarded a fresh invocation of higher prestige. Such ad#iya (sing. du#" ) may accordingly be approximated to alb as expressions of particular honour awarded by the ruler to a faithful servant, and in the manuals of secretaryship from the Ayybid and Mamlk periods (see below, section IV), the two types of honorific are often treated together. See Balog, Pious invocations probably used as titles of office or as honorific titles in Umayyad and #Abbsid times, in Studies in memory of Gaston Wiet, ed. M. Rosen-Ayalon, Jerusalem 1977, 61-8. Characteristic of the 4th/10th century was the appearance and then the growing popularity of compound honorifics including as their second elements the terms dn faith and dawla or mulk secular power, or less commonly, compounded with umma religious community and milla religion. These titles have been studied in detail by M. van Berchem, Eine arabische Inschrift aus dem Ostjordanlande mit historischen Erluterungen, in ZDPV , xvi (1893), 84-105; Caetani and Gabrieli, op. cit., i, 199 ff.; J. H. Kramers, Les noms musulmans composs avec Dn , in AO , v (1927), 53-67; and A. Dietrich, Zu den mit addn zusammengesetzten islamischen Personnamen, in ZDMG , cx (1960), 43-54. The last three of these studies contain extended lists of these types of laab , for they were to enjoy a popularity which ran for almost a millenium, and Dietrich puts the total of dn titles enumerated by Kramers and himself at 186. The dawla titles were the earliest to appear, but in the first place, were titles of great honour granted only exceptionally by the caliphs. Thus when al-Muktaf honoured his vizier al-|sim b. #Ubayd Allh of the Ban Wahb (vizier 289-91/902-4) by giving to him one of his daughters in marriage, he also awarded him the title Wal 'l- Dawla friend of the state as a special mark of intimacy (Sourdel, Vizirat, i, 356). In 330/941-2 the 0amdnid of Mawil Ab Muammad al-0asan received from al-Mutta, as a reward for his services against the amr al-umar" Muammad b. R"i, the laab of Nir alDawla, whilst his brother Abu 'l-0asan #Al of Aleppo obtained that of Sayf al-Dawla. Shortly afterwards, in 334/945, the Daylam Byids entered Ba9dd and took over the office there of amr al-umar", and the three Byid brothers Amad #Al and 0asan b. Bya received the honorifics of Mu#izz | [V:622a] al-Dawla, #Imd al-Dawla and Rukn al-Dawla respectively. All the subsequent Byid amrs obtained from the caliphs titles of this type, sometimes with a greater degree of elaboration, e.g. the #A'ud alDawla wa-T3 al-Milla of Fan-9usraw, the 9araf al-Dawla wa-Zayn al-Milla conferred on Abu 'l-Fawris 9rzl by the caliph in 376/986, and the Bah" al-Dawla wa-4iy" al-Milla wa-9iy9 al-Umma of Ab Nar Frz b. #A'ud al-Dawla (see

Mez, Die Renaissance des Islms, 133, Eng. tr. 135; L. Richter-Bernburg, Amir-malikshhnshh: #A'ud al-Daula's titulature re-examined, in Iran , Jnal. of the BIPS, xviii [1980],; and bah" al-dawla in Suppl.). It may be noted, in connection with the mention of Muammad b. R'i [see ibn r"i ], that it is in the first half of this century that we have the formal constitution of the office of amr al-umar" [q.v.] as the concomitant of caliphal decadence. In fact, this title, or a similar from like kabr al-umar", appears in the later years of the 3rd/9th century in reference to the commander-in-chief of the caliphal armies; but only with the caliphate of al-R' (322-9/934-40) did the holder of the imrat al-umar" achieve a commanding grip on affairs, so that it seemed natural for the Byids to step into the office a few years later (see Tyan, op. cit., 531-41). Especially interesting, in the light of the significance of the Byids and the dawlat alDaylam or reign of the Daylams for an attempted reconstruction, in Islamic form, of the ancient kingship of Iran, is their assumption of the ancient imperial title 9han9h emperor of emperors. #A'ud al-Dawla used it widely and with great pride in his titulature and protocol, and indeed the title became so characteristic of the Byids that the 9aznawid historian Bayha actually refers to them as the 9han9hiyn (Ta"r9-i Mas#d , ed. 9an and Fayy'1, 41, 400, 438), but it seems possible that use of the title goes back to the very beginnings of the Byids' seizure of power in western Persia and #Ir, and was begun by #A'ud al-Dawla's father Rukn al-Dawla or even before that by his uncle #Imd al-Dawla. Its assumption without caliphal permission clearly indicates a claim to temporal power by the Byids independent of any caliphal act of delegation; but after #A'ud al-Dawla's death, his weaker and squabbling successors were compelled to seek caliphal support and hence pay more respect to the caliphal fiction by seeking validation for the title 9han9h directly from the #Abbsids. As is well-known, al-|"im complied over this when asked by 3all al-Dawla, but when in 429/1038 the title was introduced publicly into the Ba9dd 9uba for the first time, a near-riot ensured, and the approval of the fuah" of the three law schools of the East, the 0anafs, the 0anbals and the 9fi#s, had to be sought before this ostensibly blasphemous (at least in the eyes of the pious) title could safely be re-introduced from the pulpits (see H. F. Amedroz, The assumption of the title 9han9h by Buwayhid rulers, in Num. Chron., Ser. 4, vol. v [1905], 393-9, and the detailed and important study of W. Madelung, The assumption of the title Shhanshh by the Byids and The reign of the Daylam (Dawlat al-Daylam), in JNES , xxvii [1969], 84-108, 168-83). Until approximately the time of the Mongol invasions, 9han9h remained an exalted regnal title, e.g. the Ism#l Grand Master of Alamt #Al" al-Dn Muammad III (618-53/1221-55) is called al-Mawl al-A#am 9han9h al-Mu#aam in one inscription (van Berchem, | [V:622b] pigraphie des Assassins de Syrie, in JA , Ser. 9, vol. ix [1897], 453-501). Subsequently, however, it declined from being a laab into an ordinary ism amongst Persian princes and those Turkish ones who at times adopted ancient Iranian nomenclature. As we have seen above in regard to dual titles, the Byids likewise secured imposing honorifics for the great viziers and secretaries who served them. Notable here is the use of the element kf capable one, and al-Brn singles out such alb of Byid officials as Kf 'lKuft, al-Kf al-Awad and Awad al-Kuft as nothing but one great lie (al-A9r al-biya, 134, tr. Sachau, The chronology of ancient nations, 131).

The Smnids [q.v.] in distant Transoxania and then 9ursn were abstemious in the use of honorifics, and al-Brn, op. cit., praises them for this, comparing them favourably with the Byids. The epithets of al-Sa#d, al-Sadd, al-0amd, al-Ri', etc., seem to have been applied to them retrospectively after their deaths, and the only clear instance of a laab used by one of the Smnids in his own lifetime is that of the last of the line, Ism#l b. N (d. 395/1005), who styled himself al-Muntair the one rendered victorious [by God], perhaps as a hopeful omenvainly, as events turned out for the restoration of his dynasty's fortunes. Yet although the Smnids, in practice independent, were usually punctilious in their deference to the #Abbsids, they did in the second half of the 4th/10th century confer alb unilaterally on their great commanders; thus N b. Manr (365-87/976-97) gave the title of Nir al-Dawla to the commander of the Bu9r 9z s, and the Smnids' Turkish commanders in 9ursn received similar titles, e.g. 0usm al-Dawla for T9 03ib and #Amd alDawla for F"i 9a (see Bosworth, The titulature of the early Ghaznavids, in Oriens, xv [1962], 214-15). The successors of the Smnids in eastern Persia and Af9nistn, the 9aznavids [q.v.], departed, on the other hand, from the path of Smnid simplicity, and from the time of the line's founder, Sebktigin, onwards, regularly sought from the caliphs numerous honorifics. Those of Sebktigin were Mu#n al-Dawla and Nir al-Dawla or Nir al- Dn wa 'l-Dawla, perhaps commonly abbreviated to Nir al-Dn (cf. S. M. Stern, in Paintings from Islamic lands, ed. R. Pinder-Wilson, Oxford 1969, 14-16; the question of the exact form of the laab is of some importance for the first appearance of the compound dn -type honorifics, see below). By the time of Ibrhm b. Mas#d (451-92/1059-99), the sultans bore a dazzling array of honorifics, in this case ahr alDawla and some twelve others, as well as the ceremonial designation (e.g. for the coinage) of al-Suln al-Mu#aam/al-A#am highly-exalted sultan, a designation probably adopted under Sal3 influence (details in Bosworth, op. cit., 215 ff., and idem, The later Ghaznavids: splendour and decay, the dynasty in Afghanistan and northern India 1040-1186, Edinburgh 1977, 55-6). Compound honorifics in dn appear at a slightly earlier date than the dawla ones, and the award of Nir al-Din wa 'l-Dawla to Sebktigin would appear to be the earliest instance of a dn title; it is certainly the only laab of this type listed by al-Brn in his $9r , 133-4, tr. 130-1, under the heading the holders of alb granted by the caliphal court, the remainder being mainly compounded of dawla , milla and umma . (Kramers thought that the linking of dn and dawla in titlessee further on this coupling, belowstemmed from a long Islamic tradition | [V:623a] in which these two strands of human existence were regarded as interdependent in this temporal life and their fortunes inextricably interwoven; he also thought that the combination of ideas came incontestably from Persia, see Les noms musulmans composs avec Dn , 61.) In the 5th/11th century, usage of the two types of laab , in dawla and in dn , fluctuates, but the coming of the Sal3s gave a great impetus to the spread of those of the second type. According to the Sal3 historical sources, when o9rl appeared in Ba9dd in 449/1058 for the second time, the caliph al-|"im bestowed on him the honorifics of Rukn al-Dawla pillar of the states and Malik al-Ma9ri wa 'l-Ma9rib king of the east and west (see Bosworth, in Cambridge history of Iran , v, 47); but in practice, o9rl was generally referred to, in the short period of life remaining to him and also retrospectively, as Rukn al-Dn. Dawla titles seem to have fallen into

disfavour under the Sal3s, for the subsequent sultans of the 6th/12th century favoured titles with dn or, instead of the frequent alliterative coupling of dn and dawla in a title, a further alliteration of dn and duny ; Malik 9h had the honorifics 3all al-Dawla (whence the name of the 3all era [q.v.]) and Mu#izz al-Dn wa 'lDunya, whereas his son Muammad had that of 9iy9 al-Dn wa 'l-Duny. Whether the over-shadowing of the dawla titles had any theological significance, as Kramers suggesteda preference for those in dn stemming from the strongly orthodox religious atmosphere of the Sunn reaction against political 9#ism, and an avoidance of the non-|ur"nic term dawla , with its connotations of the changes and vicissitudes of blind fateis unproven, but seems unlikely. But if dawla titles became less popular, the Sal3s had no hesitation in the use of compound titles with mulk for their viziers and high military commanders (e.g. #Amd al- Mulk for Ab Nar Kundur [q.v.] and Nim al-Mulk [q.v.] for Ab #Al s); and the #Abbsids themselves came to imitate the Sal3s in the bestowal of titles expressing the idea of secular power, e.g.alMutaf's award of the title Suln al-#Ir, Malik al-3uy9 to his vizier #Awn alDn Ibn Hubayra [q.v.] in 549/1154 as a reward for expelling the Trkmens from Wsi. Sal3 practice was also the model for the 9wrazm-9hs of the 6th/12th and early 7th/13th centuries; see L. Richter-Bernburg, Zur Titulatur der ]wrezm-he aus der Dynastie Antegns, in Archaeologische Mitteilungen aus Iran , N.F. ix (1976), 179-205. The 9# Fimids, the Sal3s' great opponents in the struggle for influence in the Syrian Desert region and its fringes of Syria and al-3azra, did not have this reluctance to use dawla titles, at least for their viziers and officials. A standard formula for their viziers was al-Wazr al-A3all most exalted vizier, awarded for instance to Ya#b b. Killis by the caliph al-Mu#izz in 368/979, but at this same time, compound dawla and other titles appear for the Fimids' viziers, e.g. Amn al-Milla for Ab Muammad b. #Ammr, and for their commanders, governors and vassal princes, e.g. Sayf al-Dawla for the Zrid Ysuf Buluggn in 361/972, and similar titles for Ysuf's successors Bds and al-Mu#izz b. Bds (see H. R. Idris, La Berbrie orientale sous les Zrdes X e-XII e sicles, Paris 1959, ii, 509; when al-Mu#izz in 433/1041 transferred his allegiance to the #Abbsids, the Fimid caliph al-Mustanir awarded the laab of his former vassal, 9araf al-Dawla, to al-Mu#izz's kinsman, the 0ammdid | [V:623b] al-| "id, who returned to the Fimid allegiance); and Murta' 'l- Dawla for the 0amdnid commander Lu"lu"s son Manr in Aleppo in 399/1008 (see M. Canard, Histoire de la dynastie des H'amdanides, i, 710). An individual feature of Fimid titulature for their viziers was the use of compound titles in amr al-mu"minn , e.g. -af Amr alMu"minn wa-9liatuhu for Abu 'l-|sim Amad al-3ar3ar", and Muaf Amr al-Mu"minn for Ab Manr -adaa b. Ysuf al-Fall. Very soon the titulature of these Fimid officials became remarkably luxuriant and pompous, heralding the later verbosity of Mamlk titulature in Egypt; thus in 447/1050-1 the vizier Ab Muammad al-0asan al-Yzr enjoyed the titles of al-Wazr al-A3all alAwad al-Makn, Sayyid al-Wuzar" wa-T3 al-Afiy" wa-|' 'l-|u't wa-D# 'lDu#t, #Alam al-Ma3d, 9liat Amr al-Mu"minn, to which were later added alNir li 'l- Dn, 9iy9 al-Muslimn, the ensemble denoting the wide extent of his powers, not merely as a vizier but also as chief ' and chief d# . But in general, Fimid procedure over the grant of honorifics was on similar lines to that of the #Abbsids, as is attested by the texts of such awards (called technically kutub al-tanwh documents conferring eulogy) quoted by al-|ala9and from the Mawdd

al-bayn of the Fimid author #Al b. 9alaf (a work which was long believed lost, but which has recently turned up in Istanbul, see A. H. Saleh, in Arabica, xx [1973], 192-200, and ibn 9alaf in Suppl.). Here a stereotyped formula is set forth, in which the grant of titles is accompanied by other favours such as the gift of a standard, a sword and a fine mount ( -ub al-a#9 , viii, 341). As in the Sunn world, the awarding of honorifics was often proclaimed urbi et orbi by the Fimids in Cairo, either before the caliphal palace or from the mosque pulpit; thus al-0usayn b. 3awhar was in 390/1000 honoured by al-0kim, receiving a robe of honour and having his newly-acquired title of |"d al-|uwwd supreme commander read out from the minbar (al-Marz, 9ia , ii, 15; see further on Fimid titulature in general, 0asan al-B9, al-Alb alislmiyya fi 'l-ta"r9 wa 'l-wa9"i wa 'l-9r, Cairo 1958, 65 ff., 92 ff.). After the Sal3 period, sc. after the 6th/12th century, the dominance of dn titles was firmly established, not only for rulers and their servants, but also, by what must have been unilateral adoption or else by the general consensus among religious groups, for outstanding spiritual leaders, -f 9ay9 s, etc., e.g. Na3m al-Dn Kubr, Muy 'lDn Ibn al-#Arab, 3all al-Dn Rm and Mu#n al-Dn 1i9t [q.vv.]. In any case, by this time all pretence at the caliph's being the sole dispenser of these honours had been abandoned. A consequence of this was that honorifics began to be adopted according to a method of rough conformity with a person's original ism . Al-|ala9and has a passage on this custom in -ub al-a#9 , v, 488-90, under the heading of Fi 'lalb al-mufarra#a #al 'l-asm", and the practise clearly dates back to early Mamlk times, if not before. Thus among the Turkish mamlks, #Alam al-Dn went with the name San3ar, 3aml al-Dn with A |u9, 0usm al-Dn with 0asan or 0usayn, #Al" al-Dn with #Al, T3 al-Dn with Ibrhm, etc. Even eunuchs had their characteristic combinations of laab and ism , e.g. 9u3# al-Dn with #Anbar, as had the Coptic officials of the administration in Egypt, e.g. Ta 'l-Dn with Wahba (cf. a similar list in al-Suy's | [V:624a] Risla f ma#rifat al-ul wa 'l-kun wa 'l-asm" wa 'l-alb, ed. -alh al-Dn al-Muna33id, Une importante Risla de Suy , in MFOB , xlviii [1973-4], 352-4, and also 0asan al-B9, op. cit., 103-5). This type of usage was carried over into the Ottoman empire, especially amongst the #ulam" and fuah", e.g. Badr al-Dn with Mamd, and the already-mentioned couplings with 0asan or 0usayn, #Al and Ibrhm; the alliterative effect often achieved was obviously a factor favouring the adoption of several of these (cf. F. Babinger, in Isl. , xi [1923], 20 n. 3). Whilst the #Abbsid caliphate was still a living organism (i.e. till the Mongol sack of Ba9dd in 656/1258), the granting of honorific titles remained, at least in theory, a jealously-guarded privilege of the caliphs. Local rulers or provincial governors who maintained the caliphal fiction of Sunn constitutional theory, that all executive authority derived ultimately from the caliph, sedulously sought a grant of honorifics at the outset of their reigns of governorships; and titles expressing personal closeness or a special relationship to the caliph, such as Mawl Amr al-Mu"minn, Wal Amr alMu"minn and |asm Amr al-Mu"minn, were especially sought after (early examples of the designation mawl amr al-mu"minn , say before the beginning of the 3rd/9th century, probably simply expressed a relationship of subordination and dependence, clientship or wal", between the caliph and its bearer, rather than being the grant of an official title; O. Grabar thinks that the attribution of this phrase to the governor of Egypt Amad b. ln [q.v.] in 265/878-9 still expressed dependence rather

than a title of honour, cf. his The coinage of the lnids , ANS Numismatic notes and monographs no. 139, New York 1957, 39-40). Recognition by the caliph, involving an investiture charter ( #ahd , man9r ), plus the other insignia of power such as honorifics, a richly-caparisoned charger and banners, might give a contender for power in a disputed succession the edge over his opponent. In 421/1030 the 9aznawid prince Mas#d b. Mamd hurried eastwards from western Persia to Af9nistn in order to confront his brother Muammad, who had been proclaimed sultan by the army in 9azna. At N9pr he received from the caliph al-| dir an investiture diploma for the 9aznawid empire plus a strong of honorifics, nu#t-i suln as Bayha calls them, Nir Dn Allh, 0fi #Ibd (or #Ubbd) Allh, alMuntaim min A#d" Allh and ahr 9alfat Allh Amr al-Mu"minn. Mas#d ordered that details of the award should be proclaimed and publicised in the towns of 9ursn, and it proved to be a valuable propaganda weapon in his successful wresting of the sultanate from Muammad later that year (Bayha, cited in Bosworth, The titulature of the early Ghaznavids, 224-5, and idem, The Ghaznavids, their empire in Afghanistan and eastern Iran 944-1040, Edinburgh 1963, 54). Not surprisingly, the #Abbsids held on their privilege of granting these titles in the Sunn world for as long as possible. The caliph personally was the fount of honours, and the precise form in which they were granted had to be rigorously observed; the Mamlk author Ibn Fa'l Allh al-#Umar [see fa'l allh ] states, concerning the correct form of address used by kings for governors and lesser rulers. that there was an inflexible rule in ancient times that no king was ever addressed except by the precise honorific granted to him from the caliphal | [V:624b] dwn , with no addition or omissions, bi 'l-na min 9ayr ziyda wal na (al-Ta"rf bi 'l-muala al-9arf, Cairo 1312/1894-5, 86-7). Conversely the unilateral and unauthorised assumption of alb by a person was an act of lse-majest, a virtual declaration of rebelion against the caliph or sovereign, as happened in 9ursn during the late 4th/10th century; the ambitious military commander of the Smnids, Ab #Al Sm3r, in 381/991 rose against his master N b. Manr, appropriated all the revenues of 9ursn and styled himself (talaaba) Amr al-Umar", al-Mu"ayyad min al-Sam" the heavenly-guided supreme commander (#Utb-Mann, al-Ta"r9 al-Yamn, Cairo 1286/1869, i, 155). Since the granting of such titles and honours created status and prestige for the recipients, it was natural that the #Abbsids, in the period of penury into which they had fallen by the early 4th/10th century, should expect a return for this services; during the Byid period in particular, when the caliphs were reduced to subsisting in straitened circumstances as pensioners of the Byids, this sale of honours, normally in return for presents, became all the more vital for them. The grants seem to have become in time regulated by something like a fixed tariff. In Bayha's Ta"r9-i Mas#d , 293, there is an account of the detailed discussion at Mas#d of 9azna's court in 422/1031 about the presents to be sent to the new caliph al-|"im, from whom the sultan expected confirmation of his territories plus a grant of fresh alb and other insignia of royalty; much of this discussion revolved round what was the usual rasm or practice here, with the adducing of precendents from the -affrid period. Inevitably, voices were raised against the overlavish granting of honours and titles, with a consequent cheapening in their value. Already the poet and littrateur Ab Bakr

Muammad b. al-#Abbs al-9wrazm (d. 383/993 or 393/1003 [q.v.]) had complained in a satire. What do I care that the #Abbsids have thrown open the gates of kun and alb? They have conferred honorifics on a man whom their ancestors would not have made doorkeeper of their privy. This caliph of ours has few dirhams in his hands, so he lavishes honorifics on people. (al-9a#lib, Yatmat al-dahr , Cairo 1375-7/1956-8, iv, 230; cf. Mez. Die Renaissance des Islms, 78-9, Eng. tr. 86-8). Hill al--bi" (d. 448/1056) has a long passage in his Kitb al-Wuzar", ed. #Abd al-Sattr Far3, Cairo 1958, 166 ff., lamenting the changes between the time of the viziers Ibn al-Furt and #Al b. #^s (the latter of whom had refused to increase the designation of a certain governor above the simple wish May God exalt him!, although threatened for his obduracy), and even between the time of the viziers of #A'ud al-Dawla and -amm al-Dawla, and the position at the end of his own lifetime. His main gravamen is that social and functional differentiation becomes impossible when titles lose their real meaning, and his conclusion is that Inevitably, official positions have declined in status where they have been reduced to one level [in titulature], and have become cheapened when they have all been made equal. They no longer possess any glory which one can admire, nor any splendour to be prized. Indeed, I have heard our master the caliph al-|"im bi-amr Allh may God prolong his reign!says that there is no designation | [V:625a] left for a deserving person. Hill's contemporary al-Brn likewise moralisingly observes that when the #Abbsids started rewarding their adherents with vain dawla titles, extending even to triple ones, their empire foundered: In this way, the matter became utterly opposed to common sense and clumsy to the highest degree, so that he who mentions them gets tired before he has hardly begun, he who writes them loses his time and writing, and he who addresses [people] with them runs the risk of missing the time for prayer ( al-$9r al-biya, 132, tr. 129). By the end of the 5th/11th century, the process of debasement had gone even further in the Muslim East, and the Sal3 vizier Nim al-Mulk complains volubly, in the section of his Siysat-nma on titles (ch. xl, ed. H. Darke, Tehran 1340/1962, 189-200, tr. idem, The book of government or rules for kings, London 1960, 152-63) that There has arisen an abundance of titles, and whatever becomes abundant loses value and dignity, and that In these days, the meanest official gets agry and indignant if he is given less than seven or ten titles. In particular, Nim al-Mulk denounces the confounding of dawla titles, formerly reserved for military commanders and the Turks, with the mulk and other titles used by viziers, governors and other civilian and religious officials and dignitaries, so that there results the absurdity of a Turkish general, illiterate, tyrannical and totally ignorant of the 9ar#a , being given titles like T3 al-Dn crown of the faith or Mu#n al-Dn succourer of the faith. He places the time when the floodgates were opened in the Sal3 empire to the indiscriminate and incongruous granting of honorifics as being the years after Alp Arslan's death (sc. after 465/1072 and the accession of Malik 9h; these strictures on the trends of the latter sultan's reign must be regarded as from a later hand than Nim al-Mulk's, perhaps from that of his copyist Muammad Ma9rib). 3. The Muslim West. Most of what has been said so far relates primarily to the central and eastern lands of Islam, sc. Egypt and the lands further east. The vogue for honorifics followed a rather different course in the Muslim West. In general, their use was less developed in the more puritanical West, where there was a tendency to regard elaborate and fancy names

and titles as effete and Persianising phenomena. Hence the term alb ma9riiyya is not infrequently used by Ma9rib writers in disparaging references to them, e.g. by alMaar and by the Ma9rib traveller to Egypt and Syria just before the Ottoman conquest, #Al b. Maymn al-Idrs. This last shows himself, in his opuscule Bayn 9urbat al-Islm bi-wsiat infay al-mutafaiha wa 'l-mutafaira min ahl Mir wa 'l9m wa-m yalh min bild al-#A3am (written in 916/1510, see Brockelmann, I 2, 152, S II, 153), as particularly severe against the habit of the religious scholars (the first of the two classes mentioned in the title of his work) of taking honorifics like 9ams alDn and Zayn al-Dn, which he stigmatises as bid# heretical and 9ayn devilish, in preference to the plain good old sunn names Muammad and #Umar, and of insisting on the use of these honorifics in addressing them (see Godziher, #Al b. Mejmn al-Magrib und sein Sittenspiegel des stlichen Islam , in ZDMG , xxviii [1874], 306-10 = Gesammelte Schriften, vi, 14-18). Already the Idrsid 9arfs of Morocco, with their claims to #Alid descent, had styled themselves imms, and the 9ri3 Rustamids of Thart | [V:625b] utilised not only this last designation but those of caliph and amr al-mu"minn also; these powers, of course, rejected the legitimacy of #Abbsid rule, whereas the A9labids of Ifriya, faithful in theory at least to the Ba9dd connection, never took such liberties in matters of titulature. The rise of the Fimid da#wa in North Africa during the early years of the 4th/10th century nevertheless introduced fully to the Ma9rib a titulature with messianic implications and one wholly opposed to the #Abbsid moral and constitutional position (see above, Section II), and the new trend inevitably had repercussions in Muslim Spain. A dynasty like the Spanish Umayyads, which had to defend itself on the cultural and ideological planes against both the distant #Abbsids in Ba9dd and, more pressingly, against the aggressively 9# Fimids (on the propaganda offensive of the Fimids directed against the Spanish Umayyads and its effects, see M. Canard, L imprialisme des Fatimides et leur propagande, in AIEO Alger , vi [1942-7], 162-9), could not but be influenced by the political and religious propaganda value of honorific titles. When the greatest sovereign of the family, #Abd al-Ramn III, adopted after his victory at Bobastro over Ibn 0afn in 315/927 the laab of al-Nir li-dn Allh, together with the designations of caliph and commander of the faithful, he thus placed himself firmly within what had now become the mainstream #Abbsid tradition of theologically-oriented caliphal titulature. His lesser successors studiously followed his example till the end of the dynasty in 422/1031, so that we find e.g. Hi9m II alMu"ayyad, Sulaymn al-Musta#n, etc., as did the 0ammdids who alternated with the last fainant Umayyads, e.g. al-|sim al-Ma"mn (see Lvi-Provenal, Hist. Esp. mus., i, 132-3, ii, 21, 115-16). The mulk al-aw"if who followed them during the period of the disorder and fragmentation in the middle decades of the 5th/11th century followed suit, with a disparity between the actual extent of their authority and the grandiloquence of their titles which attracted the satirical or ironic comment of contemporaries. Thus Ibn 9aldn in two places in his Prolegomena quotes verses of the panegyrist of the Zrids, Ibn Ra9 [q.v.], What makes me feel humble in Andalus is the use of the names Mu#taim and Mu#ta'id there. Royal epithets (alb mamlaka ) not in their proper place, like a cat that by puffing itself up imitates the lion. ( Muaddima , ed. Quatremre, i, 281, 412, tr. Rosenthal, i, 316, 470, attributing the verses to Ibn 9araf [q.v.]; al-Mu#taim and al-Mu#ta'id were honorifics adopted by princes of the 0ammdids of Malaga and the #Abbdids of Seville respectively). In matters of titulature, the Narids of Granada demeurrent fidles la tradition orientale

adopte par les Umayyades de Cordoue, fonde sur l'autorit absolue du souverain et sa caractre semi-religieux (R. Ari) Like their contemporaries amongst the rulers of the Muslim West, they used the title of Amr al-Muslimn, known from Almoravid times (see below), although the Mamlk chancery in Egypt simply addressed them in official documents as -ib 0amr" 9arna , according to Ibn Nir al-3ay9, cited by al-| ala9and in -ub al-a#9 , vii, 413. Certain Narid rulers assumed, as Ari implies, alb of the usual theocratic pattern, such as the dynasty's founder, Muammad I, called al-9lib bi'llh, and on returning from a successful expedition against Castile, Muammad V assumed that of al-9an | [V:626a] bi'llh (Ari, L'Espagne musulmane au temps des Nasrides (1232-1492), Paris 1973, 185-7). Yet although the trends of the East filtered through to the Iberian peninsula in considerable measure, the effect of these trends tended to operate at the highest level only, that of the monarchs themselves. The Spanish Muslim sovereigns were much more careful over the bestowal of honorifics to their servants, ministers and generals than were the eastern dynasts. Ibn Ab #$mir of course assumed the title of al-Manr [q.v.], for which his role in the state at the end of the 4th/10th century befitted him, on his return from the expedition against Leon in 371/981, and he was followed by his son #Abd al-Malik, styled al-Muaffar [q.v.]. Other isolated instances occurred, such as the award by al-0akam II of the title 9u 'l-Sayfayn in 363/974 to the general 9lib on his victorious return from a campaign in the Ma9rib, and the bestowal of the title 9u 'lWizratayn in 367/978 to both 9lib and Ibn Ab #$mir, one borne previously only by the general charged with defence of the Spanish frontier against the Christians; and over three centuries later, the great vizier of the Narids, Ibn al-9ab [q.v.], was to enjoy the alb of Lisn al-Dn and 9u 'l-Wizratayn (see Lvi-Provenal, Hist. Esp. mus., ii, 194, 213, 215-16, 228-9). North Africa remained faithful for a longer period to the puritanical ideals of an equalitarian, earlier form of Islam, in its disapproval of pompous titles, once the interlude of the rise of the Ism#l Fimids was over and the domination of Mlik orthodoxy was re-established under the Almoravids. The Almoravids originally recognised #Abbsid authority, but to mark the reality of their own power in NorthWest Africa adoptedwhether of their own accord or with the approval of Ba9dd is unclearthe title of Amir al-Muslimn. Thus there was created, from the constitutional point of view, what van Berchem conveniently called a sub-caliphate, whose rulers recognised an authority higher than their own and did not therefore adopt a titulature proclaiming their total independence and non-recognition of any superior power. The Almohads, however, came to power in the middle years of the 6th/12th century on a wave of messianic enthusiasm and under a charismatic leader, the Mahd Muammad b. Tmart [see ibn tmart ], and took up again in some measure the pattern of titulature instituted in the Ma9rib two centuries before by the Fimids. During Ibn Tmart's lifetime, his lieutenant #Abd al-Mu"min was styled the Mahd's 9alfa and Amr alMu"minn, sc. of the Almohad faithful, and on Ibn Tmart's death, he became the imm of the community, with the title of al-|"im bi-amr Allh; from the reign of Ab Ysuf Ya#kb al-Manr (580-95/1184-99) onwards, the Almohad sultans are found with honorifics of the familiar theocratic pattern (see van Berchem, Titres califiens d'occident, propos de quelques monnaies Mrinides et Ziyanides, in JA , Ser. 10, vol. ix [1907], 263-79, and the important section on the title Amr al-Mu"minn in Ibn 9aldn, Muaddima , i, 408-14, tr. i, 465-72).

The pattern of titulature was in this way established for the sovereigns of the three successor-states in the Ma9rib and Spain to the Almohads, sc. the Narids (for whom see above), the Marnids of Morocco and the 0afids of Ifriya. The 0afids' eponymous ancestor was the Mahd Ibn Tmart's celebrated companion and partisan, 9ay9 Ab 0af #Umar, and the 0afids continued to use in | [V:626b] reference to themselves the term al-Muwaidn those who proclaim God's unity (an assumption admitted by the secretaries of the Mamlk chancery in Cairo, who used the titles Za#m al-Muwaidn chief of those proclaiming God's unity and |udwat al-Muwaidn exemplar of those ... in addressing them, -ub al-a#9 , vi, 51, 65). The most significant factor in the pattern of 0afid titulature, as it evolved in the 7th/13th century, was their claim to the caliphate, put forward by the second ruler of the dynasty, Ab #Abd Allh Muammad (647-75/1249-77) in 650/1253. This was buttressed firstly by the attempts of the family's apologists to impute to them a |ura9 descent from #Umar b. al-9ab (whose kunya had been Ab 0af), so that the sultans proudly termed themselves ibn al-9ulaf" al-umar" al-r9idn; and secondly by the recognition of the 9arf of Mecca and, briefly, of the Mamlks of Egypt even, after the extinction of the #Abbsid caliphate in Ba9dd by the Mongols and before its revival in Cairo. Whence also the 0afids' adoption of the characteristically caliphal designation of Amr al-Mu"minn (although their rivals, the Mamlks, would only allow them the subcaliphal one of Amr al-Muslimn), and of theocratic alb on the exact #Abbsid pattern (al-Mustanir and al-Mutawakkil being especially favoured by various members of the dynasty), in order to demonstrate the genuine nature of their caliphal claims and the continuity of their titulature with the older #Abbsid practice. Looking forward to what will be said below about the Ayybid and Mamlk usage of such titles, we may also note that the 0afids, deeply involved in struggles with Christian powers such as Spain and France, seeking a foothold in North Africa, assumed titles reflecting their roles as leaders in the holy war, such as al-Mu3hid f sabl Allh he who fights in God's way, and allowed themselves to be addressed impersonally by titles of respect like al-Mam al-#al/al-a#l and al-0a'ra al-#aliyya (the latter expression being used by the Bey of Tunis until the proclamation of the republic there as the equivalent of the European diplomatic forms His Majesty, Son Altesse, etc.). See on all these questions of 0afid usage, von Berchem, Titres califiens d'occident, 283-93; R. Brunschvig, La Berbrie orientale sous les 0afides des origines la fin du XVf sicle, Paris 1940-7, i, 40, ii, 7-17. The Marnids for long accepted the supremacy of the already-established 0afids, hence they (and also their neighbours of Tlemcen, the #Abd al-Wdids or Zayynids) normally used the lesser title of Amr al-Muslimn rather than the fully caliphal one of Amr al-Mu"minn; many of the Marnid sultans also bore theocratic-type alb like the 0afids. But there were episodes when certain sultans did assume the higher title. For a short period of 9 month in 708/1308-9, Abu 'l-Rab# Sulaymn adopted it on the coins which he issued conjointly with his Narid ally Muammad III, apparently as an act of defiance to the 0afids. Furthermore, Ab #Inn Fris al-Mutawakkil (74959/1348-59) seems to have employed the title as a lever to secure the deposition of his father #Al, who had already styled himself in his official documents Amr alMu"minn and |"id al-Muwaidn, Almohad titles par excellence (and also Baiyyat al-Salaf al-Karm in allusion to the Berber Marnid's pretensions to an Arab genealogy). The mention in Ibn 9aldn, Muaddima , i, 414, tr. i, 472, of the Zanta rulers, sc. the Marnids, using the title Amr al-Mu"minn must be a reference to this

episode, though subsequent | [V:627a] sultans reverted to the lesser title Amr alMuslimn. The whole question of this alternation of titles was examined in great detail by van Berchem in his Titres califiens d'occident, 245-335. With the decline of the Marnids and their kinsmen the Wasids, the Sa#d 9arfs from southern Morocco rose to power in the early decades of the 10th/16th century on a wave of Moroccan enthusiasm for the expulsion of Turkish Algerian influence and of renewed Islamic maraboutist fervour for 3ihd against the Portuguese encroachers on the Moroccan coastlands. In an atmosphere which was thus fed on religious enthusiasm and popular messianic expectations of a new defender of the Muslims, it is not surprising that the first of these 9arfs of the Ss, Muammad (d. 924/1517-13) assumed the chiliastic titles of al-Mahd and al-|"im bi-amr Allh, reminiscent of Fimid and then Almohad usage. Similarly religiously-motivated honorifics were adopted by several of his successors, e.g. Muammad al-Mutawakkil, Amad alManr, etc. One should also note the very characteristic forms of address used by both the Sa#d sultans and their #Alaw successors, those of Mawly/Mly my master and Sayyid/Sd my lord, as also by other high dignitaries, princely and religious, in the Ma9rib. The form Mawln our master had been used by the Narids during the 8th/14th century, and the Christians of Spain in the 9th/15th century often referred to the ruler of Granada as Muley (Ari, L'Espagne musulmane au temps des Narides, 187); whilst both Mawln and Sayyidun had been early used by the 0afids in their official documents. The form with the first person singular pronoun affix, mawly > mly apparently appeared amongst the 0afids in the course of the 8th/14th century, but is only at first attested in Christian sources, e.g. the Muley Bolabes = Abu 'l#Abbs Amad II al-Mustanir of a Latin document of 1391 (cf. Brunschvig, 0afides, ii, 15-16). 4. The post-caliphal period. With strengthening of the grip of Turkish and Kurdish dynasties over the central and eastern lands of Islam from the 5th/11th century onwards, regal, military and ministerial titulature increased in complexity and grandiloquence. After the extinction of the Ba9dd caliphate and the establishment by the Mamlks of a puppet line only of #Abbsids in Cairo, the granting of titles became in practice the responsibility of the Mamlk chancery or Dwn al-In9" , so that this office became the concentrationpoint of a great deal of expertise in these questions. In any case, the correct ordering and recounting of all the various components of the titles of rulers and dignitaries had always been vital in epistolary and other official usages. The various manuals for secretaries and officials, stretching back to Fimid times but reaching their full florescence in a great document-producing civilisation like that of the Mamlks, devote much space to forms of address and titulature. In the most monumental of these manuals, al-|ala9and's -ub al-a#9 , the first bb of the third mala (v, 423-506, vi, 1-188, cf. W. Bjrkman, Beitrge zur Geschichte der Staatskanzlei im islamischen gypten, Hamburg 1928, 110-13) is devoted to the topic of names of all kinds, but with special reference to alb, which are traced from the origins of Islam to the author's own time, the 9th/15th century. Al-|ala9and has several lists of titles, e.g. of these which he calls honorifics of more recent times, alb muda9a, such as n"ib , s , mu9rif , aw3, ustd al-dr, bundudr , dawdr , amr al-9ur, | [V:627b] etc. this last class of titles being descriptive of offices rather than alb in the true sense of honorifics as we have been discussing them.

Of greater interest for our present purpose are al-|ala9and's numerous pages on the protocol of correct address, when addressing the caliph or sultans downwards. Many delicate and subtle distinctions are stressed here: thus al-ma3lis al-sm/al-smiyy (with y" mu9addada) is a higher designation than al-ma3lis al-sm (with single y" ); al-ma3lis al-a'" is higher than al-ma3lis al-' , and al- ma3lis al-'aw higher still ( -ub al-a#9 , vi, 141 ff.; cf. M. Gaudefroy-Demombynes, La Syrie l'poque des Mamelouks d'aprs les auteurs arabes, Paris 1923, Introd. LXXXII ff.). Even heathen rulers and notables were not to be denied their honorifics, though these were naturally on a lower level than those accorded to Muslim equivalents; the Mamlk chancery had, of course, a sphere of diplomatic contacts embracing many non-Muslim powers, from the Christian empire of Byzantium and the Latins of the Western Mediterranean to the still-pagan Turkish and Mongol rulers of Inner Asia. As an example, we might cite the titles used in addressing the Doge of Venice (Dk alBunduiyya): al-Dk al-3all, al-Mukarram, al-Muba33al, al-Muwaar, al-Baal, al-Humm, al-4ir9m, al-9a'anfar, al-9ar, Ma3d al-Milla al-Narniyya, Fa9r al-#^sawiyya, #Imd Bani 'l-Ma#mdiyya, Mu#izz Pp Rmiyya, -dk al-Mulk wa 'l-Saln N.. If such titulature was used for infidels, the luxuriance of contemporary practice for Islamic addressees may be judged! Only some of the salient features of Ayybid and Mamlk honorific titulature can be mentioned here; the existing documentation is so rich that a whole monograph could easily be written on the topic, and indeed, much of the material used by 0asan al-B9 for his book derives from these two periods and from the Syrian and Egyptian milieux; such material from the Mamlk period has further been used to good effect by Muammad Bir al-0usayn in his study, based in the first place on coin legends, alKun wa 'l-alb #al nud al-Mamlk al-Bariyya wa 'l-Bur3iyya f Mir wa 'l9m, in al-Mawrid, iv/1 (1975), 55-104. All the Ayybid sultans, and following them the Mamlk ones, bore honorifics of the alDuny wa 'l-Dn patterncontinuing here Sal3 practiceand these appear in inscriptions and official documents, although for less formal usage a shorter form in alDn only seems to have been current. Especially characteristic of the Ayybids was the use of an honorific composed of al- Malik plus a laudatory epithet (e.g. al-Malik alKmil, al-Malik al-Mu#aam) beginning with -al al-Dn's title of al-Malik al-Nir bestowed on him by the Fimid caliph al-#$'id when he appointed -al al-Dn as his vizier in succession to Shrkh in 564/1169. Titles like these had been known in the Fimid caliphate for some time, and al-0fi's vizier Ri'wn b. Walak9 had already in 531/1137 borne the titles of al- Sayyid al-A3all al-Malik al-Af'al. The inscription on the 9n al-#Aaba to the south-east of Lake Tiberias by its founder, #Izz al-Din Aybak, describes him as al-Malik al-Mu#aam connected with al-Malik al-Mu#aam, sc. with 9araf al-Dn #^s b. al-Malik al-#$dil Sayf al-Dn, at that time (610/1213-14) governor of Damascus for his father and not yet an autonomous Ayybid prince; these titles were not therefore confined under the Ayybids to reigning princes only (van Berchem, Eine ara- | [V:628a] bische Inschrift aus dem Ostjordanlands, 89). From the Ayybids, titles of this type spread to the Mamlks, and were used by the sultans, e.g. al-Malik al-Mu#izz for Aybak and al-Malik al-|hir and then al-hir for Baybars. Such titles were also adopted unilaterally by presumptous, often rebellious, amr s and governors, e.g. that of al-Malik al-Mu3hid assumed by #Alam al-Dn San3ar al0alab after the murder of |uuz in 658/1260, and the fashion spread to powers

dependent upon or culturally influenced by the Ayybids and Mamlks, such as the Raslids of Yemen, whose rulers, from al-Malik al-Manr Nr al-Dn #Umar onwards in 626/1229, all had honorifics of this type (see al-B9, 498 ff.). The titulature of the Mamlk rulers and of their amr s was particularly complex. The title Suln [q.v.], though certainly known in the Ayybid period, had not been widely used by the Ayybid monarchs, but was now extensively adopted by the Mamlk ones. Each of the sultans bore honorifics of the characteristic Ayybid type compounded with al-Malik , as discussed above, and also titles in al-Duny wa 'l-Dn (see also above). But because of their military slave origin, the Mamlk sultans and amr s usually further bore special nisba s relating to their ethnic or local origins, their early professional training or their affiliation to the household of their masters. Thus sultan Bar [q.v.] had the nisba of al-Yalbu9w because he had been the mamlk of the general Yalbu9 al-#Umar, and Baybars [q.v.] that of al--li from his original master, the Ayybid al-Malik al--li Na3m al-Dn Ayyb; the general 0usm al-Dn zdemr was called al-Mu3r from the slave dealer who had sold him; and both the sultan |alwn [q.v.] and the amr 9ams al-Dn Sonur were called al-Alf because they had been bought for 1,000 (alf) dnrs. Although these names are in form technically nisba s, they were not regarded as in any way derogatory, but were, rather, a source of pride to the holders and may in this wise be regarded as honorific titles. One class of laab borne by some early Mamlk sultans may be characterised as quasiterritorial or quasi-ethnic, sc. those titles in which the ruler claims lordship over particular regions and/or peoples. Already al-Malik al--li Na3m al-Dn Ayyb had grandiloquently styled himself 9ahriyr al-9a"m Suln al-#Arab wa 'l-#A3am -ib al-0aramayn al-9arfayn Malik al-Barrayn wa 'l-Barayn Malik al-Hind wa 'lSind wa 'l-Yaman Malik -an#" wa-Zabd wa-#Adan Sayyid Mulk al-#Arab wa 'l#A3am Suln al-Ma9ri wa 'l-Ma9rib ( RCEA , xi, no. 4308), and in certain inscriptions of the Mamlks Baybars and |alwn we find headship over al-#Arab wa #l-#A3am extending to al-Turk and even al-Daylam (= the Mongols here?) (ibid., xiii, no. 4817, etc.). A notable feature of the Mamlk age was the strongly orthodox Sunn atmosphere, now that the very seat of the #Abbsid caliphate had been transferred to Cairo and the Mamlk rulers had become the principal defenders of Islam. This stress on orthodoxy appears naturally in the Mamlks' external policy, against such assailants of the Dr alIslm as the Mongols and the Christian Franks and Armenians, and in their internal policy as repressions of Muslim sectaries like the Nuayrs and Ism#ls. Under the stimulus of an increased religiosity, both in official theological circles and in the sphere of popular religion and mysticism, the duty of 3ihd was exalted. Whence the frequency in Mamlk titulature of designations like al-Mu3hid, | [V:628b] alMu99ir, al-Murbi, al-9z, al-Mu9z, etc., though these had already appeared under the Sal3s, the Atabegs and the later Fimids as a reaction to the landings of the Frankish Crusaders (cf. al-Mu3hid as a title of the Brid Atabeg u9tigin in a Damascus inscription of 524/1130, RCEA , viii, no. 3034, and also as a title of #Izz alDin Aybak in the 9n al-#Aaba inscription, van Berchem, op. cit., 101-2). The proximity now of the seat of the caliphate and this atmosphere of religious exaltation and bellicosity probably gave an impetus also to the increased popularity of a type of laab already well-known, that compounded with one of the titles of the caliph or sultan, and expressing close dependence on the supreme ruler, the enjoyment of his favour or

support for him and the furtherance of the faith. Thus -al al-Dn, at the time of his recognition as ruler by the #Abbsid caliph, adopted the title of 9all Amr alMu"minn, and others of this type include 9iat Amr al-Mu"minn, #Umdat alMulk wa 'l-Saln, Nurat al-Islm wa 'l-Muslimn, etc. Those titles which included as one of their elements the caliphal title par exellence, Amr al-Mu"minn, were naturally the highest-regarded, and al-|ala9and arranges the different forms which this class of title took in a hierarchy of status. |asm Amr al-Mu"minn is the highest, and may only be borne by the sultan's sons or used in correspondence with certain neighbouring Muslim princes; #A'ud Amr al-Mu"minn is the highest title which can be used for the sultan's provincial governors; Wal Amr al-Mu"minn can be used by high civil officials and by religious scholars, and ranks above -af/-afwat Amr alMu"minn; and so forth ( -ub al-a#9 , vi, 108-9). The type of honorifics classified by al-B9, op. cit., 83 ff., as those indicating place and status. alb makniyya. were used as indications of reverence and humility in addressing or referring to the great. They had already been used in the heyday of the #Abbsids, for in the vizierate of Ibn al-Furt there had arisen the practice of addressing the caliph indirectly as al-9idma, in effect, the one to whom service is due, and Hill al--bi" states that what had originally been just a formula of urba , ingratiation, soon became a sunna , compulsory practice (cf. Tyan, Institutions du droit public musulman. i. Le califat, 488). By the time of the later #Abbsids, we find the caliphs regularly referred to in epistolary style (e.g. in such sources as Ab 9ma and the |d al-F'il) by such circumlocutions as al-3nib al-9arf, al-Mawif, al-9arfa, Mam al-Rama, etc. The Byid and Sal3 usage of al-0a'ra (see above) is clearly a precursor of these expressions, although by the Mamlk period, al-0a'ra had declined from being a form of address suitable for caliphs, as in al-0a'ra al-Smiya, into being used in addressing civil officials, infidel foreign rulers and the Coptic Patriarch in Egypt, according to al-|ala9and, -ub al-a#9 , v, 498. These honorifics indicating place or status enjoyed a great expansion in Ayybid and Mamlk times. That of alMa3lis spread under the former dynasty downwards from the sovereign to the great men of state, so that by al-|ala9and's time it was regarded as essentially a title for the men of the sword and the pen, but somewhat below al-3anb. Hence towards the end of the Ayybid period, the ruler tended to adopt instead the forms al-Mam al-#$l or al-Mam al-A9raf. This usage was followed by the Mamlks, so that Ibn 99, for instance, says in | [V:629a] his Ma#lim al-kitba that al-Mam and al-Maarr are the highest alb and are exclusively royal (ibid., vi, 495-6, where it is also stated that the reference in such titles is to the seat of the ruler's power or his capital). Van Berchem's opinion was that the study of mediaeval Islamic honorifics was only of value for the study of administrative institutions, and that these titles only had historical significance in so far as they were linked with specific officeslose Ehrentitel haben so gut wie keinen Werth (op. cit., 105). As already noted above, Kramers combatted this negative view, suggesting that the nature of these honorifics reflected the religious and cultural atmosphere of their time, e.g. that the later predominance of dn titles over dawla ones coincided with the Sunn reaction against political 9#ism and against external Christian pressure. Whether certain dn honorifics did owe their popularity in the Iranian world to the fact that they resembled traditional names, e.g. Fard al-Dn and Fardn/$frdn, Bah" al-Dn and Behdn, and |iym al-Dn and Kmdn (Les noms musulmans composs avec Dn , 63-5), seems impossible to prove or disprove.

5. The period of the great empires. The use of honorifics continued in the great empires of later mediaeval Islam, sc. those of the Indo-Muslim sultans, the -afawids and the Ottomans, almost down to modern times. The titulature of the first Muslim dynasties to be permanently established in the northern Indian plain, the Slave Kings of Dihl and their successors, inevitably followed grosso modo the pattern set by their original master, the Af9n 9rids [q.v.], who had in their turn continued in the ways established by the power which they had overthrown in the later 6th/12th century, the 9aznawids (for 9aznawid titulature, see above, 3, and for that of the 9rids, the information given in the abat-i Nir of Minh3-i Sir3 3z3n [q.v.], who is always careful to detail the titles of his 9rid forebears, as also those of the Dihl sultans contemporary with him). The Slave Kings, essentially the Turkish military commanders of the 9rid sultan 9ihb al-Dn or Mu#izz al-Dn Muammad (d. 602/1206 [see 9rids ]), followed their old masters in favouring on the whole alb in dn , whence |ub al-Dn Aybak, 9ams al-Dn Iltutmu9 [q.vv.], etc. However, as both 3z3n's information and the contemporary inscriptions show, there were many variants and elaborations. Thus Aybak appears in an inscription of the |uwwat al-Islm mosque in Dihl as |ub alDawla wa 'l-Dn, Amr al-Umar", whilst 3z3n gives Iltutmu9's laab in full as 9ams al-Duny wa 'l-Dn. Basking in the glory of their extensive military conquests, various sovereigns of this period conceived of themselves as following in the footsteps of Alexander the Great. Already the 9rid Mu#izz al-Dn Muammad is described on the |ub Minr [q.v.] as Iskandar al-9n, a title imitated e.g. by #Al" al-Dn Muammad 9h 9al3 of Dihl (695-715/1296-1316) on his coins, with such variants as Iskandar al-Zamn the Alexander of the age. The geographical and ethnic extent of the empire ruled by these Turkish commanders is indicated by Iltutmu9's adopting later in his reign (in an inscription of the Hns mosque) the title Mawl Mulk al-Turk wa 'l-#A3am, whereas previously he had styled himself (on the |ub Minr) by the conventional, but by then obsolete title for an eastern Islamic potentate of | [V:629b] Mawl Mulk al-#Arab wa 'l-#A3am. Like other Turkish ruling dynasties of the East, being newcomers into the Islamic society and polity and as yet uncertain of their place within these last, the Indo-Muslim rulers sought to validate their rule by expressing their loyalty to the #Abbsid caliphs (who were of course after 659/1261 puppets under the control of the Egyptian Mamlks). Iltutmu9, in the last decades of the independent #Abbsids of Ba9dd, usually styled himself Nar or Nir Amr al-Mu"minn, but in one instance at least, |asm Amr al-Mu"minn, the title favoured by the 9rids to demonstrate their fidelity to Ba9dd. Later in the century, 9iy9 alDn Balban (664-86/1266-87 [q.v. in Suppl.]) followed the same tradition when he styled himself Yamn 9alfat Allh as well as Nir Amr al-Mu"minn. See J. Horovitz, The inscriptions of Muhammad ibn Sm , Qutbuddin Aibeg and Iltutmi9 , in Epigraphia Indo-Moslemica, 1911-12, 12-34; G. Yazdani, The inscriptions of the Turk Sulns of Delhi-Mu#izzu-d-dn Bahrm , #Al"u-d-dn Naru-d-dn Mamd , 9iy9u-d-dn Balban and Mu#izzu-d-dn Kaiqubd, in ibid., 1913-14, 13-46; Z. A. Desai, The inscriptions of the Mamlk Sultans of Delhi, in Epigraphica Indica, Arabic and Persian Supplement, 1966, 4-18.

The Turco-Mongol successors of these first Turkish and Af9n Indo-Muslim rulers, the Mu9als, brought to India Tmrid traditions in using the grant of titles and other marks of honour to strengthen the loyalty of their own Turkish commanders and to win over other groups, such as the great Af9n chiefs. Bbur mentions that, in India, permanent designations (muarrar 9iblar) were given to highly-favoured amrs, such as A#am Humyn, 9n-i 3ahn and 9n-i 9nn (Bbur-nma, tr. Beveridge, 537). Bbur's son Humyn followed a careful policy in the award of titles appropriate to services rendered or expected; thus the supreme distinction of Amr alUmar" was bestowed on Amr Hind Beg, an old commander of Bbur's who had fought at Pnpat [q.v.] in 932/1526 and who was, moreover, allowed the signal honour of sitting with the emperor in formal court sessions. Under Bbur's immediate predecessors in India, the Lds, the titles of nobility had been (in ascending order) Malik, Amr and 9n. In the course of the 10th/16th century, the title of Malik fell out of fashion and that of Beg, one of prestige under the first two Mu9al rulers, subsequently declined in favour of 9n, so that under Akbar, Beglar Begi was a lower title than that of 9n-i 9nn. This last was the highest title of all, held e.g. by the young Akbar's atl or guardian Bayram 9n (d. 968/1561 [q.v.]), together with that of Amr al-Umar". Other titles tended to be associated with specific affairs or functions; thus that of $af 9n was mostly conferred on civil officials acting as wazr or as wakl of the royal household, hence mainly on Persians; whilst Akbar conferred the Hindu title of r3 not only on the hereditary successors to princely power but also on faithful Indian servants like the master-gunner Sabbahan. See Radhey Shyam, Honours, marks and titles under the Great Mughals (Babur and Humayun), in IC , xlvi (1972), 101-17, and idem, Honours, ranks and titles under the Great Mughals (Akbar), in ibid., xlvii (1973), 335-53. As the political and military power of the Mu9als shrank in the post-Awrangzb period, the conferring of titles became more and more widespread by the | [V:630a] emperors and by provincial Muslim dynasties, so that their social value declined; hence today, old titles like Mrz, 9n and Beg have in the modern subcontinent become nothing more than the equivalents of western surnames. In -afawid Persia, one notes first of all, in connection with the strongly 9# basis of the state and the theocratic nature of the early 9h's authority, a fondness for names and titles expressing devotion to or dependence upon either some venerated figure of 9#ism, such as #Al or his sons al-0asan and al-0usayn, or upon the sovereign himself, considered as the vicar on earth of God or the Imm s. In pre--afawid times, there had occasionally been used by rulers in Persia names compounded with the Persian word banda slave, devotee, e.g. the Mongol Il-9nid Muammad 9udbanda l3eyt (the laab being assumed when l3eyt became a Muslim; his pro-9#i sympathies should perhaps be noted here). Under the Turkmen -afawids, the equivalent Turkish word ul was commonly used, as in #Al-|ul, Imm-|ul, ahmsp-|ul, -af-|ul, etc., especially in regard to military commanders and governors, although the 9hs themselves retained simple regnal names. The usage of these titles in ul was imitated in Muslim India by certain of the South Indian sultans who were 9# in faith and strongly under -afawid cultural influence, e.g. the ruler of the |ub-9hs [q.v.] in Golkonda, Muammad-|ul b. Ibrhm (988-1020/15801612).

Whilst the 9hs themselves remained modest over the use of personal alb, their subordinates enjoyed a rich titulature. It is under the early -afawids, apparently towards the end of 9h ahmsp I's reign ca. 976/1568-9, that the characteristic -afawid title for the wazr , that of I#timd al-Dawla [q.v.] trusty support of the state appears; this title is much distorted in the travel accounts of contemporary western visitors to Persia, e.g. the Athemadeulat of Du Mans and the Etmadowlett or prime minister of John Bell. The late -afawid administrative manual Ta9kirat al-mulk (ca. 1137/1725) gives detailed information on this latter official and on the other important figure of the |urBa9 [see rc] (who was, in early -afawid times, before the establishment of a regular, standing army, virtually the commander-in-chief, with the title of Amr alUmar"), here called the Rukn al-Salana al-|hira, and on a host of lesser officials. The top fourteen officials of the administration had the title of #l-3h exalted in rank, and there were groups of officials with the title of muarrab al-9n confidant of the supreme ruler because of their special closeness to the throne, and with that of muarrab al-a'ra confidant of the royal presence. The first group included the head of the palace eunuchs, the royal physician ( akm-ba9 , the court astrologer (muna33im-ba9), the controller of assay (mu#ayyir al-mamlik), the state secretary, who drew the royal u9ra [q.v.] ( mun9 'l-mamlik), the keeper of the seal (muhr-dr) and the keeper of the ink-holder (dawt-dr). The second group, somewhat lower in status, included senior harem attendants, aides-de-camp (yaswuln), the heads of various departments of the royal household and workshops (buytt), including the master of the mint ('arrb-ba9), etc. (Ta9kirat al-mulk, ff. 8b, 12b, 30a-55a, tr. Minorsky, 44, 46, 55-69). Under the |3rs, the title of I#timd al-Dawla for the wazr declined in currency, being replaced by that familiar in Ottoman usage also (see below), -adr-i A#am. There was also a great expansion of | [V:630b] honorifics in dawla , mamlaka , salana , etc. for the numerous princes of the |3r family and for other great officials, a process which the Amr Kabr [q.v. in Suppl.] endeavoured in the mid-19th century years to check, but one which continued with little abatement till the end of the |3rs in 1925. This rich array of titulature, with by now elaborate forms of address extending down the social scale as low as mere village headmen, inevitably came under fire from the reforming Ri' 9h Pahlaw. Muammad Ri' had indeed already in the Fifth Ma3lis, as Sardr Siph and before he had overthrown the |3rs, abolished the honorific titles which had been sold for the personal enrichment of the 9h and court officials, even though this act had been an infringement of the monarch's prerogative. On 2 August 1935 there was issued a decree on the abolition of titles and on the terminology of social intercourse. The royal family was to receive new titles, with the 9h himself to be A#l-yi 0a'rat-i Humyn 9han9h; high officials were to be addressed just as 3anb, and the old titles of amr , beg , 9n , mrz, etc. were to be abolished. In fact, although these reforms were honoured in the press and in public announcements, the old titles continued very much in common and spoken usage (see P. Avery, Modern Iran , London 1965, 267, 273; D. N. Wilber, Riza Shah Pahlavi: the resurrection and reconstruction of Iran , Hicksville 1975, 167, 171). When western-type surnames were introduced, some people turned the old alb which went back to |3r days into family names, e.g. in the cases of Dr. Muammad Muaddi, Prime Minister 1951-3, formerly Muaddi al-Salana, and his contemporary the statesman and former Prime Minister in 1921 Amad |awm, formerly |awm al-Salana.

A centralised and bureaucratic institution like the Ottoman empire, with from the late 8th/14th century onwards extensive diplomatic contacts, firstly with the Muslim beyliks of Anatolia and the Turkmen powers of the East, and then with the Christian states of the West, increasingly affected by Ottoman expansionism, evolved a complex and elaborate chancery procedure in which the careful recounting of honorific titles played a vital role. The immense bulk of surviving Ottoman diplomatic and administrative documents would make feasible a highly detailed study of this titulature, a task which remains however to be done. For the moment, it may be noted that Feridun Beg devotes the opening pages of his great collection of correspondence to an exposition of the alb of the various classes of addressee, from the sultan at the top down to civil and military officials and members of the religious institution within the empire, and also of the alb to be used in communicating with dependent rulers such as those of the TransDanubian principalities and with foreign potentates like the Doges of Venice (Mn9e"t al-seln, Istanbul 1274/1857, i, 2-13). Also, L. Fekete devoted a section of his Einfhrung in die Osmanisch-Trkische Diplomatik der Trkischen Botmssigkeit in Ungarn, Budapest 1926, pp. XXXII-XXXVI, to an exposition of honorifics as found in administrative and diplomatic documents of the 10th/16th and 11th/17th centuries concerning relations between the Porte and local officials in Hungary or between the sultans and the Christian monarchs in adjoining lands. The luxuriance of the titulature of, for instance, Sleymn the Magnificent is seen in a letter of this sultan from 972/1565, where in the intitulatio of the document | [V:631a] Sleymn describes himself as Sultn-i Saln-i 9ar wa 9arb, -ib-|irn-i Mamlik-i Rm wa #A3am wa #Arab, |ahramn-i Kawn wa Makn, Narmn-i Maydn-i Zamn wa Zamn, A Deizi wa |ara Deizi wa Ka#ba-yi Mu#aama wa Madna-yi Munawwaran wa |uds-i 9arfi wa Ta9t-i Mir Ndira-yi #Ar wa Wilyet-i Yaman wa #Adan wa San#"n wa Dr al-Sadd Ba9dd wa Bar wa Lasn wa Mad"in An9n-Rawn wa 4iyr-i 3az"ir wa $9arby3n wa Da9t-i |pa wa 4iyr-i Ttr wa Krdistn wa Lristn wa Kulliyyan Rm ^l wa Anl wa |aramn wa Afl wa Bu9dn wa Angars memleketlerini wa bunlardan 9ayr ni mamlik wa 'iyr #am al-i#tibr Pdi9h wa Suln Suln Sleymn 9n b. Suln Selm 9n (ibid., p. XXXII). The honorifics of the sultan's subordinates were naturally less florid, but considerable care was taken to differentiate niceties of rank, so that a ' with a stipend of less than 150 ae s was addressed as |udwat al-|u't al-Islm [sic], #Umdat Wult al-Anm, whereas a ' of 150 ae s or more could add to the above titles that of Mumayyiz al0all #an al-0arm (ibid., p. XXXIV). The grand vizier was from the time of Sleymn onwards awarded the designation of -adr-i A#am most illustrious of the high dignitaries, and this title remained in use all through the Ottoman sultanate's existence, surviving the reforms in the bureaucracy of the Tanmt [q.v.], the last -adr-i A#ams being Dmd Ferd Pa9a [q.v.] (till October 1920) and his successor Amed Tewf Pa9a (till November 1922) who served Meemmed VI Wad al-Dn. The vizier had several other epithets of distinction, such as sm , af and #l, and he was entitled to the same form of address as the Khedives of Egypt in the 19th century, dewletli fe9metli (see further adr-i a#zam). It was during the Tanmt period of the mid-19th century that some attempt was first made at rationalising and restricting the unchecked growth of titulature, as part of the institutionalisation on western lines of the old Ottoman bureacracy. It seems that the traditional titles were now bureaucratised. Thus Redhouse in his Turkish and English

lexicion, s.v. bey (375a), has 5. The title given [to] the sons of Pashas, and a few of the highest civil functionaries, to military and naval officers of the rank of colonel or lieutenant colonel, and popularly, to any persons of wealth, or supposed distinction; s.v. pa9a (434a) he has an explanation of the modern military and naval positions, of general officer and flag rank, entitled to use this designation; and s.v. vezr (2136a) he has the definition a civil state functionary of the highest rank, with the title of pasha. Hence in the biography of Fu"d Pa9a [q.v.] given in Ibnlemin Mahmud Kemal Inan's Osmanl devrinde son sadrazamlar 4, Istanbul 1969, i, 159, he is referred to merely as Efendi in all official documents until he achieved the rank of vizier in 9a#bn 1271/May 1955 and thereby acquired the title of Pa9a. See for this period, S. Kekule, ber Titel, mter, Rangstufen und Anreden in der offiziellen osmanischen Sprache, Halle 1892, and the entries in M. Z. Pakaln's Tarih deyimleri ve terimleri szl, Istanbul 1946, s.v. elkb- resmiye, rtbe, mlkiye, vezir, etc. It seems that these bureaucratic gradations in titulature introduced under the Tanmt continued under the Young Turk rgime in the early 20th century, but, as in the parallel case of Persia under Ri' 9h Pahlaw, Kemal Atatrk's secularising | [V:631b] and reforming policies did not allow these titles to continue in official and public usage. According to the Law no. 2590 of 26 November 1934 Concerning the abolition of appellations (lkap) and titles (nvan) such as Efendi, Bey and Pasha, these modes of address were swept away, and religious titles such as Hac, Hafz and Molla were also banned (Kazim ztrk, Son degiiklikleriyle gerekeli anayasa 2, Ankara 1975, 306; Blent Dver, Trkiye cumhuriyetinde lyiklik, Ankara 1955, 175). Instead of Bey and Hanm, Bay and Bayan were introduced for Mr and Mrs. But as in Persia, old conventions and speech habits die hard, and in popular speech, the old title survive: Paa for generals, whether active or retired; Efendi for artisans and non-Muslims; stad for craftsmen, artists, etc.; Hoca for teachers, secular and religious; and so forth (cf. G. Lewis, Turkey 3, London 1965, 110-11). SIKKA (A.), literally, an iron ploughshare, and an iron stamp or die used for stamping coins | (see Lane, Lexicon, 1937). From the latter meaning, it came to denote the result of the stamping, i.e. the legends on the coins, and then, the whole operation of minting coins.

1. Legal and constitutional aspects.

As in the Byzantine and Ssnid empires to which the Arab caliphate was heir, the right of issuing gold and silver coinage was a royal prerogative. Hence in the caliphate, the operation of sikka , the right of the ruler to place his name on the coinage, eventually became one of the insignia of royal power, linked with that of the 9uba [q.v.], the placing of the ruler's name in the bidding prayer during the Friday congregational worship.

This right of placing the ruler's name on the coinage did not appear immediately in the Islamic state. As is well known, up to the caliphate of the Umayyad #Abd al-Malik [q.v.] at least, the former Byzantine and Ssnid money continued to circulate; and when the new holders of power within the conquered lands finally placed their own names on newlyminted coins or counterstamped them on older coins, this was not a sign of a prerogative reserved to the caliphs. Provincial governors like Ziyd b. Abhi , al0a333 , #Ubayd Allh b. Ziyd [q.v.], etc., minted coins bearing their own names only. Even when the use of the caliphs' names on coins spread, certain provincial governors continued to follow their own local minting practices; thus at the end of the 1st century A.H., the governor of North Africa Ms b. Nuayr [q.v.] still minted coins of his own, with legends in Latin. Also notable, during the period from Mu#wiya to #Abd al-Malik, was the appearance of effigies of the caliphs on coins, and when the rulers' names appeared, these were often followed by the titles of 9alfa or amr almu"minn . Some #Abbsid coins did not always have the caliph's name on them, but might be minted by the designated heir to the throne or wal 'l#ahd or by a caliphal minister. But it became more and more general for the caliph's name to take precedence, usually with their honorifics or

laab s [q.v.] also. With the break-up of caliphal unity, provincial governors began to mint their own coins, placing their own names on them but usually continuing to place first the name of the reigning caliph as a witness to their theoretical subordination to the universal caliphate. Of course, when dynasties arose in deliberate defiance of or emnity to the #Abbsids , as was the case with the Spanish Umayyads and the Fimids of North Africa and Egypt, their coinage was a completely independent one, with their own names only inscribed on the coins. (C.E. Bosworth)

2. Coinage practice.
In Lane's Arabic-English lexicon, the origin of the word sikka is given as sakk, originally a ploughshare, or a nail, pin or peg of iron, thus sikka , an engraved piece of iron, a die for striking coins, hence maskk , plural maskukt , coined money. In its literal meaning, sikka refers to coinage dies in a mint, in early days made of bronze rather than iron, which tended to shatter under the repeated blows of the hammering process that was used to transfer the inscriptions on the die to the metal blank or planchet. For the purposes of this section of the article, however, sikka is | discussed in its figurative sense, the right of a Muslim ruler to have his name inscribed on the coinage (see above, 1). From its origins in classical antiquity until today, manufacture of money, and the standards controlling it have been under governmental supervision. The manufacture of coin was an important source of revenue for the government which derived from the fees, or seignorage, charged by the mint for converting unrefined metal into coin. The government stamp on the metal served as a guarantee of its purity, and as a permit for it to become legal tender within the area of authority where it was issued. In the city states of antiquity, the coinage was first identified by images of local gods and other symbols, and was often guaranteed by the names of moneyers. Under the Roman and Parthian Empires, and later the Byzantines and Ssnids , local coinages were swept away, and replaced by those whose principal feature was the ruler's bust, often with his name and titles,

and thus monarchial coinage became the rule throughout the Mediterranean and Iranian world. In the time of the Prophet Muammad the 0i3z had no indigenous coinage of its own, and its monetary stock was composed of whatever coins were earned through trade or pilgrimage receipts. These were Byzantine gold and copper coins, Ssnid silver, and a miscellany of older coins which had remained in circulation long after the states which issued them had passed into history. The rapid spread of Islam, however, resulted in the acquisition of large quantities of Byzantine and Ssnid coins which fuelled the economy of the newly-conquered territories. The Byzantine money came mostly from outside the territories conquered by the Arabs, although there was a long-established Byzantine mint in Alexandria, and another in Jerusalem operational ca. A.D. 609-15. In the Ssnid lands in the east, however, the Arabs acquired control of many local mints. The silver coinage struck in them bore the name and bust of the ruler, the mint mark and the regnal year of striking. Because the Arabs had no coinage of their own, and the populations of the lands they conquered belonged to two empires with very different monetary systems, they took the pragmatic step of adopting both systems to avoid disrupting the local economy and antagonising their new subjects. The earliest dateable Islamic coins are silver drachms, or dirhams, bearing the name and bust of the last Ssnid ruler Yazdigird III (11-31/632-51) with the legend bism Allh in the obverse margin and on the reverse the mintmark and the date 20, his last regnal year, which corresponded to the year 31 A.H. Yazdigird's name and bust were then replaced by those of 9usraw II (590-628), which became the model for the remainder of the Arab Ssnid series. It soon became the custom for local Muslim governors to replace the name 9usraw with their own names in Pahlaw script. The dates on these coins, however, are often difficult to elucidate because in many instances it is uncertain whether those

above the year 31 were continuations of Yazdigird's regnal years or the actual Hi3ra years of striking. Outside of the former Ssnid territories, the picture is far less clear. It is not known when Islamic coinage began in the former Byzantine lands, because none of the coins in circulation there were dated. Some authorities have argued that it started soon after the Arab conquest, while others have dated its inception to the early years of the caliph #Abd al-Malik b. Marwn (65-86/684-705). In either case it is clear | that the Arabs began to strike copper fuls in long-dormant Syrian and Palestinian mints, with designs based on Byzantine prototypes, often giving the names of the towns in both Latin and Arabic. Occasionally, they bear the phrase bism Allh to give them a specifically Islamic character. Mints did not usually share the same designs, which emphasised the local nature of each issue. None bore the name of a caliph or local governor. It can thus be said with some certainty that the idea of sikka as a prerogative of caliphal sovereignty had not yet developed in the early years of the Islamic community. The situation changed significantly after #Abd al-Malik b. Marwn defeated the anti-caliph #Abd Allh b. al-Zubayr [q.v.] in 73/692. This victory enabled him to direct his attention to the creation of institutions which would serve the needs of the Islamic community and strengthen centralised Umayyad rule over the empire. Several experiments were made to reform the coinage, which are dealt with in some detail in the articles dnr , DIRHAM and FALS . It should be noted that #Abd al-Malik introduced a series of copper fuls showing a standing figure of the caliph drawing a sword in defence of the Muslim community with the legend li-

#Abd Allh #Abd al-Malik Amr alMu"minn , For the Servant of God #Abd al-Malik Commander of the Faithful. This is the only instance where an Islamic ruler adopted the style of the imperial Byzantine coinage for use among the Muslims. Although these fuls are undated, they may be attributed to the years 74-7, because they are linked stylistically to standing caliph dnrs which bore these years of striking in their legends.

#Abd al-Malik's coinage reform of 77/696-7 removed all images, names and titles from the dnr in favour of legends drawn from the |ur"n , and this model was applied to dirhams in 79/698-9. The only human name to appear in the legends was that of Muammad , which implies that, as in the frequently used laudation al-Mulk li' llh , Sovereignty belongs to God, the right of sikka was vested in the hands of God and of His Messenger. While gold and silver were given this distinction, it was not always the case for the copper coinage where the names of a caliph or governor were occasionally used to indicate the name of the local issuing authority. This usage should not be confused with the right of sikka per se, but only as a means of holding a local governor responsible for coinage issued within the area of his jurisdiction. Despite their differing characters, none of the succeeding Umayyad caliphs altered the legends on the precious metal coinage, which suggests that it satisfied both the spiritual and economic needs of the Muslim community. The revolutionaries in the late Umayyad period made a few alterations to the standard Umayyad dirham (no dnrs are known from this time). Those of both

#Abd Allh b. Mu#wiya and Ab Muslm [q.vv.] and their lieutenants bore an additional | legend: |ul l as"ala -kum #alay -hi a3r
an

ill 'l-mawaddata fi 'lurb Say: 'I ask of you no recompense for this other than the love of kin' ( |ur"n , XLII, 23). This was obviously intended to provide divine sanction for 3a#farid and #Abbsid claims to the caliphate. There were also 9ri3ite issues which bore their rallying cry l ukma ill li' llh authority belongs to God alone. There is a third type of revolutionary issue, which was the only known post-reform dirham struck in the Umayyad period to bear the name of someone other than the Prophet Muammad . This was issued in the name of 3uday# b. #Al alKirmn , and carries the additional legend: mimm amara bi-hi al-

Amr alKirmn b. #Al authorised by the Amr alKirmn b. #Al . Because the #Abbsids based their claim to the caliphate on their close relationship to the Prophet, they replaced the Srat alI9l , which was used by the Umayyads basically as an irritant to the Christians, with Muammad rasl Allh . Thus it could be argued that the original #Abbsid sikka was in the name of the Prophet. They did, however, change the way in which the caliph was named. The Umayyad caliphs were known by their proper names and those of their father, e.g. #Umar (II) b. #Abd al#Azz or Hi9m b. #Abd al-Malik, followed by the caliphal title Amr alMu"minn . The early #Abbsid caliphs became known by their unya

, #alam and laab , e.g. Abu 'l#Abbs #Abd Allh alSaff and Ab 3a#far #Abd Allh alManr , but neither of these names is known to appear on their coins. In 145/762, however, alManr granted his son alMahd the right of responsibility for the silver dirham coinage of 9urasn and Armenia. The wording of this privilege copied the style of legend used on some of the copper coinage: amara bi-hi alMahd Muammad b. Amr alMu"minn . The name of a local governor, al0asan b. al|ataba , was also found on a dirham of Armenia dated 154/771. In the reign of alMahd , 158-69/775-85, the ruler's style regularly appeared on dirhams in the form li 'l9alfa al-

Mahd and rarely with his name Muammad . Two of his sons were occasionally granted responsibility for dirhams in the form mimm amara bi-hi Ms wal #ahd alMuslimn for his heir, and mimm amara bi-hi Hrn b. Amr alMu"minn for the future alRa9d . The names of governors also appeared on the dirhams more frequently. During his brief rule (169-70/785-6) alHd was referred to either as li 'l9alfa alHd or li 'l9alfa Ms . AlRa9d 's earliest dirham coinage from alHrniyya in 170 and 171 called him by his first throne name, alMar' The Approved One: li 'l9alfa alMar' mimm amara bi-hi Hrn

Amr alMu"minn . The caliph's name then made its first brief appearance on a few rare dnrs of 170 and 171 in the form mimm amara bi-hi #Abd Allh Hrn Amr alMu"minn , where #Abd Allh was used in its titular form as it had been on the coins of #Abd al-Malik. Between 170 and 187/786-803, while alRa9d was under the tutelage of Abu 'lFa'l 3a#far alBarmak [see ALbarmika ], an extraordinary variety of coinage was issued. The gold dnrs of Egypt carried the names of its governors #Al , Ms , #Umar , Muammad , Dwd and Ibrhm

, then that of its honorary governor 3a#far (alBarmak ) and finally 9lid . Dnrs issued in #Ir between 177 and 187 bore the legend mimm amara bi-hi alAmn Muammad b. Amr alMu"minn . The silver coinage was far more complex, sometimes naming the caliph as either Hrn or alRa9d , but often not mentioning him at all. AlAmn was usually called wal #ahd alMuslimn , and his younger brother alMa"mn the second heir, wal wal #ahd alMuslimn . 3a#far 's name appeared either alone after that of the caliph and his heir, or with the names of local governors. This coinage is particularly valuable for historians because the governors' names provide a chronology for the period which would otherwise have escaped

posterity. Presumably they were granted the privilege of placing their names on the coins when they received their commissions from the #Abbsid chancellory headed by 3a#far alBarmak . After the latter's execution in 187/803, alRa9d curbed this practice, and most of the coinage recovered its former anonymity, particularly in mints such as Madnat alSalm , alRfia and alMuammadiyya | which were under direct caliphal control. The conflict that erupted between alAmn and alMa"mn after alRa9d 's death in 193/809 was reflected in the coins they each struck. No specialised, systematic study has been made of the coinage of these two rulers, which is the most complex in the history of the Islamic world, because by this time responsibility for the sikka had become highly decentralised, and indeed fragmented. For example, after the year 145/762 the Umayyad rulers of Spain were striking conventional, anonymous Umayyad dirhams. In the Ma9rib the Idrsids and other local rulers placed their own names on the coinage without any titles. During alRa9d 's rule, the province of Ifriya had fallen into the hands of the A9labids , who became its hereditary governors. They retained the design of the early #Abbsid dnr

, but differentiated it by adding the governor's name and the dynastic symbol 9lib to its legends. The province of Egypt, which alMa"mn acquired in 196/812, now became the western boundary of the #Abbsid caliphate. From then until 213/829 the names of provincial governors appeared on the Egyptian coinage, usually with that of the caliph. Between 198 and 211 Syria was controlled by Muammad b. Bayhas, who placed the caliph's name above his on the dirhams which he struck. Madnat alSalm ( Ba9dd ) was held by alAmn until 198/813, when it fell to the forces of alFa'l b. Sahl 9u 'lRi"satayn [q.v.]. His conquest marked a turning point for the currency, because in 198 alFa'l struck the first #Abbsid dnr to bear a mint name, sc. Madnat alSalm . More importantly for the purposes of this article, he added the word li' llh For God, above Muammad rasl Allh to the legends found on both

dnrs and dirhams. This dedication made it clear that the right of sikka was vested in the hands of God, passing through those of His messenger Muammad to the individual named as the issuing agent. This chain of authority can be seen in its most highly developed form on the dirhams struck by alMa"mn after he chose the eighth 9# Imm as his heir in 201/816. The reverse reads: li' llh; Muammad rasl Allh ; alMa"mn 9alfat Allh ; mimm amara bi-hi alAmr alRi' wal #ahd alMuslimn , #Al b. Ms b. #Al b. Ab Tlib; 9u

'lRi"satayn . The appointment of #Al alRi' as wal #ahd sparked off a Sunn revolt in Ba9dd , which was nominally led by alMa"mn 's uncle, Ibrhm b. alMahd (202-3/817-19 [q.v.]). He refrained from placing his name in full on the few dnrs attributed to him, but abbreviated it to its first and last letters, alif/mim. AlMa"mn celebrated his triumph over alAmn in 198/813 by adding a |ur"nic passage to the dirhams he struck in Marw, the seat of his government: With God is the Decision in the past and in the future; on that day the Faithful shall rejoice in the help of God (XXX, 4-5). While this passage was almost certainly chosen by alMa"mn to give immediate divine sanction to his seizure of the caliphate, with time and continuous usage it became the #Abbsid motto, and was found on all dnrs and most dirhams issued by the dynasty until its downfall in 656/1258. At first its use spread gradually, coming to the dirhams of Madnat al-

Salm with alMa"mn 's arrival in the city in 204/819, but it achieved greater prominence in 206/821 when the capital mint issued new dnrs and dirhams inscribed in a new monumental and highly legible Kfic script. On this reform coinage, which had come into general use by 215/831, alMa"mn harked back to the past by allowing neither his name nor that of any governor to appear in its legends. | Thus the sikka was once again issued only in the name of God and His Messenger. When Ab Is Muammad alMu#taim succeeded to the caliphate on the death of his brother alMa"mn in Ra3ab 218/833, he continued to strike the same anonymous coinage, but distinguished it slightly by altering the former leftward slant of the word li' llh to make it fully vertical. In 219/834, however, he introduced a new style of throne name, a participial phrase describing the caliph by his relationship to God rather than by the manner of his leadership of the Muslim community. From alSaff until alMa"mn , the laab is understood to have modified the title al9alfa , e.g. the Victorious Caliph, the Orthodox Caliph or the Trusted Caliph, but the new sikka read li' llh; Muammad rasl

Allh alMu#taim bi' llh For God, Muammad is God's Messenger, the One Who Relies on God. This new style was probably chosen because it conformed to the theory that the sikka originated in and descended from God's sovereign power. This form was used by all but one of al-Mu # taim 's successors until the end of the dynasty, and was only modified for political purposes when the name of the heir was added to the legends. The practice began under al-Mutawakkil, whose son was first named Ab #Abd Allh b. Amr alMu"minn , and then received his later throne name alMu#tazz bi' llh b. Amr alMu"minn on his father's dnrs and dirhams. It was taken further when the feeble caliph alMu#tamid divided jurisdiction between his son and heir 3a#far in the West and his powerful brother Ab Amad ala in the East. The heir was first named 3a#far

on his coinage, and later alMufawwi' il ' llh , while ala was always known as alMuwaffa bi' llh . After alMuwaffa defeated the Zan3 rebels he added another title to the coinage struck under his jurisdiction: alNir liDn Allh , alMuwaffa bi' llh . He subsequently included the name of his heir, Amad b. alMuwaffa bi' llh , who became known as alMu#ta'id bi' llh after his father's death, in the year before he succeeded Mu#tamid as caliph. Throughout the latter part of the 3rd/9th-10th century the unity of the #Abbsid state was breaking down because of the rise of powerful, virtually independent local rulers who emphasised their status by adding their names to both the coinage and the 9uba . Even the caliphs had occasionally honoured individual wazr s on their own coinage, but never in their own names. For example, al-

Mu#tamid included the title 9u 'lWizratayn to honour -#id b. Ma9lad in 270/883; alMutaf Wal al-Dawla to honour Abu 'lHuayn al|sim b. #Ubayd Allh in the year of his death, 291/903-4; and alMutadir #Amd al-Dawla to honour alHuayn b. al|sim , the son of the caliph's wazr , on some of his coinage dated 320/932. The local rulers, however, used only their own #alam without any titles on the coins which, in theory, they struck on behalf of the caliph. This practice started in Egypt and Syria in 265/879 when Amad b. ln placed his name below that of the caliph in the reverse field. In the East, it began somewhat earlier when the first -affrid ruler added his own name Ya#b to the coinage (ca. 259-65). Before long the practice became universal, and whether by usurpation or grant from the caliph, the presence of names on the coinage came to be seen as a right that could be exercised by any serious rebel, semi-autonomous local governor or faithful ally of the #Abbsid

caliphate. This adds an extra dimension of interest to the study of the series for the historian and numismatist, because new | coins fill in gaps in our knowledge which existing textual sources may be unable to do. In the words of Stanley Lane-Poole in his Fasti Arabici: The coins of the Muslim East do not so much recall history as make it... If the complete series of coins issued by every Muslim state was preserved, we should be able to tabulate with the utmost nicety the entire line of kings and their principal vassals that have ruled in every part of the [Muslim Community]... to draw with tolerable accuracy the boundaries of their territories at every period. While in theory the right of sikka flowed downwards from God, through the Prophet, to his vicegerent the caliph, and from him to his vassal/ally, and ultimately perhaps to the latter's heir or an important governor, in practice it now moved in the opposite direction. The local strong man who controlled the mint defined his political and even religious position by acknowledging only those overlords who were valuable to his status, or by choosing |ur"nic and other legends that defined his allegiance in the Sunn 9# divide. No detailed account of the sikka in such cases can be given here, but for illustrative purposes examples are drawn from the principal Islamic dynasties which are not discussed elsewhere in this Encyclopaedia. Until 297/909 there was only one caliphate in the Islamic community, but in that year #Abd Allh al9# proclaimed the Fimid claimant #Abd Allh alMahd bi' llh Amr al-

Mu"minn at |ayrawn in Tunisia. The statement on his sikka: alImm alMahd bi' llh #Abd Allh Amr alMu"minn prompted the Umayyads of Spain to revive their claim to the Sunn caliphate. After 316/928, #Abd alRamn III issued a re-designed coinage placing his name in the reverse field, alImm alNir liDn Allh #Abd alRamn Amr alMu"minn , which paralleled that of his Fimid rival. In later reigns this order was reversed, e.g. alImm Hi9m Amr alMu"minn al-

Mu"ayyad bi' llh . Still later, the Spanish coinage often incorporated the title and name of the chief minister as well as that of the caliph, e.g. al03ib #Abd al-Malik. Other names also appeared, often those of wazr s or masters of the mint. In such instances, however, these men should not be considered as the holders of the sikka , unlike in the East where it was usually the lowest-ranking name who actually controlled the currency. This is well illustrated by the coinage issued during the crisis in the #Abbsid caliphate, when its erstwhile vassals brought about its prolonged eclipse. In 329/940 the Amr alUmar" , Abu 'l0usayn Ba3kam was able to have his name included on alR' 's dnrs and dirhams beneath that of the caliph, where he was described simply as mawl client. On the accession of alMutta , his name appeared in full: Abu 'l0usayn Ba3kam Mawl Amr alMu"minn . The sikka then reverted to the caliph and his heir alManr . In 330/942 Ab

Muammad al0asan , the 0amdnid ruler of Mawil , was appointed Amr alUmar" with the title Nir al-Dawla. The following year, his brother's name was added to the legends below that of the caliph's heir: Sayf al-Dawla Abu 'l0asan , and that of the senior amr below the caliph's: Nir al-Dawla Ab Muammad . In 333-4/945 the name of the Amr alUmar" alMuaffar Ab Waf" ( Tzn ) appeared on coins of alMustakf , who very exceptionally called himself Imm al0a alMustakf bi' llh . Shortly after this, he was forced to cede Ba9dd to Buwayhid control, which ended both

#Abbsid independence and his life, but not before he had transformed the three sons of Buwayh from Amad , #Al and 0asan into Mu#izz , #Imd and Rukn al-Dawla. | For a time, this style of laab was the highest form of title attained by a secular ruler in the East. The Buwayhid sikka can be difficult to determine, but the general principle to follow is to go from one side of the coin to the other starting with the name of the caliph, usually found in the reverse field below Muammad rasl Allh and then to work downwards from the highest-ranking amr to the lowest, and thus arrive at the individual who actually exercised the right of sikka . The next round of inflation in coinage titulature was set off when the caliph al"i# li' llh invested #A'ud al-Dawla as supreme secular ruler in 367/977. He now styled himself al-Malik al#$dil #A'ud al-Dawla waTa3 al-Milla Ab 9u3# . On other coins struck immediately before his coronation he was described as alAmr

al#$dil and al-Malik al-Sayyid. Before long all the ruling Buwayhid amr s had royal titles and laab s in both the al-Dawla and al-Milla forms, and often in an al-Umma form as well. Bah" al-Dawla then assumed a superior laab in the alDn form calling himself The Just King of Kings and 9h of 9hs . His sikka thus read Malik alMulk , 9han9h , |iwm alDn , Abu 'lNir , Bah" al-Dawla waDiy" al-Milla wa9iy9 al-Umma. Titular excess reached its highest point under the Buwayhid ruler of Frs , Ab Kl3r (415-40/1024-49), who was one of the greatest coiners in Islamic history. Following his investiture as Amr alUmar" in 435/1044, his sikka read 9han9h alMu#azzam

, Malik alMulk , Muy Dn Allh wa9iy9 #Ibd Allh wa|asm 9alfat Allh Ab Kl3r . After his death, laab s in the Allh , al-Milla and al-Umma forms went out of fashion, and those remaining were usually shortened to the alDuny wa 'lDn form. Between 449 and 541/1057-1146 the Almoravids or alMurbin [q.v.] in the Ma9rib struck a plentiful gold and silver coinage acknowledging the #Abbsid caliphate, but never naming the caliph individually. He was referred to as alImm , #Abd Allh , Amr al-

Mu"minn , and in later years the epithet al#Abbs was sometimes added. The rulers, who were known simply as alAmr Ab Bakr b. #Umar , alAmr Ysuf b. T9u fn , etc., later adopted the sub-caliphal title Amr alMuslimn . #Al b. Ysuf named two successive wal #ahd s, Sr b. #Al between 522 and 533/1128-39 and T9ufn b. #Al (533-7/1139-43). The same style of titulature was used by the remaining Almoravid rulers, T9ufn , Ibrhm and Is . When the Almohads or alMuwaidn

[q.v.] dynasty seized power in Morocco in 540/1146, they altered their sikka radically. It was based on the belief that the sect's founder, Muammad b. Tmart [q.v.], whose followers called him alMahd , could purify Islam of its corruptions. After Ibn Tmart 's death the sect was led by his most capable disciple #Abd alMu"min [q.v.] who, after his defeat of the Almoravids, introduced a new style of coinage unlike any found elsewhere in the Islamic community. Although nominally Sunn in allegiance, the Almohads made no reference to the #Abbsid caliphate, and removed the traditional mint and date formula from the legends, which were inscribed within a new square in circle design. They did, however, take great delight in titulature and genealogy, which somewhat makes up for the lack of mints and dates. A sample sikka on a dnr of Ab 0af #Umar (646-65/1248-66) illustrates this: in reverse square, alMahd Imm al-umma, al|"im bi-Amr Allh , al9alfa alImm , Ab Muammad #Abd alMu"min ibn

#Al Amr alMu"minn , in reverse segments, Amr alMu"minn , Ab Ya#b Ysuf ibn al- | 9alfa , and in obverse segments, Amr alMu"minn alMurta' li-Amr, Allh Ab 0af b. alAmr , alhir Ab Ibrhm , ibn al9alfatayn . The same square in circle design, with similarly convoluted legends, was used by the 0afids , Marnids and Ziynids . Before leaving this type, there is the sikka found on a dnr of the last Narid ruler in Spain, Muammad

XII, Boabdil, who lost Granada to the Christians in 897/1492. It reads: #Abd Allh , al9lib bi' llh , Muammad b. #Al b. Sa#d b. #Al , b. Ysuf b. Muammad b. Ysuf b. Nar , ayyadahu Allh wanaara -hu with the Narid motto l 9lib ill Allh repeated four times in the margin. Turning to the coinage of the Fimid caliphs [see fimids ], it should be recalled that the first ruler's coins bore the legends of a first period #Abbsid dnr , with no more than his name to distinguish them from the previous

A9labid dnrs . His successor, however, changed the design and calligraphy on the coinage and used as his sikka Muammad Abu 'l|sim alMahd bi' llh al|"im bi-Amr Allh Amr alMu"minn . The coins of the third caliph still showed no overt signs of the Fimid leadership of the Ism#l 9#s : Abd Allh Ism#l alImm alManr bi' llh , Amr alMu"minn . Their real religious feelings burst forth with the first coinage of the fourth caliph alMu#izz . On the obverse, the Kalima was augmented by the sentence wa#Al b. Ab lib

waiyy rasl n"ib alfa'l wazaw3 alzahr" albatl And #Al b. Ab lib is the Nominee of the Prophet, Most Excellent Representative, and Husband of the Radiant Chaste One. On the reverse the caliph styled himself #Abd Allh Ma#add Ab Tamm , alImm alMu#izz liDn Allh , Amr alMu"minn , muy sunnat Muammad , sayyid almursaln , wa-

wri9 ma3d ala"imma almahdiyyn Revivifier of the Sunna of Muammad , Lord of the Transmitters and Heir to the Splendour of the RightlyGuided Imms . This coinage is said to have caused serious problems for the government because most of their subjects were Sunn by persuasion, and a less inflammatory legend was quickly substituted in its place, The Imm Ma#add summons (to belief in) the Unity of God, the Everlasting. When al#Azz succeeded alMu#izz , he introduced the phrase #Abd Allh wa-waliyyuhu The Servant of God and His Companion to the Fimid sikka , preceding the unya and #alam which were followed by the caliphal title and Amr alMu"minn . Only al0kim regularly placed the name of an heir on his coinage. Breaking all dynastic conventions, he disregarded the claim of his eldest son alhir and chose a distant cousin and nonentity for the position, #Abd alRam , who appears on the sikka as wa-

#Abd alRam wal #ahd alMuslimn . In Ism#l thinking the use of alMu"minn was limited to those who held Ism#l beliefs, while the general community were alMuslimn . Through his actions it would appear that al0kim saw himself as the last Imm . It is said that he secretly appointed a great religious scholar to lead the Ism#l community as wal #ahd alMuslimn , while #Abd alRam was to have exercised overt political leadership of the Muslim community. Other Fimid coins highlight the problems in the disputed succession after al$mir 's death in 524/1130. He left no direct male heir, but one of his consorts was said to be pregnant by him. The wazr alFa'l Ab #Al

Amad , who seized control of the government, first struck dnrs in the name of the expected Imm , alImm Muammad Abu 'l|sim alMuntaar li-Amr Allh Amr alMu"minn , but when the heir failed to materialise he is said to have discarded his allegiance to the Sevener Fimids and revealed himself as a Twelver 9# . He then proclaimed his beliefs | by striking coins in the name of the Twelfth and last Imm , Muammad alMahd , with himself as his lieutenant and viceroy, alImm alMahd , al|"im bi-Amr Allh Hu33at Allh #al 'l#limn , n"ibuhu

wa9alfatahu alFa'l Ab #Al Amad . After alFa'l 's overthrow the new ruler first proclaimed himself wal #ahd , because of his collateral claim to the caliphate, #Abd Allh wa-waliyyuhu, Abu 'lMaymn #Abd alMa3d wal #ahd alMuslimn , and then, casting aside his scruples, as caliph, alImm #Abd alMa3d Abu 'lMaymn al0fi liDn Allh Amr alMu"minn .

Following Saladin's accession to power after the last Fimid caliph in 567/1171, he took the cautious first step of striking his sikka in the name of the #Abbsid caliph: alImm al0asan , alMusta'" bi-Amr Allh , Amr alMu"minn , and his overlord Nr alDn Mamd b. Zang , the ruler of Aleppo, then, after the latter's death, in his own name: al-Malik alNir -al alDuny wa 'lDn Ysuf b. Ayyb . This style of sikka became the standard form for subsequent Ayybid rulers. After the death of the last Ayybid ruler of Egypt, al-Malik alMu#aam 9iy9 alDuny

wa 'lDn Trn 9h b. Ayyb in 648/1250, power was briefly held by one of the rare female rulers in the Muslim world, 9a3ar al-Durr [q.v.]. Her sikka read: alMusta#imiyya , al-liiyya , Malikat alMuslimn , Wlidat al-Malik alManr , 9all Amr alMu"minn . Lane-Poole deduced from this legend that she was formerly in the harem of the caliph alMusta#im , who presented her to al-li Ayyb , that she became Queen of the Muslims and mothered a prince, alMalik alManr , whom she termed Friend of the Commander of the Faithful. The second Mamlk ruler, Aybak, concealed himself behind two fictive rulers, first alA9raf Ms , al-Malik alA9raf Abu 'lFat Ms

b. al-Malik al-li Ayyb , and then the last powerful Ayybid : al-Malik al-li Na3m alDn Ayyb b. al-Malik alKmil , Aybak. The succeeding rulers maintained the Ayybid style of sikka until Baybars gave refuge to an #Abbsid prince who fled the Mongol sack of Ba9dd . In return for his name on the sikka, alImm alMustanir bi'llah Abu 'l|sim Amad b. alImm alhir Amr alMu"minn , the newly-recognised caliph granted Baybars the style alSuln al-Malik alhir Rukn alDuny wa 'lDn Baybars |sim Amr

alMu"minn . For the remainder of Mamlk rule, the conjoint title alSuln al-Malik remained as the royal protocol. Those rulers who were not themselves of royal descent were often identified by an epithet which indicated their original royal master; Baybars I and |alwn called themselves al-li after al-li Ayyb , while Kitbu9 , L#n and Baybars II were known as alManr after alManr |alwn . All the later Bar Mamlk rulers were descendants of |alwn , and carefully recorded their genealogy in their sikka . The Bur3 Mamlk rulers continued the same form of royal protocol as the Bars , but space limitations and, usually, a lack of royal descent kept it relatively brief. Two typical examples are alSuln al-Malik alA9raf Abu 'lNr |"it

By , #Azza naruhu and alSuln al-Malik alhir Ab Sa#d |nawh , #azza naruhu . The coinage does, however, hold one curiosity. In 815/1412 one of the #Abbsid caliphs in Egypt was elected to the sultanate as a political expedient. He styled himself either as alImm alA#am or alSuln al-Malik alMusta#n bi' llh Abu 'lFa'l al#Abbs Amr alMu"minn . Elsewhere, the founder of the Raslid dynasty of Yemen, alManr #Umar , initially struck his coins in the name of his nominal Ayybid

overloads, who had previously ruled the country. In 634/1236-7 he began to coin in his own name: al-Malik alManr | Abu 'lFat #Umar b. #Al . He followed the Ayybid convention of acknowledging the spiritual overlordship of the #Abbsid caliphate, and further emphasised his Sunn allegiance by becoming the first ruler to incorporate the names of the first four Orthodox Caliphs into the coin legends. This innovation was followed by a second when his son alMuaffar Ysuf became the first to style himself alSuln al-Malik as early as 648/1250, well before Baybars received the conjoint title in 659/1261. For Sal3 titulature on the sikka , see sal3ids , VIII. Among their successors, the Atabegs of Eastern Anatolia and Western Persia usually acknowledged the #Abbsid caliph as head of the Islamic community. As a reflection of the general insecurity of the age, each ruler was faced with the problem of how to express on his coinage the network of feudal allegiances and alliances which would maintain his security, and the coins provide a useful record of the many twists and turns in the political and military history of the time. A few examples will illustrate this. On a typical dnr of the Zangids of Mawil dated 616/1219-20, the sikka read Nir alDn Atabak b. #Izz

alDn b. Arsln 9h ( Nir alDn Mamd , son of #Izz alDn Mas#d , son of Nr alDn Arsln 9h ), on the obverse al-Malik alKmil referred to alKmil Muammad , the Ayybid ruler of Egypt, and on the reverse al-Malik alA9raf referred to alA9raf Ms , the Ayybid ruler of the 3azra and immediate neighbour of Nir alDn Mamd . On a dnr

of Mawil , struck after the Mongol conquest of Ba9dd , Badr alDn Lu"lu" [q.v.] was quick to recognise the new order in #Ir : Mngke |"n ala#am 9udbanda -yi #lam , Pdi9h r -i zamn , ziydat #amatahu , and on the reverse al-Malik alRam Badr alDn Lu"lu" . After Lu"lu" 's death in 657/1258, his son first struck coinage in the name of Mngke as above, naming himself al-Malik al-li Rukn alDuny wa 'lDn Ism#l . Then in 659, just before his downfall, he miscalculated by repudiating Mongol overlordship and struck dnrs in the name of Baybars and the #Abbsid caliph in Cairo: al-

Imm alMustanir bi'llah Amr alMu"minn alSuln alA#am al-Malik alhir Rukn alDn |sim Amr alMu"minn . The early Mongol Il 9ns of Persia inscribed their sikka in Uy9ur script: The coinage of (name) the Great 9n 's Viceroy, and under 9zn Mamd By God's Power, 9zn 's coinage. His successor l3eyt ( l3ayt ) introduced an important innovation to his first coinage. To satisfy what was probably a felt need to define his stance on the Sunn 9# divide, he incorporated the names of the first four Orthodox caliphs in the legends: Ab Bakr , #Umar ,

#U9mn and #Al , around the Sunn kalima . l3eyt then proclaimed his conversion to 9#ism by adopting the 9# kalima with the names of the Twelve Imms surrounding it. His new sikka may have been intended to quell any controversy over this move: Struck in the Days of Prosperity of our Master the Grand Sultan, Ruler of the Necks of the Community l3ayt Suln . Defender of the World and Faith, the Servant of God, Muammad , May God Preserve his Sovereignty. The later Il 9nid rulers returned to Sunn beliefs, and placed the name of the Orthodox Caliphs on all their coinages. No other names appeared besides that of the ruler, even in the cases of the last Il 9ns , who exercised no actual power in the state whatever apart from being named in the 9uba and sikka . One extraordinary exception to this practice is found on the coinage of Tmr Gr9n , or Tmr Lang. | When he seized control of Transoxiana in 771/1369-70 he did not depose the 1a9atay 9ns from their position as its nominal rulers. Between 771 and 790/136988, the name of Suyr9atmi9

appeared above that of Tmr , and between 790 and 800/1388-98 that of Mamd . Tmr called himself Amr Tmr Gr9n , but his successor 9h R9 employed the usual Persian style: alSuln alA#am 9hr9 Bahdr 9allada Allh mulkahu wasulnahu . For the Ottomans' and -afawids ' sikka , see #o9mnl . IX, and afawids . 6. After the fall of the -afawids in the part of Persia which came under the rule of the Htak Af9ns , the 9# kalima was replaced by the Sunn one on the coinage struck by 9h Mamd (1135-7/1722-4) and

A9raf (1137-42/1724-9). The sikka was now often expressed through the use of Persian couplets which bore the name of the ruler in elaborate and often playful wording. Because of the many puns and multiple layers of meaning which can be read into these distichs, they lose most of their sense in translation. They were obviously intended for the happy few who had the necessary education and means to appreciate them. On some of his coins the Af9arid Ndir 9h (1148-60/1735-47) gave himself the title Suln Ndir 9allada Allh mulkahu, on others he used distichs. Karm 9n Zand struck no coinage in his own name, but employed the invocation Ya Karm ! in its place. R.S. Poole explains the background in Coinage of the Shahs of Persia: The Zand and |3r 9ns before Fat #Al 9h did not assume full rights of sovereignty. Their money shows the position they took. The founder of each line first struck money in the name of 9h Ism#l III; then Karm 9n Zand, as wakl , struck in the name of

Imm Muammad alMahd , also using the invocation y Karm ! alluding to his own name. Muammad 0asan 9n |3r similarly coined in the name of Imm #Al alRi' . Evidently, they had no official 3uls . The later Zand 9ns , at least in some cases, had a 3uls . But on their money they assume no regal titles; there was still a -afawid heir. The principle of Karm 9n is not deviated from except in the appearance of the names without titles of his first successor Abu 'lFat and his last one Luf #Al; #Al Murd and 3a#far used allusive invocations ( Y #Al ! and Y Imm

3a#far -di !), while -di repeated that of Karm 9n . Similarly, $9 Muammad 9n |3r struck in the name of both Imms and was content with an allusive invocation (Ya Muammad !) even after he had conquered his rivals, and as sole prince had a 3uls . Probably this was because a -afawid prince, Suln Muammad Mrz , had been proclaimed by him in Tehran in 1200/1786 and was still living, although not in Persia.

Fat #Al 9h made an extraordinary innovation. Before his 3uls , he issued royal money under his name Baba 9n with the title of suln . On his later coinage he styled himself as alSuln b. alSuln Fat

#Al 9h |3r , sikka Fat #Al 9h 9usraw -ibirn or sikka Fat #Al 9h 9usraw Ka9warsitan . Muammad 9h used the title 9han9h Anbiy Muammad . Nir al-Din 9h usually placed alSuln b. alSuln Nir alDn 9h

|3r on his hammered coinage, while on some of his high denomination, machine-struck coins there was room to inscribe alSuln alA#am al9n alFa9m -ibirn Nir alDn 9h |3r . Similar styles were used by the last three |3r rulers, Muaffar alDn , Muammad #Al and Amad 9h . The sikka of the Dihl Sultans [q.v.] varies considerably, but the usual style followed the pattern of a coin of Mamd 9h (644-64/1246-66): alSuln al- | A#am Nir al-Dunya wa 'lDn Abu 'lMuaffar

Mamd ibn alSuln . One of the many sikkas of Mbarak 9h (716-20/1316-20), who regarded himself as both a religious and secular ruler, read alImm alA#am 9alfa Rabb al#$lamn |ub al-Dunya wa 'lDn Abu 'lMuaffar Mbarak9h alSuln b. alSuln alW9i bi 'llah Amr alMu"minn . The most complex coins in the series were struck by Muammad 9h II (725-52/1325-51), with over fifty varieties recorded. One group was struck in the name of his father, whom he very likely murdered: alSuln al9hid al9hid al9z 9iy9 al-Dunya wa 'lDn Abu 'l-

Muaffar Tu9lu 9h alSuln . On others he described himself as alMu3hid f Sabl Allh Muammad b. Tu9lu 9h; alW9i biTa"yd alRamn Muammad 9h alSuln; al#Abd alR3 Ramat Allh Muammad b. Tu9lu ... Still others he struck exclusively in the names of two #Abbsid caliphs in Cairo, alMustakf and al0kim .

The titles used by the founder of the Mu9al dynasty in India, Bbur (932-7/1525-30), were strongly influenced by his neighbours the 9ibnids of Transoxania. They were Sunn in character, and usually included the names of the four Orthodox Caliphs with their epithets. He often styled himself alSuln alA#am al9n al-Mukarram hir alDn Muammad Bbur Pdi9h -i 9z ; his son, Nir alDn Muammad Humyn (937-63/1530-56) used a similar style. The third ruler, Akbar (9631014/1556-1605), employed three different styles for his sikka . The first, alSuln alA#am 3all alDn Muammad Akbar Pdi9

-i 9z appeared in the early years of his reign. The next issue was anonymous in the strict sense of the word, but the legend Allh Akbar 3alla 3alluhu God is Most Great, May His Greatness be Glorified has caused many Westerners to assume that Akbar confused himself with God. It is more likely, however, that he placed this invocation on the coinage to draw attention to his newly established Tawd -i ilh Akbar 9h Akbar 9h 's Doctrine of the Unity of the Divine Being. The third type was an early instance of the use of Persian couplets in the coin legends. This may have been adopted in order to avoid placing the kalima on the coinage of a ruler who was not devoted to the practices of traditional Islam. One such example read The sun-shaped die of Akbar is the honour of this gold, while the light of the sun remains an ornament to the earth and sky. The coinage of Akbar's son 3ahngr (1014-37/1605-28) was certainly among the most artistic of any Muslim ruler. Elegant distichs, superb calligraphy and figural designs, combined with careful striking, have made his name famous as the master coiner of the age. Each issue seems to have been an occasion for fresh legends and designs, but on many his sikka read Nr alDn 3ahngr 9h [b.] Akbar 9h . His successor 9h 3ahn (1037-68/1628-58) reverted to a more traditional style of coinage, where the kalima and the four Orthodox Caliphs returned to the place of honour, and the ruler was styled

9ihb alDn Muammad -ibirn al9n 9h 3ahn Pdi9h -i 9z . The accession of the austerely religious Awrangzb (1068-1118/1658-1707) brought about the near permanent banishment of the kalima from the Mu9al coinage. Like the Ottomans and the -afawid 9h Ism#l II (984-5/1576-7), he believed that the profession of faith would be profaned if it fell into the hands of the unbelievers. This was quite opposite to the early Muslims' view that coins carrying texts from the |ur"n acted as missionaries of the Faith. Most of Awrangzb 's coins bore the couplet Struck coin in the world like the shining sun (for gold) or moon | (for silver) 9h Awrangzb #$lamgr . The reverse inscription referred to the ruler's regnal year and became virtually invariable: The year of accession associated with prosperity. The coins of the later Mu9als either bore titles, as on the coinage of 9h 3ahn , or couplets in the style of

Awrangzb . They retained their pride and claims to greatness until the end of the dynasty. The sikka of the last Mu9al ruler Bahdur 9h II (1253-74/1837-58) read Abu 'lMuaffar Sir3 alDn Muammad Bahdur 9h Pdi9h -i 9z . Today the use of the traditional sikka has virtually come to an end. The last ruler to place the kalima on his coinage was the Imm Amad (1367-82/1948-62), ruler of Yaman, who styled himself Amad 0amd alDn Amr alMu"minn alNir liDn Allh Rabb al#$lamn . His successor the Imm

Badr struck a token coinage in exile which did not circulate in the Yaman. Now the only countries which use a royal style on their coins are: Moroccoal0asan al9n al-Malik alMa9rib ; Su#d Arabiaal-Malik Fahd b. #Abd al#Azz alSu#d 9dim al0aramayn al9arfayn ; #Umn |bs b. Sa#d Suln #Umn and Brunei(in Latin characters) Sultan Hassanal Bolkiah. Elsewhere in the Muslim world, the coinage is issued in the name of the state or central bank. It has been entirely secularised and shorn of all its past associations, and is no more than a bland reflection of today's political realities. (R.E. Darley-Doran)

3. The Maria Theresa thaler.

From the mid-18th century and even amongst Bedouin and in remote parts of Ethiopia at the present time, these thalers have been used as a conventional, albeit unofficial, means of exchange, throughout the Arabian peninsula and in the Sudanic belt, and as far eastwards as the Maldives and Indonesia. Since the Empress Maria Theresa's death in 1780 restrikes bearing that date have been issued at different times from official mints because of continued demand for one reason

or another: Rome (1935-7), London (1936-61), Bombay (1941-2), Birmingham (1949-55), Brussels (1935-7), and Paris (1937-59), and still continuously from Vienna since 1961. In addition, counter-marked official and unofficial issues have been made in the Azores, Loureno Marques, Pemba, Djibuti, Bab al-Mandab, the |u#ay State of 9ihr and Mukalla, Na3d (Ibn Su#d , ante 1916 until 1923), 0i3z (under 0usayn , 9arf and then King of Mecca, 1916-20), the Maldives and Madura in Indonesia. In Western Africa, issues have crossed the Sahara from the Sudanic belt as far as Timbuktu, Nigeria and Dahomey. In these regions both the British and French authorities demonetised them in 1930. They have been used not only as a means of exchange but also for feminine decoration, and especially for bridal costumes; they have also served as a convenient source of bullion for manufacturing silver jewellery. Dubious restrikes have also been attributed to Florence, Leningrad, Marseilles, Utrecht and Venice. The first German crowns were the silver guldiners issued by Archduke Sigismund of the Tyrol in 1486, whose coinage was imitated by a number of German princes. The first thalers properly so-called originate from the discovery in 1518 of a silver mine by a Count von Schlick at Joachimsthal, on the border of Bohemia and Upper Austria. He obtained a licence to coin in silver, and made his first issues in 1525, denominating them thalers, an abbreviation of the toponym of origin. Variant spellings of this term occur in a number of European languages and in Amharic; in the Netherlands it became daalder, contorted into the American dollar. In Arabic, however, they are called ir9 , pl. ur9 . F.W. Hasluck has described in detail the extreme remissness of the [Ottoman] Turkish Government in the matter of coinage. The quality of metal was notoriously bad, and fluctuated in quality, and neither the actual quantity of money circulated nor the denominations

provided were sufficient for trade. Not only the treasury but also provincial Pa9as debased the currency by the ancient double-weights trick, taking in good money at a premium and then reissuing it heavily alloyed. | Thus foreign merchants trading within the Ottoman Empire imported their own currencies for sound business reasons. These were principally from Venice, Spain, the Austrian Empire, Germany, Poland and the Netherlands. England was exceptional in forbidding the export of bullion, and generally employed Netherlands currency. An attraction for Ottoman subjects was the consistency and unvarying fineness of the coin. It was impossible to clip it because of a collar, or an inscribed, milled or patterned edge. There was also a constancy of decoration and imagery. Within the Austrian Empire, archdukes, archbishops and others issued crowns of a fixed type, and for Ottoman subjects, the imagery itself was a guarantee of genuineness. Few of them, indeed, could read the Latin inscriptions. Spanish pieces of eight were first struck in 1497, and immediately entered into competition with the preceding currencies. In 1518, following its conquest, Mexico issued silver coinage, and then Peru after the conquest of 1524. Silver coinage of Spanish origins became even more plentiful after the discovery of the rich mines of Potosi in 1545. This was the situation for something like a century, until, in the mid-17th century, the Spanish royal ordinances attest a scandalous falsification of the silver moneys coined in our Peruvian mints. It led to the total demonetisation of Spanish currency in the British colonies in the latter half of the century. In 1728 the millesimal fineness had been lowered from 930.5 to 916.6, and the weight reduced; in 1772, when a massive recoinage was carried out, the fineness was further reduced to 902.7. The first Maria Theresa thaler was issued in 1751, the year of her accession. Ever since it has been consistently of 833.3 millesimal fineness, 1.553 ins in diameter, and weighing 433.14 gr. The legend is abbreviated, shown here by capital letters: obverse: Maria THERESIA Dei Gratia Romanorum IMPeratrix HUngariae et BOhemiae REGina; and, reverse: her coat-of-arms borne by a double-headed eagle, a decoration that could have appealed since it first occurs in Islamic numismatics in Artuid coinage, from an emblem depicted on a Byzantine tower restored by the Artuids at $mid

, Turkey, with their inscription dated 605/1208-9. The quarters display the arms of 1. Hungary; 2. Burgundy; 3. Bohemia; and 4. originally Upper Austria, but of Burgau in the restrike issues. The arms of Austria display a single-headed eagle only. The inscription reads: ARCHIDux AUSTriae DUX BURGundiae COmes TYRoli 1780. In the centre of the field is a shield of pretence bearing the arms of her husband, Francis, initially Duke of Lorraine, and after 1751 Duke of Milan and Holy Roman Emperor. On the edge of the flan is the inscription: JUSTITIA ET CLEMENTIA, with various decorative symbols, being the motto of her reign, making clipping impossible. In 1764 the Gnzburg mint was opened specifically to mint thalers for the use of bankers from Augsburg engaged in the Ottoman trade. Already in 1751 those destined for Turkey were controlled by a monopoly. The 583,250 pieces coined in 1751 had increased to 1,360,597 by 1757, and to more than 2 million by 1764. Such was the demand that issues were also authorised from Kremnitz and Karlsburg, and, later, Milan, Venice and Prague. By 1767 the traveller Carsten Niebuhr found them in Yemen. By the time that Maria Theresa died in 1780 it had become plain that coins bearing her bust were valued above all others in Arabia and Yemen. Thus in 1781 a bank- | ing firm sent a consignment of bullion to the Gnzburg mint requesting thalers with the date 1780. Permission was given, but after 1866 Vienna held a monopoly. In the first years of the present century some 46 million pieces were minted. It was the loss of her Italian territories in 1866 that caused Austria to reserve to herself the sole right to mint thalers. It was at this moment that Sir Robert Napier (later Lord Napier of Magdala) was preparing an expedition to Ethiopia to rescue beleaguered British diplomats and missionaries held by the Negus. The Vienna mint provided five million thalers; the British were well aware that no other currency could be acceptable in Ethiopia. It was a presage, but not foreseeable. In 1935 Mussolini determined to conquer Ethiopia, and, on 9 July 1935 succeeded in wresting the right to mint thalers from Vienna. It was an intolerable position for Britain. Not only was Britain pledged to Ethiopian independence; she also had commitments to Aden and the Arabian peninsula, as well as the Persian Gulf. The matter was resolved by an international commission of jurists, who ruled that the effigy on the thaler was of a person already dead for 150 years, who had been sovereign of a state that had disappeared in 1918. The successor state had twice introduced new currencies, finally the schilling, in 1924. It was in this way that the Tower Mint, in London, was enabled to mint more than 16 million pieces in 1941 when Britain invaded Ethiopia in order to restore the Emperor to his throne. Dies were also sent to Bombay, 8 million pieces being minted in 1940-1, and 10 million in the following year. Supplies were also needed for the Arab lands, and

Birmingham also minted further supplies, some of which inexplicably reached Hong Kong. Small numbers were also manufactured in Brussels and Paris. There were also unofficial mints. In the 0a'ramawt the present writer was able to pick up some fractions of thalers which had been manufactured locally as small change, and which were known as alKf coins from a well-known family of Sayyids. This accords with a remark made by Sir Richard Burton in 1872, of the situation at Zanzibar in 1857, that there are no mints, of which some sixteen exist at Maskat private shops to which any man may carry his silver, see it broken up, and pay for the coinage whatever the workmen may charge. He says that a clutch of currencies was to be found there: German crowns or Maria Theresacoined in Milan, known as Girsh Aswadas opposed to the Spanish or Pillar dollar Girsh Abyaz, or Abu Madfa# 'Father of Cannon' from the columns, and Girsh Maghrabi. Also Mexican dollars... In 1811 Captain Smee R.N. had reported that Spanish dollars were commonly current. So far no work has been done to identify the purely local manufactures. Regoudy is able to report a veritable curiosity, of a trouvaille of 672 Maria Theresa thalers confiscated by the French authorities from smugglers who were operating for the Front de Libration Nationale in Tunisia on 30 May 1959. They were chiefly restrikes from Rome, London, Bombay, Paris, Brussels, Vienna, Prague and Venice. Some 60% came from Rome and Bombay, suggesting that the trouvaille may have been formed in the 1940s in the Horn of Africa, only eventually to find its way into the hands of FLN arms dealers. The late Francesco Carbone, when he served in the Italian legation in Yemen from 1931 until 1961, | first in -an#" , then in Ta#izz , assembled a remarkable collection of thalers. Apart from 1780 restrikes, eight pieces of Maria Theresa antedated 1780, and two of her husband, Francis III Stephen; there were a further forty-eight pieces in the name of Francis I, dating between 1810 and 1830, together with one only of Francis Joseph I, of 1853. The collection was wholly random. During the whole period of Carbone's residence, the 1780 Maria Theresa thaler was in normal circulation, brought up from Aden in conveniently packed boxes. The ur9 minted in the name of the

Imm Yay b. Muammad never sufficed for local needs. Carbone thus abstracted the pieces not bearing the date 1780, replacing them in the Legation account with conventionally-accepted restrikes. The earlier group helps to illustrate the early popularity of the thaler and to show that pieces minted after 1780 not bearing Maria Theresa's name were none the less acceptable. H.G. Stride's statement that Maria Theresa died in 1780 and that all thalers issued subsequently bore this date would appear to be incorrect. In 1961 the Yemeni Government enquired of the British Legation in Ta#izz what the cost of purchasing one million Maria Theresa thalers would be. A quotation was passed to them: the cost of the silver was about five shillings, and the charge of the Royal Mint for manufacture at 16 per thousand pieces, the insurance and freight to be borne by the purchasers. The Yemeni Government did not proceed with the purchase. Ordinarily supplies of fresh thalers were introduced into circulation by Aden banks and merchants, whenever the cost of the silver, the minting charge, and insurance and freight were sufficiently below the exchange rate of the thaler to allow the bank a profit on the transaction. It remains to mention what best may be described as a medal in the Carbone collection. It is a copy in gold about 1 m thick of a thaler issued by Francis Joseph I from the Vienna mint in 1898. It was specially minted in -an#" at the mint there [see riyl ] by command of the Imm Yay on the occasions of his visit to King #Abd al#Azz Ibn Su#d and his pilgrimage to Mecca. Only a few were minted, for the Imm to give as presents to his friends, of whom Signor Carbone was one. al0I$Z 0I$Z , the birthplace and still the spiritual centre of Islam, is the

north-western part of the Arabian Peninsula north-western part of the Arabian Peninsula. As the site of the Ka#ba, as the home of the Prophet Muammad and the scene of Allh's revelations to him ( manzil al-way ), and as the capital district of the early Islamic state, al-0i3z is for Muslims as much the Holy Land (al-bild al-muaddasa) as Palestine is for Jews and Christians. Muslims are, in fact, even more zealous in guarding the inviolate character of their chief shrines; the areas surrounding Mecca ( Makka) and Medina (al-Madna al-Munawwara) are sacred preserves ( aram s) which only Muslims are allowed to enter, and restrictions have often been placed on the penetration of non-Muslims into other parts of al-0i3z. While agreeing in general that al-0i3z means the barrier, the Arabic sources differ in interpreting its application. The commonest view is that the barrier is the mountain chain of al-Sart [q.v.] separating the lowland of al-9awr or Tihma [qq.v.] along the Red Sea from the interior uplands of Na3d [q.v.]. Another view holds that the barrier stands between al-9a"m in the north and al-Yaman in the south, and modern geological research has shown that the mountains of these two regions lie beyond the Arabian Shield [see (3azrat) al-#arab ] to which al-Sart belongs. The concept of al-0i3z as an obstacle also derives from the fact that much of its area is covered by lava tracts ( arra s [q.v.]), which make it a black barrier (Yt, s.v.). Among the best known arra s in the early Islamic period were those bearing the names of Layl, Wim, al-Nr, and Ban Sulaym. Further research needs to be done on identifying these and on determining the correct forms of the modern names. On a visit to Tabk, for example, the author learned that the arra s to the south are called al-R.ht (vowelling uncertain) and #Uwayri', not al-Ra and al-#Uwayri' as often given on maps. No substantial agreement exists on defining the geographical limits of al-0i3z. Although Tihma is, strictly speaking, not a part of al-0i3z, it is often included in the region. Mecca in the hills has been called Tihmiyya, and Medina half Tihmiyya and half 0i3ziyya. In the east al-0i3z is sometimes carried as far as Fayd near A3a" [q.v.] and Salm, but this is an extreme interpretation, as is the one that extends al0i3z northwards into Palestine. The most circumscribed version of the northern extent excludes Madyan and its hinterland 0ism from al-0i3z. In the south al-0i3z once marched with al-Yaman, but in recent times #Asr [q.v.] has been interposed between the two. As treated in this article, al-0i3z corresponds in general to the Western Province of the modern Kingdom of Saudi Arabia. For descriptive purposes al-0i3z may be divided into three sections: northern, central, and southern. The central section, by far the most important for the history of Islam, is dealt with first. The central section central section may be taken as bounded in the south by the lands in the vicinity of al"if, Mecca, and 3udda [qq.v.], and in the north by the lands in the vicinity of Medina and Yanbu# [qq.v.]. From the verge of Medina a vast arra runs along al-Sart for about 300 km. almost to Mecca. | [III:362b] The old road from al-"if went north to the valley of al-Na9la al-Yamniyya, down which it descended towards Mecca. In this valley was |arn al-Manzil, the early mt

[see irm ] for pilgrims from southern Na3d and Oman; the present mt is a place called al-Sayl al-Kabr in the same valley. In al-Na9la al-9a"miyya was 9t #Ir, the mt for pilgrims from northern Na3d and Iraq coming along Darb Zubayda, the route which the consort of Hrn al-Ra9d provided with cisterns and other amenities. 9t #Ir is often mentioned as the limit of al-0i3z in this direction. A paved highway now winds down the mountains from al-"if directly to Mecca, so that the long loop to the north is avoided. The two Na9las, now called simply al-Yamniyya and al9miyya, empty into Wd Fima (classical Marr al-ahrn), the fertile bed of which is crossed by the road from Mecca to 3udda. Throughout the history of Islam travellers between Mecca and Medina have had a choice between two ways, one following the coast (al-ar or al-Darb al-Suln) and the other running east of the great arra (al-ar or al-Darb al-9ar), with variations in the itinerary for each way. Before the introduction of motor vehicles those choosing al-ar al-Suln usually by-passed 3udda in order to save time. Three hours out of Mecca were the domed tomb and mosque of the Prophet's last wife, Maymna [q.v.], in Sarif, where she was married. North of Wd Fima the road went through #Usfn, the scene of the Prophet's raid on the tribesmen of Liyn [q.v.]. Next the road traversed the cultivated area of 9ulay set back some distance from the coast. Not far beyond al-|a'ma the Red Sea would be sighted. Rbi9, though a port, had no proper harbour; ships anchored well away from the shore and transferred their cargoes to local sailing craft. As the mt for pilgrims coming overland from Syria, Egypt, and alMa9rib, Rbi9 succeeded the now ruined village of al-3ufa, which lies in a valley reaching the sea just south of Rbi9. Pilgrims coming down the Red Sea enter into irm as their ship passes Rbi9. North of Rbi9 is the reputed burial place of the Prophet's mother, $mina, at al-Abw" [q.v.], now called al-9urayba. From Rbi9 secondary routes ran northwards through the mountains to Medina, providing a more direct but more difficult approach than al-ar al-Suln, which continued to hug the coast. From the port of Mastra an alternate route known as the detour (al-Malaff) turned inland, but the main road did not do so until it reached Badr 0unayn [q.v.], where the Prophet humbled |uray9 on the battle-field. The road from Yanbu#, which has taken the place of al-3r [q.v.] as the principal port for Medina, joins the road from the south at Badr, whence al-ar al-Suln ascends Wd al-afr" towards Medina. In this valley #Abd Allh b. Sa#d of Na3d won a signal victory over Amad sn and his army from Egypt in 1226/1811. Now that an asphalt highway joins Mecca and Medina via 3udda, Rbi9, and Badr, it is easier and faster not to take the short cuts through #Usfn and the mountain passes farther north. The usual course for al-ar al-9ar runs northwards down #A 9t #Ir [see al#a ]. Sometimes it goes through the old oases of 09a and -ufayna on the eastern edge of the arra , and at other times it passes a little to the east of them. The oasis of al-Suwriiyya (modern al-Suwayriiyya), also on the eastern edge of the arra , is even farther off the road. North of the modern mine of Mahd al- | [III:363a] 9ahab, now abandoned, al-ar al-9ar proceeds for a space down another valley named al#A south-east of Medina, which is different from the blessed valley of al-#A west of the city.

The main route from Medina to Na3d forks just after the oasis of al-0ankiyya, one branch continuing eastwards to al-|am and the other heading northwards to 0"il . The main route east from Mecca (Darb al-0i3z) now runs from al-Sayl al-Kabr via al-|#iyya and al-Dawdim to al-Riy', replacing the old pilgrim trail via al-|unuliyya and al-|uway#iyya. The northern section northern section of al-0i3z may be taken as extending to the boundary between Saudi Arabia and Jordan, which stretches from a point south of al-#Aaba [q.v.] over to the range of al-ubay. The occupation by the state of Israel of a position on the Gulf of al#Aaba has made it impossible for pilgrims to follow the old overland route from Sinai via al-#Aaba. Among the small ports are Hal and Man on the Gulf of al-#Aaba, and al-Muwayli, 4ab, al-Wa3h, and Amla3 (orthography uncertain) on the Red Sea. From al-Wa3h tracks cut across the mountains to meet the interior highway at or near al-#Ul. During the past century the heaviest traffic in the northern section of al-0i3z has been over the routes east of al-Sart, first the old Syrian pilgrim road through Tabk and al#Ul [qq.v.] and then the 0i3z Railway [q.v.], which in most places followed the pilgrim road closely. The railway was damaged during the First World War, and reconstruction did not begin until 1383/1964. In the meantime a paved highway was being built north from Medina to Tabk and the Jordan boundary. The highway runs through 9aybar and Taym" [qq.v.], both of which lie a considerable distance east of the pilgrim road and the railway. The southern section southern section of al-0i3z has higher mountains, more rainfall, and much more cultivation than the other two sections. A road parallels the coast from 3udda through the ports of al-L9, al-|unfu9a, and 0aly (0aly Ibn Ya#b [q.v.], once regarded as the southern limit of al-0i3z) to al-|ama, now reckoned as the beginning of Tihmat #Asr. The lower parts of the valleys that flow sea-wards furnish good areas for agriculture. A highland road from al- "if leads to the oasis of al-9urma (not 9urma, as shown on the map in EI 2, i, 708) on the far side of the range of 0a'n (this range is often given as the limit of al-0i3z in these parts). Another highland road links al-"if with Turaba (or Taraba), also on the other side of 0a'n, and a third takes a more direct course to B9a [q.v.], beyond which lies Ta9l9 at the southeastern end of al-0i3z. Eastward-bound travellers use tracks from the border areas to al-Riy', Wd alDawsir, and other places in Na3d. Agriculture in the highlands is concentrated in the eastern oases and along the crest of al-Sart. The map in EI 2, i, 891, shows the principal tribes in al-0i3z in the time of the Prophet, apart from |uray9 of Mecca and al-Aws and al-9azra3 [qq.v.] of Medina. These three have long since disappeared from al-0i3z as important tribal aggregations, their members having been absorbed into the broader Islamic community. The populations of Mecca, Medina, and 3udda have become cosmopolitan; many nonArabs are resident there, and the importance of tribal connexions is dwindling.

Dispersed remnants of |uray9 remain in al-0i3z, mainly as small tribes or clans claiming descent from | [III:363b] the Prophet [see 9arf ], some of whom have adopted the Bedouin way of life. |uray9 has been the mother of sovereigns and dynasties, producing the R9idn Caliphs, #Abd Allh b. al-Zubayr, the Umayyads, the #Abbsids, and a number of lesser ruling houses in al-0i3z itself, foremost among them being the H9imids [q.v.] of Mecca. Other noteworthy survivors of |uray9 in al0i3z are Ban 9ayba [q.v.], the hereditary custodians of the Ka#ba. The map also does not show the three Jewish tribes of the Medina area, Ban |aynu#, Ban |uraya, and Banu 'l-Na'r [qq.v.], all of whom seem to have disappeared leaving no trace. From the central section of al-0i3z the tribes of Sulaym and Hill [q.v.], a branch of Hawzin, took part in the mass migration of Bedouins to Egypt and on to al-Ma9rib in the 5th/11th century. Their place has been occupied by 0arb [q.v.] as the dominant tribe in the area between the two Holy Cities. In the vicinity of Mecca and al-"if four ancient tribes still exist: Hu9ayl [q.v.], from whose ranks sprang an array of poets; 9af [q.v.], the early masters of al-"if; Fahm, the tribe of the brigand bard Ta"abbaa 9arran [q.v.]; and Sa#d b. Bakr, the tribe that is reported to have introduced the young Muammad, the future Prophet, to Bedouin ways while he was in the care of the nurse 0alma [q.v.] (the people of Sa#d are today all settled in villages). Among the more modern tribes are the 3adila along the coast south of 3udda and #Adwn in the mountains south of al-"if, whose chief, #U9mn al-Mu'"if, played a prominent role in the struggle involving the 9arf 9lib of Mecca, the House of Su#d, and Muammad #Al of Egypt during the early 13th/19th century. In the northern section of al-0i3z the tribes of #U9ra and 3u9m [qq.v.] have vanished, their ranges now being occupied in a general way by al-0uwayt [q.v.] towards the coast and Ban #Aiyya in the interior. Muzayna [q.v.], Fazra, and Sa#d Hu9aym have dissolved, but Bal [see EI 1 and EI 2, s.v. al-balaw] and 3uhayna [see EI 2, Suppl.] remain as flourishing entities based respectively on al-Wa3h and Yanbu#. The tribes shown on the map for the southern section of al-0i3z have given way to the great confederations of Zahrn and 9mid [qq.v.] in the highlands and numerous other tribes in their neighbourhood and in Tihma. The borderlands between al-0i3z and Na3d are occupied by elements of the modern tribes of Muayr, #Utayba, the Bum, and Subay# [qq.v.], who have replaced such older tribal groups as 9aafn and Hawzin [qq.v.]. In the far south is the tribe which bears the hoary name of |an [q.v.]. As the history of al-0i3z is intimately bound up with the history of Mecca and Medina and the many other places and the various tribes referred to above, it will not be recounted here. Suffice it to say that since the beginning of time al-0i3z was the official name of an independent polity for less than ten years, under the reign of King al-0usayn b. #Al from 1334/1916 to 1343/1924. Since 1344/1925 the whole of al0i3z has belonged to the domains of the House of Su#d.

Always a poor land inhabited by a people who chafed under the restraints of law and order, al-0i3z is now in many ways coming upon brighter times. The rapaciousness of the tribes, which for centuries made the overland pilgrimage a perilous undertaking, has been curbed and intertribal feuding | [III:364a] brought to an end. Impressive improvements in communications by land, sea, and air; closer ties with the outside world; and developments in education, public health, and other fields are bringing an easier life to many of the inhabitants. The revenue received by the Saudi Arabian government from the petroleum industry has freed al-0i3z from its former dependence on the bounty of Muslims abroad. |URAY , the tribe inhabiting tribe inhabiting Mecca Mecca in the time of Muammad and to which he belonged; the name, which may be a nickname, is mostly (e.g. Ibn Hi9m, 61) said to come from taarru9, a coming together, association; but it is also possible (cf. abar, i, 1104) that it is the diminutive of ir9 , shark, and it could then be a totemic name like Kalb, etc. (A man called | uray9, other than Fihr, is mentioned in Nasab |uray9 , 12.7-9.) The tribe is taken to consist of the descendants of Fihr, and he himself is sometimes spoken of as |uray9; but the name is mostly used only of the tribe, which is reckoned among the northern Arabs (cf. #adnn and al-#arab , 3azrat , vi, p. 544b). The ancestry of Fihr is given as: b. Mlik b. al-Na'r b. Kinna b. 9uzayma b. Mudrika b. Ilys b. Mu'ar b. Nizr b. Ma#add b. #Adnn. The following table shows the main subdivisions of |uray9 (for a fuller version, cf. Watt, Muhammad at Mecca , 7); an asterisk indicates those commonly spoken of as clans in Muammad's time. Tree Fihr, Murib, al-0ri9*, 9lib, Lu"ayy, al-Adram, #$mir*, Mu"ays, 0isl, Nizr, Ka#b, #Ad*, Huay, Murra, #Amr, Sahm*, 3uma*, Taym*, Yaaa, Ma9zm*, Kilb, Zuhra*, |uayy, #Abd (|uayy), #Abd al-Dr*, #Abd Manf, #Abd al-#Uzz, #Abd 9ams*, Nawfal*, H9im*, al-Mualib*, Asad*, Umayya, Rab#a Fihr is said to have been the leader of men of Kinna, 9uzayma and other tribes in fighting to defend the Ka#ba against an attack by Yaman tribes, but the sanctuary and various privileges con- | [V:434b] nected with it remained in the hands of the tribe of 9uz#a. There was probably little, if any permanent settlement at Mecca it this period. The descendants of Fihr lived in scattered groups among their relatives of Kinna. A change came about through |uayy [q.v.], in the sixth generation from Fihr. He collected together the scattered groups of his kinsmen of |uray9 (and this may be the occasion of their receiving the name), and, with help from certain men of Kinna and |u'#a (normally resident in Syria), he wrested the possession of the sanctuary from 9uz#a and became virtual ruler of Mecca. He is also said to have assigned quarters of Mecca to the various groups of |uray9; those in the area round the Ka#ba were known as the | uray9 of the Bi, and included all the descendants of Ka#b b. Lu"ayy, and perhaps some others, while those on the outskirts were the |uray9 of al-awhr. In so far as this assignment of quarters implied permanent settlement, it must have been made possible by the development of trade. Certainly by 600 A.D. the leading men of |uray9 were prosperous merchants who had obtained something like a monopoly of the trade

between the Indian Ocean and East Africa on the one hand and the Mediterranean on the other. They organised caravans which went to the Yaman in winter and to Gaza and Damascus in summer, and they were involved in mining and other activities along these routes. Victory in the war of the Fi3r [q.v.] against Hawzin and 9af brought the routes of the Na3d under |uray9. In particular, they gained control of the trade of the town of al- "if, which for a time had been a rival. Since al-"if [q.v.] was much higher and cooler than Mecca, many |uray9 acquired estates there. They had also stations in other parts of Arabia. The |uray9 | [V:435a] became renowned for their ilm [q.v.] steadiness or absence of hotheadedness, and in practice this meant that they placed business interests first and maintained a measure of unity despite their rivalries. On the death of |uayy (probably in the first half of the 6th century A.D.), his powers and rights passed to #Abd al-Dr, but after a time he was challenged by #Abd Manf, and this led to a division of |uray9 into two rival groups. #Abd al-Dr was supported by Ma9zm, Sahm, 3uma and #Ad, and these were known as the Alf or Confederates. The opposing party, known as the Muayyabn, the Perfumed, consisted of #Abd Manf, Asad, Zuhra, Taym and al-0ri9 b. Fihr. It is impossible to know how long these groups remained effective. There is a reference to Muayyabn in a letter from Muammad to some men of 9uz#a written in 8 A.H. (al-Wid, ed. Marsden Jones, ii, 750). Long before this, however, about 605 A.D. (Munamma, 46), the Muayyabn had been replaced by a new confederacy, the 0ilf al- Fu'l [q.v.] (exact meaning uncertain), which consisted of H9im, al-Mualib, Asad, Zuhra, Taym and perhaps al-0ri9 b. Fihr. The essential change here is that #Abd Manf has split into four parts, of whom two, H9im and al-Mualib, have remained with their former allies, while the other two, #Abd 9ams and Nawfal, have abandoned them. This may mean that #Abd 9ams and Nawfal had become strong through commercial succes. It is clear that #Abd 9ams, though not fully identified with the Alf, had developed close business relations with them. In all the stories of the pre-Islamic period there is admittedly a legendary element, but the main outline of events appears to be roughly correct, even if most of the dating is uncertain. A man did not cease to belong to his clan when he became a Muslim, and in many of the events of Muammad's career, and of the period up to 132/750, the influence of old rivalries or alliances can be seen. As late as the reign of Mu#wiya I, an appeal against injustice was made by al-0usayn b. #Al (of H9im) to the 0ilf al-Fu'l, and was given a whole-hearted response from men of Taym, Zuhra and Asad (Ibn Hi9m, 86 f.). The leaders of the revolt defeated at the Battle of the Camel in 656 were from Taym and Asad. Of course, there were changes in the relative power and wealth of the clans. The clan of #Abd 9ams rose to pre-eminence through the Umayyad dynasty, since Umayya was a son of #Abd 9ams; and the early 9#s and then the #Abbsids represented their old rivals, H9im. Under the first four caliphs and the Umayyads, men of |uray9 played a very important role in the organisation and administration of the empire, and without their skills in these fields the empire would probably not have endured. On the death of Muammad, the Anr wanted one of them to be head of the community of Muslims, but they were persuaded by #Umar to accept Ab Bakr as 9alfa , on the grounds that only a |ura9 could hold together the federation. There are indications that the Anr continued to feel strongly about this point for some decades (cf. Watt in MW , xlii (1952), 161, 164). A ad9 came to be generally

accepted, however, that the imm s are from |uray9 (e.g. Ibn 0anbal, Musnad , iii, 129, 183; iv, 421; cf. T. W. Arnold, The caliphate, Oxford 1924, 47). The Qur"nic verse IX, 100/101 was also quoted in support. When the Sunn theory of the caliphate was formulated, it was generally insisted that the caliph or imm should be from |uray9 (e.g. Ibn Ab Ya#l, abat al- 0anbila , Cairo 1952, i, 26, 34; ii, 21; al- | [V:435b] Mward, Akm , 5; Ibn 9aldn, Muaddima , i, 350-4; Eng. tr., i, 396-401). A few scholars, however (e.g. 4irr b. #Amr), held otherwise, and most of the 9awri3 considered that a pious and upright man of any origin could be imm (al-A9#ar, Maklt, 461 f.; al- Ba9dd, Ul al-dn , Istanbul 1928, 275-7; rtc.). This matter caused a little difficulty in more recent times when the caliphate was claimed by Mu9al emperors and Ottoman sultans, but most #ulam" found a way of circumventing the difficulty (Arnold, Caliphate, 162, 175). The geographer al-Hamdn (d. 334/945) mentions small groups of |uray9 in various parts of Arabia, possibly remnants of trading stations (ed. D. H. Mller, 119, 122, 165, 194, 258); while al- Ya#b (d. 284/897) reports a group of |uray9 near -ayd" in the Lebanon (Buldn, Leiden 1892, 327). At the present day there are many |uray9 living as Bedouin in the neighbourhood of Mecca, while in Mecca itself the key of the Ka#ba is held by a clan of |uray9 called 9ayba. The nisba is |ura9, but this was not much used in the heartlands of the caliphate in the early centuries; if any nisba was used it was mostly that from a clan. After a time, some men seem to have prided themselves in descent from |uray9. The following are some examples: one in #Ir, 7/13th century; six in Egypt and Syria, 8/14th, 9/15th, 11/17th, 12/18th centuries; two in North Africa, 7/13th and 9/15th centuries (GAL, ii, 110, 111, 112, 449; GAL S, i, 298 foot, 537, No. 20, 609; ii, 58; Ibn 9aldn, Eng. tr., iii, 126n.). In Pakistan, etc. it is common at the present time in the form Qoreshi. NISBA (a.), the adjective of relation adjective of relation formed by the addition to a noun of the suffix -iyy un in the masc. sing., iyyat un in the fem. sing., -iyyna in the masc. pl. and -iyyt un in the fem. pl. As a result of the increasingly frequent omission of the tanwn, the long syllable of the masc. sing., henceforward in the final position, is abbreviated to *-iy, and subsequently this diphthong is reduced to the vowel -, transliterated thus but further abbreviated to -, in pronunciation. A different, no longer productive, nisba formation is the pattern fa#li n/al-fa#l, fem. fa#liya tun: yamni n, from al-Yaman, sha"mi n, from al9a"m, tahmi n, from Tihma tu. 1. In Arabic morphology In general, the formation of these adjectives is a simple matter, the suffixation taking place directly without modification of the vocalisation or consonantal structure of the nouns to which it is applied: 9ams sun, 9ams solar; amarmoon, amar lunar; Mir Egypt, MirEgyptian, etc. It should be noted, however, that in certain cases alterations occur for which the grammarians have been at pains to codify rules. Only the most frequent modifications will be cited here: omission of the t" marba: Bara; transformation of the final - ( or ) into -aw-: duny world, dunyaw material, etc.; ma#n sense, etc., ma#naw semantic, etc., even after omission of the final -t" marba: nawt nucleus, nawaw nuclear; similarly the feminine | [VIII:54a] ending -" u is transformed into -w: ar"desert, arw belonging to the desert. There is a tendency to amplify short words by reinstating (or adding) a

third radical (w or y): ab father, abaw paternal, a9 brother, a9aw fraternal, dam blood, damaw sanguine, etc.; an h also appears sometimes: 9afat un lip, 9afaw/9afah labial. A w is even substituted for y in araw (instead of *ary) rustic, from arya village. The internal vocalisation is modified in a number of nisbas formed from proper nouns of the pattern R1aR2R3, R1aR2R3a, R1uR2ayR3 and R1uR2ayR3a: Balaw, from Baly, Madan, from al-Madna(but also Madn), |ura9, from uray9 , and Muzan, from Muzayna . The two forms with or without -- > -a- also exist as a means of avoiding confusion: 3azar, from al-3azra Mesopotamia, but 3azr insular, from 3azra island. Since the Middle Ages, but especially in modern times, the nisba in the feminine has served to create a host of abstract nouns, apparently to be formed at will according to requirements: insn man, insniyya humanity; ta#br expression, ta#briyya expressiveness. There is also recourse to the intensive suffix -n: nafs soul, nafs psychological, nafsn psychic, for example. Finally, certain particles and pronouns are used to support relative adjectives and abstract nouns: kayfa how, kayf qualitative, etc., kayfiyya modality, etc.; kam how much, kamm quantitative, kammiyya quantity; huwa he, huwiyya identity; an I, anniyya egotism. In theory, a relative adjective is never formed from a plural (l yunsab #al 3am#) but even in the earliest times this rule enunciated by the grammarians was already being circumvented: A#rb Bedouins, A#rb bedouin; Ba"imarshes in the vicinity of Bara, Ba"i, etc. Since mediaeval times, usages of this type have proliferated, especially for the formation of nouns of profession: kitb book, pl. kutub, kutub bookseller; alongside fara' specialist in far"i' [q.v.], the form far"i' is also encountered. In certain cases, the plural appears to be artificial: ma9zan [q.v.] government of Morocco, has no plural in this sense, but ma9zin > m9zni, pl. m9zniyya, denotes a horseman paid by the state; similarly, kafta skewers (no pl.) gives kaf"it seller of skewers, etc. Finally, it should be noted that in names such as 9aw, the suffix is not that of the nisba , but the personal pronominal affix of the first person. (Ed.) Ed(s). 2. In Arabic nomenclature In nomenclature, the nisba or noun of relation is one of the components of the mediaeval Arabic proper name. Its function is to express the relation of the individual to a group, a person, a place, a concept or a thing. It is most often preceded by the definite article al-. Numerous nisbas are employed in the contemporary period in the function of family names. In general, the individual who is the subject of a reference in a mediaeval Arabic biographical register possesses among the various elements of his namealong with ism , kunya , laab [q.vv.], professional designationsone or more nisbas which testify to inherited or acquired characteristics, to his path through life, geographical as well as

intellectual, to his religious opinions and to the links that he has with his contemporaries. Inherited, the nisba relates the in- | [VIII:54b] dividual to a group, such as tribe, tribal subdivision, dynasty, family, eponymous ancestor, etc.; to a place, such as a country, region, city, village, quarter, street, etc.; or even to a nickname or a professional designation handed down by his ancestors. Acquired, the nisba takes into account the activity of the person: it originates with the names of places in which he has been resident, those of persons with whom he has established favourable links, the ideas which he has defended and his beliefs. Alternatively, the nisba may refer to quoted remarks or to a physical peculiarity. The following are examples of nisbas which denote the connection to a tribe: al-Kind of the tribe of Kinda; to an ancestor: al0usaynthe descendent of al-0usayn; to a place: al-Dima9the Damascene; to a school of thought: al-Mlik the disciple of the Mlik legal school; to an event: alBadr he who took part in the battle of Badr. There are also examples of nisbas which are rare, if not unique, and are analogous to nicknames; nisbas which denote a connection with a text: a person bears the nisba al-#Antar because he has copied the Srat #Antar (F. Rosenthal, A history of Muslim historiography 2, Leiden 1968, 47); connection with a poetical work: one who knew by heart the Mamt of al-0arr is called al-Mamt (G. Gabrieli, Il nome proprio arabo musulmane, in Onomasticon arabicum, introduzione e fonti, Rome 1914, 205). Nisbas derived from professional designations should be considered separately, in that their termination in - appears to be optional: the cotton trader is called, apparently arbitrarily, al-kan or al-kan. Other professional designations appear only with the - termination, such as al-aydaln, the chemist. Role and limits of the nisba In the earliest Arabic inscriptions, written in Sabaic script, the term 9he of ... was used to signify the relationship of the member of a tribe to his group (see Ch. Robin, Les plus anciens monuments de la langue arabe, dans l'Arabie antique de Karib"l Mahomet. Nouvelles donnes sur l'histoire des Arabes grce aux inscriptions , in REMM, lxi, 114-15). Subsequently, the nisba had the function of indicating to which tribe an individual belonged, either through his origins (ar an) or through links of clientage, for example in the capacity of mawl [q.v.]. This tribal nisba implicitly contains the genealogy of the tribe. Having in one's name an element such as al-Kind signifies belonging to the tribe of the Ban Kinda, with its eponymous ancestor, its achievements, its history and its territory which forms a part of the dr al-Islm [q.v.]. It is also to the dr al-Islm that the nisbas refer which are acquired by individuals on the basis of geographical names. It may in fact be stated that the names listed by the biographers do not contains nisbas formed on the basis of the names of places which do not belong to the dr al-Islm . If an individual changes his abode, like the scholars who are identified by the sources as having travelled in search of knowledge, henceforward his nisbas, formed on the basis of the names of places in which he has resided, may be added to his name (a citizen of Damascus who goes to Ba9dd will be called alBa9dd the Ba9dd on his return; while in Ba9dd he would be known by the name of al-Dima9, the Damascene. On his death, a biographer could preserve in the wording of the name of this person both these nisbas: al-Dima9 (with the added detail: al-Dima9al-al, originally from Damascus), al-Ba9dd. But if he leaves the dr al-Islm , to travel for example to China (al--n), India (Bild al-Hind) or to

Asia Minor (al-Rm), countries which belong to the dr al-arb [q.v.], | [VIII:55a] he will not bear the nisbas al--n, al-Hind or al-Rm except in cases where these are employed as nicknames (see Ibn al-A9r, al-Lubb, ii, 64: he is called al-nbecause he has returned from China and he spends his time copying Chinese characters). The individuals recorded in the biographical sources with the nisbas al-n, al-Hind or al-Rm are natives of these countries; they are not, as a general rule, travellers who have become long-term residents in these countries, for in such cases the biographer would have described them as nazl, followed by the name of the place in question. In the context of the dr al-Islm , two further aspects of the process of formation of nisbas should be noted: (a) Within the confines of the dr al-Islm , there are some quasi-mythical regions such as 9 ursn, the cradle of -fism. The nisba 9ursn is found in the names of scholars who are not natives of this region, who have not even visited it, but who seek to ally themselves with -f masters, claiming a spiritual heritage emanating supposedly from 9ursn (see Les Cent et une Nuits, tr. M. Gaudefroy-Demombynes, Paris 1911, 3; J. Sublet, Le voile, 169, with a further example: the nisba al-|ayrawn, which could represent the Far West). (b) In the spiritual centre of this dr al-Islm are the holy cities of Islam, Mecca and Medina, the names of which can only be used in the form of a nisba in specific circumstances. Performing the Pilgrimage does not confer the right to call oneself Makk or Madan. One who resides as a guest-scholar in a mosque or an educational establishment is entitled to the epithet mu3wir[q.v.] or 3rAllh . Only those who are natives or established citizens of these places may use these nisbas which, furthermore, have become (without the article, such as Makk and Madan) what are known as proper nouns, ism #alam [q.v.], borne primarily, so it seems, by Sunns living in a 9# milieu who are anxious to affirm their orthodoxy (see Sublet, Le voile, 99102, 170-1). Also worthy of note are isms in the form of a nisba without the article, such as Bal9 and Br (cf. Ibn al-A9r, al-Lubb, i, 140, 161). In the Mamlk period, nisbas have a specific role in the composition of the names of the Mamlk slaves who, originally, have only an ism . They acquire a nisba formed on the basis of the name of the merchant who has imported them (for example, Azdamur alMu3r, see Ibn al-Dawdr, Kanz al-durar (Die Chronik des Ibn al-Dawdr), ed. Munajjid-Roemer-Haarmann, Cairo 1960 ff., ix, 71). When circumstances require it, the addition of one or more nisbas deriving from the name of the master who gives them their freedom is possible (the sultan Baybars I, for example, bore the nisbas al--li al-Na3m, which derived from the name of his master (al-Malik) al--li Na3m (alDn Ayyb). In Ayalon, Names, titles and nisbas of the Mamluks, in IOS , v (1975), 189-231, there is a list of these nisbas which were to be replaced, in the Circassian period, by the expression min followed by the name of the master (for example, mnby min |naw). The feminine nisba

The nisbas of women whose names are recorded in the mediaeval biographical sources are masculine or feminine, the two forms being capable of co-existing in the same name, according to whether the biographer considers them as forming part of the patrilineal genealogy or as elements of the woman's name. The order in which he writes the elements of the name, and in particular the kunya , seems to have a bearing on the gender of the nisba or nisbas. For ex- | [VIII:55b] ample, where the kunya is placed at the beginning of the name, as in Umm al-9ayr wa-tusamm Sa#da bint Muammad b. 0asan al-abar al-0usayn al-Makk, the nisbas are in the masculine form, being a part of the patrilineal genealogy. In the alternative formula, the kunya is placed after the genealogy and before the nisbas, as in Sa#da bint Muammad b. 0asan Umm al-9ayr al-abariyya al-0usayniyya al-Makkiyya; the nisbas placed after the kunya composed with Umm are in the feminine (see especially the volume devoted to female biographies by al-Sa9w, al-4aw" al-lmi# li-ahl al-arn al-tsi#, Cairo 1934, xii). These feminine nisbas are seldom likely to supply information regarding the places visited by the women; some women performed the Pilgrimage, but they travelled far less than men, and if they were scholars, men tended to travel to them to receive or convey ad9s and to study texts under their supervision. Children generally inherit the nisbas of their father, very rarely those of their mother. If sons or daughters are mentioned in the text of an article devoted to their mother, they are currently designated by their ism followed by the nisba most often used to designate their father or their father's family; for example: #$"i9a bint al-0arr... wa-knat Umm Amad al-0i3z (see Sublet, Le voile, 117). Composite (murakkab) nisbas Derived from composite names, of persons and of places in particular, these nisbas can have two forms: (a) A contracted form, e. g. the nisba #Ab9am corresponds to the name #Abd 9ams, #Abdal to #Abd Allh, Maras to Imru" al-|ays, Draun to the place-name Dar al-|un, Bbar to Bb alBara and Ras#n to Ra"s al-#Ayn. (b) A simple form derived from one of the two elements of the name, e.g. the nisba Mualib corresponds to #Abd al-Mualib, Bakr to Ab Bakr, Zubayr to Ibn alZubayr and Fa9r to Fa9r al-Dn. On the other hand, certain nisbas are formed on the basis of several names. In the Mu#3am al-buldn, Beirut 1979, i, 456, Ytgives the place name Ba9da9zarand. This is a fictitious name derived from a composite nisba , al-Ba9da9zarand, borne by a single individual whose origin it describes: his father was Ba9dd, his mother 9azariyya and he was born in Samarand. Two other examples given by G. Gabrieli, Il nome proprio, 20: al-abar9az is a composite of abar (of abaristn) and 9wrazm (of 9wrazm); 9af#anat is a composite of 9fi# and 0anaf, denoting one who was a 9fi# and subsequently became a 0anaf. A particular case: the fictitious nisba

Al-Suy mentions among the ten types of ansb ( nasab or genealogy and nisba ) which he describes (al-Muzhir, ii, 444-7): man nusiba il 'smihi wa 'smi abh, giving the example of the name Numayr b. Ab Numayr al-Numayr. The nasab is b. Ab Numayr, literally, son of the father of al-Numayr and the nisba al-Numayr. This is one of the formulas used to give an identity to a person born of an unknown father (the supposed father is sometimes given the name of monetary units such as Dnr or Dirham) or to an individual without a genealogy, a slave, for example. The nisba alNumayr is likewise derived from the ism ; it appears with the name of the father as a repetition of this ism . The nisba in the sources The average number of nisbas borne by an individual (scholar, man of science, soldier or prince) whose biography is recorded in the mediaeval Arab sources is five. But this does not apply to the naming of eminent persons, for whom the biographer supplies | [VIII:56a] only one or two nisbas, which often form part of the name by which the individual is best known (Sublet, Le voile, 104-7). The fragile distinction between nisba and nickname is apparent here, as in the works devoted to ansb. The latter in fact combine not only the nisbas (pl. nisab) a part of which refers to genealogy ( nasab , pl. ansb) and to the eponymous ancestor, but also laabs (nicknames) and professional designations. A specific form of biographical literature is devoted to homographic ansb. The authors experiment with possible readings of the various consonantal patterns with their vocalisations and they determine the identity of those who bear these nisbas, these laabs and these professional designations, with the object of avoiding confusion between individuals, and in certain instances the authors of these erudite works have other objectives in mind, as in the case of Ibn Mkl [q.v.]. (Jacqueline Sublet) Sublet, Jacqueline 3. In Persian and Turkish In Persian, the suffix - (MP-k) is used to form relative adjectives, but with -g/3 after the silent h" at the end of words: (a) from places, e.g. Ifahn, Dihlaw, Swa3; some apparently irregular ones go back to earlier forms of place names, e.g. Rz < Rayy, Sagz < Si3istn/Sstn. (b) from concrete nouns to form adjectives indicating function or craft, e.g. 9nag domestic < 9na, al#a3 garrison soldier < al#a , 9ikr hunter, pertaining to hunting < 9ikr. In Turkish Turkish, the suffix -li in its various realisations is used for relative adjectives of place, e.g. Izmirli, |onyal, Merzifonlu, skbl, and3i/i in its various realisations for adjectives denoting functions, professions, crafts, etc., e.g. eski3i old clothes dealer, aw3 hunter, mum3u candlemaker, bar3 coppersmith, st3 milk seller. Several of these forms have survived in the colloquial Arabic speech of such lands as Egypt and the Levant, former parts of the Ottoman empire, e.g. posta3 postman, boya3 shoe-cleaner, ahwa3 coffee-house proprietor, servant, sufra3 waiter.

| [VIII:56b] NASAB (a.) connection, pedigree, connection, pedigree, genealogy genealogy designates the most fundamental organising principle of Arab society. It would seem to be an inheritance from the earliest times. Since no convincing cognate has as yet been determined in other Semitic languages, it is not entirely implausible to suggest a prehistoric relationship between the roots s-b-b and n-s-b, unconsciously reflected in the parallelism of nasab and sabab rope, connection in the ad9(Concordance, ii, 388). Genealogy provides the historical validation of kinship and all that it involves. Kinship always dominated group life in human society and to a large extent still does today. In Arabia, the remarkable aspect of genealogy was its connection with tribes rather than individuals. With the coming of Islam and its social imperatives, combined with the enormous increase of people constituting society and the growing reduction of the Bedouin component, it could be expected that the power of kinship would change and diminish. However, kinship continued to remain a most important factor in Muslim society, for reasons such as the enduring determination of nobility with its attending privileges on the basis of tribal descent (and descent from the Prophet and #Al) and, for instance, the survival of pre-Islamic social custom, the strong trend toward heredity in the bureaucracy and in the crafts and trades, or the eventual domination of the scholarly establishment at certain periods by family relationships. This contributed to the continuous cultivation of a genealogical literature. By dint of their work, genealogists (nassba, sing. and coll., also nassb, nsib, pl. nussb) exercised a certain influence which at times made their motives suspect; it is not by chance that an emphatic remark of the Prophet that genealogists could be liars is displayed prominently in the nasab literature (e.g. Goldziher, Muh. St., i, 180, Eng. tr., i, 166; Ibn al-Kalb, 3amhara, ed. 0asan, 17; al-Bal9ur, Ansb, i, 12; Ibn 9aldn, #Ibar, ii, beginning). It appears clear that interest in genealogy somehow found written expression already in pre-Islamic Arabia, in whatever language this may have been and regardless of whether one accepts or does not accept I. Shahid's argumentation in favour of written preIslamic documents dealing with tribal relationships (forthcoming). There can be no doubt that genealogical relations were put into writing in the first century of Islam. Tribal registers for military and tax purposes are reported to have already been commissioned by #Umar [see dwn ]. For the activities of the many dimlyremembered early genealogists such as Da9fal or Ibn 9arya [q.v.], we have no reliable historical information; at any rate, they appear to have found no expression in authentic published works. But literary genealogical writingat first, it seems, dealing mostly with individual tribesis well attested among the earliest Arabic works. In the bibliographies of the historians of the first #Abbsid century (see, for instance, alHay9am b. #Ad, al-Mad"in, Muammad b. 0abb), genealogical works are preponderant. Titles including nasab , pl. ansb, were numerous; unfortunately, practically nothing is preserved in its original form. We are on safe ground with the assumption that genealogy, in pre-Islamic times and when books were published on the subject in Islam, was always a repertory of tribal lore going far beyond simple filiations and containing stories ( a9br ) of all sorts, discussing excellence [see fa'la ] | [VII:967b] and the political and legal subject of tribal rivalries [see ma9lib and muf9ara ], and, above all and most basic, featuring poetry. The so far earliest preserved small monograph by Mu"arri3 al-Sads (d. ca. 200/815-16), edited by -. al-

Muna33id (Cairo 1960), attests to it, and so does the large standard work, which was never surpassed, the 3amharat al-nasab compiled by Hi9m b. al-Kalb (d. 204 or 206/819-21) on the basis of information gathered by his father Muammad b. al-Kalb (see al-kalb ; recent editions by Suhayl Zakkr (Damascus 1983, vol. i only?) and N3 0asan (Beirut 1407/1986); 0asan has also edited Nasab Ma#add wa 'l-Yaman alkabr(Beirut 1408/1988) and plans to edit Yt's Muta'ab of the 3amhara. While the authorship of Nasab Ma#add goes back to Ibn al-Kalb, its identification as the missing part of the 3amhara is doubtful. Scattered quotations of earlier works in the preserved literature hardly allow their reconstruction, as each work's character was governed by the extent to which the various ingredients were represented in it, which holds good also for the subsequent literature (see 9kir's introduction to his edition of al-Zubayr's 3amhara, 4-6). Government records such as those referred to probably contained only barebones genealogy without any further material, but they were not meant to be literary or scholarly efforts. With the 3rd/9th century, the pre-occupation with genealogy was firmly entrenched as part of the literary and historiographical heritage. Apart from family history, it also became a strictly scholarly undertaking as attested, among other things, by the fact that specialised scholars not only wrote on the subject but also copied older works with great care and thus preserved some of them for posterity. Adab works such as Ibn #Abd Rabbih's #Id included a special long chapter on genealogy (which was a subject of much interest in Spain). Genealogy as the prime expression of historical consciousness in Arabia and the source of information for early Muslim history maintained its alliance with historiography. Already al-Zubayr b. Bakkr's 3amharat nasab |uray9 (vol. i, ed. Mamd M. 9kir, Cairo 1381) could be called a work of a9br rather than ansb (Ta"r9 Ba9dd , viii, 469, cf. similarly Ibn al-A9r, Lubb, introd., of alSam#n, Ansb). From near the end of the century, al-Bal9ur's Ansb is indeed more historical/biographical than genealogical. Vice-versa, family history and individual biography tried to present as much genealogical information as possible, and much of 9# historiography would naturally rest in a genealogical framework. Although quite different from genealogy, another term from the same root, nisba , pl. nisab [q.v.], the naming after an ancestor or tribe but also after a locality or profession, was seen as the same thing. #Umar is supposed to have spoken contemptuously of the native inhabitants of #Ir as leading their nasab back to their villages (#Id, iii, 312; for the significance of the passage, see R.P. Mottahedeh, in IJMES , vii [1976], 161-82), but for the lexicographers as, for instance, al-Azhar, Tah9b, xiii, 14-15, the extended understanding of nasab/nisba was taken as a fact, as it was in the basic reference work on nisbas by al-Sam#n. As bibliographical references show, in the 3rd/9th century, if not earlier, attempts began to be made to present the essence of genealogical works in tree form (ta93r, mu9a33ar), see al-Hamdn, Ikll, ed. Lfgren, i, 6, l. 8, and Lfgren, in Studi Orientalistici ... G. Levi Della Vida, Rome 1956, ii, 94. This procedure became frequent in the course of time, see Rosenthal, A history 2, 97-8. Ibn al-Fuwa, for | [VII:968a] instance, describes it in connection with al-Zubayr b. Bakkr's 3amhara, i, 105, 321, 379. Ibn 9aldn, op. cit., ii, 14, praises graphic representation in tree form (with its central #amd column) as providing an immediate understanding of genealogical relationships; the Bl edition reproduces these trees in print.

The Muslim approach to the eternal debate about the greater worth, for society and individual, of either noble descent or personal qualities is succinctly expressed by asab wa-nasab [q.v.]. The two words were originally allied, since noble descent and noble qualities were celebrated in Arabia as inseparable for true glory. In Islam, they tended to split into contrasting concepts, nasab being defined as nobility by parentage and asabas nobility in character and deeds; the lexicographers' arbitrary re-interpretation of a verse by al-Mutalammis (ed. Vollers, no. I, v. 2) as embodying the contrast is typical for the process (al-Azhar, Tah9b, iv, 329a = LA ). Islam distinctly preferred the egalitarian view (and the religious attitude later found firm support in philosophical ethics). In |ur"n, XLIX, 13, the statement declaring the best person to be the one most pious followed upon a recommendation of mutual acquaintance, something later understood as the core of nasab relationship; ignorance of nasab ultimately means exclusion from humanity (e.g. #Id, iii, 312). The reputed radical proponents of equality, the 9u#biyya/Ahl al-taswiya, supposedly argued that human beings differ in worth not through ancestors and asb but through their deeds and character qualities, their personal nobility and highmindedness (#Id, iii, 410, cf. Mottahedeh, loc. cit.). The continuing debate on the topic is marked by speculations such as that of al-Kind's pupil al-Sara9s who, as befits a courtier, favours the view that noble ancestry plus outstanding achievement constitutes double nobility, while lack of personal achievement vitiates ancestral distinction (al-Tawd, Ba"ir, ed. |', ix, 198 ff.), and it is, for instance, illustrated by al-R9ib, Mu'art, Bl 1286-7, i, 208 ff., where it is seen as a matter strictly concerning individuals and is no longer tribally oriented; noble descent has its advantages, but individual worth is decisive and may even compensate for ancestral misdeeds. The importance of blood lines in animals such as horses and racing pigeons was wellknown, and treatises were written on them [see amm and 9ayl]. G. Levi Della Vida's standard edition of Ibn al-Kalb's Nasab al-9ayl(Les livres des chevaux, Leiden 1928) has been followed by more recent printings (Cairo 1948; Beirut 1407/1987); see also Sezgin, GAS, viii, 128, 272. TA"R^ (a.) date, dating, chronology, era, then also annals, history history. | [X:258a] I. Dates and Eras in the Islamic World 1. In the sense of date, dating, etc. i. Etymology. The non-Arabic origin of this word was recognised by the mediaeval philologists, but the often-cited derivation of the participle mu"arra9 dated, from a supposed Persian compound mh-rz month-day, is naturally fanciful. In fact, it clearly belongs to the common Semitic root for moon and month; cf. Akkadian (waru, Sabaic wr, Ethiopic wr, Mehri war9, or, with the usual Northwest Semitic sound-shifts, Hebrew yrea (moon), yra(month), Aramaic yar. The root is strikingly absent in Arabic (where the meaning month is taken over by 9ahr , cognate with the South Arabian term for new crescent) except in the words for date, the consonantism of which indicates that they must have been borrowed from the South, not the North. Of the various Arabic forms, the oldest is perhaps the plural tawr9, which might be a

borrowing of an (unattested) South Arabian word for datings (to particular months), from which the singular ta"r9 (also, and perhaps older, tr9)and the verb arra9a (to date) would be back-formations. The lexicographers also adduce a form tawr9. Ethiopic trk, era, history, chronicle, which has occasionally been cited as the etymon of Arabic ta"r9, is, in fact, borrowed from Arabic. ii. Definitions. Calendars are basically of two types: lunar or non-lunar. Lunar calendars have months based on the cycle of the phases of the moon (the synodic month, ca. 29.53 days). In the Near East the beginning of the month is reckoned from the first visibility of the new crescent in the evening sky, just after sunset, and it continues until the next sighting of a new crescent, generally 29 or 30 evenings later. Provided the length of the synodic month is known, it is not difficult to construct a mathematically calculated calendar with a regulated succession of full (30-day) and defective (29-day) months (the Greeks did this in ancient times), but in fact it is impossible to say exactly when the next crescent will be visible, as this depends, among other things, on the weather. In the long run, it makes no difference whether timekeeping is based on the actual sighting of the crescent or on calculation, since the average number of days per month is in both cases the same, but in the short term there is likely to be discrepancy of a day or so between an empirical and a calculated calendar. Twelve lunar months will total an average of about 354 days and are thus roughly 11 days short of the solar (tropical or sidereal) year. The Muslim calendar ignores this difference with the result that the months move forward fairly rapidly through all the seasons. But most of the nations of the ancient world used luni-solar calendars, where the difference between the lunar and the solar year is compensated by intercalating a thirteenth month every two to three years. (For intercalation Arabic uses the Aramaic loan-word kabsa.) Non-lunar calendars are based on notional months with a fixed number of days and make no attempt to keep pace with the phases of the moon. The oldest, and probably most elegant, calendar of this type is that of the ancient Egyptians, with twelve months of exactly 30 days each and five monthless days (epagomenai), making a vague year of exactly 365 days, with months moving forward very slowly against the solar year (about one day every four years). The | [X:258b] Julian (reformed Roman, or Old Style) calendar had months of a fixed length of 30 or 31 days, apart from February, which had 28 days normally, and 29 every four years. The Gregorian (New Style) calendar, introduced in 1582 (at first only in the Roman Catholic countries) and now used throughout the world, differs from the Julian only by omitting three leap-days every 400 years. Calendars that count years from a fixed date are said to operate with an era, the notional starting-point of the counting being the epoch of that era. Thus the epoch of the Christian era is 1 January A.D. 1. Here, as often, the epoch is not identical with the date at which the era was actually introduced. iii. The Muslim calendar .

For determining their religious festivals and for the dating of official documents, Muslims use a pure lunar calendar, without any intercalation and thus independent of the seasons; but for the particularities of everyday life (agriculture, taxation, etc.) they have always relied on the solar or quasi-solar calendars described in the remaining sections of this article. The months of the Muslim year are: 1. (al-)Muarram 2. -afar 3. Rab# al-Awwal 4. Rab# al-$9ir (or al-9n) 5. 3umd "l-l 6. 3umd "l-$9ira 7. Ra3ab 8. 9a#bn 9. Rama'n 10. 9awwl 11. 9u "l-|a#da 12. 9u "l-0i3dja For the supposed meanings of the names, see below, iv. All schools of law (apart from the Ism#ls) agree that every month begins in principle with the actual local sighting of the new crescent and that consequently the duration of any given month cannot be predicted in advance. The only concession to calculation is the generally admitted norm (enshrined in a ad9) that if, due to adverse weather conditions, the moon has still not been sighted on the thirtieth evening after the beginning of Rama'n, one is still entitled to assume that the fast has ended. In other words, Rama'n cannot last for more than 30 days. This is not necessarily true of the other months, and indeed texts occasionally attest months of as many as 31 days. Only the Ism#ls assigned (and still assign) a fixed length to each of the months and dispense with ocular observation of the moon, but this is rejected by all the other schools [see in detail hill ]. Of course, Muslim scientists were aware of the true length of the lunar cycle, and for the purposes of calendrical and astronomical computations they fixed the duration of the odd months (Muarram, Rab# I, etc.) at 30 days and that of the even months at 29 days, and added an extra day to 9u "l-0i33a eleven times in 30 years to keep the average length of the months in agreement with the true synodic month. They were, however, not in complete agreement as to which years of the 30-year cycle are to be

treated in this way (the various possibilities are played through, from a purely mathematical point of view, in Schram). The years with one extra day are called (as alBrn, Tafhm, section 271, explains) the leap years (kab"is) of the Arabs, not that the Arabs ever actually used or use them, but the authors of astronomical tables need them when they construct tables on the basis | [X:259a] of the years of the Arabs. In other words, the regulated calendar is a convenient mathematical fiction and does not correspond to actual practice. There exist a large number of mediaeval and modern tables for converting Muslim into Julian or Gregorian dates and various computer programs are now available to carry out the same task, and all of these are based, as is inevitable, on this same convention. However, it is important to remember that the real calendar was based on the observation of the moon and that the conversions given in the tables are only approximate. As a rule, the dates mentioned in Islamic sources can only be converted exactly to another calendar if the source gives the day of the week as well, in which case the rough equivalent indicated by the tables can be corrected. When, for example, an author writes that he did something on Friday 8 9u "l-|a#da 437, and our tables equate this date with Saturday 17 May 1046 Julian, then we must assume that the author observed the new crescent one day earlier than the date given in our tables for the beginning of the month and correct the conversion to Friday 16 May. In cases like this, where the day of the week is indicated, it can be observed that the conversions given in the tables are likely to be wrong (rarely by more than one day in either direction) about half of the time, and modern historians should correct them tacitly (there is no need to call attention to the fact). Where the day of the week is not indicated we must normally content ourselves with a rough conversion. Since the month begins with the sighting of the moon in the evening sky the Muslims reckon the civil day to begin at sunset. Thus, in the above-cited example, Friday 8 9u "l-|a#da actually begins at sunset on what we would call Thursday 15 May and continues until sunset on Friday 16 May. Or at least that is how it is in theory. In practice, mediaeval Muslims, like people elsewhere, normally thought of the day as beginning at sunrise, as can be observed from the fact that in the canonical religious texts dealing with the times of prayer the liturgical day is, as a rule, reckoned to begin with the midday prayer (uhr), the first prayer after the rising of the sun, and not with ma9rib, the first prayer after sunset (see mt , where this fact is not, however, specifically mentioned). The days (or nights) of the month can be counted consecutively from the first to the last, but in the second half of the month the Muslims (like the ancient Greeks and Romans) also counted backwards, dating events by the number of nights supposedly remaining in the month. Thus, assuming a complete 30-day month, typical dating formulae might be: 1st: li-9urrati Ra3ab 2nd: li-laylatayni 9alat min Ra3ab 3rd: li-9al9i laylin 9alat (or 9alawna) min Ra3ab (etc.) 15th: li "l-nifi min Ra3ab 16th: li-arba#a #a9rata laylatan baiyat (or bana) min Ra3ab

17th: li-9al9a #a9rata laylatan baiyat (or bana) min Ra3ab 18th: li-9natay #a9rata laylatan baiyat (or bana) min Ra3ab 19th: li-id #a9rata laylatan baiyat (or bana) min Ra3ab 20th: li-#a9ri laylin baiyat (or bana) min Ra3ab 21st: li-tis#i laylin baiyat (or bana) min Ra3ab 22nd: li-9amn laylin baiyat (or bana) min Ra3ab 23rd: li-sab#i laylin baiyat (or bana) min Ra3ab | [X:259b] 24th: li-sitti laylin baiyat (or bana)min Ra3ab 25th: li-9amsi laylin baiyat (or bana) min Ra3ab 26th: li-arba#i laylin baiyat (or bana) min Ra3ab 27th: li-9al9i laylin baiyat (or bana) min Ra3ab 28th: li-laylatayni baiyat min Ra3ab 29th: li-laylatin baiyat (or bana) min Ra3ab 30th: li-sal9i Ra3ab But since it is not actually known in advance how many nights the current month will have, this backward-dating is purely conventional. This is noted quite clearly by mediaeval authors, for example by al--l (Adab al-kuttb, Cairo 1341/1922-3, 183), when he writes that careful people avoid dating in this way because they do not know how many nights remain, owing to the fact that the month can be either defective or complete, i.e. it can have either 29 or 30 days. See also al-|ala9and, -ub ala#9 , vi, 237-8, who says that some authorities forbid backward-dating, while others permit it only on the assumption that it implies the unspoken proviso if the month be complete. In other words, sal9 always means the thirtieth day; if the crescent is sighted after 29 days, then the day when one night remains is followed immediately by the first of the next month. It must be stressed again that with backward-dating, as with forward-dating, a precise conversion of Muslim dates is normally only possible if the day of the week is indicated in the source. Sometimes the month is divided into three segments of (notionally) ten nights each (al#a9r al-uwal, al-wusa and al-u9ar), whereby we must assume that in defective months the last ten nights were really only nine. The month can also be divided into ten segments of (notionally) three nights, each segment having a special name [see layl and nahr ]. For the contrived Ottoman system of dating by fractions, see the articles by Ritter and Dietrich.

Years are counted according to the era of the Hi3ra, which is supposed to have been introduced by the caliph #Umar in A.H. 17/638. The notional epoch of this era (1 Muarram A.H. 1) falls in July A.D. 622, but there is disagreement about the precise date, the Muslim authorities fluctuating between Thursday 15 July and Friday 16 July (see, for example, al-Brn, $9r, 30, 330); the latter date is the point of departure of the printed conversion tables. Some modern scholars have made a great fuss about this question and claim that the uncertainties involved in the conversion of Muslim dates result from the parallel use of a scientific and a popular era beginning on 15 and 16 July respectively, but this is without any foundation. It should be clear that if the beginning of every month depends on the sighting of the crescent, then the question of whether the first month of the era begins on a Thursday or a Friday makes no difference for the conversion of dates. iv. Pre-Islamic and agricultural calendars of the Arabian peninsula. The ancient South Arabian inscriptions reveal the use of a number of local calendars, but for most of these it has not been possible to reconstruct a complete series of months. It does, however, seem that at least some of these calendars followed the luni-solar principle. Thus in |atabn we find mentioned in one text the month brm, in another former brm and in a third latter brm, from which one can deduce that the normal year had a single month called brm, but the intercalated 13-month year two months with this name. For the Sabaean calendar we have at least 13 different month-names, among them a former and a latter ns1wr, but it is not yet pos- | [X:260a] sible to determine their order. On the other hand, the names of 0imyarite months of the period immediately before Islam are not only attested in inscriptions, but were also still known to Yaman authors of the mediaeval period, who list them in Arabicised forms (with variants and for the most part not vocalised, and thus cited below in Arabic script) and give a Julian equivalent for each. Some scholars have suspected that these equivalents are only approximate, but it is certainly possible that the 0imyarites of the monotheistic period did indeed adopt the Christian (Julian) calendar, assigning indigenous names to each of the Julian months; the fact that the inscriptions using these names have not revealed any evidence for intercalary months might be seen to support the latter hypothesis. One of these names (Mabkar for May) has survived in Yemen until today, though for the other Julian/Gregorian months the Latin or Syro-Macedonian names (see below, v) are now used. We give here the spellings found in the 0imyarite inscriptions (transliteration according to the system used by Sabaicists) with a specimen of the Arabicised forms and with the Julian equivalents according to the mediaeval texts; the year began (as Robin has demonstrated) with April:

In early South Arabian texts, years are not counted according to an era, but rather each year bears the name of a specifically appointed official (eponym); but from the later periods several different eras are attested, the best known of which is the so-called 0imyarite era, with a nominal epoch (according to the short chronology proposed by present author) in the spring of 110 B.C., though some have favoured a long chronology with an epoch in 115 B.C.

The pre-Islamic calendar in central Arabia, and specifically at Mecca, is not attested epigraphically, but is discussed in some detail by Muslim authors of the #Abbsid period, whose testimony, however, requires critical scrutiny. These state that the ancient Arabs used the same month-names as the Muslims, though they also record special 3hil names (see al-Brn, $9r, 61-2; an entirely different list is given by alMas#d, Mur3, iii, 423 = 1311). The philologists proposed etymologies for all of these, but it must be said that, apart from obvious cases like Muarram and 9u "l0i33a, the meanings of most of them can hardly be regarded as certain. For example, if 3umd really comes from 3amd and Rama'n from rama', then one might ask why only two months intervene between the time of freezing and of burning, and indeed the whole question of whether the central Arabian months were ever fixed to particular seasons is a vexed one. The Muslim tradition is unanimous in stating that the Arabs of the 0i3z distinguished between permitted (all) and forbidden ( arm ) months, i.e. months during which fighting is or is not permitted, the forbidden months being Ra3ab and then three months around the time of the pilgrimage (9u | [X:260b] l-|a#da, 9u "l0i33a and Muarram). A similar (if perhaps not identical) practice is attested for North-Eastern Arabia by Procopius (De bello persico ii, 16, 18) when he writes that the Byzantines were confident that the La9mid al-Mun9ir would not attack in the summer of A.D. 541 owing to the fact that the Arabs respected an armistice of two months at the time of the summer solstice; the Muslim historians, however, do not link the forbidden months to particular seasons. |ur"n IX, 36 says that the months are twelve, of which four are forbidden, and the next verse states that postponement ( nas" ) is an increase in unbelief, in which are led astray those who disbelieved, in that they declare it permissible one year, and declare it forbidden one year, so as to equalise the number of what God has forbidden, and so they declare permissible what God has forbidden. Muslim authors have disagreed over the interpretation of this difficult verse. Some maintain that nas" was a procedure by which an official (the nsi") connected with the Ka#ba cult at Mecca altered the distribution of forbidden and permitted months within a given year (or within two successive years), but say nothing to imply that this involved actual manipulation of the calendar. Others claim that the postponement of the forbidden months was the result of the fact that the pagan Arabs intercalated a thirteenth month every two years; nas" is thus in effect the old Arabic word for intercalation (kabsa), a practice which was abolished with the revelation of the above-cited verse. The latter interpretation has generally been accepted by Muslim astronomers (first, it seems, by the astrologer and charlatan Ab Ma#9ar alBal9) and has been favoured by modern scholars since Moberg (who summarises his views above, s.v. nas" ), but it is doubtful whether it reflects anything more than learned speculation by the #Abbsid authors. The former (non-calendrical) interpretation was accepted, and the evidence for intercalation rejected, with arguments that still seem largely valid, by Mahmoud Effendi and Sprenger, and it is in our opinion supported by an early Sabaic inscription from Haram (CIH 547), not precisely datable, but surely from well before the Christian era. Here the authors offer their excuses to the god 0lfn for the fact that they had not performed a certain ritual in the month mwbm, when war had forced them to flee their country, but postponed (ns1"w) it until the month #tr. As a result of this impiety, the god withheld the waters during the winter and summer growing seasons, whereupon the authors promised not to repeat their transgression in future. It is quite clear from the context that in Haram, at least, the verb ns1" has nothing to do with intercalation, but only with the moving of

cultic occurrences within the calendar itself. The rather striking similarity between the religious conceptions in this ancient inscription and in |ur"n IX, 37 (postponement is in both cases something of which the deity disapproves) makes it seem likely that this is also the meaning of |ur"nic nas" . For the observation of the passing of the seasons the Arabs used (if the evidence of preIslamic poetry is regarded) not the months of their calendar, but rather the system of lunar mansions [see anw"" and manzil]. This consists of dividing the ecliptic into 28 equal segments corresponding roughly to the location of the moon on each successive night of the sidereal (not the synodic) month. The observation of the positions of the stars belonging to each lunar mansion relative to the horizon at sunset or sunrise is a simple and accurate method of charting the passage | [X:261a] of the year, and it continues to form the principal basis for a large number of traditional agricultural calendars in the Arabian peninsula. v. The Julian calendar in the Islamic world. The Julian calendar was introduced throughout the Roman Empire and continued to be the basis for practical timekeeping in most parts of the Islamic world. From a chronological point of view, the principal difference between the various forms of this calendar concerns the point where the year begins. In the Ma9rib and al-Andalus, as in Western Europe, the year commences with January, and in modern times this form has, under Western influence, become generally accepted in the Muslim countries. The Melkite Christians (mainly in Palestine and Egypt) followed the Byzantines in beginning the year in September, while the Jacobites and Nestorians followed the Macedonian custom of beginning the year in October. Moreover, instead of borrowing the Latin names, the last-mentioned communities retained most of the month-names of the ancient Babylonian luni-solar calendar (still used, with some modification, as the religious calendar of the Jews), but they redefined these to make each one correspond to a Julian month. Thus, while Babylonian Nisannu (Jewish Nsn) is the month beginning with the sighting of the crescent at about the time of the spring equinox, in the Christianised calendar it has become equated with April. These Syro-Macedonian names are still used (now for the Gregorian months) in Western Asia, but in North Africa various versions of the Latin names have been employed. We give here those found in mediaeval texts from Spain and the Ma9rib, which apparently go back to an archaic Romance dialect. The Ottoman and modern Turkish forms derive in part from the Syro-Macedonian, in part from the Latin names.

The Eastern Christian counted years according to the Seleucid era, in the East generally, but wrongly, called the era of Alexander (Arabic: ta"r9 9i "l-|arnayn), with an epoch of 1 October 312 B.C. In Egypt and North Africa, years are often reckoned according to the Byzantine fifteen-year cycle of indictions, the first indiction beginning in (for example) A.D. 612, 627, 642, etc. The Byzantine indictions begin on 1 September, but the cycle is used also in connection with other methods of determining the New Year. In al-Andalus we find the Spanish era (ta"r9 al-ufr; for the name, see afar ), with its epoch on 1 January 38 B.C. It should be mentioned that the so-called Christian era of Western Europe was unknown both to Christians and Muslims in the mediaeval Near East.

| [X:261b] vi. The Coptic calendar. The ancient Egyptian vague year (see above, ii) was known to Muslim astronomers from its use in Ptolemys Almagest and they not rarely employed this beautifully simple calendar for their own calculations, but for everyday timekeeping it had been replaced in Egypt, long before Islam, by the so-called Coptic calendar, which is exactly identical with that of the ancient Egyptians apart from the fact that it adds a sixth epagomene every four years (more precisely, at the end of a Coptic year immediately preceding a Julian leap year), with the result that the Coptic calendar is permanently synchronised with the Julian. Thus al-Brn distinguishes between the intercalated (makbsa) year of his Egyptian contemporaries and the non-intercalated year that he knew from the Almagest . The Arabic spellings of the Coptic month-names vary somewhat, but are in any case fairly accurate representations of the names used in the -a#d dialect of Coptic; the latter are transliterated in the first column of the following table, while the last column gives the Julian equivalent of the first day of each Coptic month in a common year:

If a Coptic year begins in the August preceding a Julian leap year, the extra epagomene immediately before the Coptic New Year means that the first seven months all begin one day later than the dates given in the table (i.e. 30 August, etc.); from Parmoute onwards the Julian equivalents are as given. The Coptic Christians express years according to the era of Diocletian, which they call the era of the martyrs, with its epoch on 29 August A.D. 284. vii. In Persia. Islamic Persia inherited from the Ssnids a calendar which is functionally identical with that of the ancient Egyptians: it had a vague year of 365 days, divided between 12 months of 30 days each plus five epagomenai, in Persian generally known as the five G9s (pan3 gh), or as the stolen (mustaraa, duzd9a) days. In the early Islamic period these stood after the eighth month, but in A.D. 1007 the Zoroastrians in Western Persia moved the epagomenai to the end of the last month, a reform which was adopted, not at once but eventually, in Eastern Persia as well. The early New Persian forms of the month names are:

| [X:262a] Each of the thirty days of the month had its own special name, but these are not much used by Muslim authors, who generally count the days instead. The five stolen days also have, at least among the Zoroastrians, names of their own. The Zoroastrians employ this calendar to determine their religious feasts, several of which continued to be celebrated by Persian Muslims, in particular Nawrz or Nawrz [q.v.] (the Persian New Year, on 1 Farwardn), which in late Ssnid times fell at about the beginning of

summer, but by the end of the 10th century had moved forward to about the time of the spring equinox. In the early Islamic period, other festivals such as Mihragn [q.v.] (16 Mihr) were much observed, but later these fell into disuse among Muslims. Different forms of the Zoroastrian calendar were current in early Islamic times in Sogdiana (around Bu9r and Samarand), 9wrazm, Sstn and u9ristn; each of these had not only its own month-names, but the first two, at least, began their year five days later than the Persians (i.e. their New Years day fell on the Persian 6 Farwardn) and had the epagomenai after the twelfth month (not the eighth). In Ssnid times, years were counted from Nawrz of the accession year of the reigning monarch; in Islamic times the Zoroastrians either continued to count the regnal years of Yazdigird III, the last Ssnid (era of Yazdigird, begins 16 June 632), or from the year of his death (post-Yazdigird era, or era of the Magians, begins 11 June 652). Both of these eras are used on the Arabo-Sasanian coins of the early Umayyad period, and the post-Yazdigird era was employed by the Zoroastrian rulers in abaristn, but the Zoroastrians of the present day have retained only the era of Yazdigird. The Persian year, like the Egyptian, moves forwards against the solar year by about one day every four years, its first day corresponding always to 1 Choiak of the old Egyptian calendar. Muslim sources, from al-Mas#d onwards, claim that in pre-Islamic times the Persians used to correct their calendar by intercalating one whole month (30 days) every 120 (or 116) years, but there is no evidence that such a correction was ever actually carried out and the present author has attempted to demonstrate that this supposed intercalation is a myth invented by the Zoroastrian clergy during the Islamic period. In any event, it is uncontested that no intercalations took place after the Arab conquest. Since the beginning of the tax-year (iftit al-9ar3) in Persia still fell on 1 Farwardn, the problem arose that the landholders were soon being asked to pay their taxes before their fruits had ripened and various attempts were made to rectify the situation. Thus the caliph al-Mutawakkil proposed delaying Nawrz by 57 days, but the reform does not seem to have been carried out. Half a century later, al-Mu#ta'id ordered that from the year 282/895 the New Year was to be delayed by 60 days until 1 9urd9 (corresponding then to 11 June) and, moreover, that the Persian calendar was to be aligned permanently to the Julian by inserting a sixth epagomene every four years. Historians mention the celebration of al-Mu#ta'ids Nawrz for a century or so after the reform, but otherwise this calendar had little resonance. A similar, but more memorable reform was the 3all [q.v.] calendar instituted by the Sal3id 3all alDn Malik 9h in 471/1079: Nawrz was fixed to the vernal equinox (defined as | [X:262b] the point when the sun enters Aries), then 15 March Julian, which date is the epoch of the 3all era. The months had 30 days each and retained their traditional Persian names (some sources cite special fancy names for the months and days, but it is doubtful whether these were ever used), with Farwardn beginning at the spring equinox, and five epagomenai at the end of the year (after Isfandrmu9), but a sixth epagomene was added if the sun had not yet entered Aries by noon on the 366th day from the previous New Year. Although the 3all era is not often used for datings, the spring Nawrz has been accepted universally by Persian-speaking Muslims and the 3all months formed the basis for agricultural calendars in Persia until the recent past.

Historians of the Mongol period say that 9azan 9n instituted an ^l9n era, but give us no reliable information about its particulars. The modern Persian solar hi3r calendar was introduced in 1925. The year begins at the spring equinox and the months have modernised forms of the traditional Persian names (cited above); the first six months have 31 days each, the next five have 30 and the last (Isfand) has 29 in an ordinary year and 30 in a leap year, giving 365 or 366 days altogether. However, the rule for determining which years are leap years is complicated. The New Years day (1 Farwardn) falls in principle on 20 or 21 March Gregorian, but the discrepancy between the Persian and Gregorian systems of calculating the leap years means that some fluctuation occurs (for a table of leap years, see #Abd Allh, 373-5; Spulers one is wrong). The same calendar was adopted in Af9nistn in 1957, but instead of the Persian month-names the Arabic names for the roughly corresponding zodiacal signs (beginning with 0amal/Aries) are used. In modern Persia (and Af9nistn) years are counted according to solar years since the Hi3ra and can easily be converted to anni domini by adding 621 or (in the last two and a half months) 622. Thus the Persian year 1376 hi3r 9ams begins on 21 March 1997 Gregorian, corresponding to ca. 12 9u "l-|a#da 1417 hi3r amar. For a few years at the end of the reign of Muammad Ri' 9h, this was replaced by an imperial (9han9h) era dating from the accession of Cyrus the Great (559 B.C.), but the hi3r 9ams year has been re-established in the Islamic Republic. Although the Zoroastrian vague year appears to have gone out of circulation among Persian Muslims by around the end of the 5th/11th century, it lived on, rather surprisingly, as the basis for the navigational calendar of the Arab seafarers in the Indian Ocean. This did not use the Persian month-names but simply counted the days (from 1 to 365), beginning at Nawrz. For the moment, the oldest known attestation of this calendar is in the almanac of the Yemeni crown prince #Umar b. Ysuf (later alMalik al-A9raf), which puts Nawrz at 8 January (correct for A.D. 1269-72), and it is found also in Ibn M3id [q.v.], who says (again correctly) that in his time Nawrz fell on a Friday in 9u "l-0i33a 893 (14 November 1488) and in later authors. It seems that this navigational calendar was still in use until the very recent past, but European observers state (correctly?) that the point of departure of the day-count became fixed at the beginning of the sailing season and thus ceased to regress with the Persian New Year. viii. The tax year. In the #Abbsid chancery, tax collection was regulated by the Persian calendar, with tax years begin- | [X:263a] ning at Nawrz, but counted not according to one of the Zoroastrian eras, but by years of the Hi3ra. Since the Islamic year is shorter than the Persian vague year, the counting of taxational (9ar3) years lagged behind the years of the religious calendar and had to be corrected from time to time. For example, in the year of the Hi3ra 241 (May 855 to May 856) Nawrz fell on 21 April, that is to say, almost at the end of the religious year. Consequently, the authorities decided to call the tax year beginning on that date not 241 but 242, thus dropping one year from the taxational calendar; this is called tawl (changing one tax year to another). We have information of several other adjustments of the same type, but also of adjustments that were not carried out when due, and of times when two years had to be omitted from the 9ar3 calendar to make up for an earlier oversight. It is thus clear that the

reconciliation of the tax year with the religious year was not automatic or regular, but sporadic and by specific order of the authorities. In Egypt, the tax year began with the Coptic New Year (29 or 30 August) and similar attempts were made to keep the counting of tax years roughly in line with the years of the Hi3ra. The Ottomans based their financial year on the Julian calendar and at first retained the Byzantine New Year on 1 September, but they moved the beginning of the year to 1 March in 1088/1677. After this time each financial ( mliyye ) year had, in theory, the same number as the hi3r year during which it began, with the omission of one year in every 33, i.e. whenever the 1st of March did not fall within the religious year, the Turkish technical term for the omission being )w)9. Problems, however, arose from the fact that the Ottomans collected taxes according to the Julian year but paid salaries according to the lunar year, which led to difficulties with the bookkeeping whenever )w)9 became necessary. Eventually the system broke down: the mliyye year 1287/1871-2 should have been followed by mliyye 1289/1872-3, but for some reason the year 1288 was not left out, with the result that for the remainder of its history the mliyye calendar was out of step with the religious year. It was officially abolished with the adoption of the Western-style Christian era in 1927. ix. The Turco-Mongolian calendar . Before entering the dr al-Islm , the ancient Turks adopted a form of the Chinese lunisolar calendar, either directly from China or via the Sogdians. Later, this passed from the Uy9ur Turks to the Mongols, who in turn introduced it in their empire in Persia, where it was quite widely used, alongside the Islamic and 3all calendars. The Chinese lunar months begin not with the sighting of the new crescent, as in the Near East, but one or two days earlier with the calculated time of conjunction between the sun and the moon (the true New Moon), the year beginning with the New Moon occurring while the sun is in Aquarius. The year is kept in pace with the solar year by the periodic intercalation of a thirteenth month, which is inserted whenever two New Moons fall during the time that the sun is in one and the same zodiacal sign. The Turkish names for the first and last months derive from Sanskrit or Iranian; the other months have no proper names but are merely counted, as in Chinese (second moon, etc.). Early Turkish forms, subsequently borrowed both into Mongolian and Persian, with their approximate Julian equivalents, are:

| [X:263b] The intercalary month is called 9n ay, a borrowing from Chinese rn. Years were counted according to the twelve-year cycle of Chinese astrology, where each year of the cycle takes the name of an animal. The Mongols translated the Turkish animal names into their own language, in many cases (nos. 3-5, 8-10) using words that are, in fact, borrowed from Turkish (where, in turn, some of the names derive from Iranian or Chinese); in Persian texts the Turkish and the Mongolian names are used interchangeably (with inevitable variations in the Persian spelling); we quote the underlying Turkish and Mongolian forms.

The animal cycle continued to be used in Persia until the beginning of the twentieth century, generally in conjunction with the months of the 3all calendar. x. Muslim India . It would fall outside the scope of this article to discuss the many local calendars and eras used in the Indian subcontinent; for the Muslim scholarly reception of Indian chronological theory (not practice), we refer to the extensive discussion in al-Brns Ta m li "l-Hind. The Muslim rulers in India used the Islamic calendar for official datings, but, like their counterparts further west, employed the local calendars as a basis for tax-collecting, with the ambiguous term fal (seasonal) designating not one, but a variety of indigenous calendars. The Mu9al emperor Akbar [q.v.], as part of his religious innovations, officially replaced the Muslim calendar by his own Ilh era [q.v. in Suppl.], with its epoch at the spring equinox following his accession, i.e. Wednesday 28 Rab# II 963/11 March 1556 Julian, though the new calendar was not actually introduced until the beginning of the 29th year of his reign, when the sun entered Aries (according to the Akbar-nma, tr., iii, 644) on Wednesday 8 Rab# I 992 (11 March Julian = 21 March Gregorian 1584). The year began with the spring equinox, the months (and days) had the same names as the Persian (Zoroastrian and 3all) ones (see above, vii), but there were no epagomenai; instead, the duration of each month was fixed at 29, 30, 31 or 32 days (the details are not quite clear, but it seems that the longer months were in the spring and summer). Also, the years were counted in cycles of twelve, with the same names as the months. Thus Akbars twelfth year is year Isfandrmu9 of the first | [X:264a] cycle, his thirteenth year Farwardn of the second cycle. The later Mu9als abandoned the Ilh calendar, but continued to use datings according to their own regnal years, alongside the Islamic calendar. 2. Era chronology in astronomical handbooks. Practically all z3s [q.v.], mediaeval Islamic astronomical handbooks with tables and explanatory text, include an extensive chapter on chronology. This chapter, usually to be found at the very beginning of the z3 , contains the chronological information needed for the calculation and evaluation of planetary positions and other astronomical data from the whole mediaeval period. Thus various calendars and eras are described and methods for converting dates from one calendar into another explained. For each calendar at least the following are given: a list of the names and lengths of the months, a method to compute from a given date the number of days elapsed since the epoch and vice-versa, and a rule for calculating the week day (mad9al). A set of numerical tables facilitates the calculations involved. In addition, many z3s contain lists of festivals in various calendars, mathematical tables for determining the dates of the movable Christian feasts, and regnal lists of caliphs, kings and emperors. This article describes the aspects of era chronology that are typically found in z3s and explains the conversion of dates from one calendar into another step by step. More general aspects of calendars used in the Islamic world, including their historical development and the names and lengths of the months, are discussed in 1. above.

Depending on the context, the word ta"r9 as it is used in z3s can be translated into English as era, epoch, calendar or date. a. Calendars and eras. Table 1 shows which calendars are described in detail in a number of important z3s. For more information about the authors of these works, the reader is referred to the respective articles in this encyclopaedia or the Dictionary of scientific biography, New York 1970-80. It can be noted that three calendars occur in practically all z3s: the purely lunar Muslim (Arabic or Hi3ra) calendar, the Byzantine ( Rm ) calendar which is essentially equivalent to the Julian, and the Persian Yazdigird calendar with a constant year length of 365 days. The use of most other calendars discussed below is restricted to particular geographical regions or historical periods. Table 2 displays the precise dates of the most important epochs used in z3s as well as the differences in days between these epochs, which are needed for the conversion of dates from one calendar into another as described below. Note that, except in the | [X:265a] | [X:266a] | [X:267a] examples of date conversions, the Hi3ra dates given in this article are the common civil dates ratherthan the astronomical ones applied in most z3s. The names of calendars and epochs used in this article are literal translations of the names occurring in z3s, which in some cases may not be the historically or linguistically most correct ones. Arabic or Hi3ra calendar . Whereas in civil usage the beginnings of the Arabic months are determined by actual observations of the lunar crescent after new moon [see 1. above and ru"yat al-hill], most mediaeval astronomers used the schematic calendar with nineteen ordinary years of 354 days and eleven leap years of 355 days in a thirtyyear cycle. The number of days in the first n years of this cycle is determined by multiplying n by the average year length of 354 11/30 days and rounding the result to the nearest whole number. If for n = 15 the resulting half of a day is truncated, this leads to the set of leap years 2, 5, 7, 10, 13, 16, 18, 21, 24, 26 and 29. If it is rounded upwards, the fifteenth year of each cycle becomes a leap year rather than the sixteenth. The first variant seems to be more common in early z3s, the second one in later Persian z3s. Whereas most Muslim astronomers used the astronomical Hi3ra era based on the mean new moon of Thursday, 15 July 622 A.D., al-Brn and some others adopted the civil epoch determined by the first visibility of the lunar crescent on Friday, 16 July [see hi3ra ]. The early #Abbsid astronomer 0aba9 al-0sib based his rules and tables for the Hi3ra calendar on the actual time of the above-mentioned new moon and the traditional Babylonian value for the length of a lunation, which is slightly different from that implicit in the schematic lunar calendar. Thus he arrived at a highly unusual intercalation scheme, in which the years 3, 6, 9, 11, 14, 17, 19, 22, 25, 28 and 30 in the thirty-year cycle were leap years. Persian or Yazdigird calendar . Because of its constant year length of 365 days, the Persian calendar is very convenient for calculating planetary motions during long periods of time. For this reason, it was adopted as the base calendar in many z3s. The five extra days of the Persian year are usually called andar3a (from the Persian andar gh(n), intermediate times), mustaraa (stolen days) or lawi(appendages).

Various z3s describe the type of intercalation supposedly carried out in the ancient Persian calendar (see 1. above, section vii, and F. de Blois, The Persian calendar , in Iran , xxxiv [1996], 39-54), as a result of which the extra days were inserted after the eighth month, $bn, until approximately the year 397/1007. Starting with K9yr b. Labbn, more and more Muslim astronomers moved the extra days back to the end of the year (cf. Table 1). In all z3s the epoch of the Persian calendar is the beginning of the year of accession of the last Persian king Yazdigird III [see ssnids ]; in only a few cases are the Sogdian and Khwarazmian forms of the Persian calendar described. In most z3s the names of the 30 days of the Persian months and the five extra days are listed. Byzantine (rm) or Syrian calendar. This calendar is essentially the same as the Julian, but counts the years according to the Seleucid era, which is mistakenly named after Alexander the Great [see al-iskandar]. In z3s it usually employs the SyroMacedonian month names listed in 1. above, section v, and has Ti9rn I (October) as the first month of the year. Once every four years a leap day is inserted at the end of 9ub (February), leap years being those Alexander years which leave a remainder equal to three when divided | [X:267b] by four. Most deviations from these basic characteristics were incidental, e.g. al-Battn started the year with Ayll (September), the Latin version of the Z3 of al-9wrazm inserted the leap day at the end of Knn II (December), and in some tables the months were reckoned from $9r (March). Occasionally transliterations of the Latin month names were given. Egyptian or Coptic calendar (ta"r9 al-ib). In some z3s this designation stands for the ancient Egyptian calendar as used by Ptolemy [see balamiys ], which occurs with the Nabonassar (Bu9tanaar), Philip (bylbs, usually unvocalised in the manuscripts) and Antoninus ("ntnns) epochs. Mostly, however, it is used for the Alexandrian or Coptic variant of this calendar introduced by the Roman emperor Augustus ("9ss), which runs parallel to the Julian calendar. The five extra days of the Coptic year (six in a leap year) are called nas or epagomenai ("bw9mn"). As epochs we find the beginning of the year of accession of Augustus and that of Diocletian (dly"nws). In the first case, leap years are those which leave a remainder zero after division by four, in the second case those which leave a remainder three. Al-Battn used a Philip era for the Coptic calendar (Tuesday, 29 August 324 B.C.), which he mistook for the epoch of Ptolemys Handy tables. It may be noted that New Year in the ancient Egyptian calendar coincides with that in the Coptic calendar around the time of introduction of the latter, sc. 25 B.C. However, in a number of z3s we find rules based on the assumption that New Year in both calendars coincides at the time of the Philip epoch, 324 B.C. As a result, the Augustus and Diocletian epochs in these z3s are 75 days later than the dates shown in Table 2. 3all calendar . Whereas Islamic religious life was governed by the purely lunar Hi3ra calendar, for agricultural and taxation purposes a true solar calendar was found to be indispensable. In z3s the most extensive descriptions of such calendars concern the 3all [q.v.] (also called Malik), which was introduced in the late 5th/11th century by the Great Sal3Sultan 3all al-Dawla Malik-9h I [q.v.]. New Year in this calendar is defined as the day (from noon till noon) on which the sun reaches the vernal equinoctial point, to be determined by astronomical calculation rather than by a straightforward intercalation scheme. Here the method varies from z3 to z3

depending on solar theory and parameter values. In most cases, the times of successive vernal equinoxes are computed by repeatedly adding a constant length of the solar year to the time of the equinox of the epoch. A more accurate theory is contained in the 9n Z3 by the 9th/15th-century computational genius al-K9, who took into account the influence of the motion of the solar apogee on the solar equation and hence correctly obtained varying time spans between consecutive equinoxes. Al-K9 constructed his sophisticated table for the computation of vernal equinoxes on the basis of intervals of 33 3all years, practically equal to an integer number of days, and used parabolic interpolation to calculate the equinoxes within these intervals. Independent of the precise method of computing the vernal equinox, in all z3s the 3all year ordinarily has 365 days, every fourth or incidentally fifth year being a leap year of 366 days. In some z3s the leap year following four ordinary years occurs alternately after 25 or 29 years, in others after 29 or 33 years. The beginnings of the 3all months are said to be defined by the entry of the sun into the | [X:268a] zodiacal signs, but in practice the traditional Persian months of 30 days are used, 3all>> being appended to the original names and a sixth extra day being added in a leap year. The epoch of the 3all calendar generally is the vernal equinox of the year 471/1079. Other epochs in use for true solar calendars based on the vernal equinox are the ^l9n era of 9zn 9n [q.v.] mentioned by al-K9 (Tuesday, 12 Ra3ab 701/13 March 1302) and the era of 1ingiz 9n [q.v.] used in the z3 of alSan3ufn, written in Tibet in 1366 (Wednesday, 12 9a#bn 603/14 March 1207). Chinese-Uighur (Turco-Mongolian) calendar . Under the Mongol ^l9n dynasty [q.v.] the so-called Chinese-Uighur calendar was the commonly used one in Persia. It is prominently described in a number of z3s, mostly in Persian, the earliest of which is the ^l9n Z3 by al-s. The calendar is a luni-solar one and is very similar to the Revised Ta Ming li, the calendar of the Chinese Chin Dynasty, which was adopted by 1ingiz 9n after the Mongol conquest of northern China in 1215. With the Ta Ming li, the Chinese-Uighur calendar shares the underlying solar and lunar parameters, the era of the creation, and the general rules for determining the beginnings of years and months and the position of the leap month. Different from official Chinese calendars, in the Uighur calendar the solar and lunar equations, required for the computation of the true new moons, are calculated as parabolic functions. The determination of the lunar equation involves a period relation of ultimately Babylonian origin, which equates nine lunar anomalistic cycles to 248 days. In the Chinese-Uighur calendar, days are counted either from the beginning of the current month or in the Chinese sexagesimal cycle. Years are grouped into three consecutive cycles of 60 years which are called by Chinese words Shang-yan, Chungyan, and Hsia-yan. In the ^l9n Z3, the epoch of the Chinese-Uighur calendar is Thursday, 30 Rab# I 662/ 31 January 1264. (Various technical, philological and historical aspects of the Chinese-Uighur calendar have been discussed in E.S. Kennedy, The Chinese-Uighur calendar as described in the Islamic sources, in Isis, lv [1964], 435-43*; R.P. Mercier, The Greek Persian Syntaxis and the Zj-i Ilkhn, in Archives internationales dhistoire des sciences, xxxiv [1984], 35-60; and C. Melville, The Chinese Uighur animal calendar in Persian historiography of the Mongol period, in Iran , xxxii [1994], 83-98. The complete method of computing Uighur dates as found in the ^l9n Z3 has been laid out in B. van Dalen, E.S.

Kennedy and Mustafa K. Saiyid, The Chinese-Uighur calendar in ss Z3-i ^l9n , in ZGAIW, xi [1997], 111-52.) Others. A number of z3s contain descriptions of the Jewish luni-solar calendar. These range from simple listings of month names and elementary properties to complete sets of rules and tables for the computation of the Tishri new moon, which determines the beginning of the Jewish year. (The material on the Jewish calendar in the unique manuscript copy of the Mumtaan Z3 has been published in J. Vernet, Un antiguo tratado sobre el calendario judo en las Tabulae Probatae, in Sefarad, xiv [1954], 5978, repr. in Vernet, Estudios sobre historia de la ciencia medieval, Barcelona-Bellaterra 1979, 213-32. A treatise on the Jewish calendar by al-9wrazm has been analysed in: E.S. Kennedy, Al-Khwrizmon the Jewish Calendar , in Scripta Mathematica, xxvii [1964], 55-9*.) A couple of z3s mention the calendar of the #Abbsid caliph al-Mu#ta'id [q.v.], a modification of the | [X:268b] Persian calendar which fixes New Year in June and inserts a leap day once every four years. Its epoch is Wednesday, 13 Rab# II 282/11 June 895. Like his monumental al-$thr al-biya (tr. C.E. Sachau, The chronology of ancient nations, London 1879, repr. Frankfurt 1969), al-Brns al-|nn alMas#dcontains a wealth of chronological material, which has not yet been properly investigated (cf. Kennedy, Al-Brns Masudic Canon, in al-Abth, xxiv [1971], 5981*). Besides the common calendars discussed above, al-Brn includes details of the calendar reform by al-Mu#ta'id and of the Jewish, Sogdian/Khwarazmian and Indian calendars (on the last, see Kennedy, S. Engle and J. Wamstad, The Hindu calendar as described in Al-Brns Masudic Canon, in JNES, xxiv [1965], 274-84*). b. Days since the epoch ( al ). In order to convert dates from one calendar into another, it is necessary to calculate the number of days between the epoch and a given date. This is done by summing the days of the completed (tmm) years, the days of the completed months of the current (ni incomplete or munkasir broken) year, and the day of the current month. Conversely, it will be necessary to convert a given number of days since the epoch into the corresponding date, which is done by first computing the number of completed years and then distributing the remaining days over the months of the current year. In many z3s, the number of days in a given number of completed years is called the al (bases). The computation of the al and, conversely, of the number of completed years from a given number of days since the epoch, are discussed below, examples being given in section c., Conversions. Whereas in z3s all calculations are written out in words, here the following modern notation will be used: a denotes the al, c the number of days that have elapsed in the current year (including the current day), d the total number of days that have elapsed since the epoch (including the current day), and C the number of completed years. The following subscripts will be appended to these symbols: H for Hi3ra, A for Alexander (Seleucid era), and Y for Yazdigird. All divisions operate with whole numbers, usually discarding a possible remainder. When the remainder is used it will be denoted by r. Now, for a given number of completed years C, the al a is calculated in the following way:

It can be noted that all three operations are roughly a multiplication by the average year length. The constants 14 and 1 which are added in the calculations for the Arabic and Byzantine calendars are required in order to have the result increase by an extra day precisely when a leap year is encountered. To obtain the variant of the Arabic calendar in which the fifteenth year of each cycle is a leap year instead of the sixteenth, the constant 14 should be replaced by 15. Note that, instead of adding a constant and then discarding the remainder of the division, one may define a rounding rule for the remainder. For instance, equivalent to the above, the Arabic al can be found as (C H x 10631)/30, where a remainder from 1 to 15 (14 if the fifteenth year of each cycle is a leap year) is discarded and a remainder from 16 (15) to 29 is made into an extra day. For the Byzantine cal- | [X:269a] endar, the remainder of the division (C H x 1461)/4 should be discarded if it is equal to 1 or 2 and made into an extra day if it is 3. In most z3s the rules for the calculation of the al are not exact. Often the addition of a constant for the Arabic and Byzantine calendar is disregarded or the rounding rule presented is ambiguous, producing results which may be off by a day. Of the z3s listed in Table 1 only those by 0aba9 al-0sib and al-Brn take extra care to present exact rules. After the al has been determined, the total number of days since the epoch, d, can be computed by adding the lengths of the respective completed months of the current year and the day of the current month. Conversely, in order to calculate the number C of years that have been completed on day d reckoned from the epoch, the following rules can be used:

In this case, the operations are roughly a division by the average year length and constants are subtracted in all three calculations in order to obtain the correct answer also for the first and last day of every year. For the variant of the Arabic calendar in which the fifteenth year of each cycle is a leap year instead of the sixteenth, 16 should be subtracted instead of 15. The remaining days of the current year, c, can be found from the remainder r of the above divisions in the following way:

(the remainders of the divisions by 30 and 4 are discarded). Note that the results are equal to d - a if a denotes the al corresponding to the number of completed years C obtained above. The rules found in z3s for the calculation of the number of completed years and the day of the current year from the number of days since the epoch generally appear to be even less exact than those for the al . c. Conversions. The explanation in z3s of the conversion of dates from one calendar into another varies from convenient shortcuts for particular problems, especially in early z3s, via extensive theoretical expositions as found, for instance, with Ibn Ynus, to brief general

explanations supplementing the rules for using the tables, in particular in later Persian z3s. Using the rules presented in the previous section the conversion of dates can be performed in a general and straightforward way. First calculate the number of days from the epoch of the given calendar to the given date. Then add or substract the number of days between the given epoch and the desired epoch (Table 2) to obtain the number of days from the epoch of the desired calendar to the given date. Finally, transform these into completed years, completed months and the day of the current month in the desired calendar. Example: Convert 24 Rama'n 254 Hi3ra into the corresponding Byzantine date. [1] According to formula (1), the al of 253 completed Arabic years is (253 x 10631 + 14)/30 = 89655. (As usual, the remainder of the division is discarded.) (The same result can be obtained by taking 8 x 10631 days for the eight completed cycles of thirty Arabic years and then adding 13 x 354 for the remaining thirteen completed years plus 5 for the leap days accu- | [X:269b] mulated during these thirteen years.) Since Rama'n is the ninth month of the Arabic year and each two months number 59 days, the given date is the (4 x 59 + 24 =) 260th day of the year 254 Hi3ra. Adding this number to the al , we obtain 89915 for d H, the total number of days since the Hi3ra epoch (including the given day). [2] Now d A, the total number of days from the Alexander epoch up to (and including) the given date, is determined by adding the difference in days between the two epochs as found in Table 2 to d H: d A = 89915 + 340700 = 430615. (We thus assume that the given Hi3ra date is astronomical; in practice, this will have to be verified on the basis of the week day.) [3] The number of completed Byzantine years is obtained according to formula (6): C A = (4 x 430615 - 2)/1461 = 1178 (r = 1400), and the remaining days of the current year 1179 Alexander using formula (9): c A = r/4 + 1 = 351. Because 1179 leaves a remainder three when divided by four, the current year is a leap year and 9ub has 29 days. By noting that the first eleven months of a Byzantine leap year number 336 days (simply add the numbers of days or subtract the length of the last month Ayll from 366), we find that the desired date is the (351 - 336)th day of the twelfth month, i.e. 15 Ayll 1179 Alexander. In early z3s in particular, actual date conversions were not usually carried out by the above general rule, which may involve very large numbers, but by shortcuts based on expressions for the differences between the epochs in years and days. Examples of such expressions are: The (astronomical) Hi3ra epoch falls 932 Byzantine years and 287 days after the Alexander epoch. The Yazdigird epoch occurs 259 days after the beginning of 943 Alexander or 989 days after the beginning of 941 Alexander. Since reckoning from 943 Alexander the 1st, 5th, 9th, years are leap years instead of the 3rd, 7th, 11th, , the constant in formula (6) has to be changed to -4 in order to produce the correct number of completed

Byzantine years since the beginning of 943 Alexander. When the beginning of the year 941 is used, the formula need not be changed. Example: Convert 2 Knn II 1168 Alexander into the corresponding Hi3ra date. [1] The Byzantine New Year just preceding the Hi3ra epoch is that of the year 933 Alexander. According to formula (3), the number of days of the Byzantine years completed since the beginning of this year is (1168 - 933) x 1461 + 1)/4 = 85834. Adding to this the number of days in the current year, 31 + 30 + 31 + 2 = 94, we obtain 85928 days from the beginning of 933 Alexander until the given date. [2] The number of days from the (astronomical) Hi3ra epoch to the given date is 287 less, i.e. 85641. [3] The number of completed Arabic years is now found by formula (4): C H = (30 x 85641 - 15)/10631 = 241 (r = 7144). The days of the current year are then obtained using formula (7): c H = r/30 + 1 =239. These fill up 8 completed months (4 x 59 = 236 days), leaving 3 days in the ninth month. Thus the desired Arabic date is 3 Rama'n 242 Hi3ra. As we have seen above, the rules used for date conversions in z3s are often not exactly formulated. Therefore it is necessary to check the result by means of the week day of the given and the calculated date (see below, d.). Note that, in order to facilitate the conversion of historical dates, many z3s contain extensive regnal lists of caliphs and other rulers indicating the beginning and duration of their reigns. | [X:270a] d. Mad9il. A mad9al (pl. mad9il, literally entrance, translated as feria, nota or signum) is the week day of the first day of a year or month or of a particular date, represented by a number from 1 (Sunday), 2 (Monday), till 7 (Saturday). In some z3s, the number is given a separate name, #alma (indicator). Most z3s contain both rules and tables for the calculation of mad9il, which, as we have seen, are particularly important for checking the results of date conversions. Most simply, the mad9al of a given year is calculated by adding the al of the completed years C to the mad9al of the epoch (cf. Table 2) and casting off multiples of seven. For example, for the Hi3ra year 254 we add the al of the completed years found above, 89655, to 5 (Thursday), the mad9al of the astronomical epoch. Discarding multiples of seven we obtain 4, signifying that 1 Muarram 254 Hi3ra was a Wednesday. For the Persian and Byzantine calendars we find direct methods for calculating the mad9al based on the fact that the number of days of an (ordinary) year is 52 x 7 + 1. Thus the mad9al of a Persian year is simply found as (3 + C Y) mod 7 (m mod n denotes the remainder of the division m/n; note that 3 is the mad9al of the Yazdigird epoch). Similarly, the mad9al of a Byzantine year is obtained as (2 + C A + (C A + 1)/4) mod 7, where the remainder of the division by 4 is discarded.

In order to obtain the mad9il of the following months, two and one are added alternately in the case of the Arabic calendar, two for each month and five for the extra days in the case of the Persian calendar, and the respective month lengths minus 28 in the case of the Byzantine calendar. To obtain the mad9al of the current date, the day of the current month minus one should be added to the mad9al of the current month. After each addition multiples of seven are discarded. e. Tables. In practically all z3s, the chronological chapter contains a set of mathematical tables which facilitate the calculations described above. Among these tables the following appear to be standard: (1) Tables for the calculation of the al. Such tables display in three subtables, either in sexagesimal or in decimal notation, the number of days corresponding to groups of years (al-sinn al-ma3m#a), individual years (al-sinn al-mabsa), and months. For instance, for the Arabic calendar the number of days in multiples of 30 years and in 1, 2, , 30 single yearswill be given. Some authors of z3s combine more than one calendar in a single table by using a sexagesimal set-up with groups of 60 years. In each case, the number of days from the epoch to a given date is obtained by adding the day of the current month to the sum of the appropriate values from the three subtables. (2) Direct conversion tables. From this type of table the dates in one or more calendars corresponding to a set of (often equidistant) beginnings of years in a base calendar can be read off directly. Intermediate dates can then be found readily by adding the years and days in the desired calendar corresponding to the remaining single years and months in the base calendar, which are tabulated in separate subtables, as well as the elapsed days of the current month. For instance, when converting 24 Rama'n 254 Hi3ra into the corresponding Byzantine date, one may read directly from the table that 1 Muarram 241 Hi3ra corresponds to the 233rd day of the year 1166 Alexander. Furthermore, one finds that the first thirteen | [X:270b] years of the cycle of thirty Arabic years are equal to 12 Byzantine years plus 224 days, and the Arabic months before Rama'n to 236 days. Together with the elapsed days of the current month (24 1) this yields the (233 + 224 + 236 + 23 =) 716th day since the beginning of the year 1178 Alexander, i.e. the 351st day or 15 Ayll of 1179 Alexander. (3) Tables for the calculation of mad9il. Tables for mad9il make use of the fact that the same dates recur on the same week days after 210 years in the Arabic calendar, 7 in the Persian, and 28 in the Byzantine. The first step in the determination of a mad9al is therefore to cast off multiples of the length of the cycle concerned from the given year. For the Persian and Byzantine calendars, most z3s contain a doubleargument table from which the initial week day can be read off directly for every month of every year within the respective cycles. In the case of the Arabiccalendar, the initial week days of years 1, 2,, 210 are displayed together with a constant for every month which should be added to the initial week day of the given year. In early z3s, this type of table is often found under the name al-3adwal al-mu3arrad. Besides the types discussed here, many z3s contain special tables for more complicated calendars such as the 3all and the Chinese-Uighur, and tables offeasts and fasts in various calendars. Islamic tables for the determination of the Great Lent and

Easter have been analysed in G.A. Saliba, Easter computation in medieval astronomical handbooks, in al-Abth, xxiii (1970),179-212*. f. Various. In early z3s in particular, we find approximate methods for chronological calculations, such as tables for Arabic mad9il based on a cycle of eight years. In the manuscript of the z3 of 0aba9 al-0sib extant in Berlin, some problems are discussed of the types Find a date for which the Alexander year is equal to the Hi3ra year or Find an Alexander year in which Farwardn occurs twice, whereas alBrn presents solutions for problems of mixed dates, in which day, month, and year are given in three different calendars. For more information about z3s, the reader is referred to z3 and to Kennedy, A survey of Islamic astronomical tables, in Trans. of the American Philosophical Society, N.S. lxvi/2 (1956), 123-77 (repr. 1989). Many important investigations of parts of z3s are contained in SIES. The standard work on general mathematical chronology is still F.K. Ginzel, Handbuch der mathematischen und technischen Chronologie, 3 vols. Leipzig 1906-14, repr. Leipzig 1958. A valuable overview of era chronology in the Islamic world is presented in S.H. Taqizadeh, Various eras and calendars used in the countries of Islam , in BSOS, ix (1937-9), 903-27, and x (1939-42), 107-32. Additions to Sachaus edition of alBrns Chronology can be found in K. Garbers,Eine Ergnzung zur Sachauschen Ausgabe von al-Brns Chronologie orientalischer Vlker, in Isl., xxx (1952), 39-80. | [X:271a] Conversions of dates have traditionally been performed with the WstenfeldMahlersche Vergleichungstabellen, revised B. Spuler and J. Mayr, Wiesbaden 1961, or with R. Schram, Kalendariographische und chronologische Tafeln, Leipzig 1908. Now programs for various types of computers are available, partially from the Internet. The DOS programme CALH by B. van Dalen (version 1.2, 1997) includes most of the calendars and epochs described in this article. II. Historical Writing 1. In the Arab world. Ta"r9 is the most common Arabic word (widely used in New Persian and the Turkic languages as well) for history, historiography, in the sense of an ordered account of actual events. The word is certainly not archaic and is only clearly attested by the mid2nd/8th century, when it first appears in the titles of works recounting past events. By the late 3rd/9th century, ta"r9 had become the most common word for this genre of writing. Other terms for historiography were also current, however. Indeed, a9br (reports, narratives) may be older and in any case was very widely used down to early modern times. Until the mid-3rd/9th century, in fact, works of history were as likely to be identified by their subject mattere.g. sra(biography), ma9z[q.vv.] (campaigns of the Prophet), fut (conquests)as by a word naming the literary genre or class of knowledge to which they belonged. Before the late 19th century, ta"r9 seems to refer only to a kind of writing or knowledge, but in modern Arabic ta"r9 (like English

history, German Geschichte, etc.) is equivocal, comprising both events per se and the verbal representation of these events. The etymology of ta"r9 is uncertain, since it is not found in other Semitic languages. Rosenthal relates it to South Arabian wr9 moon and month, with a hypothetical derivation tawr9 determining dates by observing the moon (it should be also noted that the Sabaic dictionary of A.R.L. Beeston et alii, 162, further gives the meaning date for wr, and that the Arabic lexicographers list tawr9 as a variant of ta"r9, see e.g. L#A, s.v. "-r-9 and w-r-9). The original denotation of ta"r9 is in fact date or dating [see I. 1, above]. When it is attached to literary compositions, then, it should denote a chronologically-ordered account of events. The latter does, in fact, represent the earliest usage (occasionally in the expanded form al-ta"r9 #al "lsinn), but the word was rapidly extended to include any record of events or persons, however organised; e.g. by the mid-3rd/9th century we have al-Bu9rs al-Ta"r9 al-kabr, an alphabetically-arranged dictionary of ad9transmitters which contains very few dates. In view of this fluidity in language, no effort will be made in what follows to distinguish between ta"r9, a9br , and other words used for historical writing. (a) Origins to ca. 950. The bulk of early Arabic historical texts (or more precisely, texts which claim to be early) have not come down to us in their original form but are only preserved as citations and paraphrases in a corpus of digests and compilations assembled between the mid-3rd/9th century and the early 4th/10th century. (The earliest biographies of the Prophet form an important exception, since they were written or redacted in their present form by the early 3rd/9th centuryi.e. about half to three-quarters of a century before extant historical texts on other subjects.) It is true that much apparently archaic material can be dug out of the encyclopaedias and biographical dictionaries of later centuries, but this does not alter the nature of the pro- | [X:271b] blem. Given the present state of the evidence, then,we can determine what Arabic historiography had become by the end of Islams first three centuries, but recovering the earlier phases of historical thought and writing has proved an extremely elusive problem. A mountain of research on this issue has yet to produce results which command general assent. The historical works of the late 3rd/9th and early 4th/10th centuries represent the culmination of historical writing in early Islam in two respects. First, they synthesised a vast corpus of narratives which had been collected and put into circulation over the previous 200 years. Second, they defined the religious and political meaning of these narratives in a manner that later Muslims found nearly definitive for many centuries. The syntheses composed around the beginning of the 4th/10th century attained such prestige that few later historians made any effort to investigate anew the first 200 years of Islamic history; they were usually content to copy and abridge the classical syntheses, in particular the vast chronicle of Ab 3a#far al-abar (d. 310/923 [q.v.]). As a result, most of the older sources ceased to be copied or read in any systematic way, though many of them were still extant and consulted (rather haphazardly) down into Mamlk times. There is a second reason for the fossilisation of historiography dealing with the first two centuries of Islam: after the mid-4th/10th century, the issues and topics (and hence

the texts which embodied them) which had long been the focus of historiographic concern no longer seemed highly relevant. Since 132/ 750 the crucial issue for every historian had been the stance he ought to take toward the #Abbsids. Should they be presented as usurpers of #Alid (or even Umayyad) rights, as legitimate successors to an unbroken caliphal succession stretching back to Ab Bakr, or as the restorers of the purity of Muammads umma ? On ones resolution of this problem rested his interpretation of Islamic history for the century and a half before the #Abbsid Revolution. On a deeper level, the debate about the #Abbsids was a debate about the religious meaning of the whole history of the umma [q.v.]. This debate was framed in terms of a paradigm of covenant, betrayal and redemptiona paradigm which had its roots in the | ur"ns oft-repeated prophet narratives, whose guiding motif is the challenge by a prophet to his people to accept the worship of the one true God and to follow his commandments. This challenge represents Gods offer of a covenant (Ar. #ahd, m9 [q.vv.]), sc. obedience to His commandments in return for prosperity in this world and the next. The covenant is most often scornfully rejected, quickly followed by an outpouring of divine wrath. Even when it is accepted, as with the Jews and Christians, these communities quickly fall from wholehearted obedience and corrupt their religion. As the 3rd/9th-century historians understood the matter, Muammads people had been offered just such a covenant in the |ur"n. They had accepted it (not always without a struggle) and had been rewarded as no people before them. But soon even the Muslims had fallen prey to mans innate heedlessness (9afla) and ingratitude (kufr); they had betrayed their covenant and now were rent by schism and bloodshed. This fall from grace compelled thoughtful Muslims to ask whether the betrayal could be repaired and the umma be redeemed. If so, how and at whose hands could such a redemption be won? Historians also had to ask how the betrayal of the Muammadan covenant had occurred and who was responsible. | [X:272a] The official interpretation propounded by the #Abbsid rgimewhich had to justify its claim to rule as kin of the Prophet, while preserving the legitimacy of the first three caliphslaid the blame for the ummas sufferings on the corrupt and tyrannical Umayyads. But most historians, even those with close ties to the #Abbsid court, concurred that the original crisis had occurred in the reign of #U9mn b. #Affn (24-35/644-56), though they disagreed bitterly about which persons and groups bore the guilt for the catastrophe. The paradigm of covenant, betrayal and redemption was clearly formulated by late Umayyad times and was highly productive for nearly two centuries. But as #Abbsid authority was subverted in the late 3rd/9th century and collapsed in the 320s/930s, this paradigm lost much of its power. It could not bestow meaning and value on the sordid intrigues of petty dynasts and warlords. The historians of the 4th/10th century and later focused more and more on recent and contemporary events, and sought new ways of constructing and interpreting these. Some of the late 3rd/9th-century syntheses are best characterised as digests. Key examples would be three nearly contemporary works: the Ta"r9 of al-Ya#b (d. 283/897), al-A9br al-iwl of Ab 0anfa al-Dnawar (d. 281/894), and the Kitb al-ma#rifof Ibn |utayba (d. 276/889). Late examples of this group would be the Mur3 al-9ahab wa-ma#din al-3awharand Kitb al-Tanbh wa "l-i9rf of alMas#d (d. 345/ 956). The authors of the digests refine the disparate materials bequeathed them by earlier collectors into a single narrative line, with the intention of

offeringa clear and unambiguous interpretation of Islamic history. Even so, their works preserve the historical writing of earlier generations because they present their accounts in the form of (highly selective) quotationand paraphrase of their sources. Of these digests the most original and interesting are the works of al-Ya#b and alMas#d. Al-Ya#b composed the first true universal history in the Arabic language. He draws on a wide array of materials(many clearly non-Islamic) to give a culturally and intellectually oriented tableau of the pre-Islamic nations. Then with the coming of Muammad he narrows his focus to the political history of Islam, organised by caliphal reign and told from a markedly pro-#Alid perspective. In its overall structure alYa#bs digest provided the model for the universal histories of al-abar and alMas#d, though there is no evidence that either of these scholars used his work. AlMas#ds two digests are late works, abridgements or adaptations of a series of far larger and more systematic histories (all now lost) which he claims to have composed earlier in his career. They combine, in a manner never successfully imitated by any other historian, serious information with the arts and graces of adab. Al-Mas#ds Islamic history is similar in many ways to al-Ya#bs, buthis non-Islamic materials are far wider-ranging and clearly draw on a much richer array of sources. He is, forexample, the only Muslim author to give us a serious prcis of Byzantine history since the rise of Islam. It is noteworthy that Ibn 9aldn (d. 808/1406), so severe in his judgments on many historians, identified al-Mas#d as his true precursor. Without doubt the crucial historical works of this era are the massive compilations of alBal9ur (d. 279/893) and Ab 3a#far al-abar (d. 310/923). It is these which preserve for us the broadest cross-section of early Arab-Islamic historical writing, and hence they are our fundamental sources for the origins and early development of Islamic historiography. | [X:272b] Al-Bal9ur left us two major histories, both dealing wholly with Islamic times: Fut al-buldn, an account of the Arab-Islamic conquests (coming down into #Abbsid times), organised by region and oriented toward legaladministrative issues; and Ansb al-a9rf, a vast collection of political biographies of the caliphs and other notable figures of Islamic history, with the biographies grouped according to lineage. Al-abars Ta"r9 al-rusul wa "l-mulk is an enormous chronicle stretching from the Creation down to the last years of al-abars life. It is arranged by nation or people before the rise of Islam, with an emphasis on the Israelites and Persians; with the hi3ra it follows a strict annalistic framework. It represents only one facet of the uvre (in effect the retirement project) of a scholar who regarded himself chiefly as a |ur"n commentator, fah and muaddi9 . In principle, the compilations of al-Bal9ur (especially the Ansb al-a9rf) and alabar aimed to assemble all well-attested accounts pertaining to the major events and actors in Islamic history. Both scholars used the muaddi9>>s tool of the isnd to show the provenance of each account, thereby allowing the informed reader to assess its authenticity and religious soundness. Al-abar indeed deploys the isnd technique very rigorously; this and other indications suggest that he hoped to raise history to the full dignity of a religious science. But that turned out to be impossible, due at least in part to the highly disparate and uncontrollable sources which historians had to use. In contrast to the digests, these vast compilations makeno effort to construct a unified narrative ofevents. On the contrary, they consist of a series of discrete reports (Ar. 9abar , pl. a9br ) varying in length from a line to several pages. These a9br are

not linked by a narrative thread; they are simply juxtaposed end to end, each being marked off from the others by its own isnd . A compiler might select several reports pertaining to a given event, and these could variously repeat, overlap, or contradict one another. The criteria for including some a9br and omitting others are almost never spelled out explicitly; one must simply infer such criteria through context, literary structure, etc. In general, it was proper for a historian to abridge or paraphrase the a9br which he found in his sources; he might even blend several accounts together into a collective tradition so long as he did no violence to their contents. AlBal9ur paraphrases quite freely; al-abar follows the wording of his sources closely, albeit with considerable abridgment. Al-abar also chops up what were originally extended narratives into short segments, so as to juxtapose these with contrasting or parallel versions of the events being presented. None of the historians surveyed above ever intervenes in the narrative to explain its overall significance or to pass judgment on the actors. The reason for this reticence is partly grounded in early Islamic concepts of knowledge ( #ilm ). In this perspective, historical knowledge was constituted by statements which could be traced back to reliable authoritiesideally, to eyewitnesses of known veracity, but in any case to reputable persons who had obtained their information from good sources. The historians task was thus simply to determine which a9br were acceptable and to arrange these in a usable order. On another level, the events recorded by early Muslim historians were intensely controversial. Hence if they discussed these in their own words, they would inevitably be regarded as mere propagandists for one or another faction. Scholarly authority required a talent for self-effacement. | [X:273a] Insofar as they deal with the middle and later decades of the 3rd/9th century, the digests and compilations reviewed above drew, with varying skill and perceptiveness, on a variety of contemporary sources as well as personal observation. For the reigns of the #Abbsid caliphs from al-Mahd (775-85) until al-Ma"mn (813-33), they seem to have used an official court historiography as redacted under al-Ma"mn, supplemented by personal memoirs and other non-official accounts. The provenance of their information on these two periods raises many difficult issues, but there is no real question about the authenticity of the sources they used. That is, these sources do date from the periods which they recount, and they were composed by the authors to whom they are ascribed. As we move into their sources for earlier periods, however, we can have progressively less confidence on this point. For the century and a half stretching between Muammads call and the consolidation of #Abbsid power under al-Manr (r. 13668/754-75), the direct sources for the classical syntheses were a series of a9br collections (now mostly lost) compiled between ca. 750 and 850 by a number of scholars, of whom the most widely cited are Ibn Is(d. 150/ 767), Ab Mi9naf L b. Yay (d. 157/774), Sayf b. #Umar (d. ca. 180/796), al-Hay9am b. #Ad (207/ 822), Hi9m b. Muammad al-Kalb (d. 204/819), Muammad b. #Umar al-Wid (d. 207/822),Muammad b. Sa#d (d. 230/845), and #Al b. Muammad al-Mad"in (d. 225/840). The works of the earlier collectors in this group were certainly heavily redacted in the early 3rd/9th century, and it is in this form that the classical syntheses had access to them. Even so, the essential authenticity of these collections seems

uncontestable; they are, as they claim to be, substantially works compiled and edited by Ab Mi9naf, Sayf b. #Umar et al . The early 3rd/9th century also witnessed the emergence of two new genres which would have a very long lifespan in Islamic historiography: the earliest biographical compilations, in the form of abat[q.v.] or generations of notable Muslims (chiefly transmitters of ad9and other religious knowledge) from the time of the Prophet; and the annalistic chronicle, which strove to place each known event in the precise Hi3r year in which it occurred. The first extant biographical compilations already display a considerable range of subjects and emphases: the Kitb al-abat al-kabrof Ibn Sa#d, the terse Kitb al-abatof 9alfa b. 9ayy al-#Ufur (d. 241/855), and the Kitb abak al-9u#ar" al-3hiliyyn wa "l-islmiyyn by Muammadb. Sallm al3uma (231/846). The biographical dictionary immediately met a critical need in a religious and literary culture which was based on the transmission of knowledge from one person to another, and which needed to determine the biographical data, religious acceptability, and reliability of each transmitter. The new form quickly branched off in many directions, and by the end of the 3rd/9th century it was producing works as diverse as al-Bu9rs al-Ta"r9 al-kabrand al-Bal9urs Ansb al-a9rf. Until the mid-3rd/9th century, the favoured organisational scheme for general histories was by caliphal reign; this regnal or dynastic schema continued to be very widely used as well. After 9alfa b. 9ayy (al-a"r9), al-abar was the next extant historian to use the annalistic structure; much of the confusion in his volumes on the earliest decades of Islam comes from the fact that his source material was not originally dated by year, and could only be located in the right year by context. But after al-abar, the annal- | [X:273b] istic mode became and remained the most widely used organisational framework, at least among Arabic-language historians, down into the 13th/19th century. The a9br collections of the early 3rd/9th century differ from one another in many ways: subject matter, sources, organisation and handling of material, religio-political stance, etc. Nevertheless, they do have some important features in common. First, they all reflect the covenant-betrayal-redemption paradigm, thus demonstrating that this paradigm was already fully articulated by this era. Second, they embody predominantly #Ir (and for the earliest decades, 0i3z) perspectives; only fragments of a Syrian or Egyptian tradition have found their way into them. This seems natural enough for #Abbsid times, but obviously it implies serious distortions in their treatment of the Umayyads. Third, they are assemblages of a9br relating to a given set of events rather than integrated narratives of those events. Fourth, most of them are monographic rather than synthetici.e. they bring together reports concerning discrete events (such as the Battle of the Camel) or topics (the Ridda wars or the conquests in Syria). (Note, however, that the earliest synthetic history of early Islamic political history was probably Ibn Iss Ta"r9 al-9ulaf", probably a digest rather than a detailed compilation, now lost except for a few citations in al-abar and possibly a papyrus fragment.) Finally, these works use isnds (not necessarily in a rigorous way) to link their narratives to the original reporters of the events they include. In spite of the formal structure of these collections, which implies that they are merely transmitting wellattested reports by persons close to the events in question, there is ample evidence of a strong authorial (or at least editorial) hand in many of them. For example, Sayf b. #Umar explicitly blends reports from several different transmitters into a single version,

while al-Wid gives a level of detail and narrative elaboration which we seldom find in older accounts of the same events. Too much attention to questions of form, structure, transmission, accuracy, etc. can be very misleading, for many works from this period exhibit a very strong folkloric tendencyan emphasis on storytelling for its own sake, the heroic, the colourful, the supernatural and fantastic. Folkloric elements were assuredly present in the very origins of Islamic historical narrative, but the early #Abbsid period produced a number of larger works which are thoroughly imbued with folklore. The most notable of these is the Kitb al-Fut of Ibn A#9am al-Kf (who probably flourished ca. 200/815, pace M.A. Shaban). But this period must also have seen the origins of the widely copied Fut al-9msagas ascribed to al-Wid, though these emerged in their present form only during the Crusades. The narratives collected and redacted by Sayf b. #Umar display a distinctive combination of serious purpose and folkloric storytelling. In fact there is no clear line separating serious history from folklore in the materials that have come down to us; we must take both as original and integral aspects in the understanding and articulation of their past by early Muslims. It is precisely this unmistakable editorial/authorial presence which gives us pause about the way in which the late 2nd/8th century collectors used their sources, for we have no way of checking the accuracy or good faith of these collectors. It is true that the problem differs as between two distinct periods: (a) the life of the Prophet and the first seven decades of Islam, down to the end of the great civil war of 680-92; (b) | [X:274a] the Umayyad domination between 692 and the #Abbsid Revolution (747-50). In regard to the latter period, no one contends that the late 2nd/8th-century a9br collectors simply invented the stories they tell, if only because the events of late Umayyad times were too recent. On the other hand, the collectors do report these events in an intensely partisan mannernot only because some (not all by any means) had close ties to the #Abbsid court, which actively sought to poison the memory of its predecessors, but also because their informants were themselves the survivors of many bitter conflicts in Umayyad times. We simply cannot say how far or in what ways the a9br collectors of early #Abbsid times may have reshaped or elaborated the stories which they gathered. The major events at least are certainly not invented out of whole cloth, but how accurately do they reflect the original accounts of those who were involved in them? We can only guess. It is not even clear how much contemporary history was actually written down under the Umayyads; the Fihrist of Ibn al-Nadm [q.v.] gives us only a few titles dealing with the period 700-50, and these were probably composed in the early decades of #Abbsid rule. On the other hand, the Umayyad court was clearly intensely interested in the decades before the reign of #Abd al-Malik, since the events of that period were clearly essential to its claims to legitimacy. Rather than a developed (or even emerging) Umayyad historical literature, then, we have been given a very disparate body of recollections and statements, some originating within the Umayyad court, many put into circulation by its opponents. If we may assume that narratives about events under the Umayyads, however reshaped for partisan and ideological purposes, are ultimately grounded in reality and can be linked to contemporary reporters, we can be far less confident about the nature of Umayyad-era historiography on the 1st/7th century. Existing evidence indicates that formal historical study and writing began in the decades following the second civil war (680-92), and was no doubt associated with the need to recover an authentic and

authoritative past from the chaos and violence of those years. Moreover, at least some of this historical work was owed to Umayyad prompting and patronage, in particular the biographical materials on the Prophet collected by Ibn 9ihb al-Zuhr (d. 124/742) and Ms b. #Uba (d. after 141/758) [q.vv.]. It is undeniable that the Umayyad caliphs from #Abd al-Malik on took an intense interest in emerging theological issues, in law, and in investigating and defining the early past of the Islamic community. In the light of the unrelenting challenges to their legitimacy from every quarter, they had no choice. It is now quite clear that the Umayyad caliphs claimed final (though perhaps not infallible) authority in matters of the faith. They could hardly take a slighter interest in the historical process through which they had risen to supreme authority over the community. There was of course a counter-history of early Islam, and this clearly survives much more fully than the Umayyad version, due to the #Abbsid victory and their systematic purge of elements favourable to the old order. In both the pro-Umayyad and antiUmayyad versions of Islams beginnings, there are already clear traces of the covenantbetrayal-redemption paradigm. Having said all this, however, we simply do not know what the historical works of Umayyad-era historians looked like; we have fragments which may be more or less authentic, but these give us no idea of the whole collections from which they were drawn. Nor | [X:274b] can we trace even these fragments back to the earlier authorities on which they based their statements with any confidence. A coherent body of historical tradition quite suddenly emerges under the Umayyads, but we cannot identify the raw materials out of which it was formed. The primitive historiography (if we may call these accounts anything so formal as historiography) composed during the eight decades between Muammads call and #Abd al-Maliks consolidation of power poses problems which are not only far more severe but entirely different in nature. Both the accuracy and authenticity of every report attributed to this period are open to credible challenge. The problem is not that our texts are biased or partisan; it is a truism that every narrative represents events from some perspective, and hence encodes a complex body of ideologies, values, conceptual structures and cultural practices. But it is normally quite easy to decode these aspects of our texts, as Ignaz Goldziher demonstrated a century ago. The problem is rather that we do not know to what degree any given narrative transmits usable (albeit incomplete) information about the events which it claims to reportor sometimes, whether these events ever happened at all. A related problem is that of authenticity, i.e., quite apart from the issue of factuality, whether a given narrative actually goes back to its purported original reporter or whether it was first framed and circulated in a much later period. The parallels between archaic historical a9br and the origins of ad9are obvious, even though the structures and purposes of the two genres differ in many fundamental ways. In short, we can only offer conjectural reconstructions of archaic historiography. The following comments should be taken in that light. We cannot speak of formal historical researchthe systematic collecting and redacting of reports about events and persons during the first eighty years of Islam. But Muslims must have been deeply aware of the titanic events through which they were passing, and of the profound changes in political institutions and patterns of life which these events represented. 1st/7th-century Muslims had to try to make sense of all this, and inevitably they did so through the genres of verbal expression familiar to them. First of all, they could draw on the kind of oral

narrative practised among the tribes of ancient Arabia: prose recitals about memorable events, embellished and interpreted by verse (satirical, boasting or elegiac as the case might be). The point of these recitals was of course not meticulous accuracy but the values of manliness and tribal honour which they encoded. Particularly effective poetry often liberated itself from the events which had originally inspired it, and would henceforth stand on its own or be re-attached to new events. In another milieu, as Islamic practice became established among the newly converted tribesmen (and a few non-Arab converts), the mosque sermon quickly emerged as a vehicle for the interpretation of crucial events as well as a means of exhorting people to act. The themes of formal sermons were supplemented and reinforced by pious tales about the prophets and other religious heroes delivered by the u (sing. [q.v.]), both officially appointed and self-employed. The templates for historical narrative in the mid-1st/7th century were thus provided by the ayym al-#arab, ia, wa#and 9uba , and (deployed not to narrate events but to interprete them) various genres of 9i#r . Such oral expression flowed spontaneously from those who had experienced the life of the Prophet, | [X:275a] the great conquests and the bitter civil strife of those decades. It could not be monopolised by the caliphs or any other central authorities, though these lat- ter certainly tried to make their voice heard in the cacophony and to exert what little control they could. Some narratives seem to have been carefully composed from the outset, such as the pious exempla that demonstrate the stern morality and ardent faith of early Muslim heroes, or dramatic tableaux of confrontations between Muslim and infidel leaders. Precisely because thesewere effective moral tales, of course, they readily became topoi which could be attached to many different events by altering the protagonists and mises-en-scne. But even as details of fact were modified, the stereotyped plots, the lofty stylised language and the underlying religio-moral lessons would remain the same. However, even apparently sober reportage about the conquests and civil wars is full of problems. Much of this material must have been remembered and passed on by participants in these events, but there was never a unified, official record of it; on the contrary, the original accounts were generated by thousands of men caught up in confusing situations and scattered across vast areas. Hence their stories about what they had seen or heard began to disintegrate almost as soon as they were told. By #Abd alMaliks consolidation of power after 72/692, only scraps of authentic memory remained about the early conquests or the first civil war a half-century earlier. No doubt a few well-informed collectors could name the crucial events of that period, but no one could date them accurately or even specify the order in which they had occurred. As to minor or local events, it was impossible to separate fact from fiction. Often reporters could not even agree on which tribes or commanders had participated in which battles. (It must be said that these remarks are far less applicable to the life of Muammad than to the events following his death. The Prophet was a uniquely significant subject, of course, but just as important, his memory was preserved among a highly self-conscious and relatively stable groupsc. the -aba or Companions and their successors in Medinaand at least in principle these could generate and transmit a coherent, reliable story about his life and teaching. Whether they actually did so is, of course, very much in dispute.)

We have so far focused on oral materialsbattle stories, poetry, sermons, pious tales as the stuff of 1st/7th century historiography. Was there any historical writing in this period? There is no unambiguous evidence of it, but we cannot completely exclude the possibility of informal efforts to record local or tribal accounts of major events. We can imagine but cannot confirm efforts to collect the narrative tradition of 0im or Medina or Kfa or Bara, or of the tribes of Tamm or Kinda. It is probable in any case that efforts to build a written record (insofar as they existed) focused on the life and words of the Prophet. On the other hand, we do find some written materials (letters, treaties, administrative decrees) incorporated within the historical tradition, and a few documents (e.g. the famous Constitution of Medina, the -iffn arbitration agreement, the substance if not the exact words of many peace treaties) are likely to be authentic. Likewise, our texts on the plan of Kfa or the dwn of #Umar seem to reflect real administrative arrangements from Islams earliest decades. But in the final analysis we can say only this much. There was by the second civil war (62-72/683-92) a real and widely shared historical consciousness, R$W^ (a.), pl. ruwt, reciter and transmitter of poetry reciter and transmitter of poetry, as also of narrative traditions ( a9br ) and ad9[q.v.]. The term is derived from rawto bring, carry or convey water, and has been extended to carrying in a figurative sense, i.e. to bear by memory, to transmit or recite (cf. Lane, 1194). There is an intensive form rwiya, explained as copious transmitter (ka9r al-riwya), used in mediaeval sources as a synonym to rw . In modern research it is applied, as a rule, to the learned collectors of Bedouin poetry in the 8th century. The institution of the rwis the main basis for the preservation of pre-Islamic poetry. In the 3hiliyya[q.v.] poets used to have one or more rws, who learned their verses by heart, recited them in public, especially at the annual fairs, where poetic contests took place, and transmitted them to the next generation. It often happened that a rwbecame a famous poet himself. Lists of poets and their rws are known over several generations. A spectacular line, extending over two centuries, begins with Aws b. 0a3ar [q.v.], the stepfather of Zuhayr b. Ab Sulm [q.v.], who was his rw (Ibn | utayba, 9i#r , 57). Zuhayr, who also had his son Ka#b [q.v.] for a rw , figures at the beginning of the following list (A9n1, vii, 78): Zuhayr, al-0uay"a [q.v.], Hudba b. 9a9ram, 3aml [q.v.], Ku9ayyir #Azza [q.v.], who died in 105/723. He was the last to combine the function of poet and rw (9ir man i3tama#a lahu al-9i#r wa 'l-riwya, loc. cit.). From the list and similar information, it appears that transmission of poetry often took place in the same family or clan, but not necessarily, which implies that rwwas an accepted profession or semi-profession. The question whether transmission in the 3hiliyyawas exclusively oral, or whether poets and transmitters assisted their memory by writing, remains controversial. Whereas Sezgin (GAS, ii, 22-33) maintains the early use of writing in the process of transmission, other scholars have emphasised the oral character of pre-Islamic texts (M. Zwettler, The oral character of classical Arabic poetry, Columbus 1978, cf. 85-8). Since there is no conclusive evidence, one can only attempt to evaluate the known facts. In the 3hiliyya, the use | [VIII:467a] of writing, although well established for contracts, treaties or other official documents, could hardly have played a significant part in poetic transmission. It is possible that poets in contact with the courts of 0ra [q.v.] and 9assn [q.v.] were able to write, but among Bedouins that knowledge cannot have been common. Furthermore, the corpus of pre-Islamic verses presents characteristic

features of oral literature, e.g. a high percentage of formulaic expressions, semantic repetition and independence of detail, which later gave way to other stylistic features and modes of composition. Thus it is to be assumed that during the 6th century A.D. composition and transmission of poetry took place orally, which does not exclude the possibility of a rwnoting down verses as a mnemonic aid. In the course of the first Islamic century, the use of writing increased in various fields (cf. G. Schoeler, Schreiben und Verffentlichen. Zur Verwendung und Funktion der Schrift in den ersten islamischen Jahrhunderten, in Isl. , lxix [1992], 1-43). The first collections of poetry were made in the early Umayyad period, e.g. the Mu#allat[q.v.] (cf. M.J. Kister, The Seven Odes, in RSO , xliv (1979) 27-36). The poet al-Farazda [q.v.] mentions in some of his verses that he possessed books with collected poetry of other poets (The Na"i' of Jarr and al-Farazda , ed. A.A. Bevan, i-iii, Leiden 190512, i, 201, v. 57, 61). It is further reported that 3arr [q.v.] and al-Farazda used to dictate to their rws ( Na"i' , i, 430, 12; ii, 908, 2). It seems therefore, that oral transmission was at first aided, and then gradually replaced, by writing. In the final stage of poetic transmission, the early #Abbsid period, Bedouin poetry was systematically collected by learned rws like 9alaf al-Amar, 0ammd al-Rwiya and al-Mufa''al al-4abb [q.vv.]. There is ample evidence that they had written collections of poetry at their disposal, but they were still expected to know the texts by heart, and to recite them when requested (A9n1, v, 174). In addition, they used to collect information from Bedouins and to verify their knowledge by questioning them. These Bedouin informants, who were also called rw, are in part known by name (cf. Ch. Pellat, Le milieu barien et la formation de @i, Paris 1953, 137-8). Thus presumably the term rw/rwiyawas applied, as long as learning by heart and reciting of verses still played a part, even if a marginal one, in poetic transmission. Another aspect is the exact function of the rwand his relation to the poet who employed him. Since a rwoften became a poet himself, it has been assumed that he also served an apprenticeship with his poet, receiving a thorough training in metrics and the art of composition. This would imply that the institution of the rwnot only assured the preservation of poetry, but also the continuity of technical knowledge, and of the vocabulary, style, and thematic range of an individual poet. The first to consider the possibility of establishing schools of poetry was h 0usayn, who with regard to the list of rws mentioned above speaks of the poetic school (ma9hab 9i#r) of Aws b. 0a3r (Fi 'l-adab al-3hil, Cairo 1927, 161989, 270). The question has been studied with regard to the poetry of Hu9ayl [q.v.] by E. Brunlich (Versuch einer literargeschichtlichen Betrachtungsweise altarabischer Poesien, in Isl. , xxiv [1937], 201-69; cf. 221 ff.), as also by G.E. von Grunebaum (Zur Chronologie der frharabischen Dichtung, in Orientalia, N.S. viii [1939], 328-45), who established six schools of poetry in the pre-Islamic period. The assumption that rws received a thorough education, and reached a competence equal to that of | [VIII:467b] the poets they served, is further evidenced by reports that a rwwas expected to correct, to polish up or even to embellish the verses of his master. This seems to have been a common practice in the 3hiliyya(cf. Goldziher, Muh. St., ii, 82), as also in the Umayyad period. There is a story concerning 3arr and al-Farazda, whose rws were found to correct their metrical blunders (A9n1, iv, 54). In view of this and similar reports, it is easy to appreciate the exclamation of al-

0uay"a: Woe to verses in the hand of a bad transmitter! (wayl un li 'l-9i#ri min rwiyat al-s"; A9n1, ii, 59). It also underlines the difficulty, and sometimes the impossibility, for historians of literature to clearly differentiate between the work of a poet and that of his rw . S^RA (a.), a genre of early Islamic literature: Srameans way of going; way of acting, conduct, way of life (in these meanings it is almost synonymous with sunna [q.v.]); also memorable action and record of such an action. In ad9collections and books on Islamic law, the plural siyar is also used for rules of war and of dealings with non-Muslims (which are sometimes headed elsewhere under 3ihd ). Furthermore srameans epistle, pamphlet, manifesto, and last but not least: biography, the life and times of .... Ibn al-Muaffa# (102-39/720-56 [q.v.]) translated a Pahlavi history of the Persian kings under the title Siyar mulk al-#a3am. #Awna b. al-0akam (d. 147/764 [q.v.]) wrote a Srat Mu#wiya wa-ban Umayya. Abn b. #Abd al-0amd al-Li (d. ca. 200/815 [q.v.]) wrote Srat Arda9r and Srat An9irwn. For later popularised biographies of kings and heroes, see sra 9a#biyya .The sra, srat raslallh or al-sra al-nabawiyya, have been the most widely used names for the traditional account of Muammad's life and background. Martin Hinds (Ma9z and Sra, in La vie du Prophte Mahomet, 57-66; see also ma9z) has argued that the biographical material on the Prophet was transmitted during the first two centuries of Islam exclusively under the name of ma9z, whereas srawas applied only since Ibn Hi9m (d. 218/833 or 213/828 [q.v.]). This view has been challenged by Maher Jarrar (Prophetenbiographie, 1-59), who claims that ma9zis only part of the sra, the designation being used occasionally as a pars pro toto, and that the biography was already called sraby al-Zuhr (d. 124/742 [q.v.]), a central figure in the transmission of materials on the Prophet. History of the sra. In ma9z, Hinds discussed not only the designation of the prophetic biography but wrote also its early history. For that stage | [IX:661a] the present contribution has little to add. However, the most archaic layer of the biography, that of the stories of the u [see ; ia . 1], deserves a little more emphasis. An early 3rd/9th century papyrus, whose isnds go back to Wahb b. Munabbih (34-110/654-728 [q.v.]), contains a large srafragment of the early ia type. Its narrative is lengthy, no less entertaining than edifying, more often interrupted by poetry than by isnds; it has an outspoken pro#Al ring, and it contains a wealth of miracle stories (e.g. the Prophet practicing sorcery: Wahb 142, 20; Ibn Is, no. 426). A renewed study of Ibn Is has been stimulated by the publication of part of Ibn Is's ma9zin the riwya of Ynus b. Bukayr (d. 199/815; GAS, i, 289), which has been preserved in ms. Fs |arawiyyn 202, and in the riwya of Muammad b. Salama al-0arrn (d. 191/807; Ibn 0a3ar, Tah9b ix, 296) (Damascus hiriyya ma3. 110, fols. 158-174). These texts, which contain many fragments which were hitherto unknown or deviate from the familiar versions, shed new light on the transmission of Ibn Is's work. Comparisons of Ynus' versions with those of al-Bakk" (as preserved by Ibn Hi9m ), Muammad b. Salama and several others have led Sellheim, Samuk and Muranyi to the conclusion that there has hardly been any written standard text by Ibn Is himself and that we depend on his transmitters, whose texts should be studied synoptically, in all their variants.

Furthermore, Muranyi has pointed out that Ynus b. Bukayr transmitted materials which do not go back to Ibn Is at all. Ynus was a sracompiler in his own right, whose Ziydt al-ma9z was quoted by al-Bayha, Ibn Ka9r and several others. After Ibn Is, a limited, but interesting ma9zcollection was composed by Ma#mar b. R9id (d. 154/770; GAS, i, 290-1), which is preserved in the Muannaf by #Abd alRazz b. Hammm (d. 211/826; see al-an#n). Several other ad9collections have a ma9zsection, e.g. Ibn Ab 9ayba's Muannaf and al-Bu9r's -a . The fame of Ibn Hi9m, whose srais considered the most prominent, rests mainly on his selection from Ibn Is's work. The latter, by means of his Mubtada" section, had placed Muammad in the tradition of the earlier prophets, and had indeed made him the pivot of world history by adding a history of the caliphs. Ibn Hi9m, however, narrows the perspective down to Ancient Arabia. A chain of works with a limited focus on prehistory are al-Wid's (130-207/747-822 [q.v.]) K. al-Ma9z , which concentrates on the life and times of the Prophet only and displays a great interest in the chronology; the abatof Ibn Sa#d (d. 230/845), in which the srasection is preceded only by a brief survey of the prophets, and al-Bal9ur's (3rd/9th cent.) Ansb ala9rf, which outlines Muammad's ancient Arabian origins. Al-abar's (d. 310/922 [q.v.]) Ta"r9puts Muammad once again in the perspective of the history of the prophets and even the kings of Persia. The numerous later sraworks are mainly commentaries or compilations, although they contain important material from early sources. Of the late authors, the most interesting are al-Suhayl (d. 581/1185), who wrote a commentary on the sraof Ibn Hi9m, and his critic Mu9uly (d. 689/1290). Other compilers are Ibn Sayyid al-Ns (d. 734/1333); Ibn Ka9r (d. 774/1373; in al-Bidya wa 'l-nihya); Muammad b. Ysuf al--li (d. 942/1536; Brockelmann, II, 304-5); and Nr al-Dn al-0alab (9751044/1567-1635; see al-alab). | [IX:661b] For a survey of early sraworks, see GAS, i, 275-302; for late works see Kister, The srah literature , 366-7. Characteristics of sratexts. Be it under the heading ma9zor sra, in the prophetic biography very heterogeneous materials are brought together. Various intentions seem to prevail: to build up the image of Muammad in rivalry to the prophets of other communities, to depict him as a statesman of international stature, to elaborate on |ur"nic texts and create a chronological framework for them, to record the deeds of the early Muslims, to continue the genre of ayym al-#arab [q.v.] and to set standards for the new community. These intentions are striven after in a great variety of text types, of which the following survey is by no means exhaustive: (1) Stories about the military expeditions of Muammad and his companions (ma9zin the strictest sense). They form a continuation of the profane accounts of ayym al-#arab , with raids, battles, challenges, examples of bravery, exchanges of poetry and single combats. Islamic elements are, e.g., the intervention of angels in battle and the (often merely ornamental) addition of |ur"nic passages. In later centuries, the ma9zwere continued in their turn by would-be historical popular stories in which

Muammad is venerated, while #Al b. Ab lib develops into a military hero of supernatural stature. These popular stories, which were studied by Paret, can be reckoned with the sra 9a#biyya . The 7th/13th century author Abu 'l-0asan al-Bakr [see al-bakr ] played a central part in this genre, but he may well have had predecessors. (2) Accounts of fa'"il and ma9lib, which form the record of the merits and faults of clans and individual Companions of the Prophet, as well as their genealogies. Various lists are incorporated in the sra: of the first converts, of the Emigrants, the fighters in various battles, representatives of the Anr, etc. A specific type of text, to which also monographs were dedicated, is that of the aw"il [q.v.], in which is recorded who did something for the first time, e.g. Sa#d b. Ab Wa was the first to shed blood in Islam (Ibn Is, no. 194). The deeds of the Companions also became recorded in separate works, such as the abatby Ibn Sa#d; al-Ist#b f ma#rifat al-abby Ibn #Abd al-Barr (368-463/978-1070), Usd al-9baby Ibn al-A9r (555-630/1160-1233) and al-Iba f tamyz al-ababy Ibn 0a3ar al-#Asaln (773-852/1372-1449) [q.vv.]. These works show many overlaps with sratexts and should be read in combination with them. (3) Pieces of |ur"nic inspiration: tafsr, asbb al-nuzl and Midrash. Large parts of the sraare inspired by the Kur"n. They have been recently studied by J. Wansbrough. Some texts merely paraphrase a |ur"nic passage, e.g. sra XCIII in Ibn Is, no. 166. Typical for the sraare the accounts of the occasion for the revelation of certain |ur"nic passages (asbb al-nuzl). When the Prophet was mocked, for example, the verse Apostles have been mocked before you ... (VI, 10) was revealed (Ibn Hi9m, 262; cf. ibid., 272, and Ibn Is, no. 418). Many sratexts elaborate on a |ur"nic passage, in the manner of a Jewish midrash. The episode of the Satanic verses (al-abar, i, 1192-4), for example, was evoked and foreshadowed by XXII, 52: Satan casts something on the tongue of a prophet; God abrogates it and establishes His verses. In one version, this episode is presented as a sabab al-nuzl. The relationship between a |ur"nic passage and the story which pivots upon it may be quite loose. | [IX:662a] The long narrative of how |uray9 conspired at the eve of Muammad's hi3ra , and how Allh outwitted them by making them unable to see him, is built on VIII, 30: and when the unbelievers were plotting ... but God plots also, and God is the best of plotters, and elegantly incorporates XXXVI, 8: ... and We covered them, so that they could not see (Ibn Hi9m, 323-6; see also Wahb, 132-6). This story does not give the occasion for the revelation of the verses, but playfully talks about them together. The verse which forms the inspiration of a story need not even be quoted. The story about the reception of Muslim emigrants by the Negus of Abyssinia is built on |ur"n, III, 191, without any literal correspondence (cf. W. Raven in JSS , xxxiii [1988], 201). (4) Prophetic legend. As the |ur"n had done before, the sraaims at establishing the place of Muammad among the prophets, and that of Islam among the other religions.

The numerous stories which dwell upon the characteristics of prophethood react on the narrative repertoire of Judaism [see isr"liyyt ], Christianity and Manichaeism. Some examples: The twelve leaders (nuab") appointed by Muammad from the Anr at al-#Aaba [q.v.] are put on a par with the disciples of Jesus or the representatives of the tribes of Israel during the Exodus (Ibn Hi9m, 299; Wahb, 130). In the Ascension story, the rank attributed to the prophets is reflected by their places in one of the seven heavens: Muammad finds Ibrhm in the highest heaven, but Ms and #^s in the lower ones (Ibn Hi9m, 270). Even the physiognomies of the various prophets were subjected to comparative descriptions (Ibn Hi9m, 266, Ibn Sa#d, i/2, 125). The srasometimes recapitulates prophetic characteristics in general statements, which are exemplified by Muammad: there is no prophet but has shepherded a flock (Ibn Hi9m, 106); a prophet does not die without being given the choice (ibid., 1008); no prophet dies but he is buried where he died (ibid., 1019); the eyes of prophets sleep while their hearts are awake (ibid., 266; Ibn Sa#d i/1, 113). In ad9this generalising tendency becomes more frequent; cf. Wensinck, Handbook, 196-7. The sracontains stories about numerous miracles wrought by God through His Prophet, or by the Prophet himself, which served as the proofs of his prophethood, often with the intention of comparing him to the other prophets. From the 3rd/9th century onwards, these stories developed into the independent genre of dal"il or a#lm or amrt al-nubuwwa. Well-known authors in this field are #Abd al-3abbr alHama9n (d. 415/1025), Amad b. al-0usayn al-Bayha (d. 458/1066), Ab Nu#aym al-Ifahn (d. 430/1038), and al-Mward (d. 450/1058) [q.vv.]. For a longer enumeration of such works, see Kister, The srah literature , 355. (5) Written documents, including:- letters from the Prophet to foreign rulers, governors and to the Arabian tribes (e.g. Ibn Ab 9ayba, Muannaf , xiv, 336-46);- treaties, as for instance that of al-0udaybiya [q.v.] (Ibn Hi9m, 747-8);- the Document ( kitb ; wrongly called 'Constitution') of Medina (Ibn Hi9m, 341-4; Ab #Ubayd al-|sim b. Sallm, K. al-Amwl, ed. M. #Amra, Beirut 1989, 291-4) is a category in itself. It is an agreement between Muammad the Prophet and the believers and Muslims of | uray9 and Ya9rib, and those who follow them, join them, and strive along- | [IX:662b] side them, including Jewish groups (see muammad , at vol. VII, 367b, and the updated bibliography here below);- the lists which were mentioned above under fa'"il and ma9lib should in some cases be classified as documents. Lists of the first Emigrants, or of participants in certain battles, may have been taken over by the storytellers from government registers. (6) Speeches and sermons by the Prophet, e.g. his first addresses in Medina (Ibn Hi9m, 340-1); his speech at the Farewell Pilgrimage (ibid., 968-9). (7) Poetry. Story-tellers often interspersed their ma9znarratives with poetry. This has a function similar to that of speeches; it underlines a point or emphasises a dramatic moment by changing to another mode. Battling heroes exchange improvised poetry, as was the case in ayym al-#arab . This poetry is gen-erally of poor quality. Serious

poetry occurs as well, e.g. by Ka#b b. Zuhayr (his Bnat Su#d is the only ada in the sra) and 0assn b. 9bit [q.vv.]. A new kind of panegyric praises the Prophet, emphasising his mission, his spiritual qualities and those of his new religion. Certainly not all poetry ascribed to 0assn was composed by him, as Arafat has pointed out. Ibn Hi9m tends to place all occasional poetry on a certain event together, e.g. after his accounts of the battles of Badr, Uud and 0unayn, possibly because he took the narratives which he transmitted too seriously to contaminate them with doubtful verse. The simple, sometimes banal character of the poetry in the sra, as well as the often unlikely ascriptions may have led early critics to the verdict that much of it is inauthentic, i.e. not composed by the poets it is ascribed to. Ibn Hi9m expresses his doubts about authorship in many places. Ibn Sallm al3uma [q.v.] censures Ibn Is's unfamiliarity with poetry and his uncritically taking over of whatever poetry he found, be it ascribed to men who had never said a line of verse, to women or even to #$d and 9amd (abat ful al-9u#ar", ed. M.M. 9kir, Cairo 1974, 7-8, 11). Ibn al-Nadm accuses Ibn Is of having inserted poetry on request ( Fihrist 92). Authenticity. The question of the authenticity of the poetry has also been discussed by modern scholars (Kister, Monroe, Arafat), although it seems less urgent if one does not start from the assumption that the surrounding prose texts date back to the time of the Prophet. There is indeed no reason why some would not have included pieces of verse in his narrative, in order to comment on past events, or to make propaganda for certain factions of his own days. The poetry turns out to be easily interchangeable in different versions of a story. The sramaterials as a whole are so heterogeneous that a coherent image of the Prophet cannot be obtained from it. Can any of them be used at all for a historically reliable biography of Muammad, or for the historiography of early Islam? Several arguments plead against it: (1) Hardly any sratext can be dated back to the first century of Islam. (2) The various versions of a text often show discrepancies, both in chronology and in contents. (3) The later the sources are, the more they claim to know about the time of the Prophet. (4) Non-Islamic sources are often at variance with Islamic sources (see P. Crone and M. Cook, Hagarism). (5) Most srafragments can be classed with one of the genres mentioned above. Pieces of salvation history and elaborations on |ur"nic texts are unfit as sources for scientific historiographyexcept, of course, | [IX:663a] for the historiography of the image of the Prophet in the belief and doctrine of his community. The Document of Medina is generally considered authentic, i.e. dating back to the Prophet, but there is disagreement about the unity of the text and its attitude towards

(certain groups of) Jews, because the well-known Jewish tribes of Medina are not mentioned. Scholars, driven perhaps by a horror vacui, continue deriving historical facts from late sources. The last scholarly biography of Muammad is that by W. Montgomery Watt (Muhammad at Mecca and Muhammad at Medina , Oxford 1953, 1956), and a new one is unlikely to appear. G. Schoeler has recently published a monograph on the character and authenticity of Islamic tradition about the Prophet's life. To Muslims, the sra, which in the first centuries of Islam had been taken less seriously than ad9, gradually became almost a holy writ, whose reliability was accepted almost without asking questions. In reaction to the rise of historical criticism in the west, which often struck a patronising, if not resentful, note towards Islamic beliefs, some Muslims have felt the need vigorously to defend the veracity of the sra. The Life of Muammad by Muammad 0usayn Haykal (1935) is an example of an apologetic biography. A striking illustration of the attitude of modern Muslims towards the srais the scandal around the British author Salman Rushdie, who in his novel The Satanic verses (London 1988) has alluded to both traditional and self-invented details from the life of the Prophet, and has been subsequently severely attacked and threatened all over the Muslim world, notably in Iran. (W. Raven) Raven, W. | [IX:664a] The following is meant as a supplement to the bibls. in the arts. ibn hi9m ibn is ma9z muammad see also al-suhayl al-abar tafsr wahb b. munabbih #urwa b. al-zubayr al-wid al-zuhr. 1. General. GAS i, 275-302

P. Crone and M. Cook, Hagarism, Cambridge 1977 A.A. Duri, The rise of historical writing among the Arabs, ed. and tr. L.I. Conrad, introd. F.M. Donner, Princeton 1983 (translation of Ba9 f na9"at #ilm al-ta"r9 #inda 'l-#arab, with many new bibliographical references) M. Jarrar, Die Prophetenbiographie im islamischen Spanien. Ein Beitrag zur berlieferungs- und Redaktionsgeschichte. Frankfurt am Main etc. 1989 M.J. Kister, The Srah literature, in Camb. hist. Ar. lit., i, Arabic literature to the end of the Umayyad period, Cambridge 1983, 352-67 (an extended and updated version is forthcoming in L.I. Conrad (ed.), History and historiography in early Islamic times: studies and perspectives, Princeton) La vie du prophte Mahomet. Colloque de Strasbourg (23-24 octobre 1980), Paris 1983 G. Schoeler, Charakter und Authentie der muslimischen berlieferung ber das Leben Mohammeds, Berlin 1996 J. Wansbrough, The sectarian milieu: content and composition of Islamic salvation history, London 1978. 2. |ia /Wahb b. Munabbih. Wahb = R.G. Khoury, Wahb b. Munabbih, i, Der Heidelberger Papyrus PSR Heid Arab 23, ii, Faksimiletafeln, Wiesbaden 1972 (Codices Arabici Antiqui, i) Kister, Note on the papyrus account of the #Aqaba meeting, in Le Muson lxxvi (1963), 403-17 idem, On the papyrus of Wahb b. Munabbih, in BSOAS, xxxvii (1974), 545-71. 3. Ibn Is/ Ibn Hi9m (update). Ahmed Hebbo, Die Fremdwrter in der arabischen Prophetenbiographie des Ibn Hischm (gest. 218/834). Frankfurt am Main etc. 1984 Ibn Is = Srat Ibn Is al-musammt bi-Kitb al-mubtada" wa 'l-mab#a9 wa 'lma9z, ed. Muammad 0amd Allh, Rabat | [IX:663b] 1396/1976 Ibn Is, Kitb al-Siyar wa 'l-ma9z, ed. Suhayl Zakkr, Beirut 1389/1978 M. Muranyi, Ibn Isq's Kitb al-maz in der riwya von Ynus b. Bukayr. Bemerkungen zur frhen berlieferungsgeschichte, in JSAI , xiv (1991), 214-75 S.M. al-Samuk, Die historischen berlieferungen nach Ibn Isq. Eine synoptische Untersuchung, diss. Frankfurt am Main 1978 R. Sellheim, Muhammeds erstes Offenbarungserlebnis. Zum Problem mndlicher und schriftlicher berlieferung im 1./7. und 2./8. Jahrhundert, in JSAI , x (1987), 1-16.

4. Some late sracompilations. Mu9uly b. |il3, al-Zahr al-bsim f srat Abi 'l-| sim, ms. Leiden Or. 370 Ibn Sayyid al-Ns, #Uyn al-a9ar f funn al-ma9z wa 'l-9am"il wa 'l-siyar, ed. Y.B. #Abd al-Hd, Beirut n.d., and ed. M. al-#^d al-9a'rw, Medina 1992 Muammad b. Ysuf al--li al-9a"m, Subul al-hud wa 'l-ra9d f srat 9ayr al-#ibd[al-Sra 'l-9a"miyya], ed. M. #Abd al-Wid et alii, 4 vols., Cairo 13929/1972-9. 5. Legendary ma9z. R. Paret, Die legendre Maghz-Literatur. Arabische Dichtungen ber die muslimischen Kriegszge zu Mohammeds Zeit, Tbingen 1930 A. Schimmel, And Muhammad is His messenger , Chapel Hill 1985.6. Proofs of prophethood. Ab Nu#aym al-Ifahn, Dal"il al-nubuwwa, 0aydarbd 1320 Amad b. al-0usayn al-Bayha, Dal"il al-nubuwwa, ed. #Abd al-Ramn Muammad #U9mn, Cairo 1969, Medina 1389 Mward, A#lm al-nubuwwa, ed. h #Abd al-Ra"f Sa#d, Cairo 1391/1971 (Theological treatises on this subject:) Ab 0tim al-Rz, A#lm al-nubuwwa, ed. Sal al--w, Tehran 1977 #Abd al-3bbr b. Amad al-Hama9n, Ta9bt dal"il al-nubuwwa, ed. #Abd alKarm #U9mn, Beirut n.d.7. Documents. M. Gil, The Constitution of Medina. A reconsideration, in IOS , iv (1974), 44-66 M. Hamidullah, Documents sur la diplomatie musulmane l'poque du Prophte et des khalifes orthodoxes, Paris 1935 idem, Textes arabes, Cairo 1941 (repr. as Ma3m#at al-wa9"i al-siysiyya fi 'l-#ahd al-nabaw wa 'l-9ilfa al-r9ida, Cairo n.d.) M. Hamidullah, La lettre du Prophte Hraclius et le sort de l'original, in Arabica, ii (1955), 97-110 idem, Original de la lettre du Prophte Kisra illustr, in RSO , xl (1965), 57-69 R.S. Humphreys, Islamic history. A framework for inquiry, London and New York 1991, 92-8 M. Lecker, On the preservation of the letters of the Prophet Muhammad, in Conrad (ed.), History and historiography in early Islamic times U. Rubin, The Constitution of Medina. Some notes, in St. Isl. , lxii (1985), 5-23 J. Sperber, Die Schreiben Muhammads an die Stmme Arabiens, in MSOS As., xix (1916), 1-93.8. Poetry. W. #Arafat, Early critics of the authenticity of the poetry in the sra, in BSOAS, xxi (1958), 453-63

idem, An aspect of the forger's art in early Islamic poetry, in BSOAS, xxviii (1965), 477-82 Kister, The srah literature , 357-61 J.T. Monroe, The poetry of the srah literature , in CHAL, i, 368-73 M. Zwettler, The poet and the prophet. Towards understanding the evolution of a narrative, in JSAI , v (1984), 313-87. 9. Authenticity. Muammad 0usayn Haykal, 0ayt Muammad , Cairo 1935, tr. I.R.A. al-Faruqi. The life of Muammad , Indianapolis 1976 F.E. Peters, The quest of the historical Muhammad, in IJMES , xxiii (1991), 291-315 M. Rodinson, A critical survey of modern studies on Muhammad, in M. Swartz, Studies on Islam , New York/Oxford 1981, 23-85 A. Wessels, A modern biography of Muammad. A critical study of Muammad 0usayn Haykal's 0ayt Muammad, Leiden 1972. alMA$Z^ MA$Z^ (also ma9z 'l-nab, ma9z rasl allh ), a term which, from the time of the work on the subject ascribed to al-Wid (d. 207/823), if not earlier, has signified in particular the expeditions and raids organised by the Prophet the expeditions and raids organised by the Prophet Muammad Muammad in the Medinan period. The first such sortie is reported by al-Waid to have involved a party of thirty men led by 0amza b. #Abd al-Mualib, which in 1/623 briefly intercepted a |urash caravan heading for Mecca from Syria on the coastal route (other accounts differ). The last was an expedition in the direction of Syria by an army of 3,000 men under the command of Usma b. Zayd in 11/632, immediately following the Prophet's death. For al-Wid (largely followed by Ibn Sa#d), the compass of the term al-ma9z appears to have included important 9azwas (such as those of Uud, al-9anda, al-0udaybiyya, 9aybar, Mecca, 0unayn and Tabk; Badr is elsewhere styled a 9azwa, but not by alWid), small-scale sariyyas (such as that of 0amza), acts of assassination of individuals (notably Ka#b b. al-A9raf), and such other significant events of the period as the pilgrimage of 9/631, the i33at al-wad# of 10/632, and the death of the Prophet in 11/632. In this narrower sense of events in the Medinan period, principally expeditions, raids and assassinations numbering in excess of eighty, but also in a broader sense of the Prophet's biography and background, the term al-ma9z came to be rooted in the Islamic reflective tradition at an early stage and was perpetuated in later works both for the more serious purposes of #ilm , e.g. Ibn #Abd al-Barr (d. 463/1071), al-Kal# (d. 634/1237) and Ibn Sayyid al-Ns (d. 734/1334), and on a popular level, particularly with poetic embellishments reminiscent of chansons de geste (see Paret). It is this broader sense of the term ma9z which appears to have been prevalent in respect of the earliest Muslim transmissions relating to the period of the Prophet (see, for example, the comments by Sezgin, 251, 275). Moreover, while Sachau's suggestion that the term ma9z was a calque on the struggles of Christian spiritual athletes

(Das Berliner Fragment, 448) may be unconvincing, it is nonetheless apparent that it was the term ma9z that characterised the early transmissions and that the term sra emerged only later as signifying a literary genre relating specifically to the Prophet. In 1899, Hartmann (Die angebliche sra , 32) opined that Ibn Is hat keine sra geschrieben and it has since been proposed (Hinds, Ma9z and Sra ) (1) that not just Ibn Is but all transmitters and compilers before Ibn Hi9m (d. 218/834 or 213) who dealt with material about the period of the Prophet in general regarded that material as being about ma9z; (2) that Ibn Hi9m, in using the term ra, was simply introducing the bulk of the Ma9z of Ibn Is with a term which was in the idiom of his own time; and (3) that it was probably al-Wid (d. 207/823) who was the first to view ma9z and sra as fields of study which were related but could somehow be differentiateda view in which he was evidently followed by Ibn Sa#d (d. 230/845). Those scholars who concerned themselves with ma9z and pre-deceased Ibn Is (whose probable date of death was 150/767) included notably #Urwa b. al-Zubayr (d. 94/714), al-Zuhr (d. 124/742) and Ms b. #Uba (d. 141/758) (for further names and details, see Horovitz; al-Dr, Ba9 , 20-7, 61-113; and | [V:1162a] Sezgin, 276-87); in addition, a Kitb al-Ma9z is ascribed to each of seven compilers who died after Ibn Is, but before al-Wid (viz. Ma#mar b. R9id (d. 154/770), Ab Ma#9ar Na3 b. #Abd al-Ramn (d. 170/786), #Abd al-Malik b. Muammad b. Ab Bakr (d. 176/792), #Al b. Mu3hid (d. 182/798), Yay b. Sa#d al-Umaw (d. 194/809), al-Wald b. Muslim al-Umaw (d. 195/810), and #Abd Allh b. Wahb (d. 197/812), while the few works dating from the same period which contain the term sra in their titles bear no relation to the Prophet (see Hinds for references). Just how broad the scope of these early ma9z works was is difficult to determine, since (apart from a few papyri) they survive only in later recensions (in the case of the work of Ibn Is) or as fragmented citations in later compilations (in the case of the other works). While the existing recensions of Ibn Is's work are restricted in scope to the period and background of the Prophet, it has been proposed by Horovitz (The earliest biographies, 166-7) in respect of the Kitb al-Ma9z of Ms b. #Uba that work may have included material relating not only to the period of the first four caliphs but also to the Umayyad period. Although Schacht (On Ms b. #Uqba's kitb alma9z , 296) did not accept this proposal, other evidence has since become available to support the view that, in early Islamic times, the subject matter of ma9z was drawn at least from the period of the first four caliphs in addition to that of the Prophet. This evidence is to be found in the Kitb al-Ma9z contained within the Muannaf of the Yemeni muaddi9 #Abd al-Razz b. Hammm al--an#n (126-211/744-827), where the majority of the reports bear the isnd Ma#mar b. R9id from al-Zuhr and presumably reflect the view of those two authorities about what constituted the proper subject matter of ma9z. The account starts with the digging of the well of Zamzam (as does Ynus b. Bukayr's recension of Ibn Is's Ma9z), moves on to the background of the Prophet and the main events of his lifetime, and then touches on various events after the Prophet's death; these events include the bay#a of Ab Bakr at the Safat Ban S#ida, Ab Bakr's appointment of #Umar as his successor, the conflict between #Al and Mu#wiya, the 9r , the 9azwa (sic) of al-|disiyya, and the marriage of Fima. The published papyri shed little light on the general character of early ma9z works. The earliest relevant document so far known is an eight-line papyrus text dated by Grohmann to the early 2nd/8th century, which contains some details about the battle of

Badr ( Arabic papyri from ]irbet el-Mird, Louvain 1963, 82-4, 105). Kister (Notes on the papyrus text) has pointed out that the last section of the Chicago papyrus (Or. Inst. 17635) dealing with Badr, Bi"r Ma#na and the B. al-Na'r, which is dated by Abbott to the late 2nd/8th century and attributed by her to Ma#mar b. R9id (i, document 5), is textually almost identical with a passage in Ab Nu#aym's Dal"il al-nubuwwa (Hyderabad 1320, 176), where the isnd ends with Ibn Lah#aAbu 'l-Aswad (d. 131/748)#Urwa b. al-Zubayr; Kister himself is inclined to ascribe authorship (sic) of the papyrus to Ibn Lah#a (d. 174/790). The early 3rd/9th century Heidelberg papyrus on the meeting at al-#Aaba, the conference of |uray9 in the dr al-nadwa , the hi3ra , and the expedition against 9a9#am, which has an isnd going back to Wahb b. alMunabbih, is accorded the | [V:1162b] Arabic title ma9z Rasl Allh by Khoury, although this phrase nowhere appears in the text itself. It is therefore clear that only the eight-line 9irbat al-Mird papyrus can conceivably be regarded as being from the earliest (i.e. pre-#Abbsid) period, and this is obviously too truncated to provide any definitive picture of the shape of the ma9z works which are reported to have existed at that time. Otherwise, we have only citations at one remove or more in later works, and later references to the existence of early ma9z works. One such report, which has been noted in modern studies, is given by al-Bal9ur (d. 279/892), with an isnd going back to al-Zuhr, according to which the caliph #Abd alMalik, on seeing one of his sons with ad9 al-ma9z , gave orders for it to be burned and enjoined him (instead) to recite the Kitb Allh and know and act according to the sunna (Anonyme arabische Chronik, Band XI, 172; Goldziher, Muh. St., i, 206 (Eng. tr. 191-2); Abbott, i, 17). Goldziher, while remarking that there seems to be nothing against admitting the existence of such literature in early times, also comments that the text of this account unmistakeably bears the stamp of those circles who condemned unauthenticated ma9z in favour of authentically recommended traditions, and in this connection he goes on to cite the saying attributed to Ibn 0anbal that three things have no al : tafsr , the malim and the ma9z (see also Goldziher, Richtungen, 57). The purpose of, and the climate of opinion about, the study of al-ma9z from the second half of the 2nd/8th century onwards are, however, matters requiring further research and only a few brief additional comments can be made here. Ibn Is is reported to have committed his material on al-ma9z to writing for the #Abbsid caliph al-Manr, who instructed his son al-Mahd to study the subject with that compiler (Abbott, i, 89; Watt, The materials used by Ibn Isaq, 31); and Hrn alRashd's chief ' , Ab Ysuf, who had been Ibn Is's pupil (Abbott, i, 92), is said to have been interested in ma9z, tafsr and ayym al-#arab (Goldziher, Muh. St., ii, 207). It is apparent that one practical application of the study of ma9z was in that area of fih called siyar, which was concerned with the rules of war and of dealings with non-Muslims, apostates and rebels. The term siyar appears in the corpus iuris ascribed to Zayd b. #Al ( kitb al-siyar, bb al-9azw wa 'l-siyar), is supposed to have been used by Ab 0anfa (d. 150/767), was the name by which the work of al-Awz# (d. 157/774) was known to Ab Ysuf and al-9fi#, and figures in the titles of two works by al-9aybn (d. 189/805); it was, moreover, a term which, as the 8th/14thcentury lexicographer al-Fayym explains, took over from [the term] al-ma9z on the tongues of the fuah". While this seems in general to be true, the same type of material also appeared elsewhere under other rubrics: the appropriate section in the Muwaa" of Mlik is entitled neither ma9z nor siyar but 3ihd ; in the Muannaf

of #Abd al-Razz b. Hammm al--an#n the kitb al-3ihd includes the sort of material that the fuah" were styling siyar, but the work also includes a kitb alma9z (as noted above); the later al-Bu9r has separate books on 3ihd ( bb 1; fa'l al-3ihd wa 'l-siyar) and on ma9z, but he seems to be the exception in this respect among the 3rd/9th century compilers of ad9 materialthe other compilations contain books on 3ihd (and, in some cases, siyar) but not | [V:1163a] on ma9z (for references, see Hinds, Ma9z and Sra ). This declining interest in ma9z as such was accompanied by the emergent view that Ibn Is had drawn material from Jews and Christians (Yt, Udab", vi, 401; Ibn 0a3ar, Tah9b, ix, 45) and that he and [at least some] other ab al-ma9z had 9#i inclinations (Udab", vi, 400); in addition, Ibn 0anbal is reported as having said that Ibn Is's handling of isnd s rendered him unsuitable as an authority for ad9 (Tah9b, ix, 43), and the same consideration may have been behind his reported denunciation of alWid as ka99b (Tah9b, ix, 364). There remains the question of how useful the ma9z material is for the purpose of historical reconstruction of the early Islamic period. Here it may be remarked first that there are considerable chronological and other discrepancies between the accounts available; these have been discussed most recently by Jones (On the chronology ...) and Kister (The expedition of Bi"r Ma#na ). In more general terms, however, Watt sees little ground for doubting the truth of the main events of the ma9z and is of the opinion that the ma9z-material (in the special sense of the main outline of events, and omitting all anecdotes) ... is an essential foundation for the biography of Muammad and his time (The materials used by Ibn Isq, 28). On the other hand, Sellheim has attempted to discern the main tendencies of Ibn Is according to layering of sources (Quellenschichtung), involving an original layer which reflects historical reality, a further layer made up of legendary material about the Prophet, and a top layer reflecting political tendencies of Ibn Is's own time (Prophet, Chalif und Geschichte). Such an approach is not wholly incompatible with Schacht's conclusions that an isnd cannot alone guarantee the authenticity of a work ascribed to an author of the early 2nd/8th century and that in general the more perfect the isnd , the later the tradition ( JRAS [1949], 147). From this standpoint, Schacht has examined the short Ms b. #Uba text (an edition of the 4th/9th century) published by Sachau and has challenged Sachau's view that the text contains no elements from the #Abbsid period. Schacht has seen the growth in the 2nd and 3rd centuries of traditions ascribed to Ms as typical of the way in which spurious information was put into circulation, and has concluded that the whole of the standard biography of Ms in the later works is without documentary value, particularly the touching picture, taken seriously by Sachau and Horovitz, of his regular lectures to a circle of pupils in the mosque of Medina. This presupposes the concept of Medina as the home of Islamic learning, a concept which was as yet unknown to Shfi# (d. 204) (On Ms b. #Uqba's kitb al-ma9z , 299). More recently, Wansbrough, who also doubts the feasibility of historical reconstruction of the Islamic period up to ca. 200 A.H., has viewed the sra-ma9z literature (as he terms it) as an Islamic adaptation of Biblical salvation history. In his opinion the earliest expression of Islamic soteriology consisted in membership of the umma and it is in the sra-ma9z literature that the earliest formulation of Muslim identity is contained (The sectarian milieu, 89), the conceptual motive in this (as in scripture and sunna ) being polemic (103). Wansbrough sees a transition from the sra-ma9z

literature, where ecclesia is the dominant cognitive category and precedent is historically articulated, to the sunna-ad9 literature, where nomos is the | [V:1163b] dominant cognitive category and precedent is idealized and hence shorn of its historical dimension (87); for him, the Ibn Is-Wid-Bu9r development from sra to sunna marked a passage from loosely structured narratio to concise exemplum (77-8). In modification of this view, however, it should be remarked that the signification of sra was also close to that of exemplum and that the passage was rather from ma9z to sunna via sra , the sra material being essentially ma9z material viewed in a new light. It can also be noted that, while the narrowing-down of the scope of the ma9z to the period and background of the Prophet seems to have been conventional from the 3rd/9th century onwards, it is nonetheless difficult to discern a point at which al-Wid's even narrower definition gained any exclusive currency. It seems rather that the two senses of ma9z co-existed: in the broader of the senses, the term echoed an earlier scope which had been yet broader (note particularly the title of al-Kal#'s work) and seems to have been used more or less synonymously with the term sra as a genre label; the narrower sense appears to have been a more technical one, i.e. the ma9z proper, as distinct from the mab#a9, for example. The term sra , while occurring as a genre label more or less synonymous with ma9z, also came to signify pre-eminently the account of the Prophet's life and background as transmitted by Ibn Hi9m on the basis of the work of Ibn Isthe ma9z which became sra as exemplum, only to be largely superseded by the sunna-ad9 literature. ABA|$T (a.) , pl. of abaa , everything which is related to another and which is similar or analagous to it, which comes to mean a layer of things of the same sort (Flgel, Classen, 269, n. 1). From this a transition can be made to the idea of a rank, attributed to a group of characters who have played a role in in history history in in one capacity or another, classed according to criteria determined by the religious, cultural, scientific or artistic order etc. (Hafsi, i. 229;cf. al-Tahnaw, Ka99f, 917). In biographical biographical literature literature it is the book of classes of characters arranged by categories and organised into generations. A. Lexicography and literature. 1. This term does not appear in the |ur"n, but two other expressions approaching it do: aba and ibanalagous things which follow each other (in a temporal or qualitative sense) or placed on top of each other (in a spatial sense); You shall surely ride stage after stage ( LXXXIV, 19, tr. Arberry: abaan #an abain from one state to another, or from one calamity to another; see al-Tabar, Tafsr); [God] who created seven iban (the ranks or stages of the heavens, LXVII, 3; LXXI, 15). The common point | [X:8a] of reference is the idea of covering everything by something equivalent, of applying oneself to it ( ms , s.v.). The idea of equivalence is again found in abaa a similar epoch (al-arn min al-zamn). According to al-Ama#, ib designates a

group of people; for Ibn al-A#rb (d. 231/846) aba reflects a given state [or category] whatever its sort (al-l #al 9tilfih). So does abaa , according to alLay9: kna fulnun #al abatin 9att min al-duny: ay lt; K. al-#Ayn; M.-N. Khan, Die exegetischen Teile des Kitb al-#Ayn, Berlin 1994, 220, or again al-umma ba#d al-umma one community succeeding another. For Ibn Sduh, aba is agroupof people who correspond to an analogous group. The variant ib designates a vast number of people, grasshoppers,camels, etc. (LA and al--a9n, Takmila..., ed. Muaf al-0i3z, Cairo 1988, s.v.; Ibn Sduh, Mu9aa, ix, 118). According to al-Lay9, abaa , which may be aba in the masculine, is used as a unitary form of the noun of action ib. Numerous other meanings are to be found in Lane, s.v. 2. In adab and historiography, abaa is in common use in the sense of category or class, in particular of society: Ibn al-Muaffa#, Risla fi "l-aba, ed. and tr. Ch. Pellat, Paris 1976, 31; Ibrhm b. M. al-9aybn (d. 298/911) according to al-#Id al-fard, ed. Tarn, iv, 262-3; G. Makdisi, The rise of humanism, Edinburgh 1990, 233-4. As for al-3i, he uses it in the sense of degree, as in al-9akk f abatin doubt is made up of degrees ( 0ayawn , vi, 35, 37, Jhiz, Le cadi et la mouche, tr. L. Souami, Paris 1988, 74, 75); abaat ma#nh degree or level (of meaning) (op. cit., i, 10, Jhiz, 231; cf. i, 98). (Cf. Ibn 9aldn, Muaddima , 1073, tr. Rosenthal, ii, 344: abat alkalm.) What is more, in his work the meaning of social categories/classes is often associated with types of character: misers, singers, singing slave girls, traders, secretaries, Turks, etc. (Ch. Pellat,Arabische Geisteswelt, Zrich 1967, 48-9, 436 ff.; S. Enderwitz, Gesellschaftlicher Rang und ethnische Legitimation, Berlin 1979, 72-3, passim: al-3i on the Africans, the Persians and the Arabs). Finally, the notion of abaa applied to poets has been attested at least since the second half of the 2nd/8th century; see al-Ama# (d. 213/828), K. Fulat al-9u#ar", ed. Torrey, in ZDMG, lxv (1911), 495, 499. As to the following Prophetic tradition reported by Anas, it is very obviously spurious: My community will be made up of five classes: firstly forty years with charitable and pious people; they will be followed for the next 150 years by people who will live in compassion and mutual harmony; then for 160 years more there will come people who will turn their backs on each other and will separate themselves; then will come a period of scattering (har3) [and of war or of flight] and every-man-for-himself (na3). In another version it is said that each class would last for forty years and that another class would be added between year 40 and year 80 to arrive at the number of five (Ibn M3a, Sunan , 36, Fitan, no. 4058; cf. Ibn al-3awz, Maw9#t, iii, 196; idem, Tal, 714, several versions). It is possible that it may have been modelled on the following tradition: The best of men are those of my century (arn), and below them are those of the next century (al-Bu9r, 62, Fa9"il al-aba, i, tr. Houdas, ii, 583). In modern texts, the term is accepted most clearly to designate a social class, as in ir# al-abat the class struggle. B. The division into classes. 1. Origin and meaning.

For several scholars, the origin of this division in | [X:8b] Arab biographical literature is found in the criticism of tradition (Loth, 594 ff.). It has even been written that the genre of the abat was born within the framework of the ad9and is inseparable from it (Hafsi, i, 227). What supports the thesis of Hafsi isthat the first book of classes was perhaps the K. abat al-muaddi9n of al-Mu#f b. #Imrn al-Mawil (d. 184/800; Sezgin, i, 348; Hafsi, i, 241). One argument against his position would be the K. abat ahl al-#ilm wa "l-3ahl of Wil b. #A" (d. 131/748), but the subject matter is not known: was it the orthodox believers, i.e. the |adars and the ignorant, i.e. the predestinationists (Van Ess, TG, v, Berlin 1993, 137-8)? For Heffening, on the other hand, this grouping much rather owes its origin to the interest of the Arabs in genealogy and biography. Rosenthal, 93-5, for his part, considers that the division is genuinely Islamic and that it would seem to be the oldest chronological division which presented itself to Muslim historical thinking. It was the natural consequence of the concept of the Companions of Muammad, the Followers, etc., which in conjunction with the isnd criticism of traditions developed in the early second century of the hi3ra . Without denying the fundamental role which it played in the birth and development of the genre, it does not seem that it originated from the genre, as the semantic survey above (cf. Heffening) would suggest. The ideas of covering, of egality, of analogy (cf. also arn, which perhaps preceded abaa in the sense of generation, Rosenthal, 93, and which also has the connotation of analogy) and of succession which this term conveys, correspond well to the Muslim concept of the history of salvation, with the succession of pious men, beginning with the prophets, whose characters were so many models to be imitated. Even if tribal genealogy continued to exist, it gave way more and more to a particular form of spiritual or intellectual genealogy which also appeared, of course, in the ad9, the transmission of knowledge, but also in other disciplines. In addition, by the use of certain types of abat every effort was made to maintain the link with the primitive community which was widely mythologised. Finally, the fact that al-Ama# (see above) had already used the term abaa , however loosely, to compare two poets, and that al-3uma (d. 232/846) organised his abat al-9u#ar" (see Kilpatrick) according to an order which has nothing to do with religious merit, about the same epoch as Ibn Sa#d (d. 230/845) composed his own work, suggests that the genre in its origins was part of a global preoccupation of all scholars in different fields: to give to society the canons for transmitting knowledge, whether sacred or secular, and in particular by means of a biographical tool. This concern for continuity (Khalidi, 46-8, 205 and n. 50) insists at one and the same time on sacred history continued and on the equally secular aspects of the genre deeply rooted in its origins, also apparent in the genre of the aw"il [q.v.], which was attested at least since the time of Ibn 9ayba (d. 235/849; see book 34 of his Muannaf , Beirut 1995, vii, 247-76). It is not fortuitous if in Tal, 461-8, the section concerning them follows that on the abat. The interest in genealogy understood in that way was specified above, and can also be observed in the role which local stories play in the evolution of the genre, with certainly a touch of regional pride, but especially in order to justify the juridical practices in use in one place or another (Rosenthal, 94). Already | [X:9a] by this time, Ibn Sa#d had given a special place tothe grouping according to the capital cities and towns (Mecca,

Medina, Bara, Kfa), or even events (Badr) but the History of Wsi of Ba9al (d. 292/905 [q.v.]; ed. K. #Awwd, Ba9dd 1967; Rosenthal, 166-7) is essentially a work about the classes of traditionalists in this town. Later this division was extended to all sorts of persons, but generally scholars. 2. Criteria of classification. For the classification of the Companions, especially in the work of Ibn Sa#d, see aba . For the Successors, see tbi# . For both, see al-0kim al-Nsbr, Ma#rifat #ulm al-ad9, chs. 7, 14 (twelve classes of Companions, fourteen classes of Successors); al-Suy,Tadrb al-rw, 221-2, 234 ff., ch. 39-40, according to precedent; Marais, 222-4; Hafsi, i, 242-4, 236-8. It is difficult to give general criteria for classification for all the abat; four can be distinguished: moral and chronological, relationship with the Prophet for the first generations, chronological, and finally a late classification where alphabetical order is used (Hafsi, i, 234-6). For the classes of traditionalists, the encounter (luy) between master and disciple is a fundamental criterion for distinguishing between the two classes (#Umar, 51). The principles of hierarchisation and also of illustration of the forged ad9cited above, are seen in the original grouping which goes back to Ab lib al-Makk (d. 386/996). He distinguished five classes of forty years up to his era, citing five names for each one: caliph, jurist, traditionist, reader and ascetic (Tal, 714-17, takes up this classification which was continued by others until 560 A.H., perhaps some 40 classes). The organisation of works into classes did not seem very practical, as would appear in the work of al-9ahab: Ta9kirat al-uff comprised twenty-one (80 years); Ma#rifat al-urr", seventeen; Siyar a#lm al-nubal", about forty (from seven to thirty years); Ta"r9 al-Islm[i-xxvii (up to 400 A.H.). ed. #U.#A. Tadmur, Beirut 1987-92; i-iv (611-40 A.H.), ed. B.#A. Ma#rf et alii, Beirut 1988]; seventy (in general ten years). In this work he associates chronological organisation with organisation into classes, but in that way the traditional principle of the encounter is abandoned. Furthermore, in two of his works he designates each class by one of its illustrious representatives, cf. the class of al-Zuhr. Thus he continues in al-Mu3arrad f asm" ri3l K. Ibn M3a (eight classes, Ma#rf, 103; #Umar 49-50; Sezgin, i, 148; ed. Fayal al-3awbira, Riy' 1988) and in al-Mu#n f abat al-muaddi9n [Gilliot, in MIDEO, xix, no. 105, mistaken by Hafsi, 31, for Ta9kirat al-uff], where the first classes have names, e.g. the class of al-A#ma9 and of Ab 0anfa, then from the 3rd/9th centuries onwards he has recourse to the classes of twenty to thirty years. C. Works in the genre. See 033 9alfa, ed. Flgel, nos. 7879-7932. The lines which follow are the addenda (sometimes the corrigenda) to Hafsi, in particular the editions of texts which have appeared since. Philologists (Hafsi, ii, 155-61) and poets (iii, 50-61): Ibn al-Anbr, Nuzhat al-alibb" f abat al-udab", ed. I. al-Smarr" (Ba9dd 19702); Ibn |' 9uhba (d. 851/1448 [q.v.]), abat al-nut wa "l-lu9awiyyn, ed. M. 9ayy', Na3af 1974.

Readers and exegetes (Hafsi, ii, 2-7): Ibn al-3azar [q.v.],K.Ma#rifatal-urr"alkibr#all-abatwal-a#r, i-ii, ed. M.S. 3d al-0a, Cairo 1969; Dwd (M. b. #A., d. 945/1538), abat al-mufassirn, ed. #A.M. #Umar, Cairo 1972, Beirut 1983. | [X:9b] Traditionists and associates (Hafsi, i, 241-65): 9alfa b. 9ayy [q.v.]; Muslim, K. alabat (Hafsi, i, 248-9), ed. S.#A.M. al-|aza, announced in ATA , xxxv (1988), 17; Bard3 (A. b. Hrn, d. 301/816; Sezgin, i, 166; Hafsi, i, 249-50), abat al-asm" al-mufrada fi l-aba wa "l-tbi#n wa-ab al-ad9, ed. S. al-9ihb, Damascus 1987; contrary to Sezgin, i, 350, al-Azd (Ab Zakariyy" Yazd b. M., d. 334/935), K. al-abat, lost work, which is different from Ta"r9 al-Mawil , ed. #A. 0abba, Cairo 1967, 11; Ab 9ay9 (#Al. b. M. b. 3a#far, d. 369/979; Hafsi, 25), abat al-muaddi9n bi-Ibahn, ed. #A.S. al-Bundr, i-iv, in two vols., Beirut 1989; #Al b. al-Mufa''al (al-Madis al-Iskandarn al-Mlik, d. 611/1214; Hafsi, i, 256),al-Arba#n al-murattaba #al abat al-arba#n, ed. announced in ATA , xl-xli (1989), 15. 0anafs (Hafsi, ii, 11-17): Ibn Abi "l-Waf" al-|ura9 (d. 775/1373), i-v, ed. #A.M. al0ulw, Cairo 19932, see Gilliot in MIDEO, xxii, 191; M. b. #U. al-0anaf (d. 959/1551), add. Hafsi, ii, 15, n. 4: ms. Ali Emiri 2510; al-Hinn"i (d. 979/1572), abat alanafiyya: add. Hafsi, ii, 16, n. 1: Ba9dd, Awf 929-30; al-9azz (A. b. #Ak. alTamm, d. 1004/1595), al-abat al-saniyya f tar3im al-anafiyya, ed. #A.M. al0ulw, Cairo 19892 (19701). Mliks (Hafsi,ii, 9-11): #Iy'b. Ms [q.v.],Tartbal-madrik, i-viii, ed. M.T. alan3 et alii, Rabat 1966ff. (19832), preferable to the edition of A. Bakr Mamd, iiii, Beirut 1965-8; Ibn Farn, al-Db3 al-mu9ahhab, i-ii, ed. M. al-Amad Abu "l-Nr, Cairo 1972; continued by Amad Bb al-Takrr al-Tinbutkt (d. 1036/1627; Brockelmann, II, 176), Nayl al-ibtih3, ed. #A. b. #Al. al-Harlama, Tripoli (Libya) 1989. 9fi#s (Hafsi, ii, 17-24; introduction to al-#Abbd by G. Vitestam, K. abat alfuah" al-9fi#iyya , Leiden 1964, 3-5; introduction of 9n, see below under Ibn |' 9uhba): Muawwi# (#U. b. #A., d. ca. 440/1048); Abu "l-ayyib Sahl al--u#lk (d. 404/1013-14); 9n, 10, according to 033 9alfa, no. 7900; al-Subk (T3 alDn, q.v.): abat al-9fi#iyya al-kubr, i-x, ed. al-ann and al-0ulw, Cairo 196476; al-Asnaw (#Abd al-Ram b. al-0asan, d. 772/1370), abat al-9fi#iyya, i-ii, ed. #Al. al-3ubr, Ba9dd 1970-1 (Riy' 1981); Ibn |' 9uhba, abat al9fi#iyya , i-iv, 0aydarbd 1978-80, i-iv in 2 vols., ed. 0.#A. 9n, Beirut 1987; Ibn Ka9r (#Imd al-Dn, d. 774/1373), abat al-fuah" al-9fi#iyyn, with the 9aylof al-Maar al-#Ubd (d. 765/1363), i-iii, ed. M.Z.M. #Azab, Cairo 1993 (Gilliot, in MIDEO, xxii, no. 192, and corr. in MIDEO, xxiii, add. Hafsi, ii, 21: Ibn Mulain (A.0af #U. b. #A., d. 804), al-#Id al-mu9ahhab f amalat [corr. Hafsi: 3umlat] al-ma9hab, ms. DK 579 ta"r9). 0anbals (Hafsi, ii, 24-6): Ibn al-Mabrid (or Ibn #Abd al-Hd, d. 909/1503), al3awhar al-muna99ad f abat muta"a99ir ab Amad , ed. A.S. al#U9aymn, Cairo 1987 (Gilliot, in MIDEO, xix, no. 106); al-#Ulaym (#Ar. b. M. al-

#Amr (d. 928/1521), al-Manha3 al-amad f abat al-imm Amad , ed. M.M. #Abd al-0amd, Cairo 1965. Mu#tazils (Hafsi, iii, 175-6, Madelung, 330): M. b. Yazd9 al-Ifahn (last wrote 3rd/9th century; Madelung), K. al-Mabi; Abu "l-0. b. Farzawayh, a disciple of Ab #Al al-3ubb#, K. al-Ma9yi9;#Abd al-3abbr, abat al-mu#tazila (ten classes), with the addition of two supplementary classes by al-0kim al-3i9um, in Fa9l al-i#tizl wa-abat al-mu#tazila , ed. F. Sayyid, Tunis 1974; Ibn al-Murta', abat al-mu#tazila , ed. S. Diwald-Wilzer, Wiesbaden 1961. Overall, see Gilliot in MIDEO, xix, no. 56. A9#ars: Ibn Frak, K. abat al-mutakallimn, prob- | [X:10a] ably the oldest (Hafsi, iii, 180; Madelung 334), and Kaml al-Dn b. Imm al-Kamliyya (d. 864/1460; al-Sa9w, 4aw", ix, no. 259), abat al-a9#ira are not preserved (Hafsi, ii, 26; Madelung, ibid.); Ibn #Askir [q.v.], Tabyn ka9ib al-muftar, divides them into five classes. Ib's (Hafsi, iii, 176): al-Dar3n (d. 626/1229 [q.v.]), K. al-Ma9yi9 fi "lMa9rib (abat ma9yi9 al-ib9iyya), i-ii, Beirut 1974. 9#s and Zayds (Hafsi, iii, 171-5): al-Bar (Ab 3a#far, d. 280/893), K. alRi3l, ed. 3. Muaddi9 Urmaw, Tehran 1964; al-Ka99 [q.v.], K. al-Ri3l, ed. S.A. al-0usayn, Karbal ca. 1960/ I9tiyr ma#rifat al-ri3l(summary by al-s), ed. 0. Muafaw, Ma9had 1970. Ascetics and mystics (Hafsi, ii, 27-41): Ibn al-Mulain, abat al-awliy", ed. N. 9arba, Beirut 19862 (19731); al-Munw (#Abd al-Ra"f [q.v.]), al-Kawkib aldurriyya f tar3im al-sda al-fiyya (al-abat al-kubr; first complete ed.), i-iv, in 2 vols., ed. #A.-. 0amdn, Cairo 1994 (see Gilliot, in MIDEO xxiii). Physicians and sages (Hafsi, iii, 161-5): -#id al-Andalus (d. 462/1070), abat alumam, add. Hafsi,iii, 161, ed. L. Cheikho, Beirut 1912; ed. 0. B #Alwn, Beirut 1985; M.S. Khan, Q9 -#id al-Andaluss abat al-umam, in Islamic Studies, xxx/4 (1991), 517-40; missing from Hafsi are the -iwn al-ikma, wrongly atributed to Ab Sulaymnal-Si3istn [q.v.], and Tatimmat -iwn al-ikmaof ahr al-Dn alBayha [q.v.], new ed. R. al-#A3am, Beirut 1992. Others: Mlik (A. Bakr #Al. b. M., d. 453/1061; Hafsi, iii, 166), K. Riy9 al-nufs f abat #ulam" al-ayrawn wa-Ifriya, i-iii, ed. B. al-Bakk9, Beirut 1983; Burayh (#Abd al-Wahhb b. #Ar. al-Saksak, d. 904), abat al-mu3tahidn, ed. Ab #Abd al-Ramn Ibn #Al, in Rislatn li-Ibn Kaml B9 wa9kubrzdah, Cairo 1976. | [X:10b] 3azrat al#ARAB #ARAB , the Island of the Arabs, the name given by the Arabs to the Arabian Peninsula Arabian Peninsula. | [I:534a] (i) preliminary remarks

Although the Peninsula may not be the original cradle of the Arab people, they have lived there for thousands of years and regard it in a very special sense as their homeland. For students of Islam, Western Arabia occupies a unique position as the land in which the Prophet Muammad was born, lived and died. It was there that the inspiration of Allh descended upon the Prophet, and to this Holy Land come many thousands of Muslims every year from all parts of the Islamic world to make the pilgrimage to the Ka#ba, the House of Allh in Mecca (Makka), and to visit the Prophet's tomb in Medina (al-Madna al-Munawwara). The Peninsula has the shape of a rough quadrilateral with a length of c. 2200 km. from north-west to south-east and a breadth of c. 1200 km. The symmetry of the quadrilateral is marred by the bulge of Oman (#Umn) on the eastern side reaching out close to the Iranian coast. On the west, south, and east the Peninsula is clearly defined by the Red Sea (al-Bar al-Amar), the Gulf of Aden (9al3 #Adan), the Arabian Sea (Bar al#Arab), the Gulf of Oman, and the Persian Gulf (al-9al3 al-Fris). In the north, the Arabs themselves have often disagreed as to where Arabia ends and Syria (in the broad sense) begins. A vast steppe unrolls northwards from the Great Nafd with no natural feature suitable as a limit for the Peninsula. For the purposes of this article the Peninsula is considered as extending only to the borders separating Saudi Arabia and Kuwayt from Jordan and #Ir, even though these borders represent little more than artificial political concepts. This definition places the northernmost point of the Peninsula at #Unza, a low mesa in the desert farther north than either Jerusalem of #Ammn. From #Unza the borders between Saudi Arabia and Jordan, not yet fully agreed upon, reach the sea near the head of the Gulf of al-#Aaba, while the borders between Saudi Arabia and Kuwayt on the one hand and #Ir on the other run to the head of the Persian Gulf south of al-Bara. Along these eastern borders lie two small neutral zones, in one of which Saudi Arabia and #Ir and in the other Saudi Arabia and Kuwayt share undivided half interests. It is impossible to make a reasonably reliable estimate of the size of Arabia's population. All figures found in reference works are highly suspect, as none is based on proper statistics or sufficient familiarity with the whole Peninsula. In view of the extensive areas inhabited solely by scattered nomads and the relatively light density of population in most of the settled areas, one may doubt whether the total approaches 10,000,000, and it may well fall several millions short of this figure. The most densely populated country is the Yaman (al-Yaman). In Saudi Arabia the main concentrations are in a few cities of al-0i3z, the well watered mountains and plains of #Asr and its Tihma, some of the valleys of Na3d, and the eastern oases of al-0as and al-|af. 0a'ramawt and Oman both contain many towns and Bedouin tribes. Present state of knowledge. The inhabitants of Arabia have naturally always known much about the land, but each man's knowledge is restricted to a certain region, being detailed and particularistic rather than general and comprehensive. No single work in Arabic gives a full and accurate description of Arabia. The best volume in the language is still -ifat 3azrat al#Arab by al-Hamdn (d. 334/945-46), which, though rich in information, fails to provide a coherent panoramic view of the whole Peninsula. | [I:534b]

The serious scientific exploration of Arabia began with Carsten Niebuhr and the Danish expedition of 1762. While travellers of different nationalities pressed on with the penetration of the interior during the 19th century, British officers of the Indian Government undertook technical surveys of the surrounding seas and stretches of the coast. Technical surveying in the interior had to wait for the 20th century, when it began with an investigation of the southern border of the Yaman and preliminary studies for the 0i3z Railway. In recent years oil companies have surveyed large parts of Eastern Arabia, using the highly refined methods of modern geological and geophysical exploration, besides engaging in extensive reconnaissance in other regions. By 1374/1955 travellersboth Western and Arabhad visited virtually all of the remoter places, so that none of the old major mysteries regarding the surface of the land had been left unsolved. Traveller's reports, however, are often incomplete and sometimes inaccurate, and much remains to be done in checking and correlating those now available. A number of important reports remain unpublished or buried in archives. Recent years have also been seen the introduction of aerial photography as an indispensable procedure in mapmaking. By 1954 a good part of the Peninsula had been photographed for cartographic purposes, and some of the results had already been transferred to maps. Aerial photographs, however, are of maximum value only if supported by ground control, i.e., the establishment of fixed points on the ground whose relationship to the photographs is precisely determined. For much of Arabia such control is still lacking. The general outlines and main features of the map of Arabia have now been delineated with a fair degreeand in a few instances a high degreeof reliability, but years of study lie ahead before all the details can be filled in. Surveys done in the earlier days, such as those of the Persian Gulf, are now being redone in the interests of greater thoroughness and accuracy. Errors of the past, many of which have become established on maps, are being corrected, but the process is long drawn out. Arabian governments are now making available information about their countries in a growing body of official publications, and modern Arab authors keep producing books and articles dealing with different parts of the Peninsula. Interest in such diverse things as oil and South Arabian antiquities has called forth a flood of material by Western authors, part of which is sound but much of which is superficial, misleading, or flagrantly contradictory to fact. Arabic sources likewise are often unreliable, so that the student of Arabia must constantly be on the lookout for pitfalls along his path. (ii) physical structure and principal geographical features Lying between Asia and Africa, Arabia is of such size and individuality of character as almost to justify its classification as a sub-continent. Usually considered an appendage of Asia, it also joins Africa through Sinai, which, though politically a part of Egypt, is closer to Arabia in both physical environment and the nature of its human life. Before the development of rift valleys provided a bed for the Red Sea, Western Arabia formed a part of the African land mass, and the southern half of Western Arabia still has a greater affinity in many ways with | [I:535a] Somalia-Ethiopia than with Northern Arabia or the rest of Asia. Northern Arabia, on the other hand, merges imperceptibly

with Arab Asia through the Syrian steppe, and the Oman bulge contains a mountainous area closely resembling the ranges of Iran. Geomorphologically the Peninsula consists of two main provinces: the ancient Arabian Shield of igneous and metamorphic rocks in the west, and the more recent sedimentary areas sloping away from the Shield to the north-east, east, and south-east into the vast basin consisting of Mesopotamia, the Persian Gulf, and the eastern part of al-Rub# al9l. The Arabian Shield is actually only the eastern part of the Arabian-Nubian Shield, an immense mass of basement rocksgreenstones, schists, granite, gneiss, &c. which have thrust upwards to form bare and forbidding mountains, with the whole mass split into two by the rift valleys running southwards from the Dead Sea and along the course of the Red Sea. The older igneous rocks of the Arabian part represent primarily plutonic activity of the more remote past, while more recent volcanoes have blanketed the surrounding ground with fields of lava ( arra , pl. irr) often imposing in extent. Regions of igneous and metamorphic rocks may be rich in minerals and precious stones, but only insignificant quantities of these have so far been found in Arabia. To the north and south the eastern limit of the Arabian Shield lies not far inland from the Red Sea. Between these two extremities the limit sweeps around in a rough bulge reaching as far east as the vicinity of al-Dawdim, less than 200 km. west of the western wall of uway. The geomorphologically confused mountains of the Yaman, though composed of similar rocks, are physiographically highly different from the remainder of the Shield. Volcanic areas occur in the Yaman as well as in the mountains fringing the southern coast and those of the Oman bulge. Valleys drop sharply westwards to the coast plain of Tihma from the high mountains paralleling the Red Sea. The gentler eastward slope to the Persian Gulf is interrupted by cuestas in Na3d such as uway and al-#Arama, whose steep escarpments face westwards and whose backs then resume the downward trend. From the highlands of 0a'ramawt and ufr the slope southwards to the Gulf of Aden and the Arabian Sea is short, while a longer slope runs northwards to al-Rub# al-9l. The Oman bulge has a short descent north-eastwards to the Gulf of Oman and a much longer descent southwestwards to the same sand sea, though the mountains here, unlike those elsewhere near the coast, are steep on both sides, forming a hogback range. The sedimentary province consists predominantly of limestone, along with an abundance of sandstone and shale. These rocks are products of sediments left behind by seas that in the distant past spread out as far west as the Shield. The sedimentary deposits reach a depth of over several kilometers in the vicinity of the Persian Gulf. Organic matter from the plants and animals that lived in the old seas is the source of the enormous accumulations of petroleum discovered in Eastern Arabia during the 20th century. Islands. The islands, islets, and coral reefs ( 9a#b , pl. 9i#bn) off the Arabian coast increase in number as one proceeds southwards down the Red Sea. The Farasn Bank parallels the coast for nearly 500 km., its southern part including the Farasn [q.v.] Archipelago, where the largest islands on the eastern side | [I:535b] of the Red Sea are found.

Kamarn [q.v.] Island lies close to the coast of the Yaman. West of Kamarn the volcanic peak of 3abal al-ayr in the fairway of the sea is reported to have been in eruption as late as the early 19th century. Also in the fairway is al-Zuur, the highest island in the Red sea (nearly 700 m.). The island of Perim [q.v.] ( Mayyn) in the straits of Bb al-Mandab, the entrance to the Red Sea, stands nearer Arabia than Africa. The island of Suur [q.v.], c. 110 km. long and nearly 400 km. distant from the mainland on the southern side of the entrance to the Gulf of Aden, must for both political and ethnographic reasons be regarded as belonging to the Peninsula. The Kuria Muria Islands stand off the mainland in a large bay east of Ra"s Naws. The Arabic name for the group, 9riy Mriy [q.v.], is seldom used today, the more familiar names being al-0allniyya, al-0sikiyya, and al-Sawd", which belong to individual islands. Separated from Oman by a narrow channel is Mara, the only island of considerable size lying along the whole southern coast. The Arabian side of the Gulf of Oman is also almost entirely devoid of islands worthy of the name; one encounters only rocky islets standing alone, such as al-Fal north-west of Muscat, or in clusters, such as alDaymniyyt a little farther towards the west. The mountains of Oman end abruptly at the Strait of Hormuz, the entrance to the Persian Gulf, and some of the peaks detached from the main range form inhospitable islands, the northern tip of one of which is Ra"s Musandam. Ab Ms, an island in the Persian Gulf north-west of the port of al-9ria, has deposits of iron oxide which are worked commercially. Close to the southernmost shore of the Gulf are a number of sandy islands, the largest of which is Muay9i (shown on most charts as Ab alAbya', the name of its northern part). In the western half of the embayment between the Trucial Coast and the |aar Peninsula are islands presumed to be salt domes rising above the sea, among which are -r Ban Ys [q.v.], Dalm", Zarakkh, Ds, and 0ll. The main island of Barain (al-Barayn) has a scattering of attendant islets and a dependency of fair size, 0awr, which almost touches |aar. Trt, Ab #Al, and other islands hug the coast of Saudi Arabia, while al-#Arabiyya [q.v.] and al-Frisiyya [q.v.] lie out near the middle of the Gulf. The Great Pearl Banks (hayr, pl. hayart) stretch along nearly the entire length of the Arabian side of the Persian Gulf, with the richer banks in the central portion. The term 9a#b is not used for a reef in this Gulf, its place being taken by fa9t (pl. fu9t), na3wa, and u#a. A idd (pl. udd) is a sand bank, a la (pl. uwal) is a low sandy islet which may be covered at high tide, and a aris a projecting rock. Ru is the common word for a shoal, while an area of deep water15 fathoms (b#, pl. abw# or b#n, the Arab fathom being a little less than the English fathom of 6 feet) or more is called a 9ubba (pl. 9abb). The Persian Gulf is a shallow sea, with few depths greater than 90 m., in contrast to the Red Sea, the depth of which in places is in excess of 2,000 m. Bays and Coasts. The coasts of the Peninsula on the three sides facing the sea are relatively unmarked by major bendings or indentations; no other great land mass on the surface of the globe provides such a paucity of shelter for ships. The Red Sea has few bays on the Arabian side, but many narrow inlets of the type called 9arm, which penetrate | [I:536a] some distance inland and then broaden out into lagoons in which small sailing vessels can

anchor. The one good natural harbour along the southern coast is Aden. Between Ra"s Fartak and Ra"s al-0add there are four large bays, here called 9ubba (cf. the use of this term in the Persian Gulf mentioned above), but all are so open to sea that they give no protection. Muscat on the Gulf of Oman offers a hill-encircled bay large enough for steamers of medium size. Excellent harbours exist in the cliff-walled inlets in the vicinity of Musandam, but they are so hot and inaccesible from the interior that good use has never been made of them. The Persian Gulf has a proportionally larger number of bays, here called dawa, but their waters are almost without exception extremely shallow. Inlets in the Arabian shores of the Persian Gulf go by the name of 9awr, a term also used here for a submarine valley. One of the best examples of these inlets is 9awr al-#Udayd, which pierces the coast on the eastern side of the base of the |aar Peninsula. Mountains, Plateaux, and Plains. The chain or chains of mountains paralleling the coast of the Gulf of al-#Aaba and the Red Sea are known collectively as al-Sart [q.v.], though use of this name is not particularly widespread. In many places a lower range lies close to the coast and is separated by a plateau from a higher range farther inland. The average height of al-Sart is considerably below 2,000 m. Between the region of Madyan and Mecca only the famous crags of Ra'w [q.v.] west of Medina and a few other mountains reach noteworthy heights. Southeast of Mecca several peaks go up to over 2,500 m., and thence the chain rises to its greatest heights in southern #Asr and the Yaman (0a'r 9u#ayb west of -an#", c. 3,760 m.). The more precipitous western slopes are generally the higher, but many bold features are also met with along the inner eastern slopes. The range of 0a'n east of Mecca, the historic boundary between al-0i3z and Na3d, appears to have lost this distinction in the popular mind, though the dividing line is considered to be along the eastern slopes or among the foothills of al-Sart. Passes across al-Sart, called #aabain #Asr and nal in the Yaman, are few and far between, and are usually difficult of transit. Notable gaps in the chain are those leading through to Medina and Mecca. Interspersed among the mountains and occurring frequently along their eastern slopes are plateaux, among the most fertile of which are those in #Asr and those surrounding -an#" and 9amr in the Yaman. The plateaux are often capped with a bed of lava, and in places the lava has spilled down the western slopes to reach the verge of the Red Sea. The highlands of the Yaman present a steep face towards the south, the eastern stretch of which is al-Kawr, called after its indigenous tribes Kawr al-#Aw9il in the west and Kawr al-#Awli in the east. Northeast of Kawr al-#Awli is the highly dissected limestone plateau of al-3awl which is split in twain by the eastward-trending channel of Wd 0a'ramawt. The southern part of al-3awl reaches heights of nearly 2,000 m., while the higher elevations of the northern part do not greatly exceed 1,000 m. The cliffs along the edges of al-3awl are often awe-inspiring in their sheerness. Farther east in the region of ufr are the mountains of the tribe of al-|ar with peaks well over 1,500 m. in height. The growth of trees and grasses on the range is so thick that the residents often call it the Black Mountain. North-eastwards of | [I:536b] Ra"s Naws the mountains paralleling the coast begin to dwindle in size and number, and the coast from Ra"s -awira to Ra"s al-0add has generally lowlying country behind it.

Mountains reappear again overlooking the Arabian shore of the Gulf of Oman, along which the range of al-0a3ar runs from Ra"s al-0add to Ra"s Musandam. The towering peaks of al-0a3ar are in the central portion, in the vicinity of 3abal al-A9'ar, the highest exceeding 3,000 m. by a bare margin. Northwest of 3abal al-A9'ar the mountains called al-Kawr form a part of the main range, while 3abal 0aft rears its formidable hogbacked ridge in the open country west of the northern half of the range. In the interior the range of al-ubay lies in the borderland between Jordan and Saudi Arabia. Just south of the Great Nafd the parallel ranges of A3a" [q.v.] and Salm are together known as 3abal 9ammar. The hills of al-Nr lie in the central bulge of the Arabian Shield, near its eastern edge. East of the Shield a series of roughly parallel cuestas curve around from north to south, following the contour of the crystalline bulge. The most striking of these is uway [q.v.], the backbone of Na3d, with a length of c. 1,000 km. from 9a9m 3azra to 9a9m 9ama, where the sands of al-Rub# al-9l encompass its southern end. Just east of the sands of al-Dahn" is the low rocky plateau of al--ummn (classical al-ammn [q.v.]). Mesas, buttes, and ridges often rise singly or in groups above the plateaus and plains. The Bedouins use the term 3abal for rocky hillocks as well as massive mountains, and other terms in common use are 'il# (pl. 'ul# or 'il#n, a general synonym for 3abal , not necessarily a rib-shaped hill), azm (usually lower than a 3abal ), abra (pl. burn, whence the name of the great oil field of Kuwayt, al-Burn), and bar" (pl. bur), the last two being applied to hills whose sides are mottled with patches of sand. The promontories jutting out from the inland escarpments are called 9a9m (pl. 9u9m), the word for nose. Within the northern border of Arabia lies the southernmost portion of al-0amd, a stony plain stretching on northwards into the steppe, and south-east thereof is al-0a3ara, another stony plain. Among the major adabasplains with a mantle of gravelare alDibdiba in the noth eastern corner of the Peninsula and Ab Bar and Rayd south of the southern end of al-Dahn". The plain of al-3alada south-west of Rayd is completely ringed about by the sands of al-Rub# al-9l. Other plains are found along the southern and eastern edges of al-Rub# al-9l, all sloping towards the basin occupied by the sands. The coast plains in the west and south are confined within a fairly narrow space nearly everywhere by the mountains crowding down towards the sea. Tihma [q.v.], the general name for the coast plain along the Red Sea, is sometimes subdivided into Tihmat al-0i3z, Tihmat #Asr, and Tihmat al-Yaman. On the Gulf of Oman no more than faint traces of plains exist between Ra"s al-0add and Muscat, but between Muscat and 9in the plain broadens out into al-Bina [q.v.], one of the great dateproducing districts of Arabia. Salt pans are particularly common along the southern shore of the Persian Gulf, and much of the low ground in this region is covered with sand. Sandy Deserts.

Dunes may be star-shaped, dome-shaped, or crescent-shaped (the crescentic or | [I:537a] barchane dune = muaww, pl. maw). Dunes bare of vegetation are called u#s (sing. i#s, probably from classical di#), with the term na (pl. niyn) being used for the larger ones. Masses of sand may form long single or parallel veins ( #ir , pl. #ur) or more complex arrangements underlying which an orderly pattern can often be discerned. Wide expanses of ground are covered with relatively thin sheets of drift sand. Barchane dunes occur in sizes ranging from c. 1 m. to c. 200 m. in height, and the largest are several km. or more in length. Almost all of the dunes consist of pure sand, with no core of rock or other substances. The colour and composition of the sand itself vary from place to place, with the predominant colour in the interior approaching red. A sandy area is generally called a nafd (pl. pauc. naf"id, pl. abund, nifd) in the north and a ramla (pl. riml) in the south. The term #ir may be applied to a whole area containing a number of #ur, e.g., #Ir al-Mahr embraces seven major veins. As frequently happens with the Arabs, these common nouns are transformed into proper names applied to the most noteworthy examples of their categories: the northern desert known to Westerners as the Great Nafd is called by the Arabs simply al-Nafd, the whole southern desert known to Westerners as al-Rub# al-9l is ordinarily referred to simply as al-Ramla, while al-#Uray is a sandy area south of |aar. Almost all of the principal sandy deserts lie in the sedimentary province, where they curve around the central bulge of the crystalline Shield in the same fashion as the cuestas, along the western bases of which many of them lie. The two largest are the Great Nafd [q.v.], with an area estimated at c. 70,000 km2., and al-Rub# al-9l [q.v.], with an area estimated at over 500,000 km2., making the latter the largest continuous body of sand in the world. These two are connected by the long thin arc of al-Dahn" [q.v.] lying east of uway and al-#Arama. A similar arc runs west of uway between the two main sandy deserts, but its continuity is broken in several places. This lesser arc begins with #Ir al-Mahr, which leaves the Great Nafd south of the point of departure of al-Dahn" and merges into three parallel fingers of sand, which from east to west are Nafd al-9uwayrt, Nafd al-Sirr, and al-9uayyia. The southern extension of al-9uwayrt is named Nafd al-Baldn after the towns of the district of al-Wa9m lining its south-western edge. Almost connected with al-Sirr is Nafd | unayfi9a, the south-eastern end of which nestles under the western wall of uway. South of |unayfi9a occurs a major interruption in the arc, after which the sands reappear in #Ir al-Day, which ends north of Wd al-Dawsir. The principal direction in which the sands migrate is southwards; in other words, they are slowly but steadily forsaking the Great Nafd and working their way along the two arcs towards alRub# al-9l. Although on the map al-Rub# al-9l appears to have two long arms extending northwards, the western of these, al-3fra, is regarded by the Arabs as constituting a separate desert cut off from al-Rub# al-9l by the low ground of al-3awb (3awb Yabrn). The eastern of the two arms, also regarded as a separate region, penetrates deep into the hinterland of the Trucial Coast. Ramlat al-Sab#atayn south of the south-western corner of al-Rub# al-9l lies outside the system just described. Perhaps the largest accumulation of | [I:537b] sand on the Arabian Shield is #Ir Subay# in the southern part of the central bulge.

Various geographical features associated with drainage and water resources are discussed in the following section. (iii) climate, drainage, and water resources The Tropic of Cancer bisects Arabia, passing between Medina and Mecca, between the districts of al-9ar3 and al-Afl3, and between Muscat and Ra"s al-0add, so that most of the land enjoys a generally temperate climate. Even in the south, where the tip of the Peninsula approaches 12 N. lat., much of the country is sufficiently elevated to avoid the rigours of tropical heat. Only the lowlands along parts of the Red Sea, the Gulf of Aden, and the Arabian Sea have a semitropical rather than a temperate environment. Meteorological records, though improved in recent years, are still too scanty to provide a completely detailed picture of Arabian weather. The summer heat (ay) is intense throughout the Peninsula, reaching over 50 C. in the hottest places. The dryness of much of the interior makes the heat tolerable there, but along the coasts and in some of the southern highlands the humidity in summer is high and debilitating. Fogs and dews are common in the humid regions, but over Inner Arabia the sun shines the year round, obscured only by an occasional sandstorm or even rarer shower. Although not the happiest on earth, the Arabian climate has often been damned more violently than it deserves. Many days in fall and spring are fresh or mild. The winters are invigoratingly cool, with bitter cold occurring only at the higher altitudes, where snow crowns some of the peaks, and in the far north, where the winds are biting. The winds vary greatly in different parts, being subject in particular to the influence of the surrounding seas. In Eastern Arabia the wind tends to blow from the same quarter, but on occasion it suddenly shifts halfway round the full circle, the prevailing 9aml from c. NNW yielding to the kaws from c. SE. Winds whipping up into sandstorms may subside quickly or go on for days. In Na3d the wind may box the compass, with drastic changes sometimes taking place every half hour. The monsoons of the Indian Ocean reaching parts of Southern Arabia profoundly affect the character of the country and the life of the people there. Most of Arabia has been made and kept a desert by the scarcity of rainfall. In portions of al-Rub# al-9l no rain at all may fall for ten years on end, and in many other parts of the Peninsula the annual fall seldom if ever exceeds 150 mm. When rain does fall over the desert, it may come as a torrential downpour, providing enough moisture to carpet the ground with wild flowers. Periods of drought sometimes last for several years, bringing misery and even death to the people and causing some to migrate abroad. Higher areas tend to catch more rain than lower areas nearby: heavy winter rains may fall on the plateaus and plains in the north while the depression of Wd alSirn remains completely dry. Only the areas where the monsoons blow receive fairly ample rains. Although Arabia contains no large perennial rivers, in the monsoon zone water may be found throughout the year in some stretches of the valleys (called 9ayl in the southwest). A few of the valleys descending to the sea blend their fresh water with the salt, but most of them dissipate it throughout | [I:538a] their alluvial fans on the coast plains. In the dry zone rainwater from the higher areas occasionally comes down in spate through the stream channels (wd, pl. widyn, or 9a#b, pl. 9i#bn), which

otherwise contain only a few pools or none at all. These flash floods ( sayl, pl. suyl) sometimes cause great damage, and much of their precious water may flow away unused. Other floods come in sheets over flat surfaces such as gravel plains or the fans at channel mouths. Part of the water that seeps underground is recovered by man through wells and springs. Although the courses of some valleys can be traced for considerable distances, bodies of sand lying athwart them in places tend to prevent through drainage. A characteristic feature of the Arabian drainage system is the local enclosed basin, varying in size from very large to very small. Wd al-Sirn is not a true wdbut a depression c. 300 km. long and 50-70 km. broad into which many wds on both sides empty their sayls. Types of smaller basins are the 9abr", a hollow with an impervious bottom holding water for a while after rain, and the raw'a (called fay'a in the north), whose bottom does not hold water, so that wild vegetation may be fairly abundant there. Another type of basin is the salt pan or saline flat ( sab9a , pron. ab9a), which occurs with great frequency along the coasts and also in the interior, where it is fully enclosed. The eastern tributaries of Wd al-0am', which runs down to the Red Sea, originate in 0arrat 9aybar. A short distance farther east are the headwaters of Wd al-Rumah (alRumma in al-Hamdn), which through its extension al-Bin runs to the Persian Gulf basin in the vicinity of al-Bara, though the connecting link between al-Rumah and alBin is choked with sands of al-Dahn". The small area in 0arrat 9aybar between the sources of al-0am' and those of al-Rumah is the one place in the whole Peninsula from which an easy slope to the seas on both sides can clearly be discerned. Descending from the eastern slope of al-Sart, the three large valleys of Ranya, B9a [q.v.], and Ta9l9 converge on the upper reaches of Wd al-Dawsir [q.v.], which receives their waters in times of exceptional floods only to lose them again as it fans out against the sands of al-Rub# al-9l after piercing through the wall of uway. 0abawn (0abawnan in al-Hamdn) and Na3rn [q.v.] are valleys coursing eastwards to the sands which lie south of the southern end of uway. From the highlands of the Yaman the valley of al-9rid [q.v.] flows down into the basin of al-3awf [q.v.] (3awf Ibn Nir), the home of the ancient Minaeans. The mountains of the Yaman send water southwards towards the coast in the vicinity of Aden through Tuban, Ban, and other valleys. Water from Ban is used for an extensive development of agriculture at Abyan. The southern outriders of al-3awl give rise to Wd Mayfa#a and Wd 0a3ar. 0a3ar is the one truly perennial river in Arabia, but its total length probably does not exceed 100 km. Its water, part of which comes from the hot springs of al--idra in the uplands, supports cultivation in the area of Mayfa# at the river delta (not to be confused with Wd Mayfa#a to the west). Wd 0a'ramawt [q.v.], the principal artery of a great drainage system, is fed by valleys coming from both the southern and the northern parts of al-3awl, those from the south being far more thickly settled than those from the north. Just beyond the | [I:538b] town of Tarm the Valley of 0a'ramawt assumes the name of al-Masla, which it bears for the remainder of its course to the sea. Sam"il, one of the valleys flung out by the range of al-0a3ar towards the Gulf of Oman, provides passage for the main road from the coast to Inner Oman. The chief

valleys of al-Bina are named after the tribes inhabiting their banks, al-Ma#wil and others. Going up Wd al-3izy and Wd al-|awr, one comes to passes leading over the mountains to the Trucial Coast. In the region east of al-Dahn" between al-Bin and al-Sahb" the insufficiency of surface water has militated against the formation of true wds of any size. Wd alMiyh northwest of al-|af is a basin rather than a stream channel, deriving its name from the numerous wells and springs found within its confines. Other large basins are al-Far south of Wd al-Miyh and al-9a southwest of the city of Kuwayt. In the far north a series of valleys known as al-Widyn (Widyn #Anaza) runs northeastwards towards the Euphrates; among these are Tubal, #Ar#ar, and al-9urr. In Na3d a number of valleys between al-Rumah and Wd al-Dawsir cut through uway; al-#Atk [q.v.] is the northernmost of these. Wd 0anfa [q.v.], rising on the crest of uway rather than making a gap in the escarpment, twists down to the basin of al-9ar3 where several important valleys empty into al-Sahb" [q.v.], the course of which can be traced across al-Dahn" and al-3fra into the Persian Gulf basin. The valley of Birk cleaves through the wall of uway via a picturesque gorge and turns northwards under the name of al-#Am to follow a course towards al-Sahb". Arabia contains no large permanent lakes. Deep pools occur in places, with the most unusual ones being those in the districts of al-9ar3 and al-Afl3. In oases such as al0as big ponds may be formed by the run-off from irrigation. Dry lakes in the north may be filled with water over an area of 10 or more km2. after a rain. The thousands of wells ( bi"r , pron. br, pl. abyr, or alb, pl. ulbn) in the desert, some of them even in the central portions of al-Rub# al-9l, make possible the nomadic life of the Bedouins. The deepest is reported to descend c. 170 m. into the earth, and depths in excess of 70 m. are not uncommon. The wells may be steyned or unsteyned; they may be frequently visited or seldom seen by man. Other watering places are spots in the sand or in valley bottoms where exiguous water is secured by digging down a meter or more. Blowing sand rapidly fills in these shallow holes, so that finding them may tax even the navigational skill of Bedouins bred in the wild. The water in some of the desert wells is too salty for humans (such a well is called a 9awr, pl. 9rn), but camels drink it and furnish milk to sustain their masters. Around most of the flowing springs ( #ayn , pl. #uyn) oasis settlements or towns have grown up. Other communities draw their water only from dug wells, while sometimes tanks and cisterns are used to catch rainwater. The larger oases consist of several or more villages or towns grouped close together, each with its own belt of date groves. The oasis name may apply to the whole group, which may cover tens or hundreds of square kilometers, rather than to any single community within its confines, e.g., al-0as with its chief towns al-Hufhf and al-Mubarraz, and B9a with al-Raw9an and Nimrn. | [I:539a] Various methods of irrigation are used wherever there is sufficient water. Terracing is much practised in the south with water being led from enclosure to enclosure. In some regions an old system of underground aqueducts (fala3, pl. afl3) similar to the ants of Iran is common, while in others it is not known. In large oases such as al-0as and in Tihma the rules governing the distribution of water for irrigation are elaborate and

firmly fixed by custom. The building of dams, once an art in which the Arabs excelled, has been neglected in more recent times, but now, with a growing population and higher standards of life demanding an expansion of agriculture, it is being revived. (iv) political divisions Political divisions in Arabia are often ill defined. Few international boundaries have been agreed upon by the parties concerned, and none has been properly demarcated throughout its full length. A rapid survey of the main political divisions as they existed in 1374/1954-5 will furnish examples of the truth of these statements. The Kingdom of Saudi Arabia occupies the whole northern half of the Peninsulawith the exception of the small states of Kuwayt, Barayn, and |aar, and parts of Oman and a good share of the southern half as well. Stretching from the Red Sea to the Persian Gulf, it incorporates the large regions of al-0i3z [q.v.], #Asr [q.v.], and Na3d [q.v.], and also most if not all of al-Rub# al-9l. Saudi Arabia and the Yaman agreed in 1354/1936 upon a boundary running from the Red Sea coast to a point short of alRub# al-9l, but no serious attempt has since been made to extend the line southwards from this point over a gap between 100 and 200 km. in breadth. No land boundaries have been fixed between Saudi Arabia and any of the following states, all of which may be assumed to have territories abutting on the Kingdom: the Aden Protectorate, the Sultanate of Muscat, the Imamate of Oman, the Amirate of Ab ab (the southernmost of the Trucial States), and the Amirate of |aar. The boundary between Saudi Arabia and Kuwayt and the boundaries of their neutral zone have been agreed upon in a general way. [See further sa#diyya, al-afl3, al-#ri' , al-as , alyamma .] The Mutawakkilite Kingdom of the Yaman lies along the Red Sea between Saudi Arabia and the Aden hinterland. The British and Yamanite Governments have a not entirely satisfactory working arrangement regarding the boundary between the Yaman and the Aden Protectorate, and the joint commission provided for in the Agreement of 1370/1951 to demarcate boundary locations and to recommend solutions to disputes arising from conflicting positions has not yet been constituted. [See further al-yaman.] The British Crown Colony of Aden, the only possession of a Western power on the Arabian mainland, occupies a tiny area c. 160 km. east of the south-western tip of the Peninsula. Perim Island forms a part on the Colony, and Kamarn is subject to its administration. The Governor of Aden Colony is also Governor and Commander-inChief of the Aden Protectorate, which runs c. 1200 km. along the southern coast from Bb al-Mandab to Ra"s 4arbat #Al and reaches inland an undetermined distance. [See further #adan, a'ramawt .] The Sultanate of Muscat ( Masa [q.v.]) provides an outstanding example of the peculiarities of the political scene in Arabia. The ruler, who styles | [I:539b] himself Sultan of Muscat and Oman, lays claim to virtually all the territory east of the eastern edge of al-Rub# al-9l, a space roughly 1200 km. long and 500 km. broad. Within this space, however, the Sultan administers only three relatively small areas, the remaining areas coming under the Imam of Oman or other independent chieftains. The Sultan's foothold on the southern coastufr, which abuts on the Eastern Aden Protectorateis separated from the main base of his powerthe towns of Muscat and

Mara and the coast of the Gulf of Oman, including al-Binaby nearly 1,000 km. of coastline with its hinterland. Again, his domains on the coast of the Gulf of Oman are interrupted in the north by territories belonging to the Trucial States around Kalb and al-Fu3ayra before the third centre of his authority appears near Ra"s Musandam. The Sultan is of the line of $l B Sa#d, an Ib' dynasty which first came into power c. 1157/c. 1744. Unlike his neighbours on both sides, the Sultan is not formally under British protection, though he does have special ties with the British Government. Another Ib' ruler, the Imam of Oman, whose authority rests more firmly on a religious foundation than does that of the Sultan, directs the destinies of the interior region occupied by the Ib' community. No clear dividing line exists between the territories of the Imamate and the Sultanate; those of the Imam reach the crests of the main mountain range of al-0a3ar throughout much of its length, and a few of his governors (wls) are established on the seaward slopes. The Imam, whose theocratic realm is a continuation of the 9ri3 state founded in Oman c. 133/c. 750, has his capital at Nazw, and his two principal lieutenants reside at Tanf in Inner Oman and al-|bil in the district of al-9arya. Of all the major rulers in Arabia, the Imam, who maintains no formal diplomatic relations with any other power, is the most selfsufficient and the least known to the outside world. [See further #umn .] The Trucial Coast (Sil #Umn or simply al-Sil) is the southern shore of the Persian Gulf running southwestwards and then westwards for an undetermined distance towards |aar. When the Arabs living there in the early 19th century were preying vigorously on shipping in the Gulf, the region was known as the Pirate Coast; after the British forcibly stopped the marauding and imposed a maritime truce on the rulers of the ports, it came to be called the Trucial Coast. The Trucial States, all of which are in special treaty relations with the British Government, are regarded as being under that government's protection, though without having the formal status of protectorates. [See further bar fris .] The Sultan of Muscat claims a part of the oasis of al-Buraym, but Saudi Arabia challenges this claim on the basis of its own connexions with the place. Saudi Arabia likewise challenges the claim of the Trucial State Ab ab to al-3iw". Saudi Arabia claims an outlet to the Persian Gulf on the coast between Ab ab and |aar, but the British Government, which by treaty controls the foreign relations of these two states, disputes this claim. In 1373/1954 the two parties agreed to submit the dispute to arbitration. The |aar [q.v.] Peninsula, jutting northwards into the Persian Gulf about halfway between its mouth and its head, is the seat of an Amirate under the rule of $l 9n, a dynasty of recent origin, with its capital in the port of al-Dawa. The boundary | [I:540a] between the Amirate and Saudi Arabia in the vicinity of the base of the peninsula has not been agreed upon, and the Amir of Barayn claims a piece of territory around al-Zubra in the north-western part of the peninsula. The archipelago of Barayn [q.v.] between |aar and the Saudi Arabian mainland constitutes an Amirate under the rule of $l 9alfa, a family from Na3d which established itself in the islands in 1197/1783 and has ruled there ever since, with its capital in the port of al-Manma on the main island. British interests in the Persian Gulf come under the supervision of a Political Resident with headquarters in al-Manma.

Also subject to his administration are the Kuria Muria islands, which belong to Great Britain. On the Arabian mainland at the head of the Persian Gulf is the small roughly triangular Amirate of Kuwayt, partially separated from Saudi Arabia by a neutral zone and bounded on the north and west by #Ir. $l -ab, a family related to $l 9alfa of Barayn, has ruled Kuwayt for over two centuries [see kuwayt]. |aar, Barayn, and Kuwayt have all granted the British Government by treaty the right to conduct their foreign affairs and have agreed not to enter into relations with other powers without the consent of that government. Questions dealing with water boundaries and the appurtenance of a number of islands in the Persian Gulf remain to be settled between Barayn and Kuwayt on one hand and Saudi Arabia on the other. (v) flora and fauna Throughout most of the Peninsula a sharp contrast exists between the untilled stretches of desert and the green patches of cultivation in the oases. In places, particularly along the margins of the Peninsula where rain falls more frequently or where stream channels bring sufficient water down from the highlands, cultivation is more widespread, sometimes climbing the heights in skilfully built terraces and sometimes carpeting the narrow plains between the mountains and the sea. Arabia, however, boasts no endless prairies or pampas tamed by the plough, nor does it boast any rich belt of foreststhe best it can offer are the juniper woods of High #Asr. The plant beyond compare in the oases is the date palm ( na9la [q.v.]), so much in a class by itself that the Arab tends to think of it as a thing apart from all other trees. Not only is the date the most important staple food, but the branches and bark of the palm are also used in building huts, in making baskets and mats, and for a myriad other purposes. The date palm does not flourish at the highest altitudes, so that the villagers there depend on grains. In ufr and a few other spots coconut palms grow in place of or alongside the date, which is also replaced on occasion by the dawm palm (gingerbread tree). Wheat, barley, and the millets are the chief grains. Alfalfa (lucerne = att or a'b or barsm) is a common crop raised in the shade of the date palms, and cotton, rice, and tobacco are cultivated on a small scale. On high terraces in the Yaman and #Asr grows the coffee which made Mocha a goal for Western traders after the Portuguese found the way around Africa to India. Introduced only about five centuries ago into Arabia, coffee gave its Arabic name ( ahwa [q.v.]) to the world, but the world now goes to Brazil for its everyday bean, the bean of the Yaman having become an exotic luxury. On many terraces coffee | [I:540b] has yielded place to the more profitable t [q.v.], whose slightly narcotic leaves are chewed by people of all classes in the Yaman and other parts of the south. Frankincense ( lubn ) [q.v.] and other aromatics, exported to the West over two thousand years ago by the Incense Road from South Arabia to the Mediterranean, still grow in the south, especially in the land of Mahra, but as articles of commerce they are now of virtually no value. Of greater use today is indigo, much favored as a dye in the

south (the tree is called awr and the dye nl [q.v.]). Other common dyes are the yellowish wars and the reddish henna. Among the larger trees are tamariskssometimes planted in a row as a wind break or to stop the advance of drifting sandacacias, mimosas, and carobs. The jujube (Zizyphus spina christi = sidr [q.v.] in the north, #ilb in the south) bears an edible fruit, called dawm (a homonym of the name of the palm) by the Bedouins and kunr by the townsmen. The aloe and the euphorbia often grow to a considerable height, and some varieties of euphorbia closely resemble cactus. Arid though Arabia is, it is not without flowers and fruits. For roses and pomegranates al-"if is famed, al-9ar3 for watermelons (3i in Na3d, abab in al-0i3z, and dib9 in the north), and al-Buraym for mangoes (anb or hanb). Figs, grapes, peaches, bananas, and other fruits sometimes vary the monotonous diet of the townsman, but the Bedouin seldom savours anything more than his milk and dates. In the cool season the Bedouins roam far afield, sometimes going for months without resort to water wellsthe forage supports the camels, whose milk supports their masters. The most sought after plants for forage are the annuals (#u9b, pron. #i9b) grasses, wild flowers, and herbs which spring up green after a rain, especially in the rab# , the season of plenty following the first and best rains (wasm). The sands provide favorable soil for the growth of such annuals and so are reckoned by the nomads as among the most attractive types of desert terrain. Perennial shrubs and bushes (9a3ar) eaten by camels are na, 99, and saba (pron, aba), as well as others too numerous to mention. From time to time camels hanker after bushes of the category called am', a prime source of the salt needed by their system. Among the many plants falling in this category are raw9a, rim9, #ard, #u3rum, swd, 9inn, 9a' and 99 (not 9 as in classical Arabic). Dry bushes are also essential to the Bedouins for firewood (aab), among the best for this purpose being #abl, 9a', and rim9. Burning with a fragrant scent, these woods help to make the ceremony of brewing coffee for a guest at the open door of the tent one of the chief pleasures of life. The Bedouin likes truffles ( fa#) and eats other desert plants, though by preference and philosophy there is little of the vegetarian in his being. Twigs of the ark (pron. rk) are in common use as a toothbrush (miswk), and senna (san) is chewed as a purgative. Vegetation would be more abundant in the deserts were it not for the migrating dunes, some of which move 20 m. in a year. In many places, however, bushes have taken root and fixed the sand, a hummock of which is built up around each bush. An area of such hummocks may extend for many kilometers, making very rough country known as #af3a. Less difficult types of sandy terrain with vegetation are called marba9 or dikka (pl. dikk, cf. class. | [I:541a] dakk, pl. dikk; and dakdak = flat surface, sandy plain). Among animals the camel occupies a place analogous to that of the date palm among plants. The vast majority of Bedouins in Arabia depend on the camel above all other material possessions. The tribes which herd sheep rather than camels range over the steppes north of Arabia, close to the great rivers of Mesopotamia, and do not pass beyond the territory of Kuwayt in their southward migrations. Milk is the camel's most precious product, but its meat, hide, and wool are also put to good use, its dung (dimn) is collected to be burned as fuel, and the tail of a dead camel makes a strong rope.

Camels are sometimes harnessed for ploughing or drawing water from wells, and the nomads sell part of their stock to secure money for clothing and other necessities. In time of great thirst a Bedouin may slaughter a camel to drink the water stored in its stomach (kar9) and the urine in its bladder (mibwl). The general term for camels is ibil [q.v.] (often pronounced bil), with baw9 being common in the south. A riding camel is a 9all (pl. 3ay9 ); the plural rikb is used for both those that are ridden and those that are not. The most highly desired camels are the thoroughbreds (a"il), whose pedigree has been controlled and recorded over a number of generations. Many of these are from the breeds of Oman (#Umniyyt), among which the Bawin of al-Bina are particularly well known, though these have the disadvantage of wanting to drink every day and of not being adapted to rough country. The camels of the sands tend to be smaller and lighter in color than those raised in the mountains of the Yaman. Among the multitudinous names in the special vocabulary reserved for camels are ones describing beasts which graze on certain plants, e.g. hawrim (fem. sing. hrim) from the harm bush, and awrik (fem. sing. rika) from the ark tree. Along the coasts camels are often fed on dried sardines. Along with camels, most of the nomads keep sheep and goats ( 9anam ), though not in great flocks like those of the northern steppes. Sheep and goats are valued for their milk, fleece, and skins. Sheep are in demand as the pice de rsistance of the Arab banquet; even royalty can offer nothing more appetising than a young lamb ( al, pl. ulyn) basted in a pot with samn and served on a platter heaped high with rice. Samn, clarified butter for cooking and greasing made from the milk of the ewe (na#3a) or she-goat (#anz), is considered superior to 3abb from the milk of the she-camel (na) or wadak from the fat of camels, sheep, or cattle. The Arabian horse, the ancestor of the Western thoroughbred and once the pride of the Peninsula, is a disappearing strain. Few Bedouins now own horses, and the export of stock to India, Egypt, and the West, formerly an important item in the Arabian economy, has dwindled away to insignificance. An occasional man of rank still maintains a stud, but even this is likely to be neglected. The speed of the motor car has captured the Arab's fancy; cars are now used in place of horses for hunting and as cavalry in some of the Arabian military forces. Fine breeds of donkeys are raised, particularly the large white ones of Barayn and al0as. Donkeys are used for riding, drawing water, and as pack animals in the mountains, where their surefootedness makes them more reliable than camels. Cattle, which in most places are not numerous, are usually of the small humped variety, except in Suur, where the humpless kind is found. | [I:541b] The gazelle (aby), which in days past used to speed across the plains in great herds, is rapidly being thinned out by rifles in the hands of hunters hurtling by in trucks or cars. The three common types are the ri"m (pron. rm), the #ifr (cf. class. ya#fr), and the idm; the term 9azl is used only for the newly born kid. The swift greyhound ( sal ) of the Bedouins can on rare occasions outrun even the gazelle. Of the oryx (wu'ay in the south, baar wa9in the north), a larger antelope, small numbers survive in the remoter parts of al-Rub# al-9l but none or almost none is now left in the Great Nafd. The ibex or mountain goat ( wa#l or badan ) also seeks refuge in distant retreats on higher cliffs. Other large wild beasts are the hyena ('ab#), jackal (ww), wolf

( 9i"b , pron. 9b, pl. 9iyba), and cheetah (nimr). The lion has long been extinct in Arabia. In the mountains of the south baboons are common, often chattering along in troops; they are fond of raiding the millet fields. Smaller animals are the fox ( 9a#lab or 9a#l or un), the ratel (arinbn, class. aribn), the cony or hyrax (wabr), and the hare (arnab). The hedgehog (unfu9) with its short quills is much commoner than the unrelated long-quilled porcupine (n). The jerboa (3arb#, cf. class. yarb# ) hops about the desert on its long hind legs, resembling a miniature kangaroo; its cousin the 3ir9 (cf. class. 3ura9), on the other hand, runs on all fours. Snakes live in the sands and rocks, though seldom seen because of their nocturnal habits. Some are poisonous, including the horned viper, as well as a species of Arabian cobra (= Egyptian asp) and a large snake called the yaym (cf. class. aym), which the Bedouins say has the power of flying or leaping over a considerable distance. According to popular report, perhaps the most deadly of all is the ba9n, a small innocent-appearing snake living in the sands. The striped seasnakes of the Persian Gulf are poisonous, but they rarely if ever bite human beings. The two large lizards are the 'abb and the Arabian or desert monitor (waral), the first of which is eaten by all the Bedouins with relish, while the second is ordinarily shunned. Among the smaller lizards of the sands are the fierce-looking uay and the slippery sand-swimming skink (dammsa). The ostrich appears to have become extinct in Arabia during the past few years. Fragments of ostrich eggshells are often found in the desert, and the word na#mand other terms relating to ostriches occur frequently in place names. Trained falcons, often called simply uyr, are much used in the chase, their chief game among other birds being the lesser bustard (ubr). Species of the sand grouse such as the a and the 9a are too fast for trained falcons, though they can be overtaken by the wild variety. The presence of wild falcons is attested to by the number of high places called maara = nesting-place of the falcon ( ar). Among the larger birds of the desert are the eagle, the vulture ( nasr ), and the owl, while the flamingo, the egret, and the pelican are found along the coasts. Smaller birds are commoner in the cultivated regions, among them being the cuckoo, the thrush, the swallow, the wagtail, the Syrian nightingale ( bulbul ), and the hoopoe ( hudhud ). The bifasciated lark ( umm slim ) is ubiquitous in the desert, and the courser ( dara3) nearly so. The pigeons of the Great Mosque in Mecca are famous throughout Islam. The seas embracing the Peninsula are rich in fish, many of which, such as the king mackerel( kan#ad) and the grouper ( hmr) of the Persian Gulf, are tasty and nutritious, but are not eaten as much by | [I:542a] the Arabs as might be expected. Whales occasionally enter the Persian Gulf from the Indian Ocean. Both sharks and sardines are caught in great numbers off the southern coast, and the Persian Gulf produces delicious shrimps. The most disastrous plague visited upon Arabia by living creatures is that of the locusts ( 3ard ). The solitary mitigating aspect of a locust invasion is that a number of the invaders themselves are eaten by the people they afflict. Minor plagues by comparison are those of flies, camel ticks, and similar vermin, which are no worse in Arabia than in many other countries, even though the Bedouin may describe his life as all raml waaml (sand and lice). A more agreeable insect, even in spite of its sting, is the bee, kept for its honey.

(vi) Ethnography In the study of the ethnography of the Peninsula an array of formidable problems remain unsolved. Who were the first inhabitants? Did they arise from the soil or did they come from abroad? If immigrants, what was their original home? What was the environment in which they liveddid it differ greatly from the Arabia of today? What intrusive elements intermingled with the earliest dwellers as time went by? Who were the first people to deserve the name of Arab, and where did they come from? A measure of progress has been made in the attempt to elicit answers to these and similar questions, but far more work must be done before any of the more likely hypotheses can achieve the status of historical fact. Much more needs to be known about the geology and geography of the Peninsula, many promising archaeological sites need to be excavated, and an exhaustive investigation must be made of the various segments of the present population and their history. Moreover, the solution of Arabian problems may well depend to a considerable degree on the success of work relating to other areas. The problem of the identity of the Arabs, for example, dovetails inextricably into the broader problem of the identity of the Semites, the host of people speaking languages of the family to which Arabic belongs. Space does not permit a review of the numerous hypotheses receiving serious consideration with respect to the early history of man in Arabia. Suffice it to say that available evidence indicates that the highlanders of the Yaman may form the least adulterated large group anywhere in the world now representing what anthropologists call the Mediterranean race. East of the territory of these highlanders a Veddoid strain is said to appear, particularly among the tribe of Mahra and other tribes in the south speaking their own Semitic languages, which are distinct from Arabic. This Veddoid strain and other data suggest an ancient connection with lands farther east, perhaps India or Ceylon. The Bedouin of the north, to most Westerners the classic Arab type, is also basically Mediterranean, though not quite as characteristically so as the mountaineer of the Yaman. All along the coasts and with less frequency in the interior, other strains occur, sometimes in easily recognisable forms and at other times lying so far below the surface as almost to defy identification. The unraveling of these mysteries is the concern of the archaeologist and the anthropologist [cf. also | [I:544a] badw ]. More important for the student of Islam is the concept the Arabespecially the Muslim Arabhas had, and in many cases still has, of his ethnographical development, a concept so prevalent and tenaciously held that it merits the careful consideration of the anthropologist as well. The seeds of the Arab's own concept go far back into his past; how far can not be determined because of the relative lateness of the sources available, though the basic particulars of the concept had developed before the appearance of Islam. In weighing data pertaining to pre-Islamic times, however, one must use caution, bearing in mind the fact that most of the existing sources were recorded not only long after the event but also subsequent to the introduction of Islam with its new ways of looking at many aspects of life, so that the complete genuineness of these data may often be open to question. Furthermore, various refinements of the Arab concept were still being made in the time of the Prophet, and other refinements came even later. Finally, Islam with its

doctrine of the brotherhood of Muslims and the equality of Arab and non-Arab presented a fundamental challenge to the validity of the Arab concept as a guiding principle for the life of the community. Muslim genealogists have worked out an elaborate and ingenious system for the illustration and application of the Arab concept. Although this system has weaknesses obscurities in the early stages, obvious gaps, unexplained riddles, inconsistencies, and contradictionson the whole it hangs together well. Most important, its primary theses the core of the Arab concepthave been by no means the exclusive property of scholars; they have belonged to the people, and their influence on the politics and social life of Arabia has been penetrating and pervasive. According to the Arab concept, the Arabs constitute a race, not simply a community of people speaking the same language. This race is made up of innumerable men and women each descending in a direct line from one or the other of two ancestors, who probably were not closely related (the connection between these two eponyms is one of the major unresolved aspects of the system). Greater homogeneity could have been attained only by insisting on the descent of all Arabs from a single ancestor. That the Arabs recognized in their clear and undisputed tradition the duality of their origin is a significant fact, and its effect on the history of the Arabs and Islam has been farreaching. The system of the genealogists begins with a nod at those whom the Arabs regarded as the original inhabitants of the Peninsula, tribes such as #$d, 9amd, Iram, 3urhum, asm, and 3ads [qq.v.], all of which are believed to have disappeared before the beginning of Islam. Some of these, such as #$d and Iram, may well have been entirely legendary, while the historicity of others, such as 9amd, is not in doubt. Nothing certain is known about the identity of these tribes, though they are generally reckoned to have been Arabs, the Lost Arabs ( al-#arab al-b"ida). Sometimes they are even called the True Arabs ( al-#arab al-#riba), though this has little meaning, as in the Arab concept they are mainly a historical curiosity and an example of the terrible fate visited on people who heeded not their prophets. Although in later times there were men who claimed descent from these ancients or even tribes reputed to have sprung from them, the conclusion of the genealogist Ibn 0azm (d. 456/1064) was that on | [I:544b] the face of the earth there is no one whose descent from them is verified (ed. LviProvenal, 8). Disposing of the autochthons in this fashion, the Arab concept concentrates on the two great ancestors |an and #Adnn [qq.v.]and the two great divisions of the Arab race they fathered. As all men go back to Adam, these two must have been at least remotely related. The question of a closer relationship depends on whether |an was a descendant of Ism#l, who was recognised as an ancestor of #Adnn. One opinion commonly held opposes such a descent for |an, whose presumed line from Noah's son Shem (Sm b. N) is separately traced. |an's offspring are generally denominated the True Arabs ( al-#arab al-#riba or al-#arb") and #Adnn's the Arabised Arabs ( al-#arab al-muta#arribaor al-musta#riba), though the uncertainty of this classification is revealed by the existence of other versions, one of which brackets the Lost Arabs with |an as the True Arabs, while another reserves the title of True Arabs for the Lost Arabs, designating the people of |an as muta#arriba and those of

#Adnn as musta#riba . In any event, |an clearly comes out closer than #Adnn to genuine Arabness. The descendants of |an are the Southern Arabs, |ab"il al-Yaman, whose origin is traditionally assigned to the south-western corner of the Peninsula, while the descendants of #Adnn are the Northern Arabs, held to have made their first appearance in the northern half of the Peninsula. Whether this traditional division has a basis in truth is open to question. Certain data, for example, suggest that Saba" came from the north into the Yaman, though in the scheme of the Arab genealogists Saba" is the greatgrandson of |an and the father of 0imyar and Kahln, the eponyms of the two main branches of the Southern Arabs. The peoples of the ancient South Arabian statesSabaeans, Minaeans [qq.v.], and otherswere regarded as descendant of 0imyar, so that 0imyar in Arabic became the comprehensive term embracing the civilisation of these states. Few of those recognised without qualification as descendants of 0imyar played an important role during the Islamic period, the centre of the stage having by then been occupied by the sons of Kahln, among whom were numbered ayyi", Ma9i3, Hamdn and al- Azd. Among the subdivisions of al-Azd were al-Aws and al-9azra3, residents of Medina who rose to fame in Islam as the Prophet's Anr. La9m, 9assn, Kinda, and other tribes of Kahln became solidly established in the north and centre long before the beginning of Islam, so that a tribal map of Arabia in the 6th and early 7th centuries reveals a curious patchwork in which the ranges of many Arabs of Southern descent lie north of those belonging to Arabs of Northern descent. #Adnn, the putative progenitor of the Northern Arabs, appears to have been even more of a misty figure than |an, so that the Northern Arabs in popular practice often trace their descent back no further than #Adnn's son Ma#add or even his grandson Nizr. Mu'ar and Rab#a, sons of Nizr, were the eponyms of the two main branches of the Northern Arabs, the descendants of a third son, Iyd, having largely sunk out of sight by the time of Islam. |ays #Ayln, one of the two major divisions of Mu'ar, was of such importance that the term |ays was often used for all Northern Arabs. This division embraced Hawzin and Sulaym, and Hawzin alone included such notable tribes as 9af and the whole group of #$mir b. -a#a#a | [I:545a] (|u9ayr, #Uayl, 3a#da, Kilb, and Hill). 9indif, the other major division of Mu'ar, numbered in its ranks Hu9ayl and Tamm and above all Kinna, the tribe of which |uray9 formed a subdivision. Although the Northern Arabs by origin lacked the same identification with Arabdom that their Southern cousins enjoyed, the fact that the Seal of the Prophets came from the Northern tribe of |uray9 has redeemed their prestige under Islam in ample measure. AZ^RAT AL-#ARAB AMMAN, JORDAN, al-#Aaba, AL-UBAY|, SINAI, Taym", THE AL 0I$Z, EGYPT, RED SEA, SUDAN, ERITREA, ETHIOPIA, Bb-al-Mandab, GULF OF ADEN, ADEN, Mocha, TA#IZZ, al-Zuur, Zabd, al-0udayda, Kamarn, ABAL AL-AYR, Faraso Archipelago, -a#da, 3ayzn, TIH$MA TIH$MA, NAR$N, Abh, #AS^R YEMEN, IDDA, MECCA, al-if, 0a'n, MEDINA, 9aybar, 0am', ARABIAN SHIELD, al-#Ul, GREAT NAFD, Wd al-Sirn, #UN$ZA, Widyn,

Badana, #Anaza, IRAQ, IRAN, BASRAH, NEUTRAL ZONE, KUWAIT, Kuwait, alBin, al-Dibdiba, NEUTRAL ZONE, ABAL AMMAR, Wd, al-Rumab, Burayda, #Umayza, AL-N^R, al-Dawdim, THE ARABIAN, #Ir, Subay#, Wd, Ranya, Wd, B9a, Wd, Wd, Ta9l9, 0abawn, Wd al-9nd, 3awf Ibn Nir, -AN#$", Ma"rib, Ramlat al-Sab#atayn, Bayn, AL-KAWR, WESTERN ADEN PROTECTORATE, La3, 0A4RAMAWT, AL-AWL, EASTERN ADEN PROTECTORATE, al-Masla, al-Mukall, 0a3ar, ARABIAN SEA, -U|UR$, |i9n, Wd 0a'ramawi, -aywn, al-3alada AL-RUB#, AL-KH$L^, Rayd, Ab Bar, alDawsir, UWAY|, AL-AFL$, Yabrn, AL-$FRA, al-Sabb", AL-DAHN$", al9ar3, AL-RIY$4, NAD, UWAY|, al-0as, PERSIAN, al-Frisiyya, al-#Arabiyya, al-ahrn, Ra's Tannra, AL-DAMM$M BAHRAIN, GULF, 0ll, al-Dawa, QATAR, Ds, Zarakkh, Dalm, -r Ban Ys, AL-AFRA, al-3iw", TRUCIAL COAST, alShria, Dubayy, Ra's Musandam, AL-BURAYMI, GULF OF OMAN, al-Fu3ayra, Kalb, 9in, Musca, Mara, MUSCAT, OMAN, FAHD, AL-AR|IYYA, Nazw, 0AAR, MA-TRA, Ra's -awira, KURIA MURIA ISLANDS, Ra's Nawa, Salla, UF$R Ra's 4arbat #Al, Ra's Farra, 50, 0, 50, 100, 190, 200, 250, 300, KILOMETERS, 50, 0, 50, 100, 150, 200, 290, 34, 36, 38, 40, 42, 44, 46, 48, 50, 52, 54, 56, 58, 60, 30, 28, 26, 24, 22, 20, 18, 16, 14, 12, 58, 56, 54, 52, 50, 48, 46, 44, 42, 40, 38, 36, 12, 14, 16, 18, 20, 22, 24, 26, 28, 30, SAUDI ARABIA From Rab#a sprang the tribes of #Anaza, #Abd al-|ays, al-Namir, Ta9lib, and the strong group of Bakr b. W"il, one of whose members was 0anfa. Well before Islam the original groups of Mu'ar and Rab#a dissolved, early folk of Mu'ar moving to the territory on the Euphrates called after them Diyr Mu'ar and early folk of Rab#a to the territory on the Tigris called Diyr Rab#a. Many of their offshoots, however, remained behind in the Peninsula: Hu9ayl in the vicinity of al-"if; Sulaym in the mountains between Mecca and Medina; Tamm and 0anfa and various members of #$mir b. -a#a#a in the center; and #Abd al-|ays in the east. An attitude of hostility between |an and #Adnn, which went far back into the past, was enhanced by the rivalry that developed between the Anr of Medina and |uray9 of Mecca, so that it became a factor of extraordinary significance in the history of the early Islamic dynasties, the effect of which extended as far afield as Spain. The struggle between South and North finally faded away into an affair of dwindling consequence with the eclipse of the Arab element in the Islamic world. Only in one section of the Peninsula#Umnhas the ancient hostility endured down to the present as a vital force. For centuries the Northerners were known in #Umn as Nizrs, and the Southerners as Yamans. As the result of a civil was there in the early 18th century, the Northerners came to be called 9firs and the Southerners Hinws, a distinction which still carries weight. A major anomaly in the system appears in the case of |u'#a. A number of tribes Bahr", 3uhayna, Bal, Tan9, Kalb, and othersrecognised a common ancestor named |u'#a, but agreement was lacking as to whether he was a Southerner or a Northerner. Some said he was a son of #Adnn, while others said he was a grandson or later descendant of 0imyar. The genealogists also resorted to the device of declaring that all the Arabs were descended from three men|an, #Adnn, and |u'#abut without the suggestion that |u'#a represented a third element, neither Southerner nor Northerner. In the conflicts between the Southerners and the Northerners during the

early period of Islam, the tribes of |u'#a tended to side with the Southerners; genealogy was used for political purposes, the attribution to |u'#a of a descent from |an through 0imyar prevailed, and the tribe of Kalb of |u'#a advanced to the fore as champions of the Southern Arabs in the days of the Umayyads. In studying the history of Arabia from #Abbsid times to the present, one encounters great difficulty in determining the links between the tribes of a thousand years ago and the tribes of today. Oppenheim, Brunlich, and Caskel in their work Die Beduinen have made the most ambitious attempt so far with respect to the tribes of northern and central Arabia, but much remains to be done in spite of the laudable degree of success they have achieved. Information on the tribes during the time when | [I:545b] the government of Islam was in or near Arabia is fairly abundant, and the same is true of the last two centuries or so, but for hundreds of years in between their story remains for the most part concealed from view. Great migrations took place of which only trifling records have been recovered. Elements broke off from one tribe to join another, or whole tribes reshuffled themselves into new groupings. Popular tradition among the Bedouins has preserved some recollection of the changes, but this tradition is often far from trusthworthy. In the 4th/10th century al-Hamdn remarked on the tendency of tribes bearing a given name to associate themselves with stronger or more renowned tribes of the same name, and this tendency still holds true. In the time of the Caliph Ab Bakr the appearance of the false prophet Musaylima among 0anfa brought this tribe into disrepute; descendants of 0anfa in Na3d today prefer to name as their ancestor Rab#a, from whom 0anfa sprang, but so many other tribes have been named Rab#a and popular knowledge of the traditional genealogical system is so scant that the result is often complete confusion. The modern tribe of al-Dawsir has a tradition that its ancestor was named #Umar; the ordinary Dawsar today glibly identifies him as #Umar b. al-9ab without knowing who #Umar b. al-9ab was. The modern tribe of Ban 9fir in al-Bina of #Umn provides an example of the often unstable status of the tribes; although the Northern Arabs of #Umn are now called 9firs after this tribe, the tribe itself is notorious for the way in which it has shifted its allegiance back and forth between the Northerners and the Southerners. Some of the great tribes of the present, such as Tamm in the centre and Hamdn in the southwest, apparently represent in a generally faithful manner the ancient entities which bore these names, though many members of each have in the course of time broken away and lost their identity, while outsiders have attached themselves to this tribe or that and become completely absorbed into the community. The modern tribe of |an may be the residue of one or more segments of the original nation of Southern Arabs, or the connexion may be even more tenuous than this, despite the fact that the Bedouins of Arabia still associate this tribe with the father of all Southerners. To follow the vicissitudes of the tribe of |uray9 since the beginning of Islam, one would have to investigateamong other thingsthe history and current status of the many thousands of real and reputed sayyids and 9arfs scattered not only throughout Arabia, but from one end of the Islamic world to the other. Members of one modern tribe may tenaciously insist on their homogeneity in descent from a single ancestor, while members of another tribe readily admit that they are a confederation of diverse elements. The tribes of al-#U3mn and $l Murra, which migrated from the vicinity of Na3rn to Eastern Arabia about two centuries ago, maintain that they share a common descent from Hamdn of the Southern Arabs

through Ym. Their physical characteristics, their speech, and other facets of their life and history lend credence to this claim. On the other hand, large tribes such as #Utayba and Muayr in Inner Arabia are closely knit composites the original components of which probably first coalesced not more than five or six centuries ago. These confederations may be transitory, e.g., the confederation of Nu#aym in #Umn appears at present to be in the process of breaking down into its two | [I:546a] main constituents, $l B 9uraybn and $l B 9mis, with the old name of Nu#aym frequently being applied to $l B 9uraybn alone, while other members of Nu#aym, living c. 500 km. to the west, are no longer in close contact with the main body. Despite all the genealogical vagaries and uncertainties, it is impressive how much importance is attached by most of the Arabs of Arabia to purity of descent. Mankind is divided into those whose race is universally recognised as purely Arab (al) and those of a lower category whose blood is mixed or impure (9ayr al). The Bedouin who knows his immediate forebears through no more than six or eight generations is still profoundly convinced of his own nobility; his membership in a tribe of acknowledged purity of descent is sufficient guarantee that the line further back is without taint. Purity of blood is preserved by strict rules governing marriage, which among the Bedouins at least are seldom violated. The distinction between pure and impure, strongest among the Bedouins, is carried over to a considerable extent into the oases and towns, particularly those away from the coasts, where many of the townspeople keep alive their sense of affiliation with one tribe or another. Other townspeople are grouped together in Na3d under the appellation of Ban 9a'r, a generic term for those whose origin can not be traced back to a specific tribe. In the desert a few nomadic tribes by general consent bear the stigma of non-Arab descent. Among these is the tribe of al--ulaba [q.v.] in the north, the physical characteristics of whose members, as well as the popular traditions regarding them, suggest an origin hidden in an unusual aura of mystery, though there is no foundation for the oft-repeated legend that they are the offspring of wandering Crusaders. Others of this category in the north are Hutaym and al-9arrt. The tribe of al-#Awzim in the east has succeeded in rising somewhat above its inferior status as a result of its prowess in battle during the past forty years in the ranks of King #Abd al-#Azz of Saudi Arabia. Along the coasts, in the seaports, and in towns not far inland are found the greatest infusions of foreign or nondescript racial elements. In some cases these are well defined types from abroad, such as Somalis and Indians along the southern coast and on the Red Sea; banians or Indian merchants are also numerous in the ports of the Sultanate of Muscat and on the Persian Gulf. In other cases people of obscure origin are classified primarily on the basis of their occupations, such as the servants in Southern Arabia called -ibyn and A9dm. Because many Muslims from distant lands desire to live and die on hallowed ground, Mecca contains a strikingly heterogeneous population, in which the so-called Javanese and Bu9ran colonies (made up respectively of settlers from Indonesia and Central Asia) are among the largest. Certain foreign elements, such as the Abyssinians from the west and the Persians from the east, have a history in Arabia going back two millennia or more, yet they have never immigrated in great force and few are the places where the majority of the population has not retained its basic Arab character, at least in such important aspects as language and religion. Other foreign elements, such as some of the Baluchis settled in the interior of #Umn, have

become so thoroughly Arabised that they are now considered by their Arab neighbors as al. | [I:546b] Racial matters in Arabia are often intermingled with religious considerations. Descendants of the Prophet, who usually bear the title of 9arf in al-0i3z and sayyid in the Yaman and 0a'ramawt, sometimes form a privileged caste in the community, while at other times they lead the life of simple nomads in the desert. The numerous sayyids of 0a'ramawt, who enjoy exceptional prestige, all claim descent from a small group of families who emigrated from #Ir to 0a'ramawt in the first half of the 4th/10th century. In #Umn the title sayyid is popularly accorded to the Sultan of Muscat, who does not claim descent from the Prophet, and in Na3d the incidence of 9arfs is remarkably low. In Eastern Arabia most of the sayyids are found among the 9#ites, a fact which prompts the Sunnite Bedouins to question the authenticity of their descent. The Jews, whose history in Arabia goes back well into the pre-Islamic period, may have been in the beginning Israelites who moved southwards or Arabs converted to the Judaic religion or a combination of the two. Once fairly numerous in the south-west, almost all of the Jews have departed within the last few years for Israel. Slavery as an institution sanctioned by Islam flourished in the Peninsula until very recent times, though now it appears to be slowly dying out. The great majority of the slaves came from Central Africa, and Negro blood is found even in villages of alAfl3 in the heart of Arabia. Like other Islamic lands, Arabia has remained uncursed by a colour bar, and emancipated slaves have on occasion attained positions of influence in society. Another Negro element exists in the so-called Takrina, who come halfway across Africa, often on foot, to make the pilgrimage; some of these stay on to eke out a living in the Holy Land, where their huts stand in the outskirts of 3idda. Although migrations of persons and tribes from place to place within the Peninsula and from the Peninsula to the fertile lands farther north have been common throughout the centuries, only a relatively small proportion of the Arabs of Arabia have shown a fondness for crossing the seas to settle in foreign lands. Chief among these have been the people of #Umn, who since ancient times have moved down along the coast of East Africa and into southern islands such as Zanzibar, and the people of 0a'ramawt, many of whom have more recently established themselves in the Indonesian Archipelago, the Malay Peninsula, and India, where they have been influential in the domains of the Nim of 0aydarbd. Arabs of Eastern Arabia have moved across the Persian Gulf to occupy much of the Iranian coast, and seafarers from the Yaman have founded tiny colonies in such distant spots as Cardiff in Wales. (vii) history 1. Pre-Islamic Arabia before the First Millennium B. C. The Arabian Peninsula has as yet no history earlier than the first millennium B. C., though future investigations will certainly bring many new facts to light. Excavations have been few and limited in extent, and even the surface in many regions has not been scrutinised by trained searchers.

Scattered finds indicate that the Peninsula was inhabited in both Palaeolithic and Neolithic times, but nothing is known about who the people were or where they came from. The problem of the site of the original home of the Semites is still a matter of speculation. The Semitic nomads who began filtering into the Fertile Crescent from the adjacent deserts in the fourth millennium B. C. relied chiefly on the donkey, a beast not as well adapted as the camel to wide ranging in waterless tracts. The cuneiform inscriptions of Mesopotamia contain numerous references to Magan, Melu99a, and Dilmun, places which may have lain in Arabia, though much of the geography of the time remains | [I:547b] vague. The Egyptian records relating to Punt are similarly imprecise. Egypt's connections with Sinai and the Red Sea are very ancient, and the availability of frankincense in Southern Arabia led to indirect or even direct intercourse at an early period. A development of vast importance in the later history of Arabia and the Islamic world occurred, probably in the early second millennium B. C., with the devising of a system of alphabetic writing from which later Semitic alphabets, including South Arabic and North Arabic, derived. Tribal migrations about which little is yet known took place inside Arabia; in this millennium many of the sons of |an may have gone south to their new homes. The last centuries of this millennium were a time of change, with the Iron Age beginning in the Near East and the Semitic Aramaeans entering the Fertile Crescent in strength. The domestication of the camel appears to have been achieved during this period in Arabia, the first contribution of the Peninsula to the material progress of mankind. Arabia during the First Millennium B. C. The tenth chapter of Genesis, believed to belong to about the 10th century B. C., mentions Joktan and Hazarmaveth, who may be identified with |an and 0a'ramawt. In the same century Solomon sent vessels in the Red Sea from the port of Ezion-geber, while his caravans traded with Northern Arabia. The location of Ophir, from which Solomon received gold and other products, continues to be a mystery. From the 9th century on, Assyrian and Babylonian inscriptions make frequent mention of the Aribi, camel-owning inhabitants of Northern Arabia who paid tribute to the masters of Mesopotamia. In recent years knowledge of the ancient civilisation of Southern Arabia has expanded tremendously. So many new inscriptions and other traces are coming to hand that current conclusions must often be regarded as tentative. An intensive review of the chronology is in progress, with the general tendency favoring a downward revision of dates. Available information suggests that organised states came into being in Southern Arabia during the second half of the first millennium B. C. The four chief statesSaba" of the Sabaeans, Ma#n of the Minaeans, |atabn, and 0a'ramawtthrove on agriculture and commerce. The Mrib dam in Saba" was the most imposing structure in an elaborate system of irrigation. For centuries the Southern Arabian merchants monopolised the frankincense trade and controlled traffic between India and the West, sending their goods by overland routes which traversed Arabia from south to north. Colonies were established in Northern Arabia, and evidence of business activity has been found in Egypt, the Aegaean, and the Persian Gulf region. Strong

Graeco-Roman influence on Southern Arabian culture is shown by archaeological discoveries. Southern Arabians migrated to Abyssinia, to which they gave its name, and their influence reached along the eastern coast of Africa. Many impressive buildings in Southern Arabia were temples dedicated to pagan deities. The earlier rulers of Saba", who bore the title of Mukarrib, combined the functions of prince and priest; later they gave way to the more secular rule of kings. [For details see al-yaman.] In the north, Aramaean influence was strong in the oasis of Taym", briefly the capital of the Neo-Babylonian Empire under Nabonidus (regn. B. C. 556-539). Dedan, near modern al-#Ul, became the center of a culture now called Liynitic, using an | [I:548a] alphabet derived from South Arabic. 9amd, mentioned as a tribe in an Assyrian inscription of the 8th century B. C., held Egra (al-0i3r or Mad"in -li) just north of Dedan. The recent finding of widely dispersed 9amdic inscriptions has raised new questions regarding the spread of this derivative of the South Arabic script and those who used it. After the Persian capture of Babylon in B. C. 539, a short-lived satrapy called Araby was created in Northern Arabia. Darius I (regn. 521-485), who sought to stimulate trade via the Persian Gulf, sent out Scylax of Caryanda, who sailed from India to the northern end of the Red Sea. The world's knowledge of Arabia increased through Alexander's expeditions and the reconnaissance of the Persian Gulf carried out by Nearchus the Cretan. Alexander died in 323 just as he was planning the circumnavigation of the Peninsula and the subjugation of its peoples. Not long afterwards the Greek naturalist Theophrastus wrote an account of Southern Arabia and its products. The Ptolemies of Egypt, who often pursued a forward policy in the Red Sea, threatened the trade monopoly held by the Arabs, while the Seleucids of Syria promoted the use of the northern routes from India. The establishment of the Parthian state in the mid-3rd century B. C. weakened the Seleucids, but Antiochus III was still strong enough to conduct an expedition in 205-204 against Gerrha on the Arabian shore of the Persian Gulf. Late in the millennium the Nabataeans, a people of Arab stock with their capital at Petra, began playing a considerable role in the affairs of Syria, and Arabs appeared as rulers in various places in the Fertile Crescent, such as Charax Spasini at the head of the Persian Gulf. Arab vassal chiefs enjoyed a large measure of autonomy under Parthian rule, and the immigration of Arabs into Mesopotamia went steadily on. Towards the end of the 2nd century B. C. Eudoxus of Cyzicus sailed from Egypt to India, and in time Westerners learned the secret of using the south-west and north-east monsoons for voyaging across open water. The growing competition of the West seriously undermined the commercial dominance of the Southern Arabians, in whose homeland radical changes were taking place. An important event near the close of the 2nd century, later taken as the starting point of the Sabaean era, has been plausibly connected with the assumption of royal power in Saba" by the mountain tribe of Hamdn. Both the kingdoms of Ma#n and |atabn came to an end in the 1st century B. C., and the |atabnian capital Timna# in Bayn was destroyed. Rome, which had made a client state of Petra in B. C. 60, coveted the wealth of Arabia Felix. Augustus sent the

Prefect of Egypt, Aelius Gallus, supported by Nabataeans from Petra, on a long march in B. C. 24 towards the incense country, but the expedition, finding the deserts inhospitable and its Arab allies treacherous, did not get beyond Saba". [For details see al-yaman.] Arabia during the First Six Christian Centuries. About A. D. 50 an unknown author wrote in Greek the Periplus of the Erythraean Sea, an invaluable account of trade in the Red Sea and along the southern coast of Arabia. The King of 0a'ramawt in his capital 9abwa controlled the whole territory from Bayn in the west to ufr in the east, while the King of Saba" and of 9 Raydn (a recently assumed title) sat in afr in the mountains of the Yaman, where the power of 0imyar was growing. | [I:548b] In A. D. 105 or 106 the Roman province of Arabia was created in the old Nabataean domain, stretching from Ayla ( al-#Aaba) in the south to al-Namra in the northeast, with its capital first at Petra and later at Bostra. Merchants were encouraged to trade via the Red Sea through the port of Ayla, and Bedouin raids were warded off by the building of a limes along the desert borders. Roman knowledge of the Peninsula in the mid-2nd century was summarized by the geographer Claudius Ptolemy. Arda9r I, the first Ssnid (d. A. D. 241), is said to have founded a city in Eastern Arabia and to have induced the tribe of al- Azd to settle in #Umn. Ssnid authority on one flank of Northern Arabia and Roman authority on the other were challenged by the Arab rulers of Palmyra, but the Roman Emperor Aurelian defeated Queen Zenobia and captured her desert stronghold in 272. Something of the old glory of Saba" and 9 Raydn was regained by 9ammar (or 9mir) Yuhar#i9, who signified his triumphs about the end of the 3rd century by adding the names of 0a'ramawt and Yamanat to his royal title. His reign was followed by a relapse into weakness, during which Na3rn on the northern border was besieged by the La9mid Mar" (= Imru") al-|ays, extravagantly described as King of all the Arabs in the oldest North Arabic inscription known (al-Namra 328). Later Kings of Saba" made their title even longer by appending and of their Arabs in the mountains and the lowlands. One of the most obscure periods in Arabian history fell in the 4th and 5th centuries. The decline and impoverishment of the Roman Empire affected the Peninsula, where urban civilisation waned and the simpler ways of nomadism attracted more adherents. Christianity with its promise of a better life in the hereafter made headway in Arabia as elsewhere. The Arabs proved particularly susceptible to the doctrines of Nestorianism, coming from Mesopotamia, and Monophysitism, coming from Egypt and Abyssinia. The Abyssinians occupied the Yaman for a brief period in the 4th century, with #6zn, the first Christian King of Aksum, proclaiming himself ruler of 0imyar, Raydn, Saba", etc. 9pr II (regn. 310-79), called 9 'l-Aktf by the Arabs, subjugated Eastern Arabia; the Ssnid yoke was later removed, only to be reimposed shortly before the dawn of Islam. Judaism also made a successful appeal in Arabia, among its reputed converts being the King of Saba" in the early 5th century, Abkarib As#ad, known to Arab tradition as Tubba# As#ad Kmil, and one of its centres being the oasis of Ya9rib (later Medina).

Both the Ssnids and the Byzantine successors of Rome found it necessary to protect their territories from the unruly folk of Arabia by relying on buffer states ruled by Arab princes, the La9mids [q.v.] standing guard on the edge of Mesopotamia and the 9assnids [q.v.] shielding Syria. The two client states, like their suzerains, often came into conflict. In the first half of the 6th century al-0rith b. 3abala, the greatest of the 9assnids, proved stronger than al-Mun9ir b. M" al-Sam", the most famous of the La9mids. In the late 5th century the chief of the Southern Arab tribe of Kinda [q.v.], 0u3r $kil al-Murr, assumed the leadership of a confederacy of tribes in Central Arabia, but this loosely knit Kingdom of Kinda lasted only about half a century before it was overthrown by al-Mun9ir the La9mid. In the 6th century Southern Arabia lay open to | [I:549a] attack by the Christian Kings of Aksum and the Ssnid 9usraw I An9irwn (regn. A. D. 531-79). Persecution of the Christians of Na3rn by the Judaising Arab 9 Nuws [q.v.] led to a new Abyssinian occupation of the Yaman c. 521. The Abyssinian Abraha [q.v.] as ruler of the Yaman carried out the last repair of the dam of Mrib before its final abandonment, marched into the heart of Na3d on a campaign against the Arabs of Ma#add, clients of the La9mids, and, according to Islamic tradition, undertook an unsuccessful expedition against Mecca in the Year of the Elephant (c. 570). Under 9usraw the Persians evicted the Abyssinians, and the Yaman was Persian territory at the rise of Islam. Mecca, a town of some antiquity on the main route paralleling the Red Sea, achieved greater prominence and prosperity in the late 6th century, aided by foreign domination of the Yaman and chaotic conditions along the northern routes resulting from the long drawn out wars between Persia and Byzantium. The Meccan merchants of |uray9 showed astuteness and industry in profiting from their participation in international trade. The last centuries of this period gave birth to the form of Arabic now called classical, the dialectal sources and the exact process of the development of which remain uncertain. Used by the poets of the 3hiliyya, many of whom were Bedouins and some Christians or Jews by faith, this language became the instrument of expression for the supreme masterpiece of Islam, the |ur"n, and the great works of Arabic literature in succeeding ages (see #arabiyya). 2. Islamic Middle Ages Muammad and the Rise of Islam (A. D. c. 570-632). About A. D. 570 Muammad [q.v.] b. #Abd Allh of |uray9 was born in Mecca, then a principal centre of pagan worship. Only traditional accounts survive of Muammad's early years, during which he became well acquainted with the tribal structure of both urban and nomadic life and saw something of the world outside Arabia while accompanying merchant caravans to Syria. About 610 he received his first revelation; two or three years later he began preaching in public, after which the nature of Islam was elaborated upon in a series of revelations during the rest of his career as God's Messenger and Prophet.

The men in authority in Mecca did not welcome Muammad's message. A small body of Muslims went into exile in Christian Abyssinia; later the whole Muslim community migrated northwards from Mecca to Ya9rib, an event taken afterwards as having marked the beginning of the Islamic era (A. H. 1/A. D. 622). During the ten years Muammad maintained his capital at Medina, he erected a state guided in all its functions by the precepts of Islam. Two revolutionary concepts emerged which transformed the face of Arabia. The |ur"n, as emphasised by the divine revelations of which it consisted, was Arabic, a standard under which all Arabs could unite. Arabia had never before known an entity larger than relatively petty states or independent tribes and tribal confederations, usually at logger-heads with each other if not openly at war. At the same time, the |ur"n and Islam were not limited to the Arabs: the |ur"n is a revelation to all men, and under Islam the noblest man is the most Godfearing, not the one of highest lineage. This | [I:549b] universal appeal opened the way for Islam to go far beyond the borders of Arabia. Muammad's efforts during the Medinan period were devoted in large measure to settling affairs with Mecca, which was finally incorporated in the Islamic state in 8/630. Before this a fair number of tribes had been won over to Islam, but the great flood of applications to join Islam from tribes all over the Peninsula did not come until 9/630-1, the Year of the Delegations. Muammad died in 11/632, before there had been time to anchor the |ur"nic religion in the hearts of all who had taken the name of Muslim. Neither had there been time to carry Islam abroad, though a halting attempt had been made in that direction, and the moment was indeed ripe for shattering the fragile shells of Byzantine and Ssnid defences in the Fertile Crescent. The First Three Caliphs (11-35/632-56). Soon after Ab Bakr (regn. 11-13/632-4) succeeded Muammad as head of the Islamic state, many tribes reasserted their independence, with prophets in several cases preaching doctrines contrary to Islam. Ab Bakr reacted vigorously, dispatching Muslim columns to Central Arabia, Barayn, #Umn, and the Yaman. When 0a'ramawt, which held out the longest, was subdued, the Arabian Peninsula for the first and last time in history was effectively united throughout its length and breadth. The other great achievement of Ab Bakr's brief rule was the inauguration of the grand programme of Muslim conquests outside Arabia. After invading #Ir 9lid b. alWald marched across the Syrian Desert in 13/634 to participate in a victory over the Byzantines. The conquests started by Ab Bakr were carried forward with verve during the rule of #Umar (13-23/634-44). #Ir was taken from the Ssnids, and Arabs from both the Northern and the Southern tribes peopled the newly founded military settlements of alBara and al-Kfa. After a decisive victory over the Byzantines at al-Yarmk and the capture of Jerusalem, #Umar came to visit this holy city, the first journey of a Caliph beyond the confines of Arabia. Islam next advanced into Egypt, the occupation of which brought about stronger economic and cultural ties with Western Arabia. Although #Umar is reputed to have ordered the expulsion of all Christians and Jews from the Peninsula, numbers of them lived on there for a long time to come.

In the days of #U9mn (regn. 23-35/644-56) of the House of Umayya, wealth and luxury abounded in Medina and Mecca, into which poured booty from the lands recently subdued. #U9mn had no ear for the voice of Ab 9arr decrying the decay of the stern and frugal virtues of earlier Islam. Even more dangerous to the future of Arabia and Islam was the rift developing between the most powerful figures in the state, which led to the murder of #U9mn in Medina. The Struggle over the Caliphate (35-73/656-692). The rift in high circles widened into a chasm when #Al, Muammad's son-in-law and cousin, came to the fore as Caliph on the death of #U9mn. Muammad's wife #$"i9a and his Companions al-Zubayr and ala rose in opposition to #Al, who left Medina to march against them in 36/656. In the Battle of the Camel #Al overthrew his rivals and won #Ir, only to find himself faced with a more formidable adversary in #U9mn's Umayyad kinsman Mu#wiya, the governor of Syria. When #Al fixed his capital at al-Kfa in order to marshal strength against Mu#wiya, Medina lost | [I:550a] the preeminence it had held since the Prophet's migration. #Al's tactics against Mu#wiya so exacerbated the extremists among his own followers that they turned against him as the 9awri3. Despite the crushing victory #Al gained over these seceders at al-Nahrawn in 38/659, their party survived, Arabia long providing a fertile field for its propaganda. Mu#wiya was proclaimed rival Caliph in Jesuralem, and his forces clashed with #Al's in Western Arabia from Medina to Na3rn and the Yaman. When a 9ri3 assassinated #Al in 40/661, the #Alids set up his son al-0asan as Caliph in al-Kfa, but he soon renounced his claims in favor of Mu#wiya, who thus temporarily reunited the community of Islam. For the rest of Mu#wiya's life no serious rising took place against the new Syrian Caliphate, but resentment was stirred up by his advocacy of hereditary succession. After the accession of Yazd b. Mu#wiya (regn. 60-4/680-3), #Al's second son al-0usayn left Mecca to rally support in #Ir, only to fall a martyr at Karbal" in 61/680. His death cleared the field for a stronger candidate, #Abd Allh b. al-Zubayr, the foremost representative of the sons of the Prophet's Companions. Yazd's army defeated the rebellious Medinans in the battle of 0arrat Wim and laid siege to Mecca, Ibn alZubayr's stronghold, where the Ka#ba caught fire, but Yazd's death brought a pause in the hostilities. Ibn al-Zubayr won recognition as Caliph in nearly every quarter of Islam; in fact, had he proceeded to Syria immediately, he might well have destroyed the Umayyad power forever. While Ibn al-Zubayr lingered on in Mecca, #Abd al-Malik (regn. 65-86/685-705) of the Marwnid branch of the Umayyads gradually regained ground outside Arabia. The 9awri3, who had at first leagued themselves with Ibn alZubayr, turned against him, the 9ri3 Na3da b. #$mir of Ban 0anfa making himself master of much of Arabia, only to be overthrown by another 9ri3, Ab Fudayk. #Abd al-Malik gave al-0a333 b. Ysuf command of an army which captured Mecca in 73/692 after a long siege. Ibn al-Zubayr fell in the struggle, leaving the Holy Land of Islam in the hands of the Umayyads. Another Umayyad army marched to Eastern Arabia and put an end to Ab Fudayk. Arabia under the Umayyads (73-132/692-750).

 The Umayyads of Syria regularly appointed governors for Medina and Mecca, and exercised a measure of control, often shadowy, over other parts of Arabia. Powerful Umayyad governors of al-Bara such as al-0a333 and Yazd b. al-Muhallab made their word law in the Persian Gulf and along its Arabian shore. The Umayyad Caliphs honoured the sanctity of the Holy Cities in Arabia and lavished large sums on their shrines, even while favouring at times the claim of Jerusalem, which was easier of access, to an equal or higher rank. During much of this period Western Arabia was at peace, enjoying a prosperity such as it was not to know among the dissensions of later ages. The Umayyads developed the irrigation system, and many personages of Islam lived in their days of retirement on estates near Medina, Mecca, or al-"if. The Holy Cities became renowned not only for Islamic learning but also for indulgent living, poetry, and singing. The intense rivalry in Umayyad politics between the Northern Arabs and the Southern Arabs had its repercussions in Arabia, where Kalb, the principal | [I:550b] tribe among the Southerners, owned land in Wd al-|ur near Medina. Towards the end of the Umayyad period an alliance of 9awri3 was formed under the leadership of #Abd Allh b. Yay lib al-0a of Kinda and Ab 0amza of al-Azd. Ab 0amza took Mecca, won a victory at |udayd in 130/747, and then entered Medina, while lib al-0a supported him from their base in 0a'ramawt and the Yaman. Despite the waning might of the Umayyads, Marwn II summoned sufficient strength to overcome these 9ri3 chiefs, but only after they had contributed to his final undoing. Mecca was also used by the #Abbsids as a centre for their plot aiming at the supersession of the Syrian Caliphs. Arabia under the Early #Abbsids(132-266/750-879). The #Abbsid transfer of the Caliphate to #Ir enhanced the importance of the Persian Gulf as a seaway for trade reaching out to China and East Africa. Wares bound to and from the #Abbsid capital passed through al-Bara, while in the Gulf itself Srf on the Persian side in the 3rd/9th century became the busiest port. #Abbsid authority in Arabia kept its strength for not much over a century, during which time governors were sent to the Holy cities and the Yaman, and on occasion to the central and eastern regions. The earlier Caliphs, notably al-Mahd and Hrn, and their wives, notably Zubayda, were diligent in making the pilgrimage and encouraging their subjects to do so by improving communications and the amenities of the route. A sect of the 9awri3 known as the Ib'iyya set up its own Imamate in #Umn under al-3uland b. Mas#d of al-Azd, but an #Abbsid expedition under 9zim b. 9uzayma defeated and killed al-3uland in 134/752. Soon afterwards this Imamate was revived to endure with few interruptions for the next four centuries. #Umn, however, was an out of the way region, and the 9awri3 on the whole gave the #Abbsids little trouble. [Cf. #umn .] Taking the place of the 9awri3 as a thorn in the Caliphs' flesh were the #Alids [q.v.], both 0asanids and 0usaynids. Through skilful propaganda the #Abbsids in their campaign against the Umayyads had forestalled the #Alids and usurped the leadership

they regarded as rightfully theirs. For this the #Alids never forgave them, and one after another they contested the #Abbsid title to rule. Even though the #Abbsids themselves came from a Meccan ancestor close to the Prophet, the #Alids almost invariably found ready followers in Arabia; in the Holy Cities their rallying cry inspired the hope of regaining the place lost to Damascus and Baghdad. The first #Alid pretender in Arabia was the 0asanid Muammad al-Nafs al-Zakiyya, who appeared as the Mahd in Medina and had his claim to the Caliphate certified by no less a scholar than Mlik b. Anas, but all to no avail when he fell in 145/762 before the troops of al-Manr. A major split took place among the #Alids following the death of their sixth Imam, 3a#far al--di, c. 148/765. The main body, giving loyalty to 3a#far's son Ms alKim and five of his descendants, came to be known as the Twelvers. Others, the Seveners, advocated the cause of Ism#l b. 3a#far and his son Muammad, for which they worked, often in secret, in the movement of Ism#lism. As time went by the Ism#ls in particular tended to attract to their side the discontented and oppressed elements of society, enemies of the ruling classes. | [I:551a] Another 0asanid pretender, al-0usayn b. #Al, met a martyr's death fighting against an #Abbsid army at Fa99 near Mecca in 169/786. The #Alid cause, however, made progress in the Yaman, where it received the support of the great jurist al-9fi#, who finally won a pardon after being delivered as a prisoner to Hrn's presence. The end of the 2nd century H. saw a new upsurge of #Alid strength in Western Arabia: in Mecca the 0usaynid al-0usayn al-Afas put forward Muammad al-Db3, a son of 3a#far al--di, while the 0asanid Muammad b. Sulaymn established himself in Medina. These pretenders did not hold their ground against the #Abbsids, but greater success was achieved by Ibrhm al-3azzr, a grandson of 3a#far al--di, in the Yaman. Yielding to the tide of pro-#Alid sentiment, the Abbsid Caliph al-Ma"mn designated #Al al-Ri', the eighth Imam of the Twelvers, as his heir apparent and substituted #Alid green for #Abbsid black as the royal colour, but this change evaporated with #Al's death in 203/818. To cope with the #Alid threat in the Yaman, al-Ma"mn appointed as his governor there one Muammad, who claimed descent from Mu#wiya's lieutenant Ziyd b. Abh. Refounding the city of Zabd in 204/820 and carving out a domain for himself, Muammad established the dynasty of the Ziydids [q.v.], which, while according nominal allegiance to the #Abbsids, was actually the first of the numerous independent dynasties to spring up in Arabia as the Caliphate disintegrated. Although not a strong Caliph, al-W9i (regn. 227-232/842-847) executed a vigorous policy in Arabia. When Bedouins of Sulaym made the region around the Holy Cities unsafe with their depredations, al-Wthi dispatched the Turkish general Bu9 the Elder to bring the culprits to heel. For the next two years Bu9 campaigned against other tribes, climaxing his operations in 232/847 with a hard won victory over Numayr at Ban al-Sirr deep in the interior, after which a man of U'9 in Na3d was appointed governor of al-Yamma, Eastern Arabia, and the pilgrim route to Mecca.

Following the death of al-Mutawakkil in 247/861, the career of the #Abbsids both at home and in Arabia took a turn for the worse. The dynasty of the Ya#furids [q.v.], claiming descent from the ancient Tubba#s of 0imyar, arose in the highlands of the Yaman with -an#" as capital. 0a'ramawt secured its independence, and local rulers set themselves up in the east, where #Al b. Muammadeither a genuine 0usaynid, as he gave himself out to be, or a member of #Abd al-|aysbegan an agitation among the nomadic tribes. Another 0asanid revolt in Mecca, inaugurated by Ism#l b. Ysuf alU9ay'ir, led to the establishment under Ism#l's brother Muammad of a new state in al-Yamma, where these U9ay'irids maintained themselves until submerged by the onrush of |armaianism. Another blow was dealt the #Abbsid empire by the recalcitrant governor of Egypt, Amad b. ln, who by occupying Syria broke down the control once exercised over the tribes of the Syrian Desert. The most direct menace to the empire, however, came from the agitator in Eastern Arabia, #Al b. Muammad, who transferred his activities to Southern #Ir, where he stirred up the Zan3, the negro slaves laboring in the salt marshes, in a massive insurrection (255-70/863-83) extending as far as the Holy Cities. Ism#ls and |armaians in Arabia (266-567/879-1171). At this juncture in #Abbsid affairs | [I:551b] the rapidly spreading movement of Ism#lism (see ism#liyya ) took full advantage of its opportunities. Ism#l missionaries carried out a well laid plan of penetration, with the Persian Gulf coast and the Yaman as the principal foci for their activity in Arabia. As these two parts of Arabia remained relatively isolated from each other, the connexion between later developments in them was slight. Ism#lism was first introduced into the Yaman by Ibn 0aw9ab ( Manr al-Yaman) and #Al b. al-Fa'l in 266/879-80. Collaborating closely, these two won many followers, and #Al occupied both -an#" and Zabd for brief periods. The Ziydids and the Ya#furids fought the Ism#ls, and a new opponent arose against them in 280/893 with the arrival in the Yaman of the first Zayd Imam, al-Hd Yay, a grandson of the 0asanid al-|sim al-Rass (d. 246/860), who had fashioned legal foundations for a Zayd government closer to Sunnism than to the extreme 9#ism of the Ism#ls. The two Ism#l leaders eventually fell out, and by 303/915 both were dead, but their doctrines did not die with them. Ism#lism appeared c. 286/899 in Eastern Arabia, where under Ab Sa#d al-0asan al-3annb and his son Ab hir Sulaymn a strong state was organised. The name | armaian, the origin and meaning of which are still in doubt, remains the popular designation for this particular aspect of Ism#lism, though its application is not restricted to this region. The #Abbsids were too feeble to prevent these |armaians from sacking al-Bara and al-Kfa, and in 317/930 they entered Mecca and carried off the Black Stone to their new capital al-As" (al-0as). With the conquest of #Umn soon thereafter the |armaians held the greater part of Arabia. These disturbances prompted the 0usaynid Amad b. #^s, the most famous ancestor of the sayyids of Southern Arabia, to leave al-Bara on a migration ending in 0a'ramawt, where Ib's from #Umn then held the upper hand.

New threats to the #Abbsids came from the Byids of Iran and the I99dids of Egypt, who reached out at times to Mecca, though neither got a lasting foothold there. The Byids, who by taking Ba9dd in 334/945 assumed de facto authority over the #Abbsid realm, also brought #Umn within their sphere. Ab hir died in 332/944, and the |armaians at the behest of the Ism#l Fimids of North Africa restored the Black Stone to Mecca in 339/950-1. Under al-0asan alA#am, a nephew of Ab hir, the |armaians joined the Fimids in a pincer movement on Syria and Egypt, the former exerting pressure from the east as the latter advanced from the west. However, after the Fimids occupied Egypt in 358/969, the | armaians broke with them and sided with the Byids in resisting their designs on Syria. Damascus was captured by al-0asan in 360/971, but he was repulsed on two expeditions against Egypt before reaching the newly founded Fimid city of Cairo. Following the death of al-0asan, the |armaian government was placed in the hands of a Council of six sayyids . The Fimids won a military victory over the |armaians, but had to pay a large sum to induce them to return to al-As". The |armaians lost #Umn in 375/985-6, were checked by the Byids in #Ir and defeated in their own territory by a chief of al-Muntafi, who plundered al-|af. [Cf. also armaians.] | [I:552a] About the mid-4th/10th century the 9arfate of Mecca [for which see makka ], destined to last a thousand years, was established by a family of 0asanids known as the Mswids. The most prominent member of this family was Ab al-Fut al-0asan, who in 402/1011-2 tried to make himself Caliph, only to be thwarted by the Fimids, liege lords of the 9arfs. Contemporary with the early Mswids were 0usaynids descended from al-0usayn al-A9ar, a younger brother of the fifth 9#ite Imam, who began ruling as amrs of Medina. This line, which lasted until the 9th/15th century, came later to be known as the House of Muhann. An offshoot of Ism#lism was the Druze movement, which had its origins during the reign of the Fimid al-0kim. The Druze al-Mutan sent a letter to the |armaian sayyids of Eastern Arabia, proposing that they combine forces on the basis that they shared a common doctrine, but nothing concrete came of this. Early in the 5th/11th century the Ma#nids [q.v.] came to power in Aden and 0a'ramawt, and the Ziydds in the Yaman gave way before the Na3ids [q.v.], originally their own Abyssinian slaves. Ism#lism in the Yaman enjoyed a revival under the -ulayids [q.v.], rulers sprung from the tribe of Ym who held -an#" as nominal vassals of the Fimids, while the Zayd Imams kept their base at -a#da. In 443/1051 Nir-i 9usraw visited al-As", where he found the Council of Six still in control. The details of his eyewitness account of the |armaian state in its later days are unfortunately not supported by corroborating testimony. The 9#ism of the Byids, |armaians, and Fimids aroused a Sunnite reaction championed by the Sal3 Turks, whose leader u9ril took Baghdad in 447/1055. A Sal3 of Kirmn, |wurd |ar Arsln, brought #Umn under his sway. About this time Srf was yielding its place as the chief port of the Persian Gulf to the island of |ays, the rulers of which made themselves also lords of #Umn, where in the mid-5th/11th

century a break came in the line of Ib' Imams. For the next three and a half centuries records survive of only one Imam. The -ulayhids of the Yaman seized Aden from the Ma#nids and also expanded northwards, the authority of the Mswid 9arfs over Mecca having faded away. In 455/1063 the -ulayid #Al b. Muammad installed an agnate branch of 9arfs, the H9imids, in Mecca. Under Malik 9h in Ba9dd the Sal3s reached the zenith of their power, and thanks to him the shadowy #Abbsid of the day had lipservice paid to him in the Holy Cities as the Caliph of Islam. Malik 9h and his minister Nim alMulk concerned themselves with the affairs of the pilgrimage, spending freely to put them to rights. About 470/1077-8 the |armaians of al-As" met their final defeat at the hands of a native dynasty, the #Uynids [q.v.] of the tribe of #Abd al-|ays. There is no trace of | armaianism, left today among the Arabian people. The 9#ites of al-|af and modern al-0as, sometimes described as the remnants of the |armaians, are in fact orthodox 3a#fars of the Twelver persuasion or 9ay9s. In 461/1068-9 Aden was granted as a dowry to a remarkable woman of the -ulayid house, Sayyida bint Amad, upon her marriage to al-Mukarram Amad b. #Al al-ulay, and soon afterwards the government of the town was transferred from the Ma#nids to the Zuray#ids [q.v.], who like the -ulayids were Ism#ls of the stock of Ym. The Zuray#ids | [I:552b] ruled Aden for nearly a century, gradually acquiring a larger measure of independence. Under Sayyida, into whose hands al-Mukarram placed the authority of the state so that she was recognized by the Fimid Imam as Suzerain of the Kings of the Yaman, the -ulayids enjoyed their last days of real dominion. Her death in 532/1137-8 marked the effective end of the dynasty, the succeeding representatives of which were a feckless lot. Upon the death of the Fimid Imam of Egypt al-Mustanir in 487/1094, two parties arose among the Ism#ls which have persisted to the present day. From the party supporting al-Mustanir's eldest son Nizr descended the Isma#l Assassins of Alamt and the 93as, the head of many of whom is now the $9 Khn. The party favoring al-Mustanir's youngest son al-Musta#l Amad, allied with the -ulayids through Queen Sayyida, was strong in the Yaman. The rule of Amad b. Sulaymn, one of the greatest of the earlier Zayd Imams, ran from 532 to 566/1137-71, during which time he held -a#da, Na3rn, and al-3awf, occupied -an#" and Zabd, and made his influence felt as far north as 9aybar and Yanbu#. Like the -ulayids, the Na3ids also produced a queen to rule during the dynasty's declining years, #Alam, originally a slave girl, whose death in 545/1150-1 was followed about a decade later by the ephemeral sway of the Mahdids [q.v.], who called themselves 0imyarites and were accused of being 9awri3. The Fimids of Egypt succumbed to the Ayybids in 567/1171, and a plot to restore them was nipped in the bud in 569/1174 by Saladin, who executed the poet and historian #Umra b. #Al al-0akam of the Yaman. The center of the Musta#lian party

was transferred from Egypt to the Yaman, where it stayed until the 10th/16th century, when it shifted to India, after which a split divided the party into the D"ds of India and the Sulaymns of Southern Arabia [see bohor]. Extensive secular dominion in Arabia eluded the grasp of the Ism#ls until the reign of the Sulaymn Makramids [q.v.] of Na3rn in the 12th/18th century. Arabia in the Later Middle Ages (567-end of 9th Century/1171-end of 15th Century). The advent of the Ayybids meant the triumph of Sunnism over 9#ism in Arabia as well as in Egypt. Saladin, recognized as sovereign in Mecca, sent his brother Trn 9h to depose the third and last Mahdid and occupy the Yaman in 569/1173. During the half century or so of Ayybid rule there members of collateral branches of the dynasty sat on this southern throne. 0a'ramawt was conquered, but did not become an integral part of the Ayybid domains. Closer home the Ayybids had their hands full with the Crusaders from the West, one of the boldest of whom, Renaud de Chtillon, raided Taym", sent his men cruising against the Muslims in the Red Sea, and even thought of attacking Medina. About 598/1200 the 0asanid |atda b. Idrs moved from Yanbu# to Mecca, where he founded the dynasty of all the later 9arfs. Endeavoring to build a strong independent state in al-0i3z, he found the rivalries of the day too great to overcome. |atda died in 617/1220-1, and soon afterwards al-Malik al-Mas#d Ysuf, the last Ayybid in the Yaman, took Mecca and appointed the founder of the Raslids, who claimed descent from the 9assnids, his governor there. | [I:553a] On the other side of the Peninsula the Sal9urid Atbeg of Frs, Ab Bakr b. Sa#d, the patron of the poet Sa#d of 9rz, annexed islands in the Persian Gulf and set foot on the mainland at al-|af and al-0as. The local dynasty of the #Uynids gave way before the Sal9urid pressure and that of the tribe of #$mir of #Uayl, which supplied a new dynasty in the #Ufrids [q.v.]. Succeeding the Ayybids, the Raslids [q.v.] reigned in Ta#izz and Zabd from 625 to 850/1228-1446 as the most illustrious house in mediaeval Yaman. Islamic architecture reached one of its higher points, and scholars received the stimulus of royal approbation, some of the Raslid Sultans themselves being authors of note. Embassies came to the court from China and other distant lands. #Umar b. #Al (regn. 62647/1229-50) ruled from Mecca to 0a'ramawt, and after Hlg executed the last #Abbsid in Baghdad in 656/1258 #Umar's son Ysuf styled himself Caliph of Islam, but full enjoyment of such rank lay beyond the capabilities of the Raslid state. Baybars, the first great Mamlk Sultan of Egypt, assumed nominal overlordship of the Holy Cities, leaving Meccan affairs in charge of the 9arf Ab Numayy I Muammad (regn. 652-701/1254-1301), who strengthened the foundations of |atdan rule. Bedouins of $l Mir and other tribes roamed through the Syrian Desert, exacting large fees from pilgrim caravans and penetrating into Na3d on their raids. In Damascus the religious reformer Ibn Taymiyya (d. 728/1328) laid the theological basis for the Wahhb movement of the 12th/18th century. About the beginning of the 8th/14th century the port of Hormuz on the Persian mainland at the entrance to the Persian Gulf was moved to a nearby island, after which

it grew apace and in time surpassed its rival the island of |ays in attracting to its warehouses the merchandise of the East. Political disturbances in Mecca during the reign of the 9arf #A3ln b. Rumay9a (746-77/1345-75) provoked interference by the Mamlks of Egypt, who took the Raslid Sultan of the Yaman prisoner in a battle at #Arafa in 751/1351. Raslid fortunes were temporarily recouped by Amad b. Ism#l (regn. 803-27/1400-24), who held the Red Sea coast as far north as 0aly, but after his death the state swiftly disintegrated. The later Raslids carried on a lively competition with merchants in Egypt for Indian trade via the Red Sea. In the early years of the 9th/15th century the Ib' community of #Umn returned to its old practice of electing Imams, who succeeded one another in a series lasting over 150 years. About the same time the House of Ka9r under #Al b. #Umar set out on its long course through the tortured politics of 0a'ramawt and ufr, while 0a'ram missionaries carried the gospel of Islam into Somaliland. In the mid-9th/15th century Mni# b. Rab#a al-Murayd, the ancestor of $l Sa#d, migrated from the vicinity of al-|af to Na3d, where he settled in Wd 0anfa. In the latter half of the century A3wad $l Zmil of the 3abrid branch of the #Ufrids ruled as lord of al-|af and Barayn, making his name a byword for generosity in Eastern Arabia. Mecca prospered under the 9arf Muammad b. Barakt and the Mamlk Sultan |"itby, who erected many buildings there, while the hirids [q.v.] in Zabd and Aden supplanted the Raslids in the south. | [I:553b] 3. The Making of Modern Arabia (from the 10th/16th century to the present). In the late 9th/15th century Portuguese explorers made their way from the Mediterranean down the Red Sea, and in 903/1498 Vasco da Gama, after rounding the Cape of Good Hope, was guided to India by an Arab pilot, probably the Na3d Amad b. M3id. Portuguese vessels soon appeared in the Red Sea, and under Afonso de Albuquerque the invaders seized Arabian ports on the Gulf of #Umn and the great mart of Hormuz. Pedro, Afonso's nephew, toured the Persian Gulf in 920/1514, but Afonso died the following year without having achieved his ambitions of reducing Aden and launching an expedition against Mecca. About 912/1506-7 a new line of Zayd Imms was inaugurated by 9araf al-Dn Yay, and from then onwards the Zayds tended to fix their capital, if possible, at -an#". Coffee appears to have been introduced into the Yaman from Abyssinia about this time, and the use of t and tobacco spread among the people. Badr Ab uwayri of $l Ka9r (regn. 922-76/1516-68), whose authority in his palmier days reached from the land of al-#Awli through 0a'ramawt to Sayt, did not hesitate to offer fealty to the Ottoman Sultan. Before Badr died he lost all his territories and suffered long imprisonment at the hands of his 0a'ram enemies. Salm I, the Ottoman conqueror of Egypt in 923/1517, assumed the high title of Servant of the Holy Cities, and the reign of Sulaymn the Magnificent (926-74/152066) fenced other regions within the empire. The Portuguese in alliance with the King of

Hormuz attacked Barayn, where Murin, the uncle and successor of A3wad the 3abrid, lost his life defending the island in 927/1521. Reacting to the aggressive policy of the Portuguese, the Turks bestirred themselves in the Persian Gulf and the Red Sea. Sulaymn at Ba9dd in 941/1534 received the homage of the Arab chiefs of al-|af and Barayn, and later his troops pressed up into the mountains of the Yaman. Aden and Muscat were occupied briefly, and an Ottoman governor was installed in al-0as. For a period of some sixty years after c. 968/1560 there were no Ib' Imams in #Umn, where the secular Nabhnid [q.v.] princes in their mountain fastnesses reached the climax of their power. The slow receding of the Ottoman tide from the highwater mark reached under Sulaymn was observable in Arabia as elsewhere. The diversion of trade from the overland routes to the sea route round Africa contributed to the serious economic depression which beset the Near East during the early modern age. Besides the Austrians and other foes in Europe, the Turks had to face the -afawids, the strongest of whom, 9h #Abbs I, pursued an expansionist policy in the Persian Gulf, where he subjected Barayn in 1011/1602. In the Yaman the Zayd Imams kept alive resistance to the Turks, and al-Mu"ayyad Muammad succeeded in expelling them completely in 1045/1635. The formation of the East India Company in 1009/1600 was the prelude to a burst of activity by English traders in the Red Sea and the Persian Gulf. Allying themselves with the Persians, the newcomers drove the Portuguese out of Hormuz in 1031/1622. Once the Portuguese monopoly had been broken, the English found themselves involved in competition with the Dutch, who secured commercial | [I:554a] preeminence during the second half of the 11th/17th century. After the election of Nir b. Mur9id of the Ya#rubids of al-Azd c. 1034/1624 as Ib' Imm, this Immate remained in his family for more than a century. The Ya#rubids in their early days drove the Portuguese out of Muscat and all other pieds--terre, and in their later days extended their authority overseas to Mombasa, Pemba, and Kilwa in East Africa. 0usayn b. #Al, the third and last Pa9a of the House of Afrsiyb, under whom alBara in the early 11th/17th century had become virtually independent of Ottoman rule, incited $l 0umayd of the tribe of Ban 9lid to overthrow the Ottoman governor of al0as in 1074/1663-4. These Bedouin chiefs kept the oases and grazing grounds of Eastern Arabia subject to their will until the Wahhbs advanced to the Persian Gulf in the early 13th century H. In 0a'ramawt the Zayds of the Yaman encouraged the spread of their version of Islam at the expense of 9fi#ism. About 1070/1660 Amad b. al-0asan, a nephew of the reigning Zayd Imam, led into the main valley of 0a'ramawt a terrifying force known as the Night Flood (sayl al-layl) which undermined the position of the House of Ka9r, but Zaydism failed to secure a permanent triumph over 9fi#ism in this region. In the 12th/18th century a new era began in Arabia with the spread of the reforming movement inspired by Muammad b. #Abd al-Wahhb of Na3d. In a sense this also marked the beginning of the modern history of the whole Near East. Placing the unity of

God above all else and demanding that the popular faith be cleansed of innovations, Ibn #Abd al-Wahhb's call reverberated throughout the Islamic world from West Africa to the East Indies and moved the spirits of the modernists of the Salafiyya in Muslim countries closer than Arabia to the encroaching lands of the West. As an Arab movement opposed to the remote and vitiated rule of the Ottomans, Wahhbism [q.v.] influenced the nationalistic tendencies developing among the Arabs in the 19th and 20th centuries. Within Arabia political unity supplanted petty particularism, and orderly Islamic government functioned as it seldom had before. Soon after first preaching in public in 1153/1741, Ibn #Abd al-Wahhb concluded a basic alliance with Muammad b. Sa#d, ruler of the insignificant town of al-Dir#iyya. When Muammad died, his son #Abd al-#Azz carried on, and by 1202/1788 all Na3d had accepted the doctrines and sway of the reformers, who had withstood three expeditions directed against them by the Ism#l Makramids of Na3rn, then a power in their corner of Arabia. [Cf. also sa#dids.] In 1156/1743 the Ya#rubid line of Imams died out in #Umn while the Persians were trying to establish themselves there. Amad b. Sa#d of $l B Sa#d expelled the invaders from the Bina coast and won election as Imam. After Amad's death the electors chose his son, but he proved such an obscure figure that even the date of his death is unknown. Later rulers of $l B Sa#d [q.v.] made Muscat their capital and gave up the title of imm , calling themselves at first simply sayyid (though they claimed no descent from the Prophet) and afterwards suln . The Persians also held suzerainty over Barayn for about thirty years until the occupation of the islands by $l 9alfa in 1197/1783, since which date no part of Arabia has been subject to Persian dominion. | [I:554b] The rapidly expanding puritan state of Na3d came into conflict with the 9arfs of Mecca in a war lasting fifteen years (1205-1220/1791-1806), with the Sa#ds occupying Mecca for the first time in 1218/1803. Shortly after the death of Ibn #Abd alWahhb (1206/1792) Sa#d authority flowed eastwards to the Persian Gulf, along which it extended to #Umn. In the south the reformers reached the Yaman and 0a'ramawt, while in the north their forces threatened to overrun Syria and Iraq. The Ottoman government, unable itself to dam the flood, turned in desperation to the new Viceroy of Egypt, Muammad #Al. In the 13th/19th century foreign intervention in Arabia, both Muslim and Western, became more effective and extensive than ever before. Muammad #Al annihilated the first Sa#d state when his army captured al-Dir#iyya in 1233/1818. The British, at first welcoming and then fearing the advent of the Egyptians, carried out military actions against Persian Gulf Arabs and in Inner #Umn and occupied Aden in 1254/1839, after which their influence gradually advanced along the southern and eastern coasts and penetrated into the hinterland. Sa#d b. Suln, the most famous ruler of $l B Sa#d (regn. 1221-1273/1806-1856), wielded little or no authority in Inner #Umn, where he was hard pressed by the Sa#ds, to whom he often paid tribute. In the latter part of his reign he devoted most of his attention to his East African possessions, but five years after his death the British established Zanzibar as a Sultanate independent of Muscat. The only Ib' Imam elected during the century, #Azzn b. |ays, failed to win recognition by the British and

was overthrown in 1287/1871 after two years of rule. The Sultans who followed him depended upon British support for the maintenance of their position in Muscat in the face of the hostile Ib' tribes of the interior. During the century internecine warfare was common in 0a'ramawt, where much power rested in the hands of mercenaries imported from the mountains behind Aden, particularly of the tribe Yfi#. In 1283/1867 the |u#ays of this tribe occupied al-9ir and fourteen years later acquired full possession of al-Mukall. Proving resilient in recovering from disastrous blows struck by Muammad #Al's forces, the Sa#d state rebuilt its strength under Turk b. #Abd Allh, who fixed his capital at al-Riy', and later his son Fayal, though al-0i3z was not occupied again. Civil war between Fayal's sons after his death in 1282/1865 caused another decline in Sa#d fortunes, facilitating the reimposition of Ottoman sovereignty over part of Eastern Arabia and the rise of $l Ra9d [q.v.] of 0"il to dominance in Na3d, where al-Riy' itself was made subject. The Ottomans also reestablished themselves in the highlands of the Yaman with headquarters at -an#", but they failed to crush the resistance of the Zayd Imams. The opening of the Suez Canal in 1286/1869, making communications between Istanbul and 3idda easier and faster, helped the Turks to exercise more control in al-0i3z. $l Sa#d, thrice crushed to earth, rose once more under the leadership of Faysal's grandson #Abd al-#Azz, who took al-Riy' from its Ra9d governor in 1319/1902. #Abd al-#Azz fought for twenty years before finally overcoming $l Ra9d in the north. In 1331/1913 he drove the Turks out of al-0as and then lent the British sympathetic support during the First World War. Although the 0i3z Railway from Damascus to Medina had been inaugurated in | [I:555a] 1326/1908, the Turks had to yield Mecca when 9arf al-0usayn b. #Al, encouraged by the British, proclaimed the Arab Revolt in 1334/1916. The end of the war brought the end of Ottoman sovereignty in Arabia, the Zayd Imam al-Mutawakkil Yay b. Muammad becoming fully independent in the Yaman. In 1331/1913 a new Ib' Imam was elected in Inner #Umn in opposition to the Sultan of Muscat. Two years later the British intervened to forestall the capture of Muscat by the Imam's army. Through British mediation a treaty was concluded at al-Sb in 1339/1920 providing that the people of #Umn and the Sultan's government should abstain from interference in each other's internal affairs, but in 1373-4/1954-5 the Sultan's forces, trained and led by British officers, occupied points not held before, hemming the Imamate in on all sides. Although homage was paid to 9arf al-0usayn as King of the Arabs and later as Caliph of Islam, successor of the Ottomans, he was defeated by #Abd al-#Azz $l Su#d when war broke out between the two. Following the conquest of al-0i3z, #Abd al-#Azz annexed the territories of the minor dynasties of $l #$"i' and the Idrsids in #Asr and its Tihma, received the title of King of Saudi Arabia in 1351/1932, and defeated Imam Yay of the Yaman in a brief war in 1353-4/1934, as a result of which Na3rn was recognized as belonging to Saudi Arabia. Killed in an abortive insurrection in 1367/1948, Imam Yay was succeeded by his son Amad. Dying in 1373/1953, #Abd al-#Azz was succeeded by his son Sa#d. Thus

passed from the scene two monarchs who did far more than simply bequeath their names to the realms they wrought and guided for half a century.

NAFD

(A.), a term used to refer to certain Arabian sand seas, and derived from nafd, pl. anfd , with the meaning of a large sand dune. Nafd is used to describe a number of Arabian sand tracts (e.g. Nafd al-Sirr, Nafd al9uwayrt , Nafd alMazhr ), but the term refers particularly to the Nafd alKabr (Great Nafd ) in northern Saudi Arabia. The Nafd alKabr is the lesser of the two great sand seas of the Arabian Peninsula, the second being the so-called Rub# al9l [q.v.] in the south. The Nafd encompasses some 72,000 km2 and varies in its limits between 300 to 400 kms from east to west and, at its greatest limits, is about 250 kms from north to south. These Aeolian sands are a strong reddish colour, affected to some extent by the sandstone formations further west. The sands take on a variety of formations, including linear ridges, star dunes and compound crescentic dunes. The dunes reach as much as 100 m in height, with deep hollows ( fal3 ) which, in certain cases, reach down to the bed-rock beneath the sands. Musil called these hollows a#ara . Rainfall is very slight in the area; to the north an annual rainfall is recorded of 41 mm at Badana and | of 35 mm at Rafh"

. To the south, this rises to 102 mm at 0"il [q.v.]. However, rainfall vanishes in the sands, failing to produce pools. The Nafd is fertile in winter and spring and travellers comment on the good grazing in the dunes, including am' and yerta. According to the 9ammar , the land ceases to be Nafd where the plant raa finds the sand too shallow to put down roots. In the northern Nafd there are wells at al9a , and in the south at the oasis of 3ubba are wells and a small village. The location of the various Nafd wells used to be essential to caravans crossing the desert between 3abal A3" and al3awf . There is evidence of activity in the Palaeolithic and the Neolithic period in the sands of the Nafd and at 3ubba . 3ubba is also especially rich in 9amdic epigraphic material and rock drawings. These show animals that include bovids and human figures, and they have been termed the 3ubba -style in the region as a whole. Ptolemy in the 2nd century A.D. does not mention the sands, but he seems to locate the Sarakenoi in the Nafd , since he places them east of the Thamudenoi, and beyond the Scenitae who lived in the northern 0i3z (Ptolemy, Geography , 138-9; Musil, Northern

Ned , 311-12). Sprenger (Die alte Geographie Arabiens, 171, no. 275) and Euting (Inner-Arabien, i, 151) identify Ptolemy's Aina with the wells at 3ubba . The Nafd is generally known to the Arab geographers at the Raml #$li3 (alHamdn , i, 178, 205, 206, 222; alabar , i, 1-2, 214, ii, 2, 1212; Yt , iii, 417, 591), although other terms are also used. Ibn 0awal (ed. De Goeje, 30; cf. alIa9r , ed. De Goeje, 24 and alIdrs , fasc. ii, Sectio sexta) refer to the sands of alHabr lying near to the mountains 3abal A3" and 3abal Salm , of the ancient Arab tribe of ayyi" . AlHamdn also speaks of the sands of alHabr and alDahn" : alHabr seems synonymous with the Raml #$li3 , i.e. the Nafd , while alDahn" [q.v.] is the narrow sand running from the eastern end of Nafd to the Rub#

al9l in the south. Part of the Nafd is described as alDu33 ( Yt , ii, 554-5, iii, 713, iv, 1024), and is apparently in the western part of the sands, extending from the prominent mountains known as #Alam alSa#d towards Taym" . Another term used is alD , a sandy waste which Yt (iii, 460) places west of 3abal Salm , although 033 9alfa places the wells at alShaa in the northern Nafd within alD (tr. Norberg, ii, 208, 209). The Nafd was within the lands of the Ban Kalb, whose territories extended far to the north into Syria (alBakr , ed. Wstenfeld, i, 664). The Ramlat #$li3 was also in the lands of the Ban Butur of the ayyi" who dominated

3abal A3" and 3abal Salm to the south of the Nafd . According to alSakn ( Yt , iii, 59.), the Ban Butur lived in the Nafd in safety as no-one could reach them in the sands. In modern times, the northern desert has been largely occupied by the Ruwl" , with the area to the south tending to be occupied by 9ammar tribes. A fort at alHayyniyya on the eastern side of the Nafd was built by Ibn Ra9d in 1853. The Nafd had been included within Sa#d territory in the late 18th century, and it passed once more under $l Sa#d authority in 1921-2. Recorded itineraries generally cross from the northern side of the Nafd to the south. Musil argued that no ancient caravan routes passed from east to west. While the terrain of massive sand-dunes is difficult, it is really the scarcity of water that dictates travel in the Nafd . The caravan route via the 3awf villages in the north to

0"il passes alShaa and | 3ubba , the only wells in the central area of the sands (Palgrave, i, 91-9; Blunt, i, 187-92; Euting, i, 141-56; Polk and Mares, 173-228). Another route from the north-west was followed by Nolde, while Gertrude Bell only skirted the south-western edge of the Nafd , using the wells at Hayzn . The pilgrim road from al#Ir , the Darb Zubayda [q.v. in Suppl.], only touches the eastern side of the Nafd , approaching it in the area of al9a#labiyya . Modern asphalt roads generally have avoided the deep sands of the Nafd . However, in the course of the 1980s, a highway was started across the Nafd from 0"il to al3awf . NIGER (The Republic of Niger, La Rpublique du Niger, 3umhriyyat alNay3ar ), a modern state of West Africa, formerly the French colony of that name. The Niger Republic is, to quote Djibo Mallam Hamani (though specifically of the Ayar Massif, which fills the north of it), a carrefour du Soudan et de la Berbrie. Its geographical position on the map, and the multi-ethnic character of its societies, has had a profound effect on the Islamic life of the Nigriens throughout their history.

1. Geography and peoples

The Niger Republic covers an area of some 1,267,000 km2. However, 800,000 of these are within the Sahara, much of which is uninhabited or is uninhabitable. The bulk of the remainder of the country is Sahel. It is only along the banks of the Niger river [q.v.] (where Niamey, the

capital, is located) that there is any intensive cultivation and continuous settlement. Within the Sahara area are found the mountains of Ayar (Ar [q.v.] or Azbin). This massif extends approximately 480 km from north to south and about 240 km from east to west
AL-

RUB#
AL-

KH$L^

(A.) (Empty Quarter), a vast and inhospitable sand-sea occupying much of the south and south-east of the Arabian Peninsula. It lies approximately between 45 E. and 57 E. and 17 N. and 23 N., encompassing some 200,000 sq. miles, consisting of tracts of aeolian sands with immense dunes rising over up to 60 m, areas of gravel and limestone known as 9ua (pl. 9i ), and in the east, towards alLiw" (al3iw" ), sab9a [q.v.] at Umm alSamm and al-Kidan. It lies largely in Saudi Arabia, with its northwest limit roughly marked by the line of the 3abal uway escarpment. In the north and northeast the sands merge with the desert of 3afra near the 3abrn (or Yabrn ) oasis, and with the sands of alDahn" [q.v.] in eastern Arabia. AlDahn" in turn runs into the Nufd [q.v.] sand sea in the north of Arabia. In the southwest and the south in Yemen, the Rub# al-

9l sands approach Mrib [q.v.] and 0a'ramawt [q.v.] respectively and in the east, they reach the Liw" oasis in the United Arab Emirates and to the hinterland of #Umn . The aridity that characterises the desert today differs markedly from the environment of earlier times. In the late Miocene (6-7 million years BP [before present]), the climate was wetter with rivers flowing from the Rub# al9l into what is now the Arabian/Persian Gulf. Late Miocene elephant bones have been found in the UAE, while crocodile and turtle fossils indicate the presence of an ancient river system flowing from the Rub# al9l . Fresh-water lakes existed in the Rub# al9l between 9,700 BP and 6,390 BP, but desiccation followed thereafter, producing the harsh environment that constitutes the Rub# al9l today. In the east, the Wd Idma , Wd Habawna and Wd Na3rn drain run-off from the #Asr highlands into the western Rub# al9l

, where the waters run out in the sands. The broad bed of Sab9a Ma in western Ab Dhab marks the course of a former river system that rose in the interior, with two major periods of flow in the period 80,000-70,000 BP and again 50,000 to 25,000 BP, and a lesser flow in the period 10,000 to 4,500 BP. Some of the Rub# al9l sands may have blown in from the bed of the Arabian Gulf in the period before it was flooded as a result of sea level rises between about 18,000 BP and 8,000 BP, after the last Ice Age. A date of 23,000-17,000 BP has been given to the formation of sand dunes in the Liw" oasis on the eastern side of the Rub# al9l . The term Rub# al9l has attached itself to the entire sand-sea among Arabs and non-Arabs alike, although uncertainty exists over the origins of the name (Sir Percy Cox and D.G. Hogarth in R.E. | Cheesman, The deserts of Jafura and Jabrin, in GJ , lxv [Jan.-June 1925], 139). According to Cox, Bertram Thomas, Wilfred Thesiger and H.R.P. Dickson, the name Rub# al9l was unknown to the indigenous people living around the desert perimeter and the only general term known to most modern travellers to describe the desert was al-Ramal or alRiml (Thomas, 180; Thesiger, 37), although Cox also encountered the term Nufd

as well. Within the desert, specific tracts are identified by particular names, such as al|a#miyyt , 9uwaykla , Hawaya, Ramlat alKu9ayyib , 9aat al9ara and #Urq alAwrik . By contrast, Philby (127-32) argued that the term Rub# al9l was indeed known to the people with whom he travelled, including tribesmen from the sands. The terms alRiml or al-Ramla were used to describe the areas occupied by pastoralists. Areas of briny water were known as 9irn; and areas of better water known as Sanam. Philby regarded the waterless district specifically as the Rub# al9l , a term which he also equates with Rub# al9arab . More recently, D.P. Cole has noted that the $l Murra used the term for the region as a whole, with lesser subdivisions given their own names. The remoteness of the arduous terrain ensured that neither the Classical nor the Arab geographers had much detail to offer on the sands. Ptolemy gives the names of places, wells and mountains, but his knowledge of the interior is very limited. In the Arabic sources, part of the southern desert is termed alAf , although its application varies. Al-

Hamdn , 87; see also 127, 216) uses alAf for a valley between 0a'ramawt and Mahra. AlBakr (76) associates alAf with the region of 9ir [q.v.] in 0a'ramawt . Yt (ii, 78), on the authority of alAma# , describes alAf as a district of Arabia, placing it between Yemen and Sab in the southwest and alYamma , al9ir and #Umn in the southeast. Elsewhere, Yt associates alAf with the pre-Islamic tribe of #$d , iv, 1027, iii, 634), identifying it as a sandy district between #Umn , al9ir and 0a'ramawt . J. Halvy in 1870 refers to the desert east of Na3rn as alAf , and von Wrede in 1843 (3, 22) marks it as the desert district immediately north of the 0a'ramawt (see also Hogarth, 333 ff.). Philby, however, declared that alAf was a literary name for the sands and was not used by the local people he encountered.

Another place mentioned by the sources in the Rub# al9l area is Wabar (or Ubar: see Thomas, 161) which Yt (iv, 896) locates between Yabrn and Yemen. Wabar is said to have been cursed by Allh when its people rejected the prophet Hd , and the settlement was consumed by fire for the sins of its king, #$d b. Kinad. Philby (168 ff.) visited a place pointed out to him as Wabar and known to his guides as al0adda , but it proved to be a meteorite crater rather than a settlement. The iron-rich meteor is now in King Su#d University at alRiy' . Wabar has recently been associated with 9isur in ufr by R. Fiennes and J. Zarins, but this view is not universally accepted. A.R. alAnry has excavated a major archaeological site at |aryat alFw [see ALfw ], located on the southwestern edge of the Rub# al9l at the point where there is a break in the 3abal uway escarpment. It was probably an important town before its decline in the early 4th c. A.D., with paintings and sculpture reflecting its diverse commercial and cultural contacts with Yemen, Egypt and

elsewhere. To date, it is one of the better candidates to be related to Bedouin legends of ancient towns in the | Rub# al9l overwhelmed before Islam by sand or divine punishment. The accounts by Thomas, Philby and Thesiger show that in the heart of the Rub# al9l , no tribes permanently inhabited the sands but rather, they lived a nomadic existence on its edges, occasionally entering the inner areas in pursuit of grazing or the oryx that formerly inhabited the remoter sands. Major tribes around the Rub# al9l include $l Murra in the northeast, Ban Ys , Manir , R9id and #Awmir in the east, R9id , #Awmir , Sa#ar and Bayt Ka9r in the south, and Ym in the west. The history of these camel-rearing tribes was formerly one of raiding and feud, with the seizure of camels as a principal feature of their warfare. The inner areas of the Rub# al9l provided a relatively secure retreat after raids on the desert margins.

The obstacle to travel formed by the Rub# al9l suggests to D.T. Potts that ancient routes avoided the sands, and he suggests that travellers followed easier routes between Yemen and Na3d via alAfl3 or from Na3d through to Buraym and #Umn . Nevertheless, there are persistent indications that routes with seasonal pools exist between the southeastern coast and Yemen which are known to the tribes and are still used today. The Rub# al9l was the last part of the Peninsula to be explored, and ignorance of its vast inner areas endured until well into the 20th century. Thomas (1930-1), Philby (1932) and Thesiger (1946-7, 1947-8) all conducted major journeys of exploration in the sands, and in modern times, oil exploration and development have made it relatively more accessible. Today, a number of international boundaries in the Rub# al9l are disputed and the discovery of oil has encouraged conflicting claims, several of which remain unresolved.

NAD
(A. uplands), conventionally defined as the plateau region of the Arabian peninsula lying to the east of the Red Sea lowlands (alTihma [q.v.]) and the mountain barrier running down through the western side of the peninsula (al0i3z [q.v.]).

1. Geography and habitat.

|

The exact application of this originally topographical conception is very differently understood, and sometimes it means more generally the elevated country above the coastal plain or the extensive country, the upper part of which is formed by the Tihma and the Yaman and the lower by Syria and #Ir , or the part of Arabia which stretches from the frontiers of alYamma to alMadna and thence across the desert from alBara to Barayn on the Persian Gulf (alIa9r , Ibn 0awal ) or the territory between #Ir (al#U9ayb ) and 9t #Ir ( Ibn 9urrad9bih ) or from #Ir to alTihma ( |udma ) or the land which lies behind the so-called Ditch of Chosroes ( Kisr ) as far as the 0arra (alBhil ), or lastly, the territory between the depression of the Wdi 'l-Rumma and the slopes of 9t #Ir (alAma#

). That originally the name was applied to the plateau only is evident not only from the fact that Na3d appears in combination with various place-names; thus alAm# ( Yt , iv, 745) knows of Na3d Bar (in alYamma ), Na3d #Ufr , Na3d Kabkab (near #Araft ), Na3d Mar# (in the Yaman), alBakr (ii, 574) besides the three last named mentions Na3d al-Yaman, Yt (iv, 750-1) further mentions Na3d al0i3z , Na3d Alwa9 in the country of Hu9ayl , Na3d al9ar , and alHamdn (55) Na3d 0imyar and Na3d

Ma9i3 along with a number of places not otherwise known which are combined with Na3d . AlHamdn (177) further makes a distinction between upper Na3d ( Na3d al#Uly ) which is regarded as Na3d proper (alNa3d ) and in which he includes the district ( kra ) of 3ura9 and the town of Yabambam, and lower Na3d ( Na3d alSufl ) which is described as Ar' Na3d and with the 0i3z and al#Ar' forms Central Arabia (1,5-2, 36,18-37), the territory in which pure Arabic is spoken (136,18-137). The original meaning is also seen in the dual Na3dn i , which, it is interesting to note, is used for two mountains in the A3a" range, as well as in the place-name Na3d Mar# and in the spring pasture ground Na3dn 1 in the land of the 9a9#am mentioned by the poet 0umayd b.

9awr ( Yt , iv, 745). That the wide interpretation of the name Na3d above given is not unjustified is shown by the foundation in the second half of the 5th century A.D. by al0ri9 , chief of the Kinda, of a short-lived kingdom which extended from the Syrian limes and Medina to alYamma or from the hill of umya in the N.E. on the Wdi 'l-Rumma to 9t #Ir . At a later date, the whole of alNa3d belonged to the administrative district of alYamma ( Yt , iv, 746). The widest idea to which the name Na3d has ever been applied is probably that of the present kingdom of Saudi ( Su#d ) Arabia, which owes its origin to the Wahhb chief #Abd al#Azz b. #Abd alRamn $l Su#d

, who, as Amr of Na3d conquered alRiy' [q.v.] in 1903, was chosen as ruler of Na3d and the adjoining lands in the summer of 1921, on 10 January 1926 conquered the 0i3z and on 19 January 1927 was proclaimed king of Na3d and its dependencies at alRiy' . The northern frontier of his territories, which was delineated by treaties between the ruler of Na3d with #Ir and Great Britain on the one side (signed at #Uayr on 2 December 1922) and Na3d , Great Britain and Transjordan on the other (signed on 2 November 1925 at 0adda in the 0i3z ) runs along the neutral zone between Na3d and #Ir (29-30N. lat. and 45-46E. long.) and is then continued in a line running N. and N.W. to the intersection of 39E. long. and 32N. lat. and leaves the 3abal #Anze on its north, then S.W. to the Wd | Ra3il and passing through the S.E. the point where 38E. long. and 30N. lat. intersect. The Wd Sirn is thus still in Na3d

. This line continues towards the south from 25' to 38E. long. and crosses the former 0i3z railway towards #Aaba . The extent of the territory is estimated at 900,000 square miles. The capital is alRiy'; the more important towns are Burayda ( Berde ), #Anayza ( #Anze ), 0"il ( 0yil ), 9armala , 9ar , Ma3ma#a , 0uraymala ( 0armle ), alHufhf and al|af . The population, which with the exception of al0as with a considerable number of 9#s , has almost entirely adopted Wahhbism , belongs to the tribes of Muayr ( Mer ), 0arb , #Utayba ( #Atbe ), Subay#

, Dawsir , al#U3mn , al#Awzim , alSuhl , Ban Murra and |an [see further, su#d
,

l -]. The North Arabian Na3d forms a part of the great desert plateau, which is formed of primary rock with overlying sandstone and volcanic outbursts and has two great mountain ranges running through it; that in the north is about 64 km/40 miles long and at its northeastern end some 1,470 m/4,500 feet high, known in ancient times as 3abal ayyi" or 3abal ayyi" , i.e. A3a" and Salm (alHamdn , 125,15) is now called 3abal 9ammar or 3abal I3 ( A3a" ).

Both ranges, which rise out of a table land levelled by weathering, are of granite. The 3abal A3a" stretches from N.N.E. to S.S.W., about 56km/35 miles S.E. of it in approximately the same direction the 3abal Salm , in front of which in the S.W. lies the 3abal Ramn , while S.E. of the 3abal Salm lies the 0arra of Fayd, of volcanic origin. S.E. of this rises the sandstone plateau, overlaid with limestone, of 3abal uway ( u ) running N.W. to S.E., which forms the western declivity of a plateau which has come into existence through weathering and slopes towards the Persian Gulf on the one side and the sands of the desert of Rub# al9l on then other. It begins S.E. of the district of al|am (S.E. of the 0arra ) and stretches E. from alWa9m to al#$ri' with the town of alRiy' and then turns, west of the 9ar3 oasis, S.S.W. towards the

Wd 'lDawsir . The most important peak on this edge of the plateau is the 3abal #Ammriyya or 3idd , which towers some 152m/500 feet above the ridge usually 610-1070 m/2,000-3,500 feet high. The long southern part of the uway is intersected by numerous wd s which lead the water that falls in the rainy season to the Rub# al9l . Its most important part is the Afl3 , 64 km/40 miles long, with the oasis of Layla. Na3d is in the main steppe and desert. The Nafd and Dahn" [q.vv.] occupy the greater part of northern Na3d , while the Rub# al9l joins them on the S.E. There are no perennial streams in Na3d , so that the country has to rely upon subterranean channels of supply which are at various depths and have to be reached by wells. In the oasis of al9ar3 the wells are from 6m to 12m/ 20 to 40 feet deep, in Afl3 15 to 18m/50-60 feet, in 0"il and alRiy' about 24m/80 feet. Sometimes these springs form ponds, for example, in al9ar3 , the springs of which form three pools, the largest of which is 150 paces long and 80 broad (cf. the picture in Philby, ii, 34) while the springs of

Afl3 feed a lake nearly a mile long and a quarter of a mile broad (Philby, ii, pl. at p. 86). These supplies sometimes dry up suddenly, probably because they have found a subterranean exit, as has happened in the case of two waterholes in Afl3 and the two larger ponds in al9ar3 . The hydrographic conditions of the country are therefore exceedingly dependent on the rainfall from the summer and winter | rains. The former ( wasm or maar alayf ) fall in August and September and particularly refresh the pastures which the summer sun has dried up, while the latter produce a springlike effect in the land on which they fall. The classic phrase sa ' llh
u

Na3d an min rab# in waayf in (alBakr , ii, 627) eloquently sums up this state of affairs. Heavy rainfalls were also observed in April 1871 in the central Wdi 'l-Rumma and in May in #Anze between 3abal Salm and #Anze (1884, Ch. Huber), and Philby (ii, 10) noticed thunder showers in May as well as drizzle, while Doughty met with hail at 9abra (near #Anze ) in April. That the climate here cannot have changed very much is evident from Ibn 3ubayr , who records very heavy showers in this district in

-afar 584/April 1184. Huber met with rain in June 1884 between #Anze and Mecca, Sadlier at the end of July 1819 between al0as and 4ar#iyya heavy thunderstorms and rain, which however was described by the natives as unprecedented. Philby (i, 141, 147) records thunder and rain in December. The rainwater collects in the hollows below the thick layer of sand and enables palms to grow and also, on chemically decomposed fertile soil, wheat and barley, vegetables and fruit trees. The hot summer of course everywhere makes it necessary to water the crops from wells. On the other hand, the frequently very sudden flooding of the water-courses led in quite early times to the building of dams to hold back and store the water; such were built in the Wd 'l-Rumma at #Aneize (alBakr , i, 207; Yt , iii, 738), 4ar#iyya (alBakr , ii, 637) and on the road from alYamma to #Aneize (alHama9n , 174, 192). Doughty found remains of such dams in the 3abal A3a" . The district of al9araf is the richest part of alNa3d , and the valleys of the Wd 'l3arr and Wd

'lMiyh are celebrated for their pastures. Here the early caliphs had vast grazing grounds ( im ) e.g. in 4ar#iyya [see ALdir#iyya ], alRaba9a [q.v.], Fayd [q.v. in Suppl.], alNr , 9u 'l9ar and Na# . The most famous was that of 4ar#iyya , where the caliph #Umar I secured an area six Arab miles in diameter as pasture for 300 horses and 30,000 camels for the army. #U9mn extended this area until the diameter was ten miles. The #Abbsid alMahd abandoned it, as the policy of this dynasty was to neglect Arabia deliberately, in contrast to the Umayyads who, for example, intensively colonised western Na3d . In the 6th century A.D. Na3d was still well wooded, and al9aruba , south of the Wd 'l-Rumma, and Wa3ra were particularly celebrated in this respect, while at the present day they only possess scanty remnants of these forests. Many areas seem to have been ruined by drought or disastrous inundations (Philby, i, 115; ii, 9); the decline of alYamma is probably due to the latter cause. Crops are sometimes damaged by sharp frostsin winter (January) the temperature sometimes sinks from a maximum of 53 F. by day to below 23 and ice and snow

have been occasionally seen at the higher levelswhile the summer drought with a maximum temperature of 113 destroys the crops. The two most important wd s are the Wd 'l-Rumma about 650 miles long, which runs right across the plateau of North Arabia, rising in the 0arra of 9aybar and entering the Euphrates plain at Bara , and the Wd 'lDawsir . These have formed since ancient times the two main routes of traffic in Central Arabia. With the aim of sedentarising the Bedouin, King #Abd al#Azz Ibn Su#d established in Na3d , in the first quarter of the present century, settlements called i3ra , pl. i3ar ; on these, see ALi3ar and J.S. | Habib, Ibn Sa#ud 's warriors of Islam. The Ikhwan of Najd and their role in the creation of the Sa#udi kingdom, 1910-1930, Leiden 1978. (A. Grohmann*)

2. History.
See
,
AL-

#arab

3azrat , vii.

3. The period since the development of the oil industry.

The Kingdom of Saudi Arabia was proclaimed on 22 September 1932, and the east of the country was allotted for oil exploration to Standard Oil of California (later operating within Aramco) in 1933, thus setting the political and economic stages for the development of the country as one of the world's most important oil economies during the following decades. Saudi Arabia's greatest oilfields are found mainly in the east of the country, but Na3d , in the contemporary period named Central Province, including alRiy' , al|am and 0"il [q.vv.] districts loosely extending over the region traditionally called Na3d , | has benefited from the rapid expansion of the oil industry which began in earnest during the 1950s. Crude oil, natural gas and oil products have been made readily available in Na3d as part of a deliberate national policy. An oil pipeline was built from alDammm [q.v.] to alRiy' refinery, while during the 1970s Petroline carried crude from the east to the west coast via Na3d , making oil available for industrial and related developments on a large scale. More important than the direct effects of the oil industry was the growing amount of government expenditures within a series of economic development plans permitted by augmenting oil revenues, especially in the years after 1973. Na3d was central to this development process, containing the capital, alRiy' , and receiving earliest and most favoured treatment in the allocation of financial resources. Agriculture retains importance in the oases and smaller towns of the

Na3d . Good quality dates are grown in the oases. Other crops include wheat and barley, while the livestock holding in the region is large and increasing. A modern dairy industry containing a third of all the kingdom's cattle, is based on al9ar3 [q.v.]. AlRiy' has one of the most rapidly growing populations of the country, totalling 1.3 million in 1987. Many of the small villages around alRiy' have been absorbed into the metropolitan area. The greatest growth of the city is taking place through immigrants concentrating in the areas around the railway station and the adjacent industrial area. Recent development has been notable along the airport road and in the vicinity of the royal place at alNiriyya . Principal employers in alRiy' are the expanding government ministries and official agencies. Marked expansion is characteristic of all other towns of Na3d area. In all, the population of the Central Province was put at 3.6 million in 1985. Well over half of the Na3d 's population is urban resident, and approximately a quarter of the urban dwellers are foreign immigrants. Na3d is favoured by recent transportation development. A railway was opened between alRiy' and alDammm in 1951, while alRiy' is the central node of the trans-Arabian highway completed in 1957 and of the remainder of the national road network.

#ADN$N
, ancestor of the Northern Arabs according to the genealogical system which received its final form in the work of Ibn alKalb , about 800 A.D. The name occurs twice in Nabatean inscriptions from N.W. Arabia ( #Abd

#Adnn , #Adnon; Jaussen et Savignac, Mission Archologique en Arabie, Paris 1909-14, nos. 38, 328) also in Thamudic (Lankester Harding/Littman, Some Thamudic Inscriptions , Leiden 1952) and was taken to South Arabia along the incense-route (Corpus Inscriptionum Semit., iv, no. 808). As already noted by al3uma , abat (Hell), 5 (cf. also Ibn #Abd al-Barr, alInbh #al |ab"il alRuwh , Cairo 1350, 48), it does not occur in pre-Islamic poetry at all ( Labd , x1i, 7 is spurious), and only very rarely in early Islamic literature. This means that the name does not owe its place in the system to the conflict of parties in the Umayyad period, like Nizr and Rab#a , but is of pre-Islamic origin, although it does not spring from bedouin tradition. It may come, like other rudimentary elements of the system, from the Meccan tradition.It is noteworthy that, owing to the revival of national feeling, the name #Adnn

#UK$
, the most famous and important of all the annual fairs ( s [q.v.], pl. asw ) of the Arabs in pre-Islamic times. It was situated to the southeast of Mecca between Na9la and al"if in the territory of the tribal group Hawzin [q.v.]. It shared with two other fairs, Ma3anna

and 9u "l Ma3z , proximity to Mecca and its being held during the Sacred Months. But it was the most important of the three, and was held in the month of 9u "l |a#da , just before the start of the pilgrimage to #Araft and Mecca. The s was strategically located in the middle of the Spice Route of Western Arabia and it especially flourished in the 6th century, owing to the ByzantinePersian wars, which diverted trade from the Mesopotamian to the West-Arabian route on the eve of the rise of Islam and redounded to the great advantage of Mecca and |uray9 . The tribe of Tamm controlled some of #Uk s important functions. Although it was principally a fair for buying, selling and the exchange of various commodities, #Uk was of great significance in other non-material aspects of Arab life in pre-Islamic times, not unlike the national festival of Classical Greece, the panguriw. And it was, like other Arabian asw , a unifying force among the Arab tribes, where the Arabic literary koin received some development, where contests, literary and other, were held, and where covenants and contracts were struck, hallowed by proximity to #Araft and the sanctity of the Sacred Months. The later Islamic sources speak of arbitration conducted by the poet al-

Nbi9a al9ubyn [q.v.]. The s with its environs was the scene of some important historical events, such as the Fi3r war involving |uray9 and Hawzin . It also witnessed some historic visits, the most important of which was that of the Prophet Muammad and the Christian |uss b. S#ida [q.v.], said to have been the bishop of Na3rn , both of whom preached their respective faiths at the s . It was only natural that #Uk should have declined after the rise of Islam, which brought about a fundamental change in the relative importance of the trade routes and the role of Arabian tribes, now the sinews of the Arab conquests. Its actual demise took place in 127/745, when the 0arriyya or 9ri3ites sacked it and thus consigned it to oblivion. MAWLIDIYYA (A) (or mldiyya ; dial. mldiyya ), pl.t , a poem composed in honour of the Prophet on the occasion of

the anniversary of his birth [see MAWLID] and recited as a rule before the sovereign | and court after ceremonies marking the laylat al-mawlid. A relatively large number of mawlidiyyt are extant, drawing their inspiration from the famous Bnat Su#d of Ka#b b. Zuhayr [q.v.] so often imitated by versifiers, of whom the best known is certainly alBr (608-94/l2l2-97) [q.v. in Suppl.], whose poems enjoy a renown which has never diminished, especially the Burda [q.v.] and, to a lesser extent, the Hamziyya, which is recited in mosques and zwiya s during the month of Rab# I, between the ma9rib and #i9" prayers. Among the mediaeval authors who have left poems classifiable within the category of mawlidiyyt may be cited alBar# (5th/11th century), al-arar (d. 556/1160), Ibn al3awz (510-97/1116-1200 [q.v.]), Ibn 0a3ar alHaytam (909-74/1504-67 [q.v.]) and alBarzan3 (1040-1103/1530-91). Furthermore, it is possible to gain an overall idea of this production thanks to the four-volume collection made at the beginning of the century by Ysuf b. Ism#l alNabhn and published in Beirut in 1320/1902. In the Islamic West,

mawlidiyyt were mainly the work of court poets, but also of administrative officials and viziers for whom the composition of poems of this type constituted a part of their professional education; some well-known personalities figure among them, such as Ibn Marz (710-81/1310-79 [q.v.]), Ibn al9ab (713-76/1313-75 [q.v.]), and above all, Ibn Zamrak (733-95/1333-93 [q.v.]). Due to the occasional nature of this poetry, it is understandable that a large number of poems have not been preserved; the majority of those that survive, thanks, in particular, to alMaar (d. 1041/1632 [q.v.]) and to alIfrn (d. 1157/1745 [q.v.]), belong to a relatively short period from 761 to 768/1360-7, corresponding to the reigns of the Marnid Ab Slim Ibrhm (d. 762/1361) in Fs and of the Narid Muammad V (d. 793/1391) in Granada; to be sure, alFi9tl (956-1031/1549-1633), himself the author of at least one mawlidiyya , reproduced in his Nuzha (ed. and Fr. tr. Houdas, 149-57), these poems being composed in the reign of Amad alManr in 999/1590. Generally, the framework of the ada is respected, but adapted to suit the fundamental purpose of the poet in the sense that, while the apology of the Prophet is preceded by a nasb and a ral

, it is followed in the West by a eulogy of the sovereign which is explicable by the circumstances in which these poems were recited. The nasb contains the traditional recollection of the remains of an encampment, but the author must avoid any allusion to a woman and show the decency appropriate to the situation. He expresses on the contrary the violent passion which he feels for the Prophet, leaving some doubt as to this love, whose true mystical nature is not at all clear. The abandoned encampment is situated on the route that the poet must follow to visit the Holy Places, but, as he is very far distant from them, he calls upon a caravan guide or some pilgrims in order to ask them to bear his greetings to the Prophet and describe to him the ardour of his passion. This sentimental and moving prologue is followed by a brief lyrical expansion on the theme, or, more frequently, a narrative full of details borrowed from the traditional ral , of an imaginary journey across deserts as far as Medina. It goes without saying that this general theme undergoes numerous variants ranging from an account of the pilgrimage to Mecca to the insertion of paranetic verses or commonplaces on the flight of time, white hair, etc. The Spanish mawlidiyyt are always distinguished by a large number of descriptions. | The recollection of the Holy Places introduces the eulogy of the Prophet, which must theoretically be based on reality and never drift into hyperbole. The principal themes concern the birth, foretold by earlier prophets, the signs of prophecy visible from infancy, his mission, etc.; next, the epithets of Muammad are enumerated; then come his physical and moral portraits; and finally, the description of the miracles that he performed. In this central part of the mawlidiyya , the elements of the panegyric, expressed by means of a profusion of superlatives, are drawn from the |ur#n and ad9 as well as popular beliefs which have embellished the life of the Prophet with legendary details. It may be remarked further that the poets, idealising his image, adopt some characteristics taken from the Gospels so as to invest the founder of Islam with an aura of sanctity which makes him vie with Jesus.

After the account of the miracles, the versifiers generally express a wish to be able to visit the Holy Places, offer supplications to their saviour and invoke God's blessing on him and his Companions. This invocation, followed by a similar invocation on behalf of the sovereign and mention of the laylat al-mawlid, marks the transition to the third part of the mawlidiyya which is often as developed as the second and consists of the panegyric of the reigning prince. This part offers nothing really new corresponding to the classical mad [q.v.]. The author attributes all the virtues to the mamd , who is the restorer of the kingdom and whose arms are always victorious; but his cardinal virtue is naturally generosity, which is appealed to more or less discreetly. After the sovereign come the turns of the heir presumptive and the royal family. To conclude, the poet wishes that the prince's prosperity may endure. One can hardly expect to find much originality in these compositions crammed with rhetorical flourishes and adorned with clichs which savour of affectation and artificiality. However, the choice of images, the variety of stylistic devices, the subtle play on vocabulary and the constant appeal to the religious or literary culture of the listener, retain a certain attraction. As well as some poems in classical Arabic, there are many mldiyyt in dialect which generally contain only the eulogy of the Prophet; among those which have been preservedor those which are still composed todaysome follow the classical tradition and contain moreover the nasb and the ral , but the eulogy of the sovereign does not figure at all in them [see maln ].

#A-ABIYYA
, Arabic word meaning originally spirit of kinship (the #aaba are male relations in the male line) in the family or tribe. Already used in the ad9 in which the Prophet condemns #aabiyya as contrary to the spirit of Islam, the term became famous as a result of the use to which it was put by Ibn

9aldn , who made this concept the basis of his interpretation of history and his doctrine of the state. #Aabiyya is, for Ibn 9aldn , the fundamental bond of human society and the basic motive force of history; as such, the term has been translated as esprit de corps (de Slane), by Gemeinsinn and even by Nationalittsidee (Kremer), which is an unjustified modernism. The first basis of the concept is undoubtedly of a natural character, in the sense that #aabiyya in its most normal form is derived from tribal consanguinity ( nasab , iltim ), but the inconvenience of this racial conception was already overcome in Arab antiquity itself by the institution of affiliation ( wal" ), to which Ibn 9aldn accords great importance in the formation of an effective #aabiyya . Whether it is based on blood ties or on some other social grouping, it is for Ibn 9aldn the force which impels groups of human beings to assert themselves, to struggle for primacy, to establish hegemonies, dynasties and empires; the validity of this principle is tested firstly in Arab history, pre-Islamic and Muslim, and secondly in the history of the Berbers and other islamicised peoples: the Arab empire is the product of the #aabiyya of |uray9 , especially of the Ban #Abd Manf group, but once power ( mulk ) has been seized, the dominant group tends to detach itself from the natural #asabiyya on which it is based, and to substitute for it other forces which become the instrument of its absolutism. This extraordinary appreciation of a non-religious force as the motive power of history (the religious element only superimposes itself as a secondary element) involved Ibn 9aldn in delicate problems of reconciliation with the traditional view of Muslim history and civilisation, a view, moreover, which he supported

with whole-hearted conviction; this effort of harmonisation, apparent in more than one page of the Muaddima

MI#R$
(A.), originally designates a ladder, and then an ascent, and in particular, the Prophet's ascension to Heaven.

1. In Islamic exegesis and in the popular and mystical tradition of the Arab world.
The |ur"n (LXXXI, 19-25, LIII, 1-21) describes a vision in which a divine messenger appears to Muammad , and LIII, 12-18, treats of second mission of a similar kind. In both cases, the Prophet sees a heavenly figure approach him from the distance, but there is no suggestion that he himself was carried away to Heaven. However, it is otherwise with the experience alluded to in XVII, 1, Glory be to Him who transported His servant by night ( asr bi#abdihi layl an) from the Mas3id al0arm to the Mas3id alA which We have surrounded with blessing, in order to show him one of our signs. For this verse, tradition gives three interpretations: (1) The oldest one, which disappears from the more recent commentaries, detects an allusion to Mu ammad's Ascension to Heaven. This is the more interesting, as these traditions (alBu9r , Cairo 1278, ii, 185, Bb kna

'l-nabiyyu tanmu #aynuhu wal yanmu albuhu , no. 2; Muslim, Bl , 1290, i, 59; alabar , Tafsr 1 , xv, 3, cf. B. Schrieke, Die Himmelreise Muhammad's, in Isl. , vi [1915-16], 12, 14) retain also the original signification of the story of Ascension (A.A. Bevan, Mohammed's Ascension to Heaven, in Beihefte zur Zeitschr. fr die Alttestam. Wissensch., xxvii = Studien ... Julius Wellhausen ... gewidmet, Giessen 1914, 56; Schrieke, op. cit.). This explanation interprets the expression almas3id ala , the further place of worship in the sense of Heaven and, in fact, in the older tradition isr" is often used as synonymous with mi#r3 (see Isl. , vi, 14). One would thus have, in this verse, witness to the nocturnal ascension of the Prophet to the heavenly spheres (Schrieke, op. cit., 13 ff.; J. Horovitz, Muhammeds Himmelfahrt, in Isl. , ix [1919], 161 ff.), but a witness limited merely to an allusion to the adventure, without saying anything about the manner in which it developed. (2) The second explanation, the only one given in all the more modern commentaries, interprets almas3id ala as Jerusalem and this for no very apparent reason. It seems to have been an Umayyad device intended to further the glorification of Jerusalem as against that of the holy territory (cf. Goldziher, Muh. Stud., ii, 55-6; Isl. , vi, 13 ff.), then ruled by #Abd Allh b. al-Zubayr. Alabar seems to | reject it. He does not mention it in his History and seems rather to adopt the first explanation (see Isl. , vi, 2, 5, 6, 12, 14; alabar

, Annales, i, 1157 ff., a passage which appears to represent the historian's final verdict formed on full consideration of the evidence before him, cf. Bevan, op. cit., 57). Explanations 1 and 2 concur in interpreting #abd in XVII, 1, by Muammad , and this seems to be right ( Isl. , vi, 13, n. 6). The i3m# admitted both interpretations and, when the Umayyad version had arisen, harmonised the two by assigning to isr" the special sense of night journey to Jerusalem . The Ascension, having lost its original meaning, was altered in date, being made to fall at a later period, as appears, in fact, to have been done previously by Ibn Is in the oldest extant biography of Muammad (Bevan, op. cit., 54). The story of the night journey to Jerusalem runs as follows: One night, as Muammad was sleeping in the neighbourhood of the Ka#ba at Mecca (or in the house of Umm Hni" , Isl. , vi, 11) he was awakened by the angel Gabriel who conducted him to a winged animal called Bur [q.v.], and with Muammad mounted on this animal they journeyed together to Jerusalem. On the way thither they encounter several good and several wicked powers ( Mi9kt almab , Dihl 1268, 521-2; alBa9aw

, Mab al-sunna , Cairo 1294, ii, 179, with a harmonising interpolation) and visit Hebron and Bethlehem (alNas" , Sunan , Cairo 1312, i, 77-8; al-Nuwayri, ms. Warner 2a, p. 93, 11. 710). At Jerusalem, they meet Abraham, Moses and Jesus, of whom a description is given (e.g. alBu9r , Cairo 1278, ii, 147). The alt is performed, Muammad acting as imm and thereby taking precedence of all the other prophets there assembled. This meeting with the prophets at Jerusalem resembles and may well have been modelled on the transfiguration of Jesus on Mount Tabor (Matt. xviii, 1; Mark ix, 1; Luke ix, 28), cf. Isl. , vi, 15, and Goldziher, in RHR , xxxi, 308. (3) The third interpretation of XVII, 1, is based on XVII, 62, where ru"y vision is explained as isr" . This implies that the night journey was not a real journey but a vision. Standing at the i3r , Muammad saw Jerusalem and described it to the unbelieving |uray9ites (alBu9r , ii, 221, iii, 102; Muslim, i, 62; alabar , Tafsr , xv, 5, l. 14 ff., etc.). The story is woven into a connected whole as follows: Muammad journeys by night to Jerusalem, returns and at Mecca describes his adventures; the |uray9 disbelieve him and Muslims apostasise; Muammad seeks to defend the truth of his story, but he has forgotten the particulars; whereupon God causes him actually to behold Jerusalem (see Isl. , vi, 15-16).

In the more modern and longer narratives, the story is further amplified (see e.g. A. Mller, Der Islam in Morgen- und Abendland, i, 86). The Prophet is said to have held 70,000 conversations with God, although the whole journey proceeded so quickly that, when he returned, his bed was still warm and the water cup which he had overthrown with his foot at his hurried departure, was not yet empty. By Muslim theologians the question has been discussed, whether the isr" happened while Muammad was asleep or awake and whether it was his spirit or his body which journeyed. The orthodox opinion is that the journey was performed by Muammad with his body and awake. Alabar in his commentary (xv, 13) very decidedly supports this meaning for the following reasons: (1) If the Prophet had not been carried away in a corporeal sense, the event would afford no proof of his divine mission and those who disbelieved the | story could not be accused of infidelity. (2) It is stated in the |ur"n that God caused His servant to journey, not that He caused His servant's spirit to journey. (3) If the Prophet had been carried away in spirit only, the services of Bur would not have been required, since animals are used for carrying bodies not for carrying spirits (Bevan, op. cit., 60; Schrieke, op. cit., 13; alabar , alBay'w , and alBa9aw , Tafsr , ad XVII, 1). Mystics and philosophers often favour an allegorical interpretation (Goldziher, Geschichte der Philosphie im Mittelalter, in Kultur der Gegenwart, i/5, 319). The question of the possibility of an ascent to Heaven is several times touched on in the |ur"n . In XL, 38, Fir#awn gives Hmn orders to build a palace so that he can reach the cords of Heaven and climb up to the god of

Ms (cf. also XXVIII, 3). In LIII, 38, the calumniators are asked whether they had perchance a ladder ( sullam ) so that they could hear the heavenly voice, and in VI, 35, the consequences are considered which the signs brought by the Prophet with the help of a ladder to Heaven might have on his hearers. The old poets also talk of ascending to Heaven by a ladder, as a means of escaping something one wants to avoid (Zuhayr, Mu#allaa , 54; alA#9 , no. XV, 32).

0ad9 gives further details of the Prophet's ascension. Here the ascension is usually associated with the nocturnal journey to Jerusalem, so that the ascent to Heaven takes place from this sanctuary. We also have accounts preserved which make the ascension start from Mecca and make no mention of the journey to Jerusalem. In one of these, the ascension takes place immediately after the purification of the heart (see alBu9r , -alt , bb 1; 0a33 , bb 76; Manib , bb 42; Amad b. 0anbal , Musnad , iv, 207, v, 143; alabar; Annales, i, 1157-8). In the last-mentioned passage we read: When the Prophet had received his revelation and was sleeping at the Ka#ba , as the |uray9 used to do, the angels Gabriel and Michael came to him and said: With regard to whom have we received the order? Whereupon they themselves answered: With regard to their lord. Thereupon they went

away, but came back the next night, three of them. When they found him sleeping, they laid him on his back, opened his body, brought water from the Zamzam well and washed away all that they found within his body of doubt, idolatry, paganism and error. They then brought a golden vessel which was filled with wisdom and belief. Thereupon he was taken up to the lowest heaven. The other versions of the same story show many additions and variants; according to one, for example, Gabriel came to Muammad through the roof of his house which opened to receive him; according to another, it was Gabriel alone who appeared to him and there are many similar variants. All these versions, however, put Muammad 's ascension at an early period and make it a kind of dedication of him as a Prophet, for which the purification of the heart had paved they way. Ethnographical parallels (Schrieke, op. cit., 2-4) show other instances of a purification being preliminary to an ascension. Similar stories are found in pagan Arabia (Horowitz, in Isl. , ix, 171 ff.) and also in Christian legends (op. cit., 170 ff.). Another story ( Ibn Sa#d , i/1, 143) says that the ascension took place from Mecca although it does not associate it with the purification of the heart which it puts back to the childhood of the Prophet [see alma ]. How did it come about, however, that this, obviously the earlier, tradition of Mecca as the starting point of the ascension, was ousted by the other | which made it take place from Jerusalem? The localisation of the |ur"nic alMas3id alA in Jerusalem is by some connected with the efforts of #Abd al-Malik to raise Jerusalem to a place of special esteem in the eyes of believers (Schrieke, op. cit., 13; Horovitz, op. cit., 165 ff.; idem, The earliest biographies of the Prophet and their authors, in IC , ii [1928], 35 ff.), and in any case it cannot be proved that this identification is older than the time of #Abd al-Malik. It might all the easier obtain currency as Jerusalem to the Christians was the starting point of Christ's ascension, and from the 4th century Jesus's footprint had been shown to pilgrims in the Basilica of the Ascension; and now, perhaps as early as the time of #Abd

al-Malik, that of their Prophet was shown to Muslim pilgrims (Horovitz, Muhammeds Himmelfahrt, 167-8). The idea of the heavenly Jerusalem may have had some influence on the development of the isr" legends; when Muammad meets Ibrhm , Ms and #^s in Jerusalem, the presence of these prophets in the earthly Jerusalem is not at once intelligible, but it loses any remarkable features if Bayt alMadis (Ibn Hi9m , 267) from the first meant the Heavenly Jerusalem (Horovitz, op. cit., 168, another explanation, see above). Perhaps also the phrase alla9 brakn awlahu was taken to support the reference to Jerusalem; when these words occur elsewhere in the |ur"n they refer to sites in the holy land (H. Lammens, Les sanctuaires prislamites dans l'Arabie occidentale, in MFOB , xi [1926], 72). While the stories quoted above only say that Gabriel took the Prophet up to the heights of Heaven, but are silent as to how, others add that a ladder ( mi#r3 ) was used for the ascent (see Ibn Hi9m , 268; alabar , Tafsr , xv, 10; Ibn Sa#d , i/1, 143); this ladder was of splendid appearance; it is the one to which the dying turn their eyes and with the help of which the souls of men ascend to Heaven. The ladder is probably identical with Jacob's ladder in Genesis, xxviii, 12; the Ethiopic Book of Jubilees, xxvii, 21, calls this ma#reg , and sra LXX, 3,4, calls God 9u

'lMa#ri3 to whom the angels and the spirit ascend ( ta#ru3 ). According to XXXII, 4, the amr rises to God; according to LVII, 4, and XXXIV, 2, God knows what descends from Heaven and what ascends to it, and in XLIII, 32, there is a reference to steps ( ma#ri3 ) in the houses of men. The term was already known, and is presumably taken from Ethiopic (Horovitz, op. cit., 174 ff.). Among the Mandaeans, also, the ladder (sumbilta) is the means of ascending to Heaven (Ginza, tr. M. Lidzbarski, 49, 208, 490), and there are parallels to the ladder of the dead in the mysteries of Mithras (see Tor Andrae, Die Person Mohammeds, 45; Wetter, Phos, 114, n. 2); the Manichaean #amd alsab ( Fihrist , 335, 10), by means of which the dead man is taken to the sphere of the moon is a more distant parallel (Bevan, op. cit., 59). Just as the mi#r3 is associated with the ascension, so alBur is originally connected with the night journey to Jerusalem; it found its way, however, at an early date into the legend of the ascension (see alBu9r , Manib , bb 42; Amad b. Hanbal, Musnad , iv, 207; v, 387; alabar , Tafsr , xv, 12). At the gate of each of the seven heavens through which he wanders with the Prophet, Gabriel is asked for his own name and that of his companion (alBu9r , -alt , bb

1; alabar , Tafsr , xv, 4; idem, Annales, i, 1157). After he gives these, he is next asked if Muammad has already been sent as a prophet (a-waad bu#i9a ilayhi, correction for the original a-waad bu#i9a found in alabar , Annales, i, 1158; see Snouck Hurgronje, in Isl. , vi, 5, n. 4); this also indicates that the ascension originally belonged to the | period immediately after his call (Schrieke, op. cit., 6). In each heaven they meet one of the earlier messengers of God, usually Adam in the first, Yay and #^s in the second, Ysuf in the third, Idrs in the fourth, Hrn in the fifth, Ms in the sixth and Ibrhm in the seventh heaven; there are also variations and Adam appears as judge over the spirits of the dead (Andrae, 44-5; Schrieke, 17; Amad b. 0anbal , Musnad , v, 143; cf. Apoc. Mosis, 37). Of the other messengers of God we are only toldin addition to being given a description of their personal appearancethat they greeted Muammad; Ms is an exception, who expressly says that Muammad is higher in the esteem of God than himself and that the number of his followers surpasses his own (alabar , Tafsr

, xv, 11). On another occasion, Muammad engages in a conversation with Ms after God had imposed upon him 50 alt s a day as obligatory prayers for the faithful. On Ms 's advice, Muammad asks several times for an alleviation, and each time God grants it; but when Ms says 5 alt s are still too many, the Prophet refuses to ask for less (on Genesis, xviii, 23 ff., as the prototype of this episode; cf. Goldziher, Muh. Studien, i, 36; Schrieke, 19; Andrae, 82). According to some versions, Ms dwells in the seventh heaven and the conversation seems to be more natural there. To the ascension belong the visits to Paradise and to Hell. Paradise, according to others in the first; in some it is not mentioned at all. The statements about its rivers are contradictory (Schrieke, 19; cf. kaw9ar ), the sidrat almuntah is usually placed in the seventh heaven (Bevan, 59; Schrieke, 18). In one description, Hell is put below the first heaven (Ibn Hi9m , 269; alabar , Tafsr , iv, 10). According to another, the place of punishment of the damned is on the way between Heaven and earth, and Muammad sees it on his journey to the Bayt alMadis (alabar , xv, 101, also Amad b. 0anbal , Musnad , i, 257; ii, 353; iii, 120, 182, 224, 231, 239). On the punishment in Hell, cf. Schrieke, 17; Andrae, 44; Horovitz, 173; Reitzenstein, Das mandische Buch der Grsse, 81 ff.; Lidzbarski, Johannisbuch, 98 ff.; Ginza, 183.

That Muammad appeared before God's throne in the seventh heaven and that the conversation about the obligatory prayers took place there, is already recorded in the oldest stories (see above), but only rarely do they extend the conversation between God and the Prophet to other subjects (alabar , Tafsr , xxvii, 26; Musnad , iv, 66, as a dream; Andrae, 70). But objection was raised to the assertion that Muammad on this occasion saw God face-to-face (Andrae, 71 ff.), and the question was also raised at an early date whether the ascension was a dream or a reality, whether only the soul of the Prophet was carried up or also his body (L. Caetani, Annali, Intr. 320; Andrae, 72; Bevan, 60; Schrieke, 13, n. 1). The ad9 contains, besides these, other details which Asn (Escatologia, Madrid 1919, 7-52; idem, Dante y el Islam , Madrid 1927, 25-71) discussed. In developing the story of the Prophet's ascension, Muslim writers have used models afforded them by the Jewish and Christian Apocalypses. A few features may also come from the Zoroastrians from the Arda Viraf; cf. the works already mentioned by Andrae, Bevan, Schrieke, Horovitz and W. Bousset, in ARW , iv, 136-69. Later accounts (see section 2 below). The ascension of the Prophet later served as a model for the description of the journey of the soul of the deceased to the throne of the divine judge (Asn, Escatologia, 59-60); for the -fs , however, it is a sym- | bol of the rise of the soul from the bonds of sensuality to the heights of mystic knowledge. Ibn al#Arab thus expounds it in his work Kitb alIsr" il mam alasr

(Asn, 61 ff.; Andrae, 81-2), and in his Futt , ii, 356-75, he makes a believer and a philosopher make the journey together but the philosopher only reaches the seventh heaven, while no secret remains hidden from the pious Muslim (Asn, 63 ff.). Abu 'l#Al" alMa#arr 's Rislat al9ufrn is a parody on the traditional accounts of the mi#r3 (Asn, 71 ff.). Asn in his two books quoted has dealt with the knowledge of Muslim legends of the ascension possessed by the Christian Middle Ages and their influence on Dante. In a separate work (La escatologia musulmana en la divina comedia, Madrid 1924), he has collected and discussed the literature produced by his Escatologia down to 1923; on later works, see M. Rodinson, Dante et l'Islam ..., in RHR , lxxxix (1951), 203-35. According to Ibn Sa#d , id1, 147, the isr" took place on 17 Rab# I, the ascension on 17 Rama'n . For centuries, however, the night before 27 Ra3ab a date also significant in the history of Mecca (see C. Snouck Hurgronje, Mekka, ii, 71)has been regarded by the pious as the Laylat alMi#r3 , and the eve is, like the Mawlid alNab , devoted to reading the legend of the feast (see al#Abdar , Mad9al , i, 143 ff.; G.A. Herklots, Qanoon-e Islam 2, 165; E.W. Lane, Manners and customs, London 1896, 474-6; Snouck Hurgronje, The Achehnese, i, 219; Asn , Escatologia, 97). (B. Schrieke [J. Horovitz])

2. In Arabic literature.
In writing as well as in speech, the miraculous apocalyptic phenomenon has undergone a development which it is not always convenient to confine within the limits of theological analysis. The literature of the mi#r3 must be understood as meaning all the accounts by known or unknown authors devoted to Muammad 's ascension. Unlike the idea of the isr" , recorded as being in a horizontal plane, the idea of the exaltation of a non-angelic being is not attested in the |ur"n . Thus it was at a later date that the miracle came to be recounted. We are not in a position to establish the chronology of this process. Study of the portrayals of the nocturnal journey remains to be done. Analysis of ad9 s, disentangled from theological refutation, could provide an interesting contribution to the study of the imaginary in Islam. Some accounts of the mi#r3 are by known authors, but the most widely circulated, attributed to the Imm Ibn #Abbs , cousin of the Prophet, has often been regarded as apocryphal. It remains nevertheless a decisive text. Ibn #Abbs , at once a historical and a mythical personality, has indeed become the interpreter of the community, the prototype of its expounder, since he is assigned to the origins of the group, at the moment of its foundation (Cl. Gilliot, Portrait mythique d'Ibn #Abbs , in Arabica, xxxii [1985], 127-84). This account must be viewed as a text which combines an openness to the depictions of the imagination with a respect for the basic provisions of the Law (J.E. Bencheikh, Le voyage nocturne de Mahomet, Paris 1988). |

The other authors are scholars who took over the literature of the Mi#r3 , probably so as not to allow it to develop embarrassing efflorescences. The following list enables us to verify this: al|u9ayr , Abu 'l|sim #Abd alKarm , d. 465/1073 [q.v.], whose Kitb alMi#r3 was published in 1964. He was actually an A9#ar 9fi# theologian and author of the famous Risla . alBakr , Abu 'l0asan Amad b. #Abd Allh [q.v.], a highly controversial personality who appears to have lived in the last half of the 7th/13th century. It was allegedly forbidden to read his Life of the Prophet. The manuscript of his Kitb |iat almi#r3 offers a version very close to that of Ibn #Abbs .

 al9ay , Muammad b. Amad b. #Al , d. 984/1576, 9fi# traditionist, author of alIbtih3 bi 'lkalm #al 'lisr" wa 'lmi#r3 , printed in 1970. alBarzan3 , 9fi# muft of Medina and preacher, author of a |iat alMi#r3 . His grandson, 3a#far b. Ism#l , d. 1317/1899), jurisconsult and specialist in the Sra of the Prophet, was also 9fi# muft in Medina; he is the author of T3 alIbtih3 #al 'lnr

alwahh3 fi 'lisr" wa 'lmi#r3 , printed in Cairo in 1314 with the |iat almi#r3 of his grandfather in the margin. AlBarzan3 's version differs perceptibly from that of Ibn #Abbs both in its very mannered language and in the structure of the account. We are dealing here with a very dry, spare version, which is clearly that of a scholar anxious to contain imagination within acceptable bounds. We cannot supply dates for Muammad alm alBbil al0alab , author of alSir3 alwahh3 f laylat alisr" waiat almi#r3 , printed at an unspecified date in Aleppo, which its author holds to be better than any version that had ever been made before. He states that, out of a concern to lighten the text, he omitted every isnd . He adds nevertheless that he composed it by relying on works held in esteem and famous versions of the mi#r3 . According to all the evidence, it is in fact a late version which retains the essentials of Ibn #Abbs

, enriching them with details. Furthermore, we can easily understand why the author neglects to supply his references: he is handling narratives which are, in a sense, public property. He expresses the imaginary vision of the group and stretches its credibility. The literature of the mi#r3 develops into an amalgamation of three miraculous accounts concerning the Prophet: (a) That of his purification by the angels, who open his chest and cleanse his heart of all sin. It was at a late date that this act was sometimes regarded as a kind of preparation for the ascension. The idealisation of his personality was carried to its limit. Only alBarzan3 's version mentions the opening of his chest. The other authors confine themselves to an ablution. (b) That of the nocturnal journey from Mecca to Jerusalem on alBur [q.v.]. This account was subjected to a process of amplification before being attached to the mi#r3 , although different dates were cited for the two events. In some accounts, Muammad actually meets on his journey some of those being punished in Hell (cf. alabar 's commentary on sra XVII). (c) Finally, that of the ascension properly so-called which includes the visit to the seven heavens, with a glimpse of Hell, the arrival at the Throne, the dialogue with God, the visit to Paradise and the return of Mecca. A narrative organisation was progressively established according to four essential sequences: an initial miraculous union, an initiatory raising to Heaven, a | glorifying appearance before God and a return to mankind. Elucidating the series of sequences could be of help in the search for an underlying plan providing the background for the imagination of the text. It happens that some works of a varied nature depict the same subjects as the mi#r3

account; those depictions are concerned with the constitution of the heavens and the fringes of Hell, description of the Throne, the dwellings in Paradise and angelology. These are to be found in three categories of works: (a) The |ia alanbiy" [q.v.] or legends of the prophets, the first of which are devoted to the origin of the universe. (b) General histories, whose first chapters contain numerous elements of cosmology and cosmogony. A comparative study of the use that is made of certain themes by historians and mi#r3 accounts would be very significant. (c) The resurrection literature, which brings together the texts devoted to the iyma [q.v.]. This literature of news purveying consists of edifying opuscula which make no precise reference to canonical texts and provide an account of what happens to the believer from the time of his death until his appearance before God. We find here descriptions of the angels, Paradise and Hell. Some descriptive and narrative elements, independent of one another in their origin, are joined together. The Prophet himself is included in this universal destiny and, awakened from the rest of death, he mounts alBur a second time to head for the Rock at Jerusalem and finally appear before the Lord (Bencheikh, op. cit.). The accounts of the mi#r3 and iyma are clearly related. The same working of the imagination, the same process for setting the portrayals, led to the production of the two texts. One should mention here besides that the frontiers of writing have not prevented the migration of legends. The Kitb

#A3"ib alma9lt of al|azwn [q.v.] devotes a long section to angelology; the story of 0sib Karm alDn in the Thousand and one nights contains the cosmogonic account of Buluiyya possibly borrowed from al9a#lab 's |ia alanbiy" . We are dealing here with what we should call preconstituted bodies of writings regarding Heaven, Hell, Paradise, the Throne and angels. Each of these bodies of writings has been given an independent setting following a long process of elaboration during which all kinds of pre-Islamic and Islamic themes have been gathered together. Only detailed analysis of this process will allow us to establish a reliable chronology for this literature. One should note in addition that we cannot totally isolate the mi#r3 accounts from the great visionary texts of Ibn #Arab , Ibn Sn and alSuhraward [q.vv.]. The remarkable Kitb al-Tawahhum, if it is the work of a mystic, alMusib , preserves just as many of these themes as are connected with the literature with which we are concerned (French tr. Roman, Paris 1978). But one must realise that the imaginary aspect of the mi#r3 is also nourished by the pronouncements of speech. The

[q.v.] or preacher is to be found at the heart of religious observance. With him we leave learned discourse and mystical meditation to follow the dialogue between a desire and a function: the desire of the believer who needs to believe, the function of the one who gives him something to believe and undertakes to supply him with imaginative depictions in order to do this (Bencheikh, op. cit., Introd.). The oral legend of the mi#r3 has not been col- | lected. The narrative structure laid down in the texts considered to be canonical has taken over depictions of diverse origin. The texts have been given an iconography to respond to the need of their listeners for marvels. This need the theologians regard with suspicion. The inventors of fabulous tales were pursued and treated severely by Ibn 0anbal , Ibn al3awz , al9azl and alSuy . The establishment of collections of apocryphal ad9 s for the denunciation of forgeries, if it is informative on the orthodoxy of their thought, is just as helpful in interpreting the ramblings of the imagination. In fact, the same questions posed by the theologians on the subject of the mi#r3 , have determined the direction of the flow of the imagination. There has been lively argument concerning the idea which it was necessary to have of the ascension. We must also review the possible interpretations of this miraculous deed: (a) The ascension took place in spirit; it is a question of a vision that occurred during the Prophet's sleep. In a sense it was Heaven that visited Muammad . It was an illumination, and the physical person was not concerned: the mi#r3 annuls the human condition and registers itself in an unhinging of the intellect. The soul, purified, traverses Heaven as far as God in a trajectory of which the traces remain only in the Prophetic witness.

The outburst of faith on the part of the Believer will repeat the movement. (b) The isr" was really performed by Muammad while awake, but the ascension took place only in the spirit. This artificial distinction is useful: it bears witness to a process of linking the isr" , at first independent, with the mi#r3 (cf. H. Birkeland, The legend of the opening of Muhammad's breast, Oslo 1955). (c) The ascension was really effected, body and soul, by a Prophet who was in a full state of consciousness. This interpretation lays the foundations of the miracle, which becomes a theological argument. In this way there becomes authenticated a willingness to attribute to the Prophet a dimension which goes beyond his historicity: progressively, the opening of the breast at the end of the purification, the isr" and the mi#r3 constitute a unique account which offers the advantage of crossing important zones of the religious imagination. This interpretation de-spiritualises the ascension, without removing its character of a supreme initiation. Whilst refusing to see in it an internal impetus, a wandering of the spirit, it affords free range to imaginative portrayals. The account gives formal licence to imagine the unthinkable. Revelation triumphs out of ecstasy. The brilliant but solitary illumination of an individual is abandoned for the benefit of the communal initiation of a prophet. The latter is charged to inform his people of the answers that, in the course of his journey, he is entrusted to communicate. From that time on, the difference can be seen between |ur"nic utterance reduced to a mention and the speech of literature charged with illustrating the former. A modern application of this splitting of writings is attested; in Ra3ab 1387/October 1967 Egyptian State Television broadcast a film on the

isr" and mi#r3 under the responsibility of the al-Azhar authorities. The process of image production is allowed, but controlled. The literature of the mi#r3 is then embodied in an act of adoration. For one who takes neither the path of thought nor that of spirituality, there remains the portrayal, at the risk of blasphemy. Behind Muammad , the only one authorised to travel to the forbid- | den space, imagination is set free. A whole community accompanies its prophet on his initiatory journey. Thus the act of faith is progressively established in an account destined to arouse visions. The image here is effective; it revives confidence and provokes fear. In this way the delights of Paradise materialise and the tortures of Hell are displayed to view at the cost, furthermore, of a decisive anticipation of theology, since God will only deliver His sentences on the Day of the Last Judgment. Here appear the two essential functions of the mi#r3 literature: it verifies in advance divine justice, and it responds to a deep-seated need for the marvellous. The image goes beyond efficacy in order to touch on the aesthetic. Happiness lies not only in the distant reward of virtue, it also finds its source in the spectacle offered. The text is adorned with illuminations: from alBur to the Throne of God, from the seven heavens to the sojourn in Paradise, then to Hell; the marvellous spans the space that separates thought from desire. But it asserts its humanity and does not profane the sacred. In addition, imaginative use is made of the real. The topography of the places uses language without any surprises: we climb to Heaven with a ladder, we skirt the walls, we pass through doors: here mountains, rivers, seas, gardens are lost to view. The highest place is just a throne, a common attribute of sovereignty, men's destinies are written with a pen, an ordinary instrument for writing. Gold, silver, pearls, gems abound in our markets, darkness, smoke, ice have their place in our environment (Bencheikh, op. cit.). It was probably difficult to represent the unimaginable without profaning it. Only the obligation of reserve ensured the survival of the myth. The encounter with the Creator shows us how blasphemy was avoided; all the dialogue between God and

Muammad is made up of the last verses of Srat alBaara , put alternately into the mouths of the two interlocutors. Furthermore, this is not the only way of using verses or ad9 s which are often solicited. The text then becomes a commentary full of their imagery and is legitimised. It is woven into an irrefutable writing and takes on the authenticity of a sacred iconography. For the listener, the mi#r3 celebrates the prophetic mission and offers the account of an apocalypse populated with |ur"nic resonances. Called to appear before God, the Prophet receives along his way the homage due to the last of the Messengers. Islam relies on and celebrates at the same time its primacy. It lays hold of the future and reconquers time past; Muammad passes through the heavens accompanied by all the great witnesses of the universal faith. The ritual of adoration is then written down in an outburst of beauty! The heavens are successively of smoke, copper, silver, corindum and pearl. The trajectory of the symbol accompanies the ascension journey; the beauty of the spectacle declares the proximity of God. Each year millions of Muslims devote themselves to the spectacle of their faith. Finally, how can we forget that the mi#r3 has entered into universal literature, thanks to the Book of Mahomet's Ladder. This translation in Latin of a text in Castilian, itself translated from Arabic, had indeed become famous, thanks to the furious polemic to which it has given rise; and indeed, Dante was actually able to be inspired by it for his Divine Comedy. The Editors regret that they have failed to provide for a section on the mi#r3 in Persian literature. Whilst they anticipate that they will be able to repair this omission, they would refer the reader to the article by A.M. Piemontese, Una versione persana della storia del Mi#r

, in OM lx/1-6 (1980), 225-43, which offers a translation of a passage of the Sh#ite commentary of sra xvii by Abu 'lFut alRz (ca. 480-525/1087-1131). (J.E. Bencheikh)

3.
Mi#r3 literature in East and West Africa. The celebration of the mi#r3 has here given rise to an extensive popular literature. Among the 9fi# communities, from Egypt down to the East African coastlands, the most widely-used text is alBarzan3 's |iat almi#r3 or, in a later edition, alIsr" wa 'lmi#r3 , available in Haji Mohamed's bookshop in Mombasa together with the same author's Mawlid [see MAWLID. 2. In East Africa]. In Swahili , there are two translations in prose and several versified elaborations; for the oldest known version in the ukawafi metre usually reserved for | liturgical hymns, and which can still be sung, see Knappert, Swahili Islamic poetry, iii, Leiden 1971, 227-75; the same text but with long comments and notes, is given by idem, Miiraji, the Swahili legend of Mohammed's Ascension, in Swahili , Jnal. of the Inst. for Swahili Research (Dar es Salaam), xxxvi (1966), 10556. A different poem, written before 1922, is given by idem, Utenzi wa miiraji by Mohamed Jambeini, in Afrika und Uebersee, xlviii (1966), 241-74, and another by Yusuf Ulenge (perhaps from an older oral

text) is printed in Swahili , xxxviii (1968), with the title Utenzi wa Miraji. A much longer poem, published by E. Dammann in his Dichtungen in der Lamu Mundart des Suaheli, Hamburg 1940, 1-72, was written down for Dammann by the poet Muhammad b. Abubakari Kijuma, who was probably also its author; see Muhammad b. Ibrahim Abou Egl, Life and works of Muhammad Kijuma, Ph.D. thesis, London Univ., 1984, unpubl. In the mosques of East Africa , the laylat almi#r3 is celebrated after the night worship by singing these hymns, after which the imm explains its significance to the congregation. In Hausaland (Northern Nigeria), poetic versions of the mi#r3 narrative were first discovered by M. Hiskett, see his A history of Hausa Islamic verse, London 1975, 48-62. Here the theme of the mi#r3 is part of the mu#3izt literature, poems written by scholars in praise of Muammad which can be sung or read. These are called madahu, from mad praise. They are very popular and often recited during Rama'n when the people are in a receptive mood. A detailed description can be found in the thesis of Abdullahi Bayaro Yahya, The Hausa verse category of Madahu, with special reference to theme, style and the background of Islamic sources and belief, Univ. of Sokoto, Nigeria 1987, unpubl.; it includes Hausa texts and translations, and the place of the mi#r3 as one of the themes of madahu is comprehensively discussed. In 9ana , the mi#r3 theme forms part of the Mawlid (as it does in East Africa) and is celebrated in the north of the country, in Kumasi and along the coast.

In the Gambia , the theme of the mi#r3 appears to have penetrated into the pre-Islamic Mandinka epic of Sunjata. In G. Innes' edition (Sunjata, London 1974, 156-7, l. 287), the diviner Siise, before answering the king's question, goes into retreat like a shaman and states that for forty days I saw the seven layers of the sky. This is clearly taken from the mi#r3 , in which Muammad himself claims the same; see Knappert, Traditional Swahili poetry, Leiden 1967, 152. In Peul or Fulani [see FULBE ], there is a long section on the mi#r3 in an important long poem edited by J. Haafkens in his Chants musulmans en Peul, Leiden 1983, 193-203. The long poem Busarau (from Arabic bu9ar" , pl. of ba9r , bringers of good tidings, ibid., 144-335), was heard sung and recorded from manuscripts by Haafkens in northern Cameroon, where it is extremely popular. (J. Knappert)

4.
Mir#r3 literature in Indonesia. In West Java the celebration of the mi#r3 is still very popular; it takes place in the mosque, in the home or in the langgar, the little prayer-cabin near the house. On the eve of 27 Ra3ab , or in some places even on the preceding evening, people come together in families, or invite friends, usually the men and the women separately. The men do not smoke, since | smoke is disliked by the angels. The proceedings begin after the alt al-

ma9rib and are concluded by a selataman, a festive meal, around midnight, after the alt al9" . Those who know stories will tell them, about the life of Muammad and his night journey. Customarily, one who is qualified reads from alZahr albsim f awr Abi 'l|sim , by Sayyid #U9mn b. #Abd Allh b. #$il b. Yay (Jakarta 1342/1924, pp. 80), in Malay. During the reading gaharu or aloes wood is burnt as incense. If someone is present who can read Arabic (normally only in the towns), he will be asked to read the 09iyat alimm al#rif bi"llh ta#l Abi 'lBarakt Sayyid Amad alDardr #al |iat alMi#r3 li"l #allma

alhammm barakat alanm Na3m adDn al9ay , Cairo n.d., pp. 27. This work is known generally as Dardr . A poetic rendering of the mi#r3 was written in Malay by Hadji Adam b. Hadji Kasman, the Syair Mi'raj Nabi Muhammad, Jakarta 1926. Another very popular Arabic booklet here is alBarzan3 's |iat almi#r3 . In the mosque, the preacher will begin with prayers, then launch into a naa , an admonishing sermon. He will explain the passages |ur"n XVII, 1, and LIII, 9, in the light of the well-known commentaries. He may also quote the even better known Mawlid of alBarzan3 , usually referred to as 9araf alanm (see MAWLID, and C. Snouck Hurgronje, The Achehnese, Leiden 1906, 209-14). In this Mawlid (as in most) there is a section on the mi#r3 , section XIV (see for a translation, J. Knappert, Swahili Islamic poetry, Leiden 1971, i, 57-8). The recital is terminated with the salutation to the Prophet, again from the Mawlid of alBarzan3; it begins: Y Nab , salmu #alayka

... (for a translation see Knappert, op. cit., iii, 322, v. 31). All the worshippers present sing this passage with the preacher, while standing up, as the text prescribes, out of respect for Muammad . It is customary for the imm to commemorate the mi#r3 also in his 9uba on the Friday before as well as the Friday after 27 Ra3ab . In some districts of West Java, gambling is strictly prohibited during the entire month. At night the mosque may be illuminated, and even fireworks are set off by some; houses, too, keep a lamp burning all night. The celebration may last until the alt alub ; this is doubtless in memory of Muammad 's return from the heavens before dawn (Knappert, op. cit., iii, 271, v. 96). The religious teacher (in West Java, the Kiyai) will embellish the legend, weaving moral lessons into it for the edification of his flock. In the Periangan, he will speak in Sundanese, after the prayers in Arabic; there is a Sundanese poem describing the mi#r3 which can be sung that night. Some people fast during 27 Ra3ab , others fast for nine days, not consecutively but in sets of three at the beginning, middle and end of Ra3ab , which was already a holy month in ancient Arabia. Between the two World Wars, processions came into fashion, especially among the younger generation in Indonesia. Various Islamic groups have different opinions about the desirability of celebrating the mi#r3 . Modernists explain it totally symbolically as a spiritual development. Reformists want to abolish mi#r3 celebrations as a bid#a which smacks of worshipping the Prophet rather than God alone. The conservatives however, cling to the #dat and take the mi#r3

literally, namely that Muammad saw God with the eyes of his head (alBarzan3 ). (J. Knappert)

5. The
Mi#r3 in Islamic art. In the spirit of the Second Commandment, strict Muslims have always frowned on the portrayal of humans and other living creatures in painting and sculpture, and this antipathy is naturally strongest when the depiction of the Prophet Muammad is in question. But this peculiarly Semitic prejudice was never fully shared by the Aryan Persians, and it is thus primarily in the miniature painting of Persia that we must look for representations of the mi#r3 . As Sir Thomas Arnold wrote, It was a frequent practice of the poets ... to include in the Preface ... a lyrical outburst on the theme of Muammad 's Ascension ... No incident in the religious history of Islam is more commonly represented in Muslim art than this of the Ascension of the Prophet. There were, indeed, a number of works, poetical and otherwise, devoted entirely to the mi#r3 and its attendant circumstances (see 2. above). These are represented in painting, firstly, by the fragments of a large 8th/14th century manuscript preserved in Topkap album H. 2154, the miniatures being attributed to Amad Ms , and secondly, by the celebrated Paris manuscript of 840/1436 (Suppl. turc 190) written at Herat in the Uighur language, and illustrated by one of the finest and most original sets of miniatures in the whole of Persian painting. But these are special cases, and we are more concerned here with mi#r3 miniatures occurring in manuscripts of the classical poets.

The earliest of these date from the opening years of the 9th/15th century. In the Miscellany of Iskandar Suln (British Library Add. 27261, dated 813-14/1410-11, fol. 6a) the mi#r3 is represented with the Prophet, mounted on Bur , surrounded by angels, and conducted by Gabriel, soaring above the Ka#ba enclosure at Mecca, and this scheme of representation is closely followed in a detached double-page miniature of much the same date in the Chester Beatty Library (Cat. 292. i, ii). In both these (as in the Paris manuscript of 1436) the Prophet is shown unveiled, but in the latter, as in many subsequent examples, the face has been partially obliterated and repainted. This same basic composition, i.e. with the Ka#ba below and the Prophet unveiled, is followed in main line Persian painting of the later Tmrid period, as in the Nim of 900/1494 in the British Library (Or. 6810, fol. 5b), and is continued in the Keir Collection miniature (Cat. III. 207) dated 910/1505, which is one of those added under 9h Ism#l to the Nim of Ya#b Beg A |oyunlu ( Topkap , H. 762), and may well be an early work of Suln Muammad . This latter miniature is among the most striking and original treatments of the subject, the ascent, accompanied by innumerable angels and observed by others through a circular hole in Heaven, being enclosed in a rectangular frame with the Ka#ba enclosure below and little desert villages with palm trees occupying the margins. The

mi#r3 is of frequent occurrence in 9rz manuscripts of the Tmrid period (though, curiously enough, hardly ever found in the mass of manuscripts illustrated in the Commercial Turkman style), but in | these, as in the numerous -afawid examples, the Ka#ba is omitted, only the Prophet, now veiled, and accompanying angels being shown. Perhaps the finest of all is in the great Nim of 9h ahmsp (British Library Or. 2265, fol. 195a) which is almost certainly the work of Suln Muammad , and probably represents his swan-song. Exceptional portrayals are occasionally encountered. Thus in the Topkap Nim of 844/1441 (H. 774), probably of western Indian origin, the Prophet is depicted as a golden disc inscribed with his name; and in the British Library copy of 1075/1665 (Add. 6613, fol. 3b) the ascent is made against a background of concentric circles with the symbols of the planets revolving round the sun. In post-arawid painting, the theme becomes somewhat vulgarised; Bur may sport a peacock's tail and a clumsy |3r crown, and the Prophet is sometimes reduced to a sort of shapeless bundle. But on a fine painted lacquer mirror-case of 1288/1871 in the Bern Historical Museum, Muammad Ism#l depicts the scene in traditional manner, though on a miniature scale.

MI0R$B (a.), pl. marb , the prayer niche in the mosque . Etymological origin of the word. In Islamic religious practice and in Islamic architecture, the word denotes the highest place in a mosque, a niche which shows the direction of the ibla [q.v.], or the station of the Imm in a mosque (Lane, 1865, 541). The word includes the radicals -r-b, from which comes the verb ariba , which in Form I means to be violently angry, to be affected by canine madness; in Form II to provoke, to sharpen, or to excite s.o.; in Form III to fight, to wage war with ..., or to battle with ... and in Form VI to make war, or to wage war with one another. Due to these definitions, some scholars expressed their doubts that mirb derives from the above verb. Lane put forward ... that the explanation of this is because the person praying wars with the devil and with himself by causing the attention of his heart (loc. cit.). A similar interpretation was offered by Goldziher when he suggested that the mirb was a place of struggle, a battlefield, and he referred to the Prophet Muammad who said that as blood circulates in people, likewise Satan circulates around them (Goldziher, 1872, 220). The above explanations are clearly not satisfactory. It was because of this difficulty that some scholars surmised that it was a loan-word in Arabic. Dillman tried to connect it with the Ethiopian mek w erab (Dillman, 1865. 836). The possible Ethiopian origin of the word and its connexion with mek w erab was refuted by Praetorius who, after studying some early South Arabian inscriptions, concluded that the word mirb at that time meant some kind of a building, but conceded that the origin of the word was still obscure (Praetorius, 1907, 621). Others, as e.g. Beer (1895, 19) and Daiches (1908, 637-9) tried to connect it with Hebrew orb t which occurs several times in the Old Testament and means ruins, ruined cities, ruined dwellings or even palaces or fortified buildings. This theory was again considered to be very unlikely by Nldeke (1910, 52). The majority of scholars have never doubted that the word is Arabic and, accordingly, have tried to find its provenance and original meaning by examining pre-Islamic Arab literature and one of the earliest and most important sources of the Islamic period, the | |ur"n . Rhodokanakis was one of the first scholars to study these early sources and to publish his observations in two articles. In the first article he concluded that the word in pre-

Islamic literature meant a palace, a niche, a recess or a room, a balcony or a gallery. Then he quoted a sentence from the ad9 where the word can be interpreted as sanctuary (Rhodokanakis, 1905, 296). In his second article, Rhodokanakis narrowed down the meaning of the word and suggested that it actually referred to a part of a king's or a prince's building, namely to a meetingroom, or more precisely to a throne-recess within such a room, as mentioned in |ur"n , XXXIV, 12. Such throne-recesses can be found, Rhodokanakis continued, in the Umayyad palaces such as |uayr #Amra and M9att (we can now add also 9irbat alMaf3ar ). In other verses of the |ur"n , namely in XIX, 12, it refers to a sanctuary, while in III, 36, the word is used for a lady's private chamber (see also Dozy, 1927, i, 265). Rhodokanakis mentions that in XXVIII, 21, it was not clear whether the Prophet meant a complete palace or only a chamber (Rhodokanakis, 1911, 71). Horovitz referred to some of the occurrences in pre-Islamic poetry, among them one of alA#9 's poems (alButur , 0amsa , CDIV, 4) where the word, he claimed, meant a throne-recess (Horovitz, 1927, 260). In a more recent article, Serjeant explained that the basic meaning was a row of columns with their intervening spaces. He also suggested that under the Umayyads, while retaining its other senses, it was the name given to the mara [q.v.] (Serjeant, 1959. 453). Mamd #Al 9l claimed that the ancient South Arabian mi9n was almost identical in usage with the mirb . It was a kind of mas3id or muall

[q.vv.], or even a burial place in the shape of a portico, place for prayers, and services for the dead ( 9l , 1962, 331-5). In connexion with this last interpretation, the present author in an article called attention to the fact that flat marble, stucco, stone, or faience marb strongly resemble tombstones. Tombstones from early Islamic times onward frequently depict a mirb design. He re-examined one of alA#9 's poems (alButur , 0amsa , CDIV, 4) where the word mirb occurs and suggested that it can be interpreted as a burial place, as opposed to Horovitz's explanation as a throne-recess. Other literary examples also use the word in the same context (Fehrvri, 1972, 241-54; also, idem, 1961, 32 f.). From these interpretations it would appear that in pre- and in early Islamic times, the word mirb was basically used for a special place within a palace or in a room; it was the highest, the first and the most important place. At the same time it also denoted the space between columns and was equally used for a burial place. Its architectural origin and introduction into Islamic religious practices as the most prominent feature in a mosque should be examined from these various angles. Architectural origin. In his EI 1 article on the mirb , Diez mentioned that orientalists and art historians give a twofold origin for the mirb : the Christian apse and the Buddhist niche (Diez, 1936, iii, 485). Both features were alien to the Arabs and were not required by Islamic religious practices. Thus it could never have been introduced and accepted by the early Muslims without an adequate theological explanation. As an architectural feature, the mirb is made up of three basic elements: an arch, the supporting columns and capitals, and the space between them. Whether in a flat or in a recessed form, the mirb gives the impression of a door or a doorway. | The application of this feature can be as varied as the pre- and early Islamic meaning of the word suggests. The idea of a decorated recess or a doorway in the form of what we know and accept today as a mirb goes back to remote antiquity in the Near East. In its secular sense it was used in palaces as a raised platform with a dome above supported by four columns under which the divine ruler carried out his most important functions (Smith, 1956, 197). It was a royal baldachin, the first place in a ma3lis

, a throne-recess. In its religious context it was a sanctuary, fixed or portable, under which the cult images were placed and were provided with a shelter. The tradition of these domical shelters can be traced back to some of the tent traditions of the Near East, particularly to those among the Semitic people (Smith, 1950, 43; idem, 1956, 197). Such domical tents or structures were also used over burial places. These ancient oriental traditions were later adopted by Judaism and Christianity. The direction of prayer and divine service in ancient religions, particularly in Judaism, added greater importance and widened the scope of these antique traditions. Orientation was especially important among the Semitic peoples and it was not a matter of choice. The Jews turned towards Jerusalem, and in this respect all monotheistic people looked up to the Jews and followed their practice (Krauss, 1922, 317). Early synagogues, however, had no orientation; only the prayer was directed towards Jerusalem. The Ark of the Law, the "aron haode9 , had no permanent place in the building. It was only in the second phase of the development of synagogue architecture that it received a permanent station within the building, an apse or a niche which was orientated towards Jerusalem (Sukenik, 1934, 27). The earliest known synagogue with such an apse was excavated at Dura-Europos dating from the 3rd century A.D. (Lambert, 1950, 67-72; Goodenough, 1953, figs. 5945, 599). From subsequent centuries there are several other examples known, some of these depicting the Ark in mosaic decoration on the floor of the apse (Hachlili, 1976, 43-53). The essential part of these decorations include a pair (or two pairs) of columns supporting a semicircular arch framing a conch, while below, within the columns, the Ark is shown as a pair of doors, thus symbolising a doorway, the portal of the life beyond, or the portal of the dead (Goldman, 1966, 101 f.). Christianity followed the same tradition. Early churches had an east-west orientation, the entrance was facing the east and the altar was towards the west. There was usually an apse, a haykal, with a pulpit. Churches and funerary chapels (martyria) generally had a large dome in Syria and in Palestine, not because of the structural function but rather because of the importance attached to this form (Smith, 1950, 92). The domical form with a portal below was frequently depicted on coins from the 4th century A.D. onwards (Smith, 1950, figs. 17-21), and later appears on tombstones (Goldman, 1966, 105). The form of the Christian and Coptic apse was so strikingly similar to a mirb that it was not surprising that Arabic sources mention it as a feature borrowed from Christian churches (Lammens, 1912, 246; Creswell, 1932, 98). The form was, however, not unfamiliar to the Arabs. Pre-Islamic sanctuaries in Arabia had a round tent, a ubba [q.v.] over their idols or over some of their burials (Lammens, 1920, 39-101; idem, 1926, 39-173). It seems reasonable to surmise that the Arabs wanted to preserve this ancient Semitic Arabian custom, but intended to dress it in an Islamic garb by offering a new interpretation to the pagan tradition. | By examining the life of the first Muslim community in Medina we may understand how and why this feature was adopted and introduced into Muslim religious practices. During his lifetime the Prophet was not only a religious leader, but also a statesman who used the first primitive mosque in Medina not only as a place for communal prayer but also for public ceremonies where he

received delegations and delivered judgements. He used to sit on his minbar [q.v.] which was set against the ibla wall. Thus in the strictest contemporary interpretation of the word, the place was a mirb . When the Prophet died he was buried in a room next to the ibla wall, whereby in every sense his grave was in a mirb , coinciding with the definition of a burial place. His grave was most probably marked, we may presume, with a stone which included in its decoration all the elements already known from Judaic and Christian funerary art. That this must have been the case is perhaps attested by the references given by Ibn Rusta and Ibn Baa (Fehrvri, 1972, 252). Furthermore, there is ample evidence suggesting that early marb were not recessed niches but flat panels reminiscent of tombstones. History. Primitive Islam and the Umayyad period. After the Prophet's death, in the early primitive mosques the mirb was indicated by a stripe of paint or by a block of stone embedded in the ibla wall. It appears that this practice was followed for a considerable time, since alMarz informs us that when #Amr b. al#$ built his mosque at Fus in the winter of 20/641-2, he placed no hollow mirb in it ( 9ia , ii, 247). In the Prophet's mosque at Medina, the ibla was indicated by a large block of stone which was first placed to the north, i.e. towards Jerusalem, but in the second year of the 0i3ra was moved to the south side, facing Mecca (Burton, 1893, 361). In the earliest mosque at Wsi , built by al0a333 b. Ysuf [q.v.], the excavators found no mirb

recess (Safar, 1945, 20). This was also the case in the mosque of Banbhore in Pakistan, built at the end of the 1st/early 8th century (anon., in Pakistan Archaeology, i [1964], 52). That must indicate that in both cases the ibla was most likely marked either by a flat stone or by a stripe of paint. The earliest known surviving mirb is a marble panel known as the mirb of Sulaymn in the rock-cut chamber under the |ubbat al-a9ra [q.v.] in Jerusalem (Pl. I, 1). Creswell has already indicated that this was most likely the earliest surviving mirb . His argument was based on the shape of the arch, on the primitive Kfic inscription on the lintel, and the simple scroll motif on the arch and the rectangular frame (Creswell, 1932, i, 70, pl. 120a in ii; idem, 1969, i/l, 100, fig. 374). Several other points can now be added to Creswell's remarks. The crenellations and pearl motifs on top of the panel recall pre-Islamic, i.e. Ssnid , monuments with identical decorations. The vertical bands down the pilasters are similar to those on the mosaics of the circular arcade and on the drum of the dome (Creswell, 1932, figs. 189-9, 201, 205, pls. 33b, 37b). Further evidence is provided by coins depicting mirb designs on their reverse, most likely accepting the mirb of Sulaymn as their model (Miles, 1952, 156-71; idem, 1957, 187-209, nos. 7-8; idem, 1959, 208, pl. I/1; Fehrvri, 1961, 90-105). All of these coins are attributed to the period when #Abd al-Malik introduced his financial reforms in 75/694-5. The first concave mirb , i.e. mirb mu3awwaf , was introduced by #Umar b. #Abd al#Azz , governor of Medina, when he rebuilt the Prophet's Mosque in 87- | 8/706-7 (al-

Marz , 9ia , ii, 247; Ibn Ta9rbird , alNu3m alzhira , i, 76). It was richly decorated with precious material (Sauvaget, 1947, 83-4). After that, semicircular marb rapidly spread throughout the Muslim world. Such a mirb was introduced into the Great or Umayyad Mosque at Damascus when alWald I took over the entire building and rebuilt it between 87/706 and 96/714-15. According to early accounts, it was set with jewels and precious stones. This mirb was destroyed in 475/1082 and then was subsequently rebuilt but destroyed again by fire in 1893. The third concave mirb was built in the Mosque of #Amr at Fus in 92/710. Semicircular marb flanked by pairs of columns were found in almost all of the Umayyad desert palaces (Creswell, 1932, fig. 438, pl. 120b, e; idem, 1969, i/2, figs. 538, 638, pls. 66 f., 103e and 115b). The earliest surviving concave mirb in a mosque is, according to Cresswell, in the Mosque of #Umar at Bor , built during the late Umayyad period (1969 i/2, 489, fig. 544, pl. 809).

1. Jerusalem, Mirb of Sulaymn under the

|ubbat al-a9ra , 72/962.

2. Ba9dd , The 9ak mirb , late Umayyad or early #Abbsid period.

3. Mawil , Corner mirb in the Mausoleum of Imm Yay b. |sim , 7th/13th century.

4. Aleppo, marble mirb

in the Madrasa al-Firdaws, 633/1235.

5. Cairo, stucco mirb in the Mosque of al3uy9 , 478/1085.

6. Cairo, stucco mirb in the Mausoleum of Sayyida Ruayya , 527/1133.

7. Cairo, stucco and glass mirb in the Mausoleum of Sulaymn alRif# , 690/1291.

8. Cairo, marble mirb in the Mausoleum of |alwn , 684/1285.

9. Cordova, mirb in the Great Mosque, 354/965.

10. Nyin , stucco mirb in the Mas3id -i 3mi# , second half of 4th/10th century.

11. Sar-i Pul, Af9nistn . Stucco mirb which was standing until recently in the Ziyrat of Imm -i 9urd , 6th/12th century. (Photographed and reproduced courtesy of Prof. A.D.H. Bivar).

12. Yazd, faience mirb in the Mas3id -i 3mi# .

13. |uhrd , Iran, mirb in the Mas3id -i #Al , lustre-painted and monochrome glazed tiles, 700-8/1300-8.

14. Konya, faience mirb in the Sadrettin Konevi, 673/1274-5. (Photographed and reproduced courtesy of Mr. H. Dewenter)

15. Dihl ,

mirb in the tomb of 9ams alDn Iltutmi9 , ca. 633/1235. (Photographed and reproduced courtesy of Mr. J. Burton-Page)

16. 1ar9 Af9nistn , wooden mirb in the Mosque of 9h Muy alDn , 6th/12th century. (Photographed and reproduced courtesy of Prof. A.D.H. Bivar). #Ir . The 9ak mirb in Ba9dd (Pl. I, 2) is the earliest known surviving example in the country, as it may date from the end of the Umayyad or from the beginning of the #Abbsid period (Sarre and Herzfeld, 1920, ii, 139-45, Abb. 185-7, Taf. XLV-XLVIa-d; Creswell, 1940, 35-6, fig. 26, pl. Ia-c; alTutun , 54-62, figs. 3-5, pl. 4). It was carved out of a simple block of marble in a shallow semi-elliptical form. The spiral columns are crowned by acanthus capitals which support the bell-shaped niche-head, which is framed by a frieze of acanthus leafs, followed by a narrow stripe of astragals and a band of palmettes alternating with bunches of grapes. A vertical ornamental band runs down at the back, the lower part of which is missing. The most interesting part of the design is the shell which is contained

within a horseshoe shape. The shell as niche-head occurred for the first time in a grotto at Bniys in Syria, dating from the 1st century B.C. (Wheeler, 1857, 37). Later, it was frequently used in classical art but was more popular in the eastern part of the Roman Empire. The motif was associated with funerary monuments and as such was often depicted in Jewish and in early Christian sepulchral art. The rounded shape of this mirb and the fact that it was made of marble, which was not available in #Ir , may indicate that it was imported from Syria or from the southern part of Asia Minor. As a rule, mirb niches are rectangular in #Ir and in Persia. The origin of this form may be found either in the rectangular recesses of Nestorian churches in #Ir or in the Persian wn [q.v.]. The earliest known rectangular mirb in the country survives in the fortress palace of U9ay'ir , dating from the latter half of the 2nd/8th century (Creswell, 1940, pl. 120e; alTutun , figs. 6-9, pl. 5). Marb which were erected in mosques, palaces and in private houses at Smarr mark the first turning point in the development of this feature in Islamic religious architecture. The significant role of the marb was emphasised by their size, which became considerably larger during the 3rd/9th century, e.g. in the Great Mosque of Smarr (Creswell, 1940, pl. 66c) or in the Mosque of Ab Dulaf, where the excavators discovered two mirb s, one within the other (Francis, 1947, pls. 5/1-2; alTutun , figs. 15-7, pl. 9). By then, they were more lavishly decorated with carved stucco. A large mirb with stucco decoration was excavated in the 3awsak al9n palace in

Smarr , built by the caliph alMu#taim between 224/838 and 228/842, close to his throne-room. The niche was more than one metre deep and was flanked by two pairs of columns. The walls of the niche and | the columns were coated with stucco (Herzfeld, 1923, Abb. 132; Creswell, 1940, fig. 191). Flat mirb panels were used in smaller mosques, mausolea and in private houses. The prototype for these flat marb was clearly provided by the mirb of Sulaymn in Jerusalem. Evidence for this is to be found in the mirb of the 3mi# al#Umariyya at Mawil . It is made up of six panels, the lowest central one being the earliest, probably of the 3rd/9th century, and a close copy of Jerusalem flat mirb (Sarre and Herzfeld, 1911, ii, 285-6, Abb. 275, Taf. CXXXV). Several stucco flat marb were discovered in private houses in Smarr , presenting all the three styles of the Smarr stucco decorations (Herzfeld, 1923, Abb. 167-70, 269-60, 316, Taf. LXII and XCVII). An interesting combination of flat marb can be seen in two small mausolea in Mawil , the Mausoleum of Yay b. |sim (Pl. II, 3) and in the Mausoleum of Imm #Awn alDn (al Tutun

, fig. 59, pl. 34), both erected during the 7th/13th century (Sarre and Herzfeld, 1911, 249, 263-8, Taf. CXXXV). These two marb are almost identical. They are made up of two flat panels showing the correct ibla direction. In each of these two mirb s there is a mosque-lamp hanging down from the pointed arch. Out of the later rectangular marb in #Ir , mention should be made of the main prayer niche of the Great Mosque in Mawil which appears to be a combination of flat and rectangular types (alTutun , figs. 34-6, pls. 17-9). It is flanked by a pair of octagonal pilasters decorated by intertwined scrolls and crowned by what Herzfeld called lyra capitals (Sarre and Herzfeld, 1911, ii, Abb. 230-3). The spandrels and the canopy have rich arabesque decoration. Below, at the back of the recess there is a decorated panel showing a pair of spiral pilasters on bell-shaped bases and topped by identical capitals supporting the arabesque-decorated spandrels and canopy. This mirb may also be regarded as a transitional form between the simple and multi-recessed marb that played an important role later in Persian religious architecture. The inscription round the niche bears the signature of the artist, one Muaf from Ba9dd , and the date 543/1148 (Van Berchem in Sarre and Herzfeld, 1911, i, 17; Herzfeld, 1911, ii, 216-24). There was a free-standing mirb built of stone in the courtyard of this Great Mosque in Mawil , but it was moved to the #Abbsid Palace Museum in Ba9dd . It was attributed to Nr alDn Arsln 9h I (589-607/1193-1211). It has two recesses, the outer one being rectangular in plan while the inner one set back from it is pentagonal. There is an interesting innovation here, namely, the frame is composed of small compartments (Sarre and Herzfeld, 1911, i, 18, ii, 227, Taf. V; Francis, 1951, pl. 3, no. 10; alTutun

, fig. 38, pl. 20). A similar frame, but decorated with human figures, appears around a niche that was discovered near Sin3r on the site of G Kummat and which might have been a mirb (Reitlinger, 1938, 151-3, figs. 14-7; Francis, 1951, pl. 5, no. 16; alTutun , figs. 39-9, pls. 29-30). The marble mirb of the Pan3a #Al in Mawil (built in 686/1287) can be regarded almost as a triple mirb since the central pentagonal recess is flanked by a small niche on either side. All three recesses are crowned by muarnas [q.v.] semi-domes, while each panel in the central niche is decorated by a hanging mosque-lamp (Sarre and Herzfeld, 1911, ii, 270-8, fig. 268, Taf. VII; Francis, 1951, pl. 2, no. 5). Syria and Palestine. Marb were usually concave in these countries, but flat panels were used from time to time. A small and somewhat simple marble | mirb panel decorates the first pilaster under the western portico in the Umayyad Mosque in Damascus. It is attributed to the lnid period (Creswell, 1940, ii, pl. 123c). Another flat mirb , a stucco panel, probably of the same period, is in the Mam #Abd al#Azz at al9arr . Two pairs of pilasters support the rectangular frame which surrounds the richly decorated canopy and spandrels (Herzfeld, 1910, 53-6, Abb. 18, Taf. IV-V; idem, 1923, Taf. LXXVIII; Creswell, 1940, 356, pl. 121c). Rectangular marb with stucco decoration came to light during the excavations at Meskene, ancient Blis [see maskana ]. One of these was in the central room of the Great Mosque. Another triple

mirb with a central deeper rectangular recess flanked by shallow openings was found in room no. 1, while a third one was in room no. 2 (Salles, 1939, 221-4, pls. XCIXa-b, Ca). Two stucco mirb s, almost identical in shape and decoration were discovered at Palmyra. The shallow semicircular niches are flanked by pairs of pilasters supporting round arches, with shellshaped canopies inside. The spandrels have arabesque decoration and the panels are surrounded by floriated Kfic inscriptions. Marble coating for marb was first introduced in Syria, which was always rich in this material. As one of the possible prototypes and earliest examples for these polychrome marble-lined marb , the one in the Madrasa 9dba9tiyya in Aleppo, made of polychrome stones, should be considered. An interesting innovation can be observed here: the upper part of the mirb , namely the rectangular frame surrounding the spandrels, is much wider than the lower part (Herzfeld, 1942, fig. 72; Sauvaget, 1931, 79, no. 21; Creswell, 1959, 103). The same form can be observed in the polychrome marble-lined mirb of the Madrasa Sulniyya in Aleppo, dated 620/1223 (Herzfeld, 1921, 144; Creswell, 1959, 102). A slightly earlier and similar example can be found in the Madrasa A9rafiyya (607/1210). Polychrome marble work, however, reached its apogee in the mirb of the Madrasa al-Firdaws, erected in 633/1235 (Pl. II, 4). It is the most developed and the largest of the polychrome marble prayer niches. The spandrels depict skilfully interlaced ornaments, the lines of which also form the frame of the upper part. Above there is a semi-circular panel filled by three coloured interlacing patterns and framed by an inscription. This type of marble-work found its way to Egypt and greatly influenced the decoration of later Ayybid and Mamlk marb in Cairo. An interesting example of polychrome marble-work is a small flat mirb in the courtyard of the Bimristn Nr alDn

in Damascus which was built in 549/1152. It is of white marble, but the arch and the spandrels have polychrome marble decoration. Creswell attributed it to Mamd b. Zank b. Asunur , whose name appears in an inscription on the building and the date of construction. Creswell has also suggested that this was the earliest marble mosaic work (Creswell, 1959, 202). On stylistic grounds, however, Herzfeld claimed that it must have been erected at least a century later, possibly in the late 7th/13th century (Herzfeld, 1942, 10). Egypt. The first concave mirb in the country was built by |urra b. 9ark [q.v.], governor of Egypt, in 92/710-11, in the Mosque of #Amr at Fus . The structure of the main mirb in the Mosque of Ibn ln (265/878-9) is original and so are its four flanking columns and capitals. The polychrome marble of the recess and the wooden lining of the canopy and that on the archivolt are later in date (Creswell, 1940, 348-9, pl. 122; Fattal, 1960, 22-4, pls. 1011). There | are five flat marb in the mosque, two of which are contemporary with the building. They are placed on piers in the fourth arcade of the sanctuary. One of them is badly damaged, but the other one is well preserved (Creswell, 1940, 349, pls. 123a-b; Fattal, 1960, 24-5, pls. 17, 18 and 29). The two panels must have been almost identical. A pair of pilasters on bellshaped bases and topped by identical capitals support a pointed arch, the outline of which can also be observed on the damaged panel. A row of pearl motifs provides the border for both and an inscription runs across on top. A difference can be noted in the decoration of the spandrels and in the spaces below the arches within the pilasters. In both instances the influence of Smarr is obvious, just as it is evident in the overall plan and decoration of the mosque. The main mirb of the Mosque of al-Azhar, built between 359/969-70 and 361/971-2, although several times altered and restored, still retains some of its original decoration. The canopy with its elaborate and deeply-cut floral design, the soffit of its arch covered by finely executed scrollwork, together with the floriated Kfic

inscription on the archivolt, are most probably of the same period as the mosque (Creswell, 1952, 55-6, pls. 4a, 7c; Hill, Golvin and Hillenbrand, 1976, fig. 22). This original stucco work was covered by a wooden lining until 1933 (for a picture of this, see Hautecour and Wiet, 1932, ii, pl. 91). The marble lining of the niche and the flanking columns are much later in date. One of the finest stucco marb in Cairo which survives in its original form, is that of the Mosque of al3uy9 , built in 478/1085 (Pl. III, 5). Its stucco decoration, after that of the Mosque of Ibn ln , and that of the al-Azhar, is the third outstanding example in Egypt. The design here is nevertheless richer and more refined (Creswell, 1952, 157-9, fig. 80, pls. 48c, 116a; Hill, Golvin and Hillenbrand, 80, figs. 31-2). None of the prayer niches built in Egypt in the following two centuries has ever surpassed it. The decoration of the Sayyida #$tika , built during the first quarter of the 6th/12th century, is more restrained but presents some new ideas. The frame, which is an epigraphic band, does not surround the entire niche, but only its stilted and pointed arch; then it turns at right angles and runs all around the interior. Furthermore, in the spandrels there are large fluted paterae in high relief surrounded by pearl motifs. Finally, above there is a geometric band of overlapping ovals (Creswell, 1952, 229-30, pls. 80e, 117b). Somewhat similar but more elaborate patterns appear above the marb of the Mausoleum of Sayyida Ruayya , built in 527/1133 (Pl. III, 6). The horizontal panel over these marb recall the Smarr ornaments (Creswell, 1952, 249, fig. 143, pls. 87b, 119c-d, 120a; Hill, Golvin and Hillenbrand, 82, figs. 37-8). All the recesses in this building have scalloped canopies. Three phases can be observed in the development of these canopies. In the first phase, the slightly projecting frame follows the outline of the scallop. One of the earliest examples of this is the mirb of Umm Kul9m , built in 516/1122 (Creswell, 1952, 239-40, fig. 135, pls. 82b, 118b). The second phase of the development is connected with the muarnas , where the frame of the scallop is combined with muarnas cones, placed in one, two or three lines above the other (e.g. the marb of the Sayyida Ruayya ). The third phase was used in Turkey and will be discussed further below. There are two stucco mirb

s in Cairo which present a special group. They are triple-recessed. The earlier of these two is in the mausoleum of I9wat Ysuf , built during the last | quarter of the 6th/12th century (Creswell, 1952, 235-6, pl. 118a). These recesses here are plain, but their upper parts are surrounded by a continuous band of floriated Kfic which runs all around the interior of the building. The spandrels are filled by carefullyexecuted arabesques which are comparable to earlier stucco work in Cairo and accordingly may indicate an earlier date for this mirb . The decoration of the second triplemirb , in the mausoleum of Muaf Pa9a (middle of the 7th/13th century), is so deeply cut that it looks like openwork. The recesses here have keel-arches which most probably originated in Egypt and were widely used there during the 6th/12th and 7th/13th centuries (Creswell, 1959, 178-80, pls. 57c, 107c). An interesting and somewhat bold experiment can be observed in the mirb of the mausoleum of Amad b. Sulaymn alRif# , erected in 690/1291 (Pl. IV, 7). The ibla wall has pieces of glass embedded in the stucco background and these are painted in green (Creswell, 1959, 220-1, pl. 109c; Lamm, 1927, 36-43). The experiment was not successful and was never attempted again. It has already been pointed out that the polychrome marble-work of Syria had greatly influenced mirb decorations in Egypt from the mid-7th/13th century onwards. One of the earliest examples is the comparatively simple but very large mirb in the mausoleum of Na3m alDn , built in 647-8/1249-50 (Creswell, 1959, 102, pl. 106c). A marble lining for the main mirb of the Mosque of Ibn ln was executed at the order of L3n

[q.v.] in the same year, and that of the Mosque of al-Azhar in 665/1266. The mirb of the Madrasa of |alwn [q.v.] and the almost identical one in his mausoleum, both erected in 684/1285 (Pl. IV, 8), are perhaps the most outstanding examples of polychrome marble work in Egypt. Both niches are of horseshoe shape instead of the conventional concave one. Inside of the mirb in the mausoleum there are four rows of small arcades, each crowned by a shell, while in that of the madrasa there are only two rows. The canopies and the spandrels of these two mirb s are covered with gold mosaic, showing grapes and vine leaves. The arches are also horseshoe-shaped, being made up of white and coloured voussoirs. The influence of the Ma9rib is well demonstrated here in the shape of the niches and the arches and the coloured voussoirs, while the row of small arcades reveals Syrian traditions (Creswell, 1959, 202, pls. 108b, c). Ma9rib . The earliest known surviving mirb in the Ma9rib is in the Mosque of B Fatt at Ssa in Tunisia, erected between 223/838 and 226/841. It is a low plain niche of horseshoe form with an arch of the same shape (Creswell, 1940, 247, pls. 58e, 121a; Hill, Golvin and Hillenbrand, fig. 129). In the Great Mosque of Ssa we find for the first time a mirb in front of which a dome was built (Hill, Golvin and Hillenbrand, 100-1, figs. 140-1). This example was shortly followed in the Great Mosque of |ayrawn , when it was rebuilt by Ab Ibrhm Amad II. He was also responsible for the decoration of the new mirb

. The niche has a horseshoe form flanked by a pair of marble columns crowned by Byzantine-style capitals supporting a horseshoe arch (Creswell, 1940, 308-14, fig. 232, pl. 121b; Marais, 1926, i, 19-22, 68 f., figs. 7, 36; Hill, Golvin and Hillenbrand, 92-3, figs. 96-9). The canopy has wooden panelling which was most probably added later. The walls of the niche are covered with marble openwork. The archivolt and the surface of the mirb are adorned with polychrome and monochrome lustre tiles imported from #Ir in 248/862-3 (Marais, 1928). The date of these tiles and | the introduction of lustre has recently been questioned; nevertheless, it is clear that stylistically these tiles are related to those excavated at Smarr . Another horseshoe-shaped mirb is that in the Great Mosque of Cordova (Pl. V, 9), which was erected and decorated at the order of al0akam II in 354/965 (Marais, 1926, i, 222 f., 264-66, figs. 146, 154; Creswell, 1940, 143). This mirb is remarkable in its size and in its extremely rich decoration of polychrome marble and gold mosaics. Several new decorative features which appeared here for the first time were accepted and applied to later marb in the Ma9rib . The niche itself is very spacious and high, crowned by a complete dome, the earliest such example in a mirb niche. The lower part of the niche has plain marble panels, followed by a cornice with a Kfic inscription. Above there are seven trefoil arches supported by marble columns with gold capitals. These arches are almost identical to those which decorate the upper part of the mirb . Inside the arcade there are floral decorations in Byzantine style. The horseshoe arch rests on the wall and on two pairs of marble columns, one behind the other. The archivolt is decorated with voussoir stones, all with rich floral designs, and in white and in polychrome alternately. Acanthus scrolls with a rosette decorate the spandrels. The cusped arcades inside the niche and on top of the mirb are most probably imitations of the false window-openings found over the library portal in the Great Mosque of |ayrawn (Terrasse, 1932, 110); but as Marais once suggested, they can ultimately be traced back to Syria (Marais, 1926, i, 266, n. 1, fig. 147). The mirb is one of the most beautiful examples in the whole Muslim world. Its form and decorative style has several times been copied, but its fineness and richness has never been surpassed. Certain elements in this

mirb like the horseshoe form of the recess, the cusped arches inside and on top of the mirb , the broad archivolt with voussoir stones and floral decorations became the accepted features of later marb in the Ma9rib . The mirb in the Great Mosque of Tlemcen in Algeria is one which reveals close connexions to that in Cordova. It was built in 530/1135. The decorated and cusped archivolt closely resembles that of Cordova but is more elaborate, although executed only in stucco. The arch and the spandrels are surrounded, as in Cordova, with an epigraphic band of floriated Kfic . The niche itself is pentagonal, a form that was to be frequently used in the Ma9rib and also in Turkey. The niche is flanked by an opening on either side giving access all round the mirb (W. and G. Marais, 1903, 140 f.; Marais, 1926, i, 313 f., figs. 213-4, 381-5; Hill, Golvin and Hillenbrand, 111, figs. 208-9). This mirb in Tlemcen is clearly a deliberate copy of the Cordova mirb , albeit executed in cheaper material. Some architectural and decorative details which were new elements in Cordova appeared here, but in more developed forms, and were used again in later examples. These can be best observed in the mirb of the Kutubiyya Mosque in Marrke9 , built ca. 541/1146 (Marais, 1926, i, 321 f., fig. 179; Hill, Golvin and Hillenbrand, 125: Basset and Terrasse, 1926, 119). The canopy here is decorated with a muarnas , as is the large dome in front of the mirb . The muarnas was a new feature in the architecture of the Ma9rib and was introduced there during the first half of the 6th/12th century. The horseshoe arch of the niche rests on three pairs of engaged columns. The broad and cusped archivolt is decorated with trefoil arcaded compartments. A frame filled with geometrical and star patterns surrounds the arch, | while above there are five blind window-openings with lobed arches. In the Great Mosque of Tinml in the High Atlas, the mirb

, built in 548/1153, closely resembles that of the Kutubiyya. The same arrangements, sc. a muarnas canopy inside the pentagonal niche, a similarly decorated dome in front of the mirb , and a horseshoe arch supported by three pairs of pilasters and a frame not unlike that in the Kutubiyya, can be observed here. Again there are flanking niches and an open path behind, once more presenting a free-standing mirb (Marais, 1926, i, 323, 385, figs. 181, 216-17; Hill, Golvin and Hillenbrand, 128, figs. 472-3). There is another mirb in the Mosque of Ya#b alManr (also known as the Mosque of the |aba ) in Marrke9 , which stylistically is related to this group. This is perhaps the latest example of this type. It was built in 592/1195 (Hill, Golvin and Hillenbrand, fig. 413). In southern Tunisia, at Tozeur, the mirb built in 592/1195 represents an entirely different type. The pentagonal recess is considerably smaller than any of the previous examples, and is crowned by a semidome coated with carefully-carved stucco decoration of floral designs and epigraphic bands. The horseshoe arch has an incomplete double archivolt interrupted by the attached rectangular frame. Once again the floral decoration of the wedge-shaped compartments on the archivolt recalls Cordova (Marais, 1926, i, 385-94, fig. 218; Hill, Golvin and Hillenbrand, 107, figs. 174-6). The unusual form and decoration of this mirb was due to an Almoravid patron and to the presence of Andalusian craftsmen. Iran, Afghanistan and Central Asia. During the last two or three decades, several new monuments, among them mosques and mausolea, have been discovered and excavated in this area. The mosque known as the Tr Khna in Dm9n , built probably during the latter half of the 2nd/8th century, no longer stands alone as an early mosque in Iran. The comparatively deep rectangular recess with an oblique wall at its back for correcting the ibla direction, may be regarded as the original mirb in this mosque. It has no decoration now, but we may presume that once it was coated with stucco (Godard, 1934, 226; Survey, 933-4). The recently-excavated Great Mosque in

Srf (late 2nd/8th or early 3rd/9th century) had a rectangular mirb but only the foundation walls were discovered in situ (Whitehouse, 1970, 2 f., figs. 1-2). The Mas3id -i 3mi# in Fahra3 near Yazd (probably of the 3rd/9th century) has its main mirb in the original place but the walls are of more recent date ( Pirniy , 1349/1970, 2-13; Alfieri, 1977, 65-76, pls. X, XIa, b; Shokoohy, 1978, 67, pl. 106). In the old part of the Mas3id -i 3mi# in 9rz the tilted back wall of the niche, together with the horseshoe arch and the fragmentary stucco decoration on its soffit were regarded as original and dated to the end of the 3rd/9th century (Pope, 1934, 324; idem, Survey, 1264-6, fig. 455, pls. 259A-B; idem, 1965, 80-1, fig. 75). As far as it is known today, the mirb of the Mas3id -i 3mi# of Nyin is the earliest example surviving in its original form (Pl. V, 10). It is a double-recessed rectangular niche, the prototype of which may be found in the 3awsa alKhn , mentioned above. The canopies, frame, columns and capitals are covered with stucco. Unfortunately, the decoration is missing from the lower part of the inner pair of columns and from the back wall of the niche. Stylistically, this mirb was dated to the latter part of the 4th/10th century (Flury, 1921, 230-4, 305-16; Survey, fig. 921, pls. 265B, 267, 269A-D, 511B-C; Pope, 1965, 84, fig. 76). Details of the stucco patterns | can be traced back to Smarr , but here they appear in a more developed form. Later marb were double- or triple-recessed, e.g. like that in Ma9had

-i Miriyn [q.v.] in Soviet Central Asia, dated to the end of the 5th/11th century (Kotov, 1939, 108, pls. XLV-XLVIII; Survey, 2721-5, figs. 922-4; Pugaenkova , 1958, 169-74). The decorative patterns are similar to those of Nyin . Stucco marb have also been discovered in Af9nistn . One of them was excavated in the Audience Hall in the Great Palace at La9kar Bzr [q.v.] (Schlumberger and Sourdel-Thomine, 1978, i/B, 39 f., pls. 73d, 147-9), and it was attributed to the 9aznavid period. Two small mausolea at Sar-i Pul in northern Afghanistan had fine stucco mirb s with inscriptions. These inscriptions were published by Bivar, who suggested that one of these two mausolea, namely the Imm -i 9urd and its mirb (Pl. VI, 11) cannot be earlier than 450/1058-9, while the second one in the Ziyrat of Imm -i Kaln most likely dates from the 6th/12th century (Bivar, 1966, 57-63). Unfortunately, both mirb s have disappeared during the last few years (Shokoohy, 1978, 110-11, figs. 88-94, pls. 193-4, colour pl. 47). The mirb of the Mas3id -i 3mi# of Zawra , dated 551/1156-7 (Godard, 1936, fig. 199) and those in the Mas3id -i 3mi# in Ardistn

(553-5/1158-60); Godard, 1936, fig. 195; Survey, pls. 322-4) may be regarded as combinations of flat and multi-recessed marb . The mirb of the Immzda Karrr at Buzn (now in the Archaeological Museum in Tehran, acc. no. 3268) should also be mentioned. The niche is very deep like an wn , and it is covered by a vault instead of a semi-dome. A floriated Kfic inscription in the frame gives the date of construction as 528/1143. This inscription is very interesting since the hastae of the letters end in human heads. The stucco work here is richer and more refined than in any of the previous examples, showing the wide scope of this technique (Pope, 1934, 114; Smith, 1935, 65-81; Survey, pls. 331A-C, 312A). The two most outstanding stucco mirb s in Iran are those in the Madrasa 0aydariyya in |azwn and in the Gunbad-i #Alawiyn in Hamadn . Both are remarkable not only because of their enormous size, but mainly because of the extreme refinement of the stucco decoration. The designs appear as if superimposed in two or three layers. In |azwn , the lower part of the mirb is missing, but the remaining upper half indicates that stucco work may very well have been at its finest here (Godard, 1936, 200, fig. 136; Survey, pls. 313A, 316A-C, 512D). The mirb in the Gunbad-i #Alawiyn in Hamadn perhaps does not surpass that of |azwn , but certainly comes close to it (Herzfeld, 1922, 86-99; Survey, pls. 330, 331A-B; Wilber, 1959, 151-2, fig. 116, cat. no. 55). Herzfeld and Wilber attributed this latter mirb to the Il-

9nid period, while Pope dated it to the end of the 6th/12th century (Survey, 1301; idem, 1965, fig. 186). It seems most likely that both mirb s were erected about the same time, before the Mongol invasion of Iran in 617/1220. With the coming of the Mongols, the style slowly changed, and that change is already apparent in the stucco mirb of the Immzda Abu 'lFa'l waYay in Maallat Bl , dated to 700/1300 (Wilber, 137, pl. 67, cat. no. 44), and in two others, one of them in an wn outside 9rz (Fehrvri, 1969, 3-11; idem, 1972, fig. 8) and the other one in the Prs Museum (Fehrvri, 1972, fig. 7). The finest example of Il9nid stucco mirb s are in the Mas3id -i 3mi# in Reza'iye ( Ri'"iyya ) (Wilber, 112-3, pl. 9, cat. no. 16), dated 676/1277 and in the Mas3id -i 3mi# of Ifahn , built at the order of | l3eyt Muammad 9udbanda in 714/1314 (Wilber, 141, pls. 87-8, cat. no. 48; Survey, pls. 396-397A-B; Pope, 1965, fig. 189. In both mirb

s, the stucco work is of superb quality and is executed in openwork. The gradual change in taste and style resulted in the application of two new techniques in the decoration of marb : the usage of mosaic faience and the application of lustre tiles. Mosaic faience originated in Iran, and the earliest examples are to be found in the monuments of 9ursn and Central Asia dating from the 6th/12th and early 7th/13th centuries. The technique was, however, perfected and first applied in the decoration of marb in Turkey (see below). Mosaic faience decoration in Iran was not introduced in mirb s before the 8th/14th century. The earliest known such mirb is in the Immzda 9h 0usayn in Warmn , erected ca. 730/1330 (Wilber, 1955, cat. no. 86, 177-8, pl. 184). Two other outstanding examples are in the Mas3id -i 3mi# of Yazd, dated 777/1375 (Pl. VI, 12; Survey, pl. 443; Pope, 1965, 185, colour pl. VII and fig. 246), and the second one is in the Mas3id -i 3mi# in Kirmn , dated ca. 957/1550 (Survey, pls. 401, 540). Flat marb were also used in Iran. At first, these were of carved stone or alabaster. By the 6th/12th century faience tiles were used, with the decoration in relief painted in cobalt blue or turquoise under the glaze. During the early 7th/13th century lustre painting was introduced on these tiles. Occasionally cobalt blue and turquoise colours were added. There are some ten such large lustre marb known today which were made up of several tiles. All of these were made in |9n . One of the earliest dated large lustre marb is preserved in the shrine of Imm

Ri' in Ma9had , dated Rab# II 612/July 1215 (Donaldson, 1935, 125-7; Bahrami, 1944-5, 37-8, pls. 20-1). These lustre tiles have recently been studied by O. Watson in great detail (see Watson, 1985). A somewhat unusual mirb is the Mas3id -i #Al in |uhrd [q.v.] near |9n (Pl. VII, 13). The recess and the lower part of the ibla wall are covered with monochromeglazed and with some hundred lustre-painted tiles. Although the mirb in its present form was built probably during the last century, some of its lustre tiles bear the date of 700/1300. At the back of the rectangular recess there is a monochromeglazed mirb tile, slightly tilted for correcting the ibla direction, dated 708/1308 (Watson, 1975, 59-74, pls. Ia-c, Vd). Turkey. Early marb in Anatolia were constructed of stone with a comparatively small and shallow recess which could be three-sided, such as the mirb of the Ulu Cami in Erzurum, dated 575/1179 (nal, 1968, 31, fig. 20). Others are pentagonal, like that of the #Al" alDn Mosque in Nide (620/1223), which is a double-recessed mirb with both recesses crowned by muarnas canopies (Gabriel, 1931, i, 120, pl. XXXVI). Marb with muarnas

canopies became popular in the country, and they were equally used in stone and in faience, and later in the Ottoman period in marble. Scalloped canopies were also used in Turkey, and here we find them in the third phase of their development (for the earlier two phases, see above, under Egypt); they are all double-recessed, the inner recess is crowned by a scallop, while the arch of the outer niche is cusped. One such mirb was placed in the Ulu Cami of Kzltepe (Dunaysir), dated to 601/1204 (Gabriel, 1937; idem, 1940, i, 51, ii, pls. XXXI/1-2, XXXII/1-3). The second such mirb is in the Madrasa Mas#diyya in Diyarbakir, which is similar to the previous example (Gabriel, 1940, i, 197), | ii, pl. LXXIII/3; Bakirer, Kat. no. 20, 143-4, res. 55-7, ek . 20). The second decorative technique applied for marb in Turkey was mosaic faience. There are more than twenty such marb known in the country (for a complete list of these, see Meinecke, 1976, and for tile work in Turkey, ney, 1976). The tiles were coloured in cobalt blue, turquoise, black, aubergine and white. They form epigraphic and geometric bands round the niches and cover the walls, the muarnas canopy and the spandrels. The most beautiful of these faience marb are in Konya, which was the capital of the Sal3s of Rm . One such mirb was installed in the Sadrettin Konevi, dated 673/1274-5 (Pl. VII, 14; Meinecke, 1976, cat. no. 85, i, 64, ii, 352-5, Taf. 36/4; Bakirer, Kat. no. 49, 182-3, res. 111-13, ek . 49). Towards the end of the 7th/13th century, tiles were used together with stucco and terracotta, as is the case in the mirb of the Arslanhane Cami in Ankara, built in 689/1290 (Otto-Dorn, 1957; Akurgal, Mango and Ettinghausen, 1966, 149, colour pl. on p. 132; Aslanapa, 1971, 121; Meinecke, cat. no. 18, i, 41-3, ii, 66-74, Taf. 8/3; Bakirer, Kat. no. 58, 196-8, res. 132-4, ek . 58). In the 9th/15th century, faience mosaic is replaced by the cuerda seca technique, and perhaps one of the most beautiful marb with such decoration is that of the Green Mosque in Bursa, dated 824/1421 (ney, 1976, 62; Goodwin, 1971, 66, fig. 60; Aslanapa, 1971, fig. 214). Simultaneously with

the cuerda seca, underglaze-painted tiles were also introduced during the 9th/15th century. At first, these were painted in blue and white and later in polychrome. India. The earliest known and reported mirb in India which survives is in the so-called Arhai-din-ka-Jhompra Mosque at A3mr , completed in 596/1199-1200. It reveals strong Hindu and Buddhist elements in its decoration and in its arch, which is cusped and is carved out of a single block of marble (Nath, 1978, 17, pl. XXI). Three marb in the tomb of 9ams alDn Iltutmi9 (died in 633/1235) are more in the traditional Islamic styles and in a way serve as models for later Indian prayer niches. The central mirb here (Pl. VIII, 15) is a combination of a rectangular and a flat mirb . The frame is filled by an epigraphic band. The pair of polygonal and richly decorated columns and capitals support the cusped arch. The canopy of the recess, just like that of the mirb panel at the back, is filled with plaited Kfic . In the centre of the back panel there is a rosette in relief. Such patterns can be observed in the centres of several Indian marb . It is possible that the architects tried to imitate, either consciously or not, the black meteorite stone which is in the centre of the mirb of Sulaymn in Jerusalem. One of the characteristics of Indo-Muslim architecture was to place three or more marb in the ibla wall, as it had already been done in Iltutmi9 's tomb, where this principle was applied perhaps for the first time. The lotus flower, a hanging mosque-lamp or a vase from which a scroll emerges, become permanent decorative features of the back panels of later prayer niches. Such decorations appear in the five marb of the Royal Mosque, the |ila

-yi Kuhna in Dihl , which was built by 9r 9h Sr in about 949/1542. The marb of the |ila -yi Kuhna are enormous in size. They are multi-recessed, with a cusped arch over the outer recesses, and a mirb panel at the back of each one decorated with a hanging mosque-lamp (Brown, 1964, 93, pl. LXII/2). By the late 9th/15th and early 10th/16th centuries, Persian influence becomes stronger and more apparent in Indo-Muslim architec- | ture. It was perhaps most obvious in the decoration of prayer niches. Most of the Indian elements were omitted and were replaced by Persian motifs. The cusped arch, however, remains. A good example for this strong Persian influence is the mirb of the 3mi# Mas3id in Fatpur Sikr which is decorated with polychrome inlaid stonework and which reminds us of contemporary -afawid faience-tiled marb (Brown, pl. LXXII/1). Later prayer niches in $gra and Dihl are built of marble and decorated in polychrome in the pietra dura technique. Wooden and portable marb . Marb either as large niches or small portable wooden panels appeared during the Fimid period in Egypt. They were found in excavations in Fus , but several others are known from later periods and are preserved in the Museum of Islamic Art in Cairo (Weill, 1936, pl. X). Large wooden mirb

s were also introduced and these were popular all over the Islamic world. One of the earliest of these large wooden marb is from the Mas3id -i Maydn in Abyna in Iran, dated 497/1103 (Ettinghausen, 1952, 77, pl. XII). An interesting wooden mirb was discovered in Af9nistn in the upper Logar Valley in the village of 1ar9 , in the Mosque of 9h Muy alDn . It has an overall geometrical decoration and an inscription in Kfic (Pl. VIII, 16). It is dated to the 6th/12th century (Bombaci, 1959, figs. 13-4; Rogers, 1973, 249, no. 82). The wooden mirb of the Madrasa Halawiyya in Aleppo (dated 643/1245) is considerable in size, measuring 350 cm. in height and well over 100 cm. in width. It is richly decorated with geometrical patterns and is inlaid with ivory (Guyer, 1914, 217-31; Sauvaget, 1931, no. 15).

UBA

(A.), sermon, address by the 9ab [q.v.]. The 9uba has a fixed place in Islamic ritual, viz. in the Friday-service, in the celebration of the two festivals, in services held at particular occasions such as an eclipse or excessive drought. On the Friday it precedes the alt , in all the other services the alt comes first. A short description of the rules for the 9uba according to al9rz ( Tanbh , ed. Juynboll, 40), one of the early

9fi# doctors [q.v.], may be given here. (a.) One of the conditions for the validity of the Friday service is that it must be preceded by two sermons. The conditions for the validity of these sermons are the following: the 9ab must be in a state of ritual purity; his dress must be in accord with the prescriptions; he must pronounce the two 9uba s standing and sit down between them; the number of auditors required for a valid 3um#a must be present [see alat ]. Regarding the sermon itself, there are obligatory: the amdala , the alt on the Prophet, admonitions to piety in both 9uba s, prayer ( du#" ) on behalf of the faithful, and recitation of a part of the |ur"n in the first 9uba or, according to some doctors, in both. It is commendable ( sunna ) for the 9ab to be on a pulpit or an elevated place; to salute the audience when directing himself towards them; to sit down till the a9n is pronounced by the mu"a99in ; to lean on a bow, a sword or a staff; to direct himself straightway to his audience; to pray ( du#" ) on behalf of the Muslims; and to make his 9uba short. (b.) Regarding the 9uba s on the

days of festival the same author says (42) that they are like those of the Friday-service, except in the following points: the 9ab must open the first with nine takbr s, the second with seven. On the #d he must instruct his audience in the rules for the zakt alfir , on the #d ala' in the rules for the sacrifice of this day. It is allowable for him to pronounce the sermon sitting. Regarding the 9uba s of the service during an eclipse, al9rz (43) remarks that the preacher must admonish his audience to be afraid, and in the service in times of drought he must ask Allh 's par- | don, in the opening of the first 9uba nine times, in the second seven times; further, he must repeat several times the alt on Muammad as well as isti9fr , recite verse 9 of Sra LXVI, elevate his hands and say Muammad 's du#" (which is communicated by al9rz in full). Further, he must turn towards the ibla [q.v.] in the middle of the second 9uba

and change his shirt, putting the right side to the left, the left to the right, the upper part beneath and keep it on till he puts off all his other garments. These prescriptions give rise to the following remarks. C. H. Becker was the first to point to the relation between the Islamic pulpit and the judge's seat in early Arabia. This explains why the 9aib must sit down between the two 9uba s; it explains why he must lean on a staff, sword or bow; for these were the attributes of the old Arabian judge. It is not easy to see why the 9uba precedes the services on Friday, whereas on the days of festival and the other special occasions alt comes first. 0ad9 tells us that Marwn b. al0akam was the first to change this order by pronouncing the 9uba before the performance of the alt on the days of festival (e.g. alBu9r , #^dayn , bb 6 and especially the pathetic picture in Muslim, #^dayn , trad. 9). It is also said that Marwn was the first to hold the 9uba on these days on a pulpit, the old custom being a service without minbr or a9n . According to other authorities (cf. Muslim, ^mn , tr. 78, 79, and al-

Nawaw 's commentary) the 9uba before the alt was an institution going back to #U9mn or even to #Umar . The common opinion of traditionists is, however, that it was an innovation due to the general tendency of the Umayyads to favour their own dynastic interests rather than those of religion. If this opinion should be right, the innovation as well as the holding of the 9uba in a sitting attitude may be looked upon as an endeavour to go back to the pre-Islamic judicial rites concerning minbar and 9uba . Regarding the prayer on behalf of the faithful ( du#" li 'lmu"minn ) it must be observed that in this prayer before the Fridayalt it has become customary to mention the ruling sovereign. The history of Islam is full of examples of the importance which was attached to this custom, especially in times of political troubles, the name mentioned in this du#" betraying the imm 's political opinion or position. Though it is not prescribed by law to mention the ruler's name, the suppression of the name at this occasion exposed the 9ab to suspicion. In countries where Muslims lived under non-Muslim rule, even a prayer for the worldly prosperity of the ruler could expose the 9ab to suspicion on the part of his fellow-Muslims (cf. Snouck Hurgronje, Islam und Phonograph, 13 f. = Verspr. Geschr., ii, 430 f.; idem, Mr. L. W. C. van den Berg's beoefening van het Mohammedaansche recht, in Ind. Gids, vi/1, 809 f. = Verspr. Geschriften, ii, 214 f.). The custom of mentioning the ruler in prayer is found as early as the 5th century B.C. in the Aramaic papyri of Elephantine (Pap. i, line 26; cf. also Harnack, Mission und Ausbreitung des Christentums, i, 286. Several of the characteristics of the

9uba prescribed by the doctors of the law occur also in ad9 . The 9uba s of Muammad usually begin with the formula amm ba#du (alBu9r , 3um#a , bb 29). Side by side with the amdala (Muslim, 3um#a , tr. 44, 45) the 9ahda occurs ( Amad b. 0anbal , ii, 302, 343: A 9uba without the 9ahda is like a mutilated hand. In a large number of traditions it | is stated that Muammad used to recite passages from the |ur"n (e.g. Muslim 3um#a , tr. 49-52; Amad b. 0anbal , v. 86, 88, 93 etc.). The 9uba must be short, in accord with Muammad 's saying: Make your alt long and your 9uba short (Muslim,

3um#a , tr. 47). Just like the alt the 9uba must be right to the purpose ( ad an , Muslim, 3um#a tr. 41). The audience must be silent and quiet; who says to his neighbour 'listen', has spoken a superfluous word, alBu9r , 3um#a , bb 36). The two 9uba s pronounced by the standing 9ab , who sits between them, are based on Muammad 's example (alBu9r , 3um#a , bb 27; Muslim, 3um#a , tr. 33-5; Amad b. 0anbal , ii, 35, 91, 98). During the a9n Muammad used to sit on the minbar ; the ima was spoken when he had descended (in order to hold the 9uba standing); this order was observed by Ab Bakr and #Umar ( Amad b. 0anbal

, iii, 449 bis). Neither the term 9uba nor the verb 9aaba in their technical meaning occur in the |ur"n . Even in the passage containing an admonition not to abandon the Friday-service for worldly profit, it is only the alt which is mentioned ( Sra LXII, 9-11). It would be wrong to conclude from this silence that the 9uba did not yet form a constituent part of worship in Muammad 's time. Still, it is not probable that the different kinds of service were accurately regulated from the beginning. 0ad9 has preserved descriptions showing that Muammad 's 9uba often did not have much to do with the regular sermon of later times. Ab Dwd , Kitb aldiyt , bb 13, reports, for example, that the Prophet pronounced two 9uba s at the end of a complaint raised against a collector of the zakt . Still, it is not possible to distinguish between the kinds, as may appear from the following traditions. According to one of them related on the authority of Ab Sa#d al9udr it is said that Muammad on the days of festival used to open the service with the alt

; then he pronounced the 9uba and his 9uba usually consisted in the command to participate in some mission or expedition ( Amad b. 0anbal , iii, 56 f.). A similar statement is to be found in Muslim, #^dayn , tr. 9: When Muammad had concluded the alt on the days of festival by the taslm , he remained on his feet and turned to the sitting audience; when he wanted to send a mission or when he desired some other arrangement, he gave his orders on it; he used also to say: give alms, give alms; ... then he went away. This state of things lasted till Marwn , etc.. This is a very simple description of the service and would be a considerable support to the view that a service with a fixed order only arose long after Muammad 's time. Yet it must not be forgotten that the description just translated betrays the tendency to contrast the simple service of the Prophet with the highly official style introduced by Marwn , who even had a minbar built on the muall . According to another tradition, the Prophet once interrupted his 9uba in order to reply to a stranger who had asked for instruction in the Muslim faith (Muslim, 3um#a , tr. 60); he is also portrayed as interrupting the 9uba to call out directly to a man (ibid., tr. 54-9). However uncertain the value of these traditions may be, it seems not out of place to suppose that a fixed order of service on Friday and the days of festival arose only after Muammad 's lifetime. This order reposes on three elements: the early-Arabian 9uba ,

Muammad 's sunna and the example of Jews and Christians. In his study on the history of Muslim worship | C.H. Becker endeavoured to establish a close connection between the services on Friday and the days of festival on the one hand, and the mass on the other. But this view was opposed by Mittwoch, who found in the Jewish liturgy features corresponding to a9n and ima , to the amdala , the recitation of the Tora (first 9uba ) and the recitation from the Prophets (second 9uba ). It is perhaps impossible to decide the question; probably the example of the Jewish as well as that of the Christian liturgy exercised influence on the final constitution of the Muslim worship. It is customary to pronounce the 9uba in Arabic; nevertheless, this rule is not infrequently broken in nonArabic speaking lands. The history of the 9uba in Islam remains to be written, and the study of oratory from the minbar or pulpit likewise remains to be undertaken. On the latter point, the enquirer might utilise with profit the texts (of varying degrees of authenticity) of those sermons of the Prophet given in the Sra , in the ad9 collections and in historical texts, as well as in those adab works which have preserved specimens of famous 9uba s. Collections like those of Ibn Nubta alFri [q.v.] and the specially-compiled anthologies of sermons used by the professional 9ab

s, just as the secretaries used collections of model letters, would also be found useful. Collections of this latter type are often arranged according to the calendar, i.e. there are four sermons for each month plus supplementary ones for festival days, the Prophet's birthday and his mi#r3 or night-journey (see Ahlwardt, Verzeichnis der arab. Hss., iii, 437). MINBAR (A.), the raised structure or pulpit from which solemn announcements to the Muslim community were made and from which sermons were preached:

1. Early historical evolution and place in the Islamic cult.

In contrast to the mirb [q.v.], the minbar was introduced in the time of the Prophet himself. The word, often pronounced mimbar (cf. Brockelmann, Grundriss, i, 161), comes from the root n-b-r high; it could be derived from the Arabic quite easily with the meaning elevation, stand, but is more probably a loanword from the Ethiopic (Schwally, in ZDMG , lii [1898], 146-8; Nldeke, Neue Beitrge z. sem. Sprachw., Strassburg 1910, 49). Its case is therefore somewhat similar to that of mas3id . It means seat, chair (e.g. Wstenfeld, Chron. Mekka, ii, 8; A9n 2 , xiv, 75) and is used, for example, for saddle (alabar , Gloss.) and of a litter ( A9n , xiii, 158; cf. Schwally). It is therefore identical with ma3lis (alBu9r , 3um#a , bb 23), with sarr (al-Mubarrad, Kmil , 20; A9n , iii, 3), ta9t or kurs

( Ibn alA9r , Usd al9aba , i, 214; cf. also Becker, Kanzel, 8). The use of the word for the pulpit is in keeping with its history. When the 9ab [q.v.] spoke among the Arabs, he usually did so standing (cf. Mufa''aliyyt , ed. Lyall, xci23; al3i , Bayn , Cairo 1332, i, 129, ii, 143) frequently beating the ground with bow and lance (ibid., i, 198; Labd , 7, 15, 9, 45); or he sat on his mount as did e.g. |uss b. S#ida ( Bayn , i, 25, 31, ii, 141). The Prophet did both of these things. In #Arafa he sat on his camel during his 9uba and on other occasions, when addressing the community during the early period, even as late as the day of the capture of Mecca, he stood (cf. |ur"n , LXII, 11). The people sat on the ground around him (alBu9r , 3um#a , bb 28; #^dayn , bb 6). In the mosque in Medina, he had a particular place, as is mentioned in the stories | of the introduction of the minbar . Sometimes, we are told, he stood beside a tree or a palm-tree (alBu9r , Manib ,

bb 25; ed. Krehl, ii, 400); as a rule however, beside a palm-trunk ( 3i9# , so Ibn Sa#d , i/1, 9, 10, 11, 12) and on a few occasions beside one of the pillars (alBu9ar , Manib , bb 25, ed. Krehl, ii, 401; alDiyrbakr , 9ams , ii, 75). This is undoubtedly the original tradition: the Prophet stood beside one of the palm-tree trunks used as pillars in the mosque. For beside (usually ma il ; alBu9r , Buy# , bb 32: #inda ) up against ( ma #al ; already in alBu9r , 3um#a , bb 26) is sometimes found later and for the column or trunk, we find a stump on which he sat. Various passages record how the minbar was introduced, notably the following: Ibn Sa#d , i/1, 9-12; alBu9hr , -alt ,

bb 18, 64, 91; 3um#a , bb 26; Buy# , bb 32; Hiba , bb 3; Manib , bb 25; Muslim, Mas3id , tr. 10; see also Wensinck, Handbook, s.v. Pulpit; Usd al9ba , i, 43 below, 214; Wstenfeld, Medina , 62-3; Ibn Baa , i, 275-6; the whole material is in alDiyrbakr , 9ams , i, 129, ii, 75-6, and Srat alalab , ii, 146-7. The details are variously given. The minbar , we are told, was built of arf wood or tamarisk from the woods near Medina; the builder was a Byzantine or a Copt and was called Bm or Bl , but the names Ibrhm (Usd, i, 43), Maymn , -ab , Kulb and Min are also given. He was a carpenter, but a slave of the wife of one of the Anr

or (alBu9r , Hiba , bb 3) of the Muh3irn . Others say he belonged to al#Abbs . The suggestion is sometimes credited to the Prophet and sometimes to others. The palm-trunk is said to have whined like a camel or a child when the Prophet mounted his new seat, but was calmed by stroking and kind words from the Prophet. Most stories take it for granted that the minbar was primarily intended for the 9uba ; in some it is added that the object was to enable the large assembly to hear him ( Ibn Sa#d , i/1, 10, 11). We are told also that the Prophet performed the alt on it and, during the su3d , he came down from it. He also took care that the people could see his alt and follow him (alBu9r , -alt , bb 18; 3um#a , bb 26). This last tradition however presupposes the later custom of standing upon the minbar (note that the same idea of the palm-stump occurs in 3um#a , bb 26). In this connection, it is interesting to note a tradition in Ibn alA9r according to which the Companions asked the Prophet to take up a raised position, as many wufd were coming (Usd al9ba

, i, 43). Another tradition is in keeping with this, according to which the Prophet, when he was visited by a man named Tamm , stood on a kurs and addressed him from it (ibid., 214; cf. Lammens, Mo#wia , 204, n. 5). Here we have a seat of honour on which the ruler sits. This is undoubtedly in keeping with the character of the minbar ; while the raised seat was in general use among the northern Semites, the Arabs usually sat on the ground, often leaning against a saddle. The raised seat was the special mark of the ruler or, what is the same thing, of the judge. We are told that Rab#a b. Mu99in was the first to sit on a minbar or sarr when acting as judge ( A9n 2 , iii, 3; alMarz , iv, 6-7). Al0a333 , for example, when he addressed the people (hardly in the mosque) sat on a chair which belonged to him ( kurs lahu: alabar , ii, 959) and when he tried and condemned his enemies, a sarr was erected for him (ibid., 1119); in the same way a kurs was placed for Yazd b. al-Muhallab when he issued his orders for a battle (ibid., ii, 1107; see also Becker, Kanzel, 8). If tradition usually suggests that the minbar was introduced exclusively for the 9uba , this seems to be | a somewhat one-sided view. The minbar was primarily, as Becker was the first to point out, the throne of the mighty Prophet in his capacity as a ruler. In keeping with this is the tradition that it was introduced in the year 7, 8 or 9 (alabar , i, 1591; alDiyrbakr ,

9ams , ii, 75; Usd al9ba , i, 23). The Prophet used it for the publication of important announcements, for example, the prohibition of wine. That he should also make his public speeches to the community from the new seat was only natural. His 9uba s, however, were not confined to the Friday worship, and he could still deliver a 9uba without a minbar , e.g. at the festival on the muall [q.v.], where Marwn was the first to put up a minbar (alBu9r , #^dayn , bb 6), and beside the Ka#ba after the capture of Mecca (Ibn Hi9m , 823). The Prophet's minbar is often called a#wd from its material (alBu9r , -alt , bb 64; 3um#a , bb 26). It consisted of two steps and a seat ( ma3lis : alDiyrbakr , 9ams , ii, 75; alBu9r ,

3um#a , bb 23; ma#ad : alabar , i, 1591). After the time of the Prophet, it was used in the same way by Ab Bakr, #Umar and #U9mn (see below). Its significance as a throne is seen from the fact that in the year 50, Mu#wiya wanted to take it to Syria with him; he was not allowed to do so but he raised it by 6 steps. At a later date, #Abd al-Malik and alWald are said to have wanted to take the Prophet's minbar to Damascus (alabar , ii, 92-3; 9ams , ii, 75; Ya#b , Ta"r9 , ii, 283; Ibn alFah , 23-4; Wstenfeld, Medina , 63). In the time of the Prophet, it stood against the wall so that a sheep could just get past (alBu9r , -alt , 91). In the time of alMuaddas , in the centre of the Mu9a there was pointed out the position of the old minbar , above which Mu#wiya was said to have built his new one (82; cf. Ibn 0awal , 26, and al|azwn , ed. Wstenfeld, ii, 71). According to some ad9

s, it was over the aw' of the Prophet (alBu9r , -alt f Makka , bab 5; Fa'"il alMadna , bb 5, 12 and passim). At a later date, new minbars were erected in the mosque (see Wstenfeld, Medina , 64, 96). That the Umayyads should have a minbar of their own was natural; they sat on it, just as their predecessors had done (cf. Goldziher, Muh. Stud., ii, 42). Mu#wiya took it with him on his journey to Mecca (Chron. Mekka, i, 333); he also had taken it to the festivals on the muall (alYa#b , Ta"r9 , ii, 265), just as Marwn used to do in Medina (see above); it was therefore still portable and indispensable for the sovereign when he wished to make a public appearance as such. In Ibn 3ubayr 's time, the minbar al9uba in Damascus was in the central mara ( Rila , 265). According to Ibn 9aldn , Mu#wiya was the first in Islam to use the throne ( sarr , minbar, ta9t ,

kurs ) but he is clearly not referring to the minbar of the mosque ( Muaddima , Cairo 1322, 205-6, fal 3, 37). The minbar taken to Mecca by Mu#wiya remained there till the time of alRa9d; when the latter visited Mecca on his Pilgrimage in the year 170/786-7 or 174/790-1 a minbar man9 with nine steps was presented to him by the amr of Egypt and the old one was put up in #Arafa . At a later date, alW9i made minbars for Mecca, #Arafa and Min (Chron. Mekka, i, 333, iii, 114). The Meccan minbar was a portable one. It usually stood beside the mam but was put beside the Ka#ba during the 9uba (Ibn 3ubayr , 95, 97; cf. Chron. Mekka, iii, 429). According to alBatann , this custom was kept up until Sultan Slaymn |nn (926-74/1520-66) built a marble minbar , north of the mam ( alRila al0i3ziyya , 100). It seems at first to have been doubtful whether | manbir

should be put up in the provinces or not. According to al|u'# , #Amr had a minbar made in alFus but #Umar ordered him to take it away: he was not to raise himself above the Muslims so that they would have to sit below his heels (alMarz , iv, 6-7; Ibn Ta9rbird , i, 76; alSuy , 0usn almu'ara , i, 63, ii, 135). The idea obviously was that the throne belonged to the caliph alone. After #Umar 's death, however, #Amr is said to have used a minbar (alMarz , iv, 8, 27). It stood there till |urra b. 9ark [q.v.] rebuilt the mosque. During the rebuilding, it was put in the |aysriyya , which was used as a mosque; only when the mosque was completed in the year 92/711 did |urra put up a new minbar . Tradition, however, is uncertain. The minbar removed by |urra perhaps dated from the time of #Abd al#Azz b. Marwn , who had taken it from a church or had been presented with it by the Nubian king (alMarz , iv, 8; Ibn Ta9rbird

, i, 78). |urra 's minbar remained till 379/989, when the Fimid vizier Ya#b b. Killis replaced it by a gilded one. A large new minbar was placed in the mosque of #Amr in 405/1014-15 by al0kim (alMarz , iv, 8; Ibn Ta9rbird , i, 78-9). We hear of no objections in other places to the manbir in the amr . In Mad"in as early as the year 16/637, Sa#d b. Ab Waa erected a minbar in the mosque improvised in the ^wn of Kisr (alabar , i, 2451, 9). In Bara , Ab Ms put up a minbar in the middle of the mosque. This was, however, found inconvenient because the imm had to cross from the minbar to the ibla over the necks of the (seated) believers. Ziyd

then placed the minbar against the south wall ( Yt , i, 642). On the other hand, we are told that #Abd Allh b. #Abbs (governor of Bara 36-40/656-60) was the first to mount the minbar in Bara (al3i , Bayn , i, 179). When Ziyd had to fly from Bara , he saved the minbar which he put up in his Mas3id al0uddn (alabar , i, 3414-15). The minbar was the symbol of the ruler, and the governor sat upon it as representative of the ruler. It therefore formed a feature of the Mas3id al3am#a , where the community was officially addressed. In the year 64/683-4, therefore, there were minbars in all the provinces. In this year, homage was paid to Marwn b. al0akam not only in the capital but in the other manbir in the 0i3z , Mir , 9a"m , 3azra , #Ir

, 9ursn , and other amr (alMus#d , Tanbh , 307). Special mention is made of the fact that abariyya had no minbar . In the 1st century and beginning of the 2nd one, we find the wl in the smaller towns delivering the 9uba standing, with the staff only. But in 132/749-50 the governor #Abd al-Malik b. Marwn had manbir put up in the ur of Egypt (alMarz , iv, 8, 17 ff.; Ibn Ta9rbird , i, 350-1). When the 9uba became purely a religious exhortation and the ruler was no longer the 9ab , the minbar became the pulpit of the spiritual preacher, and every mosque in which the Friday service was celebrated was given a minbar . At the same time, i.e. after alRa9d , the change was gradually completed and the preacher spoke, standing on the pulpit. 0ad9 s therefore came into existence, according to which the Prophet used to deliver two 9uba s on Friday, standing just as is done to-day (alBu9r , 3um#a , bb s 27, 30) and

#Umar (ibid., bb 2). The minbar was thus now quite analogous to the Christian pulpit. It is very probable that this latter also influenced its form. We have already noted above, regarding a minbar in the mosque of #Amr , that it was said to be of Christian origin. The same thing | came to be said of the Prophet's minbar (Wstenfeld, Medina , 63). Mu#wiya made the Medina minbar larger, while the one brought by him to Mecca had only three steps and was of course portable. We again hear of portable minbars later, which did not exclude their being large (cf. above, on the minbar of Mecca). Thus the manbir in alMa9rib are said to have been portable. Ibn al033 regards this (the oldest) custom as bid#a and therefore ascribes it to al0a333 ( Mad9al , ii, 47, 13 ff.). The oldest minbars were all of wood. There is, however, one ad9 which says that the Prophet had a kurs of wood with iron legs made for the reception of Tamm (Usd, i, 214, 8 from below; cf. Lammens, Mo#wia , 273, n. 3); it is however uncertain what relation this had to the minbar . A minbar of iron was made as early as the Umayyad period ( Ibn Ta9rbird , i, 78, 8: al-minbar aladd , probably correct in spite of Becker, Kanzel, 10, n.; cf. 79, 4, and see below); and also of stone (Goldziher, Muh. Stud., ii, 42, n. 5, with a reference to Ibn 0a3ar ); later, they were also built of brick (Wstenfeld, Medina , 64, 96). As a rule, the minbar stood against the ibla

wall beside the mirb . AlMahd had tried to reduce the manbir to their original small size (alabar , iii, 486, 12; alMarz , iv, 12, 13 ff.), but he could not arrest the development. In the larger mosques several manbir were even built. Ibn alFah , in about 300/912-13, already mentions five minbars in the mosque in Jerusalem (100, 8 f.). In the Suln 0asan mosque in Cairo, four were planned and three erected, when a minaret fell down in 762/1361 and diverted attention to other work (alMarz , iv, 117, 18 f.). The importance which the minbar already had in the time of the Prophet caused special reverence to be paid to it, and the sanctity of the mosque was concentrated round this and around the mirb . The governor of Kfa , 9lid b. #Abd Allh al|asr (105-20/723-38), received a letter of censure from the caliph because he had prayed for water on the minbar ( Kmil , 20, 15). A false oath taken on or beside the minbar of the Prophet absolutely led to hell ( Ibn Sa#d , i/1, 10, 3 f., 12, 19 f.; Ibn 0anbal

, Musnad , ii, 329; cf. J. Pedersen, Der Eid bei den Semiten, 144, 147). Legends grew up which represented the Prophet seeing into the future from the minbar (alBu9r , 3um#a , bb 29) and being able to follow the battle of Mu"ta [q.v.] from it (cf. alWaid , tr. Wellhausen, 311; Ibn Hi9m , 796) and also telling how his prayers on the minbar were specially efficacious. Just as the Ka#ba was covered ( kas ), so was the same thing done to the minbar . #U9mn is said to have been the first to cover the minbar of the Prophet with a afa ( 9ams , ii, 75, 1 from below). Mu#wiya did the same thing when he had to give up his attempt to abolish it (ibid., 76, 4; alabar , ii, 92, 4). It was not quite the same thing when al0kim rediscovered the already-mentioned iron minbar and covered it with gilded leather because it was covered with dirt (read: a9ar ) i.e. rust ( Ibn Ta9rbird , i, 79, 5 f.). Under the #Abbsids , a new kiswa was sent every year for the minbar of the Prophet from Ba9dd; the sultans later did not renew it so frequently (Wstenfeld, Medina , 64). We find other references to the covering of the minbar on special occasions (Ibn 3ubayr , 149, 16). Ibn al-

033 ( Mad9al , ii, 74) demands that the imm should put a stop to the custom of putting carpets on the minbar . (J. Pedersen)

2. Architectural features: the Arab, Persian and Turkish lands.

As noted in 1. above, the minbar was in early times used as a seat by the ruler or his governor, from which he addressed the Muslims at the Friday worship, consonant with the use of mosques in the Umayyad period as places of political assembly also (see mas3id . I. E. 1, and J. Sauvaget, La Mosque Omeyyade de Mdine, Paris 1947, 134-5, 142-4). According to C.H. Becker, the change in the purpose of the minbar from the ruler's or governor's seat to the purely religious pulpit occurred towards the end of the Umayyad period (Die Kanzel im Kultus des alten Islam , in Orientalische Studien Th. Nldeke ... gewidmet, Giessen 1906, i, 344-7). Unfortunately, we do not have any examples or even descriptions of how minbars looked during the Umayyad period. Evidently it took some time before minbars were generally in use. In 132/749-50 provincial cities in Egypt were provided with minbars by order of Marwn II, and we may therefore presume that they became standard mosque furniture in other parts of the Islamic world as well. Little is known of minbars during the #Abbsid period. It is reported that the caliph alMahd ordered Muammad b. Ab 3a#far alManr in 161/777-8 to reduce the height of minbars to make them the same size as that of the Prophet (alMarz ,

9ia , Bl 1853, ii, 247). This incident would suggest that minbars at that time were high, a possibility borne out by the fact that the Great Mosque of Smarr had, according to its ibla wall plan, a minbar which, on architectural evidence, was about 3.90 m high (J. Schacht, An unknown type of Minbar and its historical significance, in Ars Orientalis, ii [1957], 156). The only surviving minbar from the early period of Islam is in the Great Mosque of |ayrawn in Tunisia. Made of teak and measuring 3.31 m with eleven steps, it is a magnificent example of carved woodwork. It is said to have been brought from Ba9dd by the A9labid amr Ab Ibrhm Amad (242-249/856-63), and was probably completed in 248/862-3 (K.A.C. Creswell, Early Muslim architecture , Oxford 1940, ii, 314, 317-19, pls. 89, 90). It is the earliest extant example to have the basic elements of a wooden minbar , that is, a platform with steps and a portal without a door at the entrance to the steps. The framework consists of upright and transverse strips of wood with rectangular and triangular panels fitted in by the tongue-and-groove technique. The framework is decorated with vine tendrils forming circular loops enclosing a vine leaf and bunch of grapes, a composition found in the tie beams of the Dome of the Rock in Jerusalem. Most of the panels on the minbar are geometric grilles, but on the eastern side there are ten very beautiful panels carved in arabesque. The naturalistic style of the design on these ten panels and, in particular, the pine cones encircled by vines, recall wooden panels found near Ba9dd . In Creswell's view, the resemblance strongly suggests that the ten panels were carved there. E. Khnel has pointed out that their ornamentation resembles that of the Umayyad palace at Mu9att [q.v.] (Die Islamische Kunst, in A. Springer, Handbuch der Kunstgeschichte, Leipzig 1929, vi, 385). E. Diez remarked on the dissimilarity of the style of the naturalistic panels to the more abstract style of the early

#Abbsid period typified in the stucco and wood decoration in Smarr , and believed that some of the | carved strips and panels may have belonged to an Umayyad minbar before being assembled in the present structure (EI 1, MINBAR , at iii, 500). The minbar has been subjected to damage and restoration and, particularly, it must have been restored after |ayrawn was sacked by the Fimid caliph alMustanir in 441/1049-50 (H. Saladin, La Mosque de Sidi Okba Kairouan, Paris 1899, 8, 104). According to Creswell, in the repairs of 1907 the panels were replaced in a new order. The rectangular panels with geometric designs, which are of varying quality, are difficult to date, and whilst some are more recent, others appear to have been made at an early period (L. Golvin, Essai sur l'architecture religieuse musulmane, Paris 1970, 228). In early Islamic times, some minbars were movable, which would at once indicate that they were made of very light material, probably wood. Judging from the form and size of the ibla wall in the Great Mosque of Smarr , it must have had a movable minbar which was kept in a special room close to the mirb (Schacht, op. cit., 156). The minbar of the Ka#ba in Mecca was on wheels and was normally kept in the mam Ibrhm [q.v.], but was pushed out to stand beside the Ka#ba for the Friday sermon (Ibn 3ubayr , Rila 2 , ed. M.J. de Goeje, Leiden and London 1907, 95, 97). This was presumably the minbar donated by the #Abbsid caliph alW9i (227-32/841-7) (Schacht, op. cit., 157). The practice of moving minbars in and out of the assembly area has actually survived in

certain parts of the Islamic world, mainly in North Africa. Few early ones remain, but the existence of a recess to the right of the ibla wall of some Friday Mosques proves that the original minbars of these mosques were movable. The series of movable minbars begins with that of the Great Mosque of Sfax built in 235/849, which has a recess for the minbar (see Schacht, op. cit., 149 ff., for this and further examples). Since the minbar was a symbol of authority, where the governor sat as representative of the ruling power, it was therefore an important feature of the Mas3id -i 3mi# when the community gathered to be officially addressed. AlMuaddas refers to the minbar as an object of high regard in Muslim communities. A township, for instance, could only be called a city if it enjoyed the right to possess a minbar and held public assemblies. He frequently categorises towns according to whether they had a minbar or not. The significance of a minbar was such that in Iran townships fought hard for the right to have one. Several references from alMuaddas indicate that the number of minbars in a city was a sign of its prosperity (193, 261-2, 267, 273, 282, 306, 309). No minbar survives from the early period in Iran, but Ibn Fundu mentions that he saw a minbar in the $dna Mosque in Sabzawr dated 266/879 ( Tr9 -i Bayha , ed. K.S.K. 0usayn , Hyderabad 1968, 86). He also adds that the name of the ruler of 9ursn , Amad al9u3istn , who held power there during the reign of the caliph alMu#tamid [see

9u3istn ], was written on it. The earliest surviving minbar in Iran is in the 3mi# Mosque of 9u9tar and is dated -afar 445/May-June 1053 (N. Meshkati, A list of the historical sites and ancient monuments of Iran , tr. H.A. Pessyan, Tehran 1974, 109). It is an early example of a minbar decorated with star- and polygonshaped panels, filled with arabesque interlace pattern, fitted by the tongue-and-groove technique covering the sides, a type of decoration which became popular in Egypt, Syria, Turkey and other parts of the Islamic world from the 5th/11th century onwards. No other minbar with this type of decoration is known in Iran | from the Sal3 period. In central Iran, five minbars survive from the period of Sal3 rule. All of these reveal the same structure as that at |ayrawn , namely, a flight of steps with posts at their entrance leading up to the speaker's seat, which consists of a platform supported on four posts. The minbars are on a smaller scale than that of |ayrawn , but the sides, consisting of carved rectangular panels, are similar (for a detailed description and analysis of these Iranian minbars, see J. Golmohammadi, Wooden religious buildings and carved woodwork in central Iran , Ph.D. diss., Univ. of London 1988, unpubl.). One of the earliest of these five is the minbar in the Mas3id -i 3mi# in Abyna , dated 466/1077. The second, dated 543/1148, is in the Immzda Ism#l in Barz, and the third, dated 583/1187, is in the Mas3id -i P"n in Far9and . All these three minbars are in the Natanz region. The minbar in the Mas3id -i 3ami# in Muammadiyya

near N"n , and that called the -ib -i Minbar in a building attached to the 0usayniyya in Far9and , have no dates, but they may be attributed to the 5th-6th/11th-12th centuries on account of the use of the bevelled technique of carving in the arabesque decoration, and in the case of the -ib -i Minbar, the style of the Kfic inscription. A notable and important feature of these five Iranian minbars is the application of a so-called bevelled style of carving. This particular decorative technique was identified by E. Herzfeld as found on the stucco decoration of Smarr (Die Ausgrabungen von Samarra, i, Berlin 1923, 5-8, 10-14). It consists of patterns cut at a deep slant giving contrast of light and shade. The patterns, often repeated and separated by curving lines, were covered with dots, notches and slits and rows of beads or pearls were frequently used as a decorative border. While this style and technique was first used in Smarr during the 3rd/9th century, it survived in Iran, as R. Ettinghausen has pointed out, in a somewhat modified form, losing its repetitive arrangements, during the 5th-6th/11th-12th centuries (The Beveled Style in the post-Samarra period, in Archaeologica orientalia in memoriam Ernest Herzfeld, New York 1952, 76-81). This style of carving was out-moded in Egypt by the late 5th/11th century, but we can still observe it in other parts of the Islamic world right up to the end of the 6th-12th century, although the bevelling tended to be considerably shallower.

1. Minbar of the

Mas3id -i 3mi# , Abyna (Photo.: J. Golmohammadi)

2. Ceramic minbar of the

Mas3id -i Maydn , K9n (Photo.: B. O'Kane)

3. Minbar of the Shrine of al-

0usayn at #Asaln (Photo.: B. O'Kane)

4. Minbar of the

Suln 0asan Mosque, Cairo (Photo.: B. O'Kane)

5. Minbar of the

Sd #Uba Mosque, |ayrawn (Photo.: J. Golmohammadi)

6. Minbar of the

#Al ' alDn Mosque, Konya (Photo.: Mevlana Museum, Konya) The minbar of the Mas3id -i 3mi# in Abyna is perhaps one of the most outstanding works of the bevelled style still surviving in Iran. Its panels are carved with deeply bevelled patterns, including abstract leaves with spiral tips, which can be traced back to the stucco decoration of Smarr . Also noticeable on it is the use of the tongue-and-groove technique, which existed in the early days of Islam (see E. Pauty, Les bois sculpts jusqu' l'poque Ayyoubide, in Catalogue gnral du Muse Arabe du Caire, 1931, pl. II, nos. 4627, 4630, and pl. IV, no. 4739). Although the Barz minbar has bevelled panels, the decoration is mainly arabesque interlace, showing the influence of new decorative trends. This minbar is also notable for its balustrade, which is composed of a lattice grille made up of geometric patterns formed by small pieces of turned wood fixed to each other by the technique that is well-known in the ma9rabiyya [q.v.] work of Egypt; this is the earliest known dated example of such work in Iran. The existence of these minbars is significant, since they pre-date the Mongol invasion; it was previously thought that all Sal3 minbars were destroyed by the Mongols.

The bevelled style of carving can further be observed on the minbar in the Great Mosque of | #Amdiyya in #Ir , dated 548/1153 (Ettinghausen, op. cit., 74, Pl. X). Here polygonal panels set in a plain frame are decorated with semi-palmettes with spiral tips within curving scrolls. In Turkey there is a minbar from Malatya, which is now preserved in the Ethnographic Museum in Ankara (G. ney, Anadolu Seluklu mimarisinde ssleme ve el sanatlari, Ankara 1978, 115). It has small polygonal pieces set in a plain framework carved in the bevelled style. The field of the panels is made up of deeply incised small scale arabesques with bevelled surfaces. It has been attributed to the 7th/13th century, but Ettinghausen has correctly pointed out that it must be earlier, namely dating from the 6th/12th century (op. cit., 82). So far no other dated piece of woodwork carved with the bevelled style is known from the 7th/13th century; Ettinghausen's dating appears therefore justified. The structure and decoration of the sides of minbars began to change towards the end of the 5th/11th century, when there appeared a new method of construction and design. This was to cover the sides with small pieces of wood in the shape of stars and polygons. The earliest known example of this type is the minbar of the Mas3id -i 3mi# in 9u9tar already mentioned. The new composition appears in its fully developed form on the Fimid minbar of the Shrine of al0usayn at Askelon, now preserved in the Museum of Hebron (L.H. Vincent and E.J.H. Mackay, Le Haram el Khalil, spulture des patriarches, Paris 1923, 219-25, pls. XXV-XXVII). It is dated 484/1091-2. The entire surface of the sides is covered with elaborate geometrical patterns composed of small polygonal pieces of wood fitted into incised strapwork by the tongue and groove technique. The main elements of the pattern consist of hexagons, polygons and six-pointed stars. Each of the polygonal pieces is filled with interlaced arabesque designs. The carving, however, is no longer in the bevelled style, but executed in deep straight cuttings. Another interesting feature of this minbar is its balustrade grille composed as a ma9rabiyya , making it a very early dated example of such work. The present canopy and door of the minbar are later, probably of the

Mamlk period. During the Fimid period in Egypt, the system of decoration was to continue appearing in two early minbars, that in St. Catherine's Monastery on Mt. Sinai dated 500/1106, and that in the Mosque of #Amr in |s dated 550/1155 (C.J. Lamm, Fatimid woodwork, its style and chronology , in Bull. de l'Inst. d'gypte, xviii [1935-6], 78). The latter example has a pavilion over the speaker's seat, and the decoration at the back of the seat recalls a mirb . From the Fimid period onwards, the minbar developed its standard form, having a domed canopy over the speaker's seat, a doorway and decorative elements consisting of stars and polygons made up of small carved pieces of wood. This form was to become standard in Syria and Turkey as well as Egypt. A good example of this type is the minbar of the A Mosque in Jerusalem, which was donated in 564/1168 by Nr alDn to Aleppo and later taken by -al alDn to Jerusalem (ibid., 88). A popular decorative feature of the 6th/12th century onwards, inlay work of ivory and mother of pearl, appears on this same minbar (M.S. Briggs, Muhammadan architecture in Egypt and Palestine, New York 1974, 216). Later on, Mamlk minbars were noted for their elaborate inlay work, which included not only ivory and mother-or-pearls, but also ebony and bone. Such minbars are to be found in the mosques of Ibn ln and -li al"#

in Cairo (L. Hautecoeur and G. Wiet, Les Mosques du Caire, Paris 1932, pls. 82, 85). Towards the end of the Mamlk | period, we witness the decline of both carved and inlay decoration. The minbar from the Mosque of |"it -Bay, now in the Victoria and Albert Museum in London, dated 872/1468-9, is a good example of these late Mamlk works (V. and A., no. 1050-1869). In Iran, star and polygon decoration was slow to become popular. Thus the minbar of the Mas3id -i 3mi# in N"n dated 711/1311-12 has sides still constructed with rectangular panels rather than stars and polygons (M.B. Smith, The wood minbar in the Mas3id -i 3mi# , N"n , in Ars Islamica, v[1938], 21-2, figs. 1-7). Part of its carved decoration consists of chains of lozenges or leaves, filled with comma-like volutes, which reflect a style that became popular in Iran during the 8th/14th century. It also has a lattice-work balustrade with a geometrical design made up of slats. This is an early example of this type of lattice-work in Iran, which was used for screens, windows, gateways and balcony balustrades. Another outstanding minbar from this post-Mongol period in Iran is that of the Mas3id -i 3mi# of Sryn in Frs , now preserved in the ^rn Bstn Museum in Tehran (S.M.T. Muafaw , Ilm -i

Prs , Tehran 1343 9 /1964, tr. R.N. Sharp, The land of Pars, Chippenham 1978, 5). According to its inscription, it was made in 771/1369. It is distinguished by the use on its sides of the star and polygon style, which was by that time applied in woodwork in Iran. Another feature of this minbar is the distinctly floral element of its carved decoration, which was later to become characteristic of the Tmrid period. The minbar of the Mosque of Gawhar 9d in the sanctuary of the Imm Ri' in Ma9had made in 840-50/1436-46 is a fine Tmrid piece. It is distinguished by profuse ornamentation of star and polygon patterns with tendrils carved in relief in the Tmrid style (Diez, op. cit., 500). This minbar is unusual in Iran in having a canopy, in this case surmounted by a crown of stalactites (EI 1, Mirb , fig. 8, which shows the minbar ). Wooden minbars carved to a very high standard were also produced in Anatolia. Wood was plentiful there, so its use for mosque furniture is easily understandable. One of the earliest wooden minbars in Anatolia is in the Alaeddin Mosque, Konya, and is dated 550/1155 (J.H. Loytved, Konia. Inschriften der Seldschukidischen Bauten, Berlin 1907, 22-4). It is made of walnut wood, and apart from its intricately carved star and polygon decoration it has a balustrade grille with a |ur"nic inscription on the rails and a cusped arch with panels over the entrance. It bears no particular resemblance, either in structure or decoration, to the Sal3 minbars in Iran, and in fact is in the Syro-Egyptian form. Minbars of the Alaeddin type became increasingly popular in Anatolia during the 7th-8th/13th-14th centuries. A good example of these is the minbar of the Ulu Cami of Siirt, now in the Ethnographic Museum in Ankara, which is carved to a very high standard (E. Akurgal, The art and architecture of Turkey, Oxford 1980, 202). Similar minbars still kept in their original places are those of the Ulu Cami of Sivrihisar dated 670/1272, and that of the Erefolu

Cami in Beyehir dated 696-8/1297-9. The tongue-and-groove technique, which is called kndekr in Turkish, was applied to a full extent in the decoration of these minbars. It is remarkable, however, that in Anatolia a kind of false kndekr was also frequently used, most likely for the reason that it cut the cost. Large panels were carved in polygonal patterns and mounted on the skeleton structure of the minbar . Sometimes the geometric patterns were made separately and glued on to the background. Small strips of incised wood were nailed between them to | give the appearance of strapwork and also to hide the joins in the panels. This method, however, does not prevent warping, and in time slits appeared between the panels. Examples of such false kndekr technique can be seen on the minbars of the Alaeddin Mosque, Ankara (594/1197-8), the Ulu Cami, Divrii (626/1228-9), and the Ahi Elvan Mosque, Ankara 784/1382 (Akurgal, op. cit., 202). Although minbars were most commonly made of wood, they were also constructed of other materials, such as brick, ceramic and stone. AlMuaddas , 77, mentions one in #Arafa (in the 0i3z ) which was made of bricks. There is also an undated brick minbar in the 4th/10th century Tr9na Mosque in Dm9n , though it appears to be much later than the building itself. (R. Hillenbrand, The mosque in the medieval Islamic world, in Architecture in continuity. Building in the Islamic world today. The Aga Khan Award for Architecture , ed. S. Cantacuzino, New York 1985, 37). In central Iran, there are five known examples of ceramic tiled minbars dating from the period ca. 1445-1525 A.D. (B. O'Kane, The tiled Minbars of Iran , in Annales Islamologiques, xxii [1986], 133-53, pls. XXXVI-XLIII). All are decorated with variations on a design of eight and twelve pointed stars, which include patterns of light blue stems with amber and white flowers, and floral arabesques of amber and light blue on a dark blue ground. Some have inscriptions giving the name of the donor, or of holy figures or religious texts. The finest is in the

Mas3id -i Maydn in K9n and is decorated with mosaic faience of a standard far above average. One inscription on the left-hand side gives the name of the craftsman as 0aydar , the tile-cutter, and another inscription states the time of construction as being in the reign of Sultan Ab Sa#d Grgn , which has led O'Kane to date the minbar to the year before Sultan Ab Sa#d 's death in 874/1469, when he was briefly ruler of the area. The minbar of the mosque of the 9nagh of Bundirbd is the largest of the five, and is dated by O'Kane to about the time of the repair works to the mosque itself, carried out in 848/1473. These tiled minbars belong to a period of growing use of tiles and mosaic faience in Iranian architecture. The taste for them did not last for long, probably because there was a need to retain mobility in certain circumstances. There are two late examples of tiled minbars from 9wa , one in the summer mosque of the Old Arg, which is datable to the 1820s and the other in an unidentifiable building also probably 19th century. Both are low with four steps (O'Kane, op. cit., 153). There are a number of stone minbars in the Islamic world, such as those of the 9ay9 , Asunur and 9ir Mosques in Egypt. Perhaps the most famous is in the mosque of Suln 0asan dated 757-64/-1356-63 (Hautecoeur and Wiet, op. cit., 103, 300, pls. 119, 132). The Mosque of Bar

in the Cemetery of the Caliphs, dated 806-13/1403-10, has a fine stone minbar carved with intricate geometric patterns, the sides in particular having star and polygon designs similar to those on woodwork. It resembles the carved stone minbar in the Mosque of 9ay9 dated 750/1349 in having a door with a stalactite portal and canopy surmounted by an onion-shaped dome (ibid., 261, 300, 314, 334, pls. 119, 157). The Friday mosque of Hart had a marble minbar of great beauty, which now no longer exists, carved for it at the end of the 9th/15th century by the stonemason 9ams alDn ( 9ulat ala9br , part of 9
w

ndamr 's general history describing Hart , ed. G. I#timd , Kbul 1345 9 /1966, 12). A.D.H. Bivar | has drawn attention to the stone minbar of the Muaffarid Amad dated 789/1387-8 in Sr3n (see kitbt , pl. XXIII, 29). The earliest stone minbar in Anatolia is in the Alaeddin Mosque in Nide dated 620/1223. The minbar is simple with no decoration except arabesques carved on the stone balustrade (A. Gabriel, Monuments turcs d'Anatolie, Kayseri-

Nide , Paris 1931, i, 120, 122, pl. XXXVI). Marble and stone minbars were mainly popular in the Ottoman period. The Mosques of Byazd and Memed Pa9a in Amasya, both dated 891/1486, have fine marble minbars of highquality workmanship. The minbar of the latter is particularly notable for its rich floriated decoration (Gabriel, op. cit., ii, 37-8, 43, pls. VI-2, VII-2). The most interesting minbar is in the Selimiye Mosque in Edirne dated 961/1574, and is superior in its size, beauty and the quality of its workmanship. It is carved from a single block of stone, and the side is dominated by an equilateral triangle containing a sun disc. The fretted border and balustrade are composed of traditional polygonal designs and the conical canopy decorated with ceramic tiles. The stone minbar in the Sokollu Mosque in Istanbul dated 980/1572 also has a tiled canopy, as well as a lattice balustrade in stone imitating those in woodwork (G. Goodwin, A history of Ottoman architecture , London 1971, 264-5, 274, pls. 253, 261; O. Aslanapa, Turkish art and architecture , London 1971, 223, 225, pls. 174, 178). (J. Golmohammadi)

3. In India.
In the various building styles of India (as defined in HIND , vii) the typology of the minbar is very variable, from a crude construction of three simple stone steps to elaborately carved canopied structures of nine steps or more. Stone is always the preferred material, even in the brick-building region of Bengal; however, the absence of any structural minbar in many well-preserved old mosques may suggest that wooden minbars were also known, although no early examples have survived.

Dihl sultanate. In none of the earliest mosques is there an original minbar (that in the 3am#at 9na at Nizamuddin, the oldest mosque still in worship, is a modern replacement; old photographs, however, show a simple minbar of three stone slabs). This pattern is maintained up to the Lod

period, to judge by a very few extant examples in Dihl (e.g. mosque at Ba gumbad; mosque at the b"l known as R3on k b"n ); only in the special case of the #dgh attributed to Mull Ibl 9an is there a more elaborate structure, a tall stone platform level with the top of the mirb arch whence the voice of the 9ab might reach the great concourse gathered for the #d assembly. Outside Dihl itself, the minbar of the 3mi# mosque of Iri , 815/1412, is a massive stone structure of seven steps, the last extended to a square platform supported on pillars. Among the regional styles, no early mosques remain in the Pan3b . Bengal shows excellent early examples of canopied minbars; the earliest, in carved basalt, in the Ba mas3id in 1h Panu" [see panu" , 1h

] of the early 8th/14th century, has nine steps leading to a domed upper chamber, with arched openings on three sides and what appears to be a mirb representation against the western wall of the prayer-chamber. This design was followed in the great $dna mas3id of 0a'rat Panu" [see panu" ] of 776/1374-5, where as Ravenshaw's photograph shows (J.H. Ravenshaw, | Gaur : its ruins and inscriptions , London 1878) the mirb -like decoration on the western wall is carved with the representation of a hanging lamp, and the outer surface carved with geometrical diaper patterns. Similar but plainer is the minbar of the nearby |ub 9h (Ravenshaw's Golden) mosque, 993/1585. Further instances of this type occur; but there are also many simple minbars of three simple stone slabs. One late aberrant minbar , in the mosque of Muammad b. Tip Suln , 1258/1842, consists of three polished stone steps occupying half of the central mirb , space having been severely limited by the neo-Palladian design of this building. In the few remaining buildings of the 3awnpur sultanate, in 3awnpur itself, in the 3mi# mosque at Iw , and in the Ah" kangra mosque at

K9 , Banras , the minbar takes the form of a massive stone structure of nine steps up to a square stone platform, with no trace of there ever having been a canopy. The typological similarity to the Iri example mentioned above points to a geographical rather than a dynastic determinant of style. The favourite style of minbar in the Gu3art sultanate is again the massive stone nine-stepped structure, although as Amad 9h 's mosque, the earliest in Amadbd (816/1414), shows, the upper platform was covered by a canopy; the canopy may be taken entirely from a Hindu temple manapa , supported by pillaged pillars, although even when purpose-quarried stones are used they are often elaborately carved in accordance with the characteristic richness of the Gu3art style. The steps may further be enclosed by stone sides to form handrails, again with carved surfaces. In many mosques the canopies have been removed, probably when many fine stone buildings were plundered during the early years of Marh rule in the early 12th/18th century. A feature found in many Gu3art mosques is the presence of a low square platform in front of the lowest step of the minbar ; its original purpose is not clear, but it is not uncommon now to see it covered with mats and used to seat young students when the mosque is in use as a |ur"nic school. In Mlw the canopied minbar is again the preferred style, as exemplified in that of the early mosque of Malik Mu99 at

Mn , 835/1432, where the upper platform is surmounted by a square roof resting on pillars which appears to be temple spoil, with projecting eaves and a parapet surmounted by a row of shield-shaped merlons; to the west the wall takes the form of a mirb of black polished basalt, with the characteristic Mlw row of merlons in low relief. This is surpassed by the magnificient minbar of the 3mi# mosque (completed 858/1454), perhaps the finest in the subcontinent: eleven steps lead to the upper platform, originally railed on north and south; the three open sides are of the same shape as the arches of the mirb s, slightly ogival; the canopy itself has its eaves supported by sinous brackets, of the same shape as those in the 3mi# mosque of Dhr and of H9ang 's tomb in Mn; above the row of merlons there is a marble dome of the characteristic Mlw shape, i.e. stilted below the haunch by being raised on a cylindrical drum. Here, as in the Gu3art mosques, there is again a square low platform at the foot of the minbar steps. At 1ndr [q.v.] the minbar of the 3mi# mosque is typologically similar, but without the sinuous brackets and more solidly built (now whitewashed); that of the great #dgh similar but plainer, and of only eight massive steps (the even number is unusual). In the Deccan , however, the minbar is usually of | the plain pattern of three modest stone steps; so at the first Bahman mosque, the 3mi# mosque of Gulbarga (769/1367), and others in Bidar. In the massive #dgh

at B3pur [q.v.], certainly of Bahman date, the minbar has nine stone steps leading to an open platform; in the arched opening of the west wall behind it is a flight of smaller steps leading to the top of the wall. In the buildings at B3pur (and Gg ) of the #$dil 9hs , the most ornate of the Deccan styles, the minbar remains of the simple pattern of three (occasionally five) stone steps, and the same is true of the |ub 9h mosques of 0aydarbd . Throughout the Mu9al period, the minbar is of the stepped uncovered type. Sometimes, as at the 3mi# mosque in Fatpur Skr , the massive red sandstone steps have small pierced screens at their sides; in the time of 9h3ahn , when many modifications were also made to earlier buildings, the minbar is often a simple structure of three steps but built of polished, sometimes also inlaid, marble, and a few have a chair-like back slab which may carry a brief inscription. The 3mi# mosques of Dihl ( 9h3ahnbd ) and $gr each have a central platform, approached by steps, in the an

, outside the prayer-chamber, which may fulfil the functional purpose of the minbar when there is a vast concourse of worshippers to be addressed, even though there is a minbar in its normal position within the prayer-chamber. (J. Burton-Page)

4. In East Africa.
Several different types of minbar are to be found on the East African coast. One type is apparently peculiar to it. In the Middle Ages and up to the 19th century the greatest number of Friday mosques had a stone minbar consisting of two steps and a seat. At Kisimkazi, Zanzibar, there is only one step and a seat, while at Kua, Juani Island, Mafia, and at Mgao Mwanya, on the Tanzanian mainland, there are three steps and a seat. In all these cases the lowest step is very shallow, and is known in Swahili as kiapo, or place for taking solemn oaths. The person taking the oath stands on the lowest step, and touches the minbar . A brief account of Swahili oaths is given by Mtoro bin Mwenyi Bakari of Bagamoyo, Tanzania, in C. Velten, Desturi za Wasuaheli, Gttingen 1903, 273-7, but without explanation of the ritual. The later Friday mosque at Ungwana, Kenya, built ca. 1500-50, is alone in its period in having seven stone steps and a seat at the top, with masonry sides formerly surmounted by a wooden handrail. In recent times similar stone minbars have been constructed in mosques in the Lamu archipelago. Only two wooden minbars are known. This does not arise from any distase for wood, but because it is vulnerable to the white ant, ubiquitous in eastern Africa. The wooden minbar in the Friday mosque at Lamu is dated by an inscription 917/1511, and that at | Siu 930/1523, both of these being in Kenya. At Magogoni, a small village near Dar es Salaam, Tanzania, a portable minbar is used on feast days. It consists of a simple wooden upright chair constructed on a flat pedestal, the latter projecting to form a step in front, and the space between the legs being enclosed to form a cupboard. It is of recent and rough construction. In a number of Friday mosques, however, the minbar takes the form of a recess, or of a raised platform within a recess built out behind the ibla wall. It may be reached by a staircase within its recess, or by a staircase from inside the mirb . The minbar thus resembles a window on the right-hand side of the mirb

. It is sometimes provided with a balustrade for the preacher to lean on. Where the staircase leads out of the mirb , it is sometimes connected with a room or office for the use of the imm , for whom often an external door is also provided. The dating of minbars of this type is uncertain, but local tradition, which is probably correct, assigns their construction to the end of the 19th and the beginning of the 20th century.

MARW$NI
B. AL-

0AKAM
B.

ABI 'L-

#A , Abu 'l|sim and then Ab #Abd al-Malik , first caliph of the Marwnid branch of the Umayyad dynasty [q.v.], reigned for several months in 64-5/684-5. Marwn , born of al0akam 's wife $mina bt. #Alama alKinniyya , stemmed from the same branch of the Umayyad clan of |uray9 , sc. Abu 'l#$ , as the Rightly-guided caliph #U9mn , and was in fact #U9mn 's cousin. The sources generally place his birth in A.H. 2 or 4 (ca. 6236), but it may well have occurred before the Hi3ra ; in any case, he must have known the Prophet and was accounted a Companion. He became secretary to #U9mn

when he already had a considerable reputation for his profound knowledge of the Holy Book (alMad" in , in alBal9ur , Ansb , v, 125: min ara" alns li'l|ur" n ), and doubtless helped in the recension of what became the canonical text of the |ur" n in that caliph's reign [see ur" n . 3]. Also during this reign, he took part in an expedition into North Africa, and it was apparently his share of the rich plunder from this which laid the foundations of Marwn 's extensive personal fortune, invested in property in Medina; and it is further mentioned that he was for a while governor in Frs . He was wounded at the Yawm aldr , the defence of #U9mn 's house in Medina against the insurgents of the Egyptian army in 35/656, and fought at the Battle of the Camel with #$i9a and her allies [see AL3amal ], but seized the opportunity personally to slay ala , whom he regarded as the most culpable person in the murder of #U9mn . Somewhat surprisingly, he then gave allegiance to #Al after the battle. During Mu#wiya 's caliphate,

Marwn was governor of Barayn and then had two spells as governor of Medina, 41-8/661-8 and 547/674-7, alternating with his kinsmen Sa#d b. al#$ and alWald b. #Utba . It was during these years that he acquired from the caliph the estate, with its lucrative palm groves, of Fadak [q.v.], which he subsequently passed on to his sons #Abd al-Malik and #Abd al#Azz . It is possible that Mu#wiya latterly grew suspicious of Marwn 's ambitions for his family, especially as the family of Abu 'l#$ was perceptibly more numerous than that | of 0arb , Mu#wiya 's grandfather; Marwn himself had, according to alBal9ur , Ansb , v, 164, ten sons and two daughters, and al9a#lib , La" if, 136, tr. 107-8, states that he further had ten brothers and was the paternal uncle of ten of his nephews. It may have been fears of the family of Abu 'l#$ that impelled Mu#wiya to his adoption ( istil ) of his putative half-brother Ziyd

b. Sumayya [see ziyd

B.

abhi ] and to the unusual step of naming his son Yazd as heir to the caliphate during his own lifetime. There was certainly a lack (with the exception of alWald b. #Utba , Mu#wiya 's nephew) of mature, experienced Sufynids to succeed Mu#wiya , whereas at the time of the expulsion of the Umayyads from the 0i3z (see below), Marwn was the most senior of the Umayyads and the only one whom the Prophet had known ( 9ay9 kabr in the sources, probably referring as much to his prestige and authority as to his age). When the difficulties arose in 60/680 over Yazd b. Mu#wiya 's succession, involving a refusal of allegiance by the cities of the 0i3z , Marwn advised the governor of Medina, alWald b. #Utba , to use force against the rebels. After the withdrawal of the expeditionary force of Muslim b. #Uba alMurr and its return to Syria (beginning of 64/autumn 683), the Umayyads and their clients who had been previously expelled but had returned with

Yazd 's troops, comprising principally members of the lines of al#$ under #Amr b. Sa#d alA9da [q.v.] and of Abu 'l$ under Marwn , were forced by the partisans of the anti-caliph #Abd Allh b. al-Zubayr [q.v.] to abandon their properties in the 0i3z and flee to Syria for a second time. Marwn was back in Syria by the beginning of 684, and some accounts say that he went in the first place to Palmyra rather than to the court of the ephemeral caliph Mu#wiya II b. Yazd [q.v.] at Damascus. With the latter's death, and in face of the widespread support, even in Palestine and northern Syria, for a Zubayrid caliph, Marwn despaired of any future for the Umayyads as rulers, and was himself inclined to give his allegiance to #Abd Allh b. al-Zubayr. But heartened by the urging of #Ubayd Allh b. Ziyd b. Abhi [q.v.], Marwn allowed his own candidacy to go forward at the meeting of Syrians at al3biya [q.v.] convoked to hail a successor to Mu#wiya II, and with the support of the leader of 3u9m

, Raw b. Zinb# , was hailed as caliph, with 9lid b. Yazd b. Mu#wiya [q.v.] and #Amr b. Sa#d alA9da named as next heirs. With this acclamation and the support of the Kalb under Ibn Badal [see assn
B.

mlik ], Marwn was able to defeat the |ays under al4ak b. |ays alFihr [q.v.] at the battle of Mar3 Rhi [q.v.], probably to be placed in July or early August 684. Then shortly after his installation as caliph in Damascus, Marwn married Umm H9im F9ita bt. Ab H9im b. #Utba , the widow of

Yazd I and mother of his two sons; this diplomatic alliance gave him a link with the Sufynids . Marwn was now able to consolidate his position in Syria and Palestine. His short reign was filled with military activity, beginning with the expulsion of the Zubayrid governor, #Abd alRamn b. #Utba alFihr , called Ibn 3adam , from Egypt. Marwn seems to have secured that province by Ra3ab 65/February-March 685, leaving there as governor his son #Abd al#Azz . Although the sources are confused here, it seems that Marwn 's forces also repelled a Zubayrid attack on Palestine led by Mu#ab b. al-Zubayr [q.v.]. It is possible, but not certain, that a Marwnid army itself invaded the 0i3z under | 0ubay9 b. Dul3a , but was repelled at alRaba9a [q.v.] to the east of Medina. Marwn certainly took steps to secure #Ir , which had declared for the Zubayrid cause, sending an army under #Ubayd Allh b. Ziyd which by-passed the hostile

|ays centre of |irsiya in al3azra and had reached alRaa when the news of Marwn 's death arrived. This last event took place in the spring of 65/685, possibly as a result of a plague which was affecting Syria at this time. The date of Marwn 's death is variously given in the sources: Elias of Nisibin has 7 May, and the Islamic historians such dates as 3 Rama'n /13 April (alMas#d , Tanbh ) and 29 9a#bn -1 Rama'n /10-11 April ( Ibn Sa#d , al9alfa b. 9ayy , alabar ). The place of his death is given by several authorities as Damascus ( Ibn Sa#d , alabar , alMas#d , Tanbh ), but by alYa#b and alMas#d , Mur3 , as al-

-innabra on the Lake of Tiberias, a place used, it seems, as a winter residence by the early Umayyads. The length of his reign is placed at between six and ten months. Even less certain is Marwn 's age when he died; the sources make him at least 63, but he may well have been over 70. On the occasion of the successful outcome of the Egyptian expedition, Marwn had taken the opportunity to vest the succession in his own sons #Abd al-Malik and #Abd al#Azz [q.vv.], and it was accordingly the former who succeeded to the caliphate in Damascus after Marwn 's death, apparently without opposition (at least, at this moment) from the two heirs designated at al3biya , #Amr b. Sa#d and 9lid b. Yazd , but now set aside. Marwn 's life had been crowded with action, above all in its later years, filled with military campaignings and the negotiations surrounding his succession to the caliphate. He seems to have suffered severe aftereffects from various wounds, and his tall and emaciated frame earned him the nickname of 9ay bil insubstantial, gossamer-like thread (see al9a#lib , La" if, 35-6, tr. 56). His brusqueness and lack of the social graces resulted in his being described as fi9 uncouth . Later, anti-Umayyad tradition stigmatised him as ard

ibn ard outlawed son of an outlaw , associating him with his father al0akam who was allegedly exiled by the Prophet to " if, and as abu 'l3abbira father of tyrants because his son and five of his grandsons subsequently succeeded to the caliphate. But he was obviously a military leader and statesman of great skill and decisiveness, amply endowed with the qualities of hilm [q.v.] and dhiya , shrewdness, which characterised other outstanding members of the Umayyad clan. His attainment of the caliphate, starting from a position without many natural advantages beyond his own personal qualities (for he had no power-base in Syria and had spent the greater part of his career in the 0i3z ), enabled his successor #Abd al-Malik to place the Ummayyad caliphate on a firm footing so that it was able to endure for over 60 years more.
AL-

RU-$FA

, the name of several places in the Islamic world, from Cordova in the west to N9pr in the east (see Yt , Buldn , ed. Beirut, iii, 46-50). Amongst the Rufa settlements of #Ir were: 1. Rufat Abi 'l#Abbs (

#Abd Allh alSaff ), begun by the first #Abbsid caliph in lower #Ir on the banks of the Euphrates, near alAnbr [q.v.], and probably identical with that town called alH9imiyya . 2. alRufa , the name of a quarter of the city of Ba9dd [q.v.] founded soon after the caliph alManr [q.v.] built his Round City. The quarter of alRufa (whose name refers to the paved, embanked causeway across the swampy ground enclosed by the bend of the Tigris within which the quarter was laid out) was, according to the historical accounts, built by alManr on the eastern banks of the river, opposite the palace of al9uld and the Round City, for his son and heir alMahd [q.v.] when the latter returned from Rayy in northern Persia in 9awwl 151/October-November 768. It combined a palace complex, with protective rampart and moat, and an army encampment with a review ground ( maydn [q.v.]) and with various estates granted out as a"i# to members of the #Abbsid family and to the great military commanders (see Ya#b , Buldn

, 249, 251, tr. 31-2, 35-6). From this last function as a military centre, it was originally known as #Askar alMahd . Alabar (iii, 365-7, tr. H. Kennedy, AlManr and alMahd , Albany 1990, 56-9) plausibly explains that the caliph wished to separate his Arab supporting forces by the river which divided the two sides of Ba9dd , so that if one section of the army rebelled, he could call upon the forces on the opposite bank. The building of alRufa took seven years, and was not completed till 159/776, by which time alMahd had (in 159/775) succeeded to the throne. The new quarter was connected to the western side of Ba9dd by a bridge of boats, al3isr , whose obvious strategic importance was such that each end was guarded by a police post of the 9ura [q.v.]. Lassner has suggested that alManr began the construction in alRufa of a palace complex of such splendour in order to buttress his son's right to succeed to the caliphate against his nephew #^s b. Ms [q.v.], thereby asserting alMahd 's claims. As caliph, alMahd made alRufa

his official | residence, but towards the end of his reign preferred to spend much of his time at a new palace and pleasure ground, that of #^sb9 , also on the eastern side of the city but away from alRufa . His successors Hrn alRa9d and alAmn [q.vv.] chose, however, to reside at al9uld on the western side; and eventually, alMu#taim [q.v.] moved his seat to the new military centre of Smarr [q.v.] some 100 km/60 miles upstream from Ba9dd . The foundation of alRufa was the starting- point for the expansion of Ba9dd into such suburbs as 9ammsiyya to its north-east and Mu9arrim to its south; in later times, the tombs of the #Abbsid caliphs were located along the river bank above alRufa . (C.E. Bosworth) 3. In Syria. This place, distinguished as Rufat Hi9m
,

Rufat
AL-

9m

, is now a ruinous site 30 km/19 miles to the south of the Euphrates in a depression near the 3abal Bi9r , on the ancient desert route from 0im -Salamiya to alRaa or alRaba , containing the pilgrimage place of the Arab Saint Sergius, martyred here in the early 4th century, after which it was officially named Sergiupolis in Byzantine times. Archaeological excavations have shown it to be a Roman site, which in the 6th century was embellished with four churches inside the impressive rectangular city walls; among them, the basilica of the Holy Cross, founded in 559 and housing the relics, shows by inscriptions a continuous building tradition until the 12th century. Also inside the city walls, three large cisterns and an ingenious system of supply and distribution of the spring rain water remained famous throughout mediaeval sources. It seems to be referred to in the K. al#Uyn wa 'lad"i : It had been a Byzantine city of ancient foundation with cisterns and a 'water way' ( ar li 'lm" ) from the margins of the desert (in Fragmenta, 101). Outside the north gate a church or praetorium (?) is connected to the 9assnid prince alMun9ir b. al0ri9 (569-82) by an inscription, and literary sources point to the presence of the 9assanids as well: alNu#mn b. al0ri9 b. al-Ayham is mentioned as a governor there, and is said to have repaired the cisterns destroyed by a La9mid and to have constructed a large new one ( Yt , ii, 955, 784 (according to the

A9br mulk 9assn ); 0amza alIfahn , Ta"r9 , Berlin 1340/1921-2, 79 (and quoted by Ibn al#Adm , Bu9ya , i, 114), as well as Abu 'lFid" , Mu9taar , Cairo 1325, i, 73, only mention the restoration). AlAma# mentions, besides the B. 3afna of 9assn , the B. 0anfa (of Bakr b. W"il , Ibn alKalb -Caskel, ii, 156; Kala , |ab"il , i, 312-13) as inhabitants (quoted by alBakr , Mu#3am , i, 441); it is he who gives a further name of this, Rufa alZawr" (compare Bakr with Yt

, ii, 784, 955; Musil, Palmyrena, 267; the Byzantine name for alRufa , | . ml , supposed by Ibn al#Adm , i, 113, remains unexplained, cf. Ibn 9urrad9bih , 218, B.t. lmiy , 35 ml s from it?). Earlier, the city was within the region of the Tan9 (alBal9ur , Fut , 145; I. Shahid, Byzantium and the Arabs in the fourth century, Washington 1984, 405, 465), andafter them?of the Ta9lib (alabar , i, 2072; Rotter, Die Umayyaden und der zweite Brgerkrieg, Wiesbaden 1982, 131). This may have led to the occasional attribution of al- | Rufa to Diyr Mu'ar (e.g. alabar , iii, 2219). In the #Abbsid period, it came under the control of the B. 9af3a b. #Amr , a branch of the (North Arabian) #Uayl (alAma# , in Yt , ii, 284; Caskel, ii, 338; Kala , |ab"il

, i, 351); alBakr alone connects this Rufa with a verse by alA9nas b. 9ihb alTa9lib on the extended animal-hunting of the (South Arabian) B. Bahr" (? 9arak
un

lib un , 56; Yt , ii, 782). Ibn al#Adm mentions some B. -li of H9im here at the time of Hrn alRa9d ( Bu9ya , iii, 1467-8, vii, 3446). Administratively, alRufa belonged to |innasrn or Aleppo in Umayyad and later #Abbsid times (Ibn al#Adm , i, 113-14), under Hrn alRa9d it was added to the #awim province ( Ibn alFah , 111), but some transmitters were uncertain about its district (

Rufat alRaa , even alRufa in the 3azra , Ibn al#Adm , v, 2103 with correction; Ibn #Askir , iv, 259; Ibn 9urrad9bih mentions it twice, apparently with its closer neighbours, 74, and together with Blis in the #awim , 75). From Zangid until Mamlk times it seems to have mostly been known as a Christian suburb or in the district of |al#at 3a#bar , which was also called Klnks / Kallinikosin this way a remark by Barhebraeus, Chronography, ed. Budge, 120, tr. 2111, could be understood: Hi9m died in Rufa of Kallinikos (cf. ibid. tr. 2218; Syriac Chronicle, ad a. 1234, i, 215), and it would fit the reading of an Arabic inscription on the silver goblet from the Rufa treasury suggested by R. Degen, This is what Zayn alDr , daughter of ust9 Ab Durra, bestowed to the church of the protected |al#at 3a#ba [r], probably meaning alRufa (before 1243, in Ulbert, Resafa, iii, 72; for |al#at 3a#bar

as a district ( a#ml ), cf. Ibn alDawdr , vii, 283, year 624/1227). Nothing is reported on the Islamic conquest of alRufa , and the sources rather convey the impression of its lying in ruins until the building activities of Hi9m (Ibn al#Adm , i, 113). But it is mentioned as being on the march of the |ays 3af b. 0akm from the 3azra against the B. Ta9lib with their poet alA9al and their day at 3abal Bi9r in 73/692-3 (alBal9ur , Ansb , v, 329; A9n , Bl , xi, 59; Ibn al#Adm , i, 431 ff.; 9iznat al-adab, Cairo, ix, 4); and also, before 724, a Bishop Abraham of alRufa is documented (Degen 70-1, quoting Wright, Cat. of Syriac mss., ii, 796 ff.). The main information on alRufa in Muslim sources pertains to the caliph

Hi9m (105- 24/724-43), whose residence it became at least in summer and who was buried there. While these sources unanimously locate the residence in or next to this alRufa (e.g. Ibn Fa'l Allh al#Umar , Maslik , Cairo 1342/1924, 332-3), some modern authors have doubted this and identified it with the ruins of |ar al0ayr al9ar [q.v.] (Sauvaget, Remarques sur les monuments omeyyades, in JA [1939], 1- 13). This theory was finally rejected by O. Grabar (City in the desert, 1978, 1-2, 31). It is not very easy to recognise the original tradition within the several additions transmitted by the historians; the news of the death of his predecessor and the regalia were brought to Hi9m at alZaytna , where he possessed a small dwelling (duwayra), and he then rode from alRufa to Damascus (alabar , ii, 1467); perhaps because of the unexplained leap, later sources locate either the transmission of the news and the regalia to alRufa (al#Uyn wa 'lad"i , 82; Abu 'lFid" , Mu9taar , Cairo 1325, i, 203) or from alZaytna to alRufa ( Yt

, ii, 784). While O. Grabar still thinks of an identification of alZaytna with Kar al0ayr al- | 9ar (City in the desert, 13-14), a hint by Ibn Buln [q.v.] possibly gives the clue to understanding the sequence of residences: Hi9m was fleeing from the mosquitoes on the banks of the Euphrates to alRufa (in Yt , ii, 785)this fits the surroundings of the straitened dimensions of his princely residence alZaytna , perhaps in the vicinity of his further possessions near alRaa . Another often-embellished story mentions him avoiding the Syrian cities in favour of alRufa because of the plague; like other Umayyads, he fled to the desert (e.g. alabar , ii, 1737; Ibn al#Adm , i, 113-14).

Hi9m is reported to have reconstructed alRufa and erected two castles there ( arayn , alabar , ii, 1738; al#Uyn wa 'lad"i , 101; Ibn Buln , ibid.). Whether their descriptions as possessing a pool and olive yard (alabar

, ii, 1813) or as being luxurious constructions with floral paintings (Ibn al#Adm , vii, 3044; mani# here evidently not meaning the cisterns, as in ibid., i, 113) go back to eye witnesses or are literary topoi, cannot be decided. Brief archaeological soundings and a survey showed several large Umayyad structures to the south of the city (Otto-Dorn, Ulbert, Sack). The court of Hi9m was magnificent, and must have shown Persian traditions in several respects (cf. the analysis by R. Hamilton, Wald and his friends, Oxford 1988, passim). Whoever was interested in Persian topics among the early #Abbsid caliphs would also refer to accounts of Hi9m 's court (e.g. alManr , alabar , iii, 412; alMas#d , iv, 47-8, 133-4 = 2234, 2379). Even translations from the Persian seem to have originated from Hi9m 's secretaries in alRufa , Slim b. #Abd alRamn or #Abd Allh and his son 3abala (alabar , ii, 1750, 1649-50; alMas#d , Tanbh , 106, 113; Ibn al#Adm

, ix, 4143; M. Grignaschi, in BEO, xix [1967] 12-13, 24-5, 51- 2). The Arab tradition at his court was upheld by the poets (see, besides the famous competition by alFarazd , 3arr and alA9al , descriptions by Ism#l b. Yasr alNas" , in A9n , iv, 125; 9lid b. -afwn al-Ahtam, in Ibn al#Adm , vii, 3044; Abu 'lNa3m , in A9n , ix, 78 ff., Ibn al#Adm , x, 4640). And in one respect he tried to surpass pre-Islamic customs: he himself built the greatest hippodrome ( alba ) for 3,000 horses here, six bowshots long ( A9n , x, 64; alMas#d , iv, 41 = 2219; Ibn al#Adm , vi, 2858). Also, the biographies of traditionists at his court contain material on alRufa , most famous among them being alZuhr (GAS, i, 280-3), Ab Man# #Ubayd Allh b.

Ab Ziyd and his grandson Ab Muammad al0a333 b. Ysuf (Ibn al#Adm , v, 2100 ff.; alSam#n , vi, 135; Ibn #Askir , x, 669-70, iv, 259-60), 9uayf b. #Abd alRamn or Ibn Yazd al0arrn (Ibn al#Adm , vii, esp. 3265-6). Out of Hi9m 's family, his son Sulaymn stayed in alRufa until his defeat by Marwn II (alabar , ii, 1908). Hi9m 's tomb and body were desecrated under the first #Abbsid (alabar , iii, 2498-9; alYa#b , ii, 427-8). The town had an #Abbsid governor in 137/754 (al-

abar , iii, 94-5); but apart from occasional visits of a caliph, only one event is mentioned, the sack by the Carmathians at the end of 289/December 902, when the mosque, adjoining the cathedral, was burnt and the #Abbsid defender Sabk alDaylam killed (alabar , iii, 2219; Ibn al#Adm , ii, 946: better, 9ibl alDaylam; cf. the excavations by D. Sack). Eyewitnesses are few, and are repeatedly quoted in geographical sources: alAma# , who mentions merchants, rich and poor, travelling abroad and employing local Bedouins ( #arab ), a small s with ten shops | and textile manufacturing ( Yt , ii, 284-5; still in alIdrs , ed. Rome, 649, Ibn al#Adm , i, 113-14, and 033 9alfa , 3ihn num , 594, without any contemporary observations). Apart from the cisterns and the walls the city is especially famous for its Christian buildings, figuring in the mentioning of its dayr , listed separately by alBakr , Yt , and al0imyar

. This gave rise to several topoi and anecdotes of nostalgia of the Umayyads and of the monasteries, which do not seem to correspond with reality, as already noticed by Musil (Palmyrena, 268; cf. the story connected to the visit of al-Mutawakkil in 244/858, alabar , iii, 1436; al0imyar , Raw' (Beirut 1975), 253; Ibn al#Adm , i, 114, quoting the K. alDiyrt by al9im9; for the clich of the monastery, see L. Conrad, in The quest for understanding, ed. S. Seikaly et alii, Beirut 1991, 271-2). Only the Christian Ibn Buln was interested in the great church, of which he describes the external gold mosaic in 440/1048- 9 (cited in Yt , ii, 785). Judaeo-Arabic inscriptions in one building, dated 1102 and 1127, prove the presence of a Jewish community there (A. Caquot, in Syria, xxxii [1955], 70-4). Ibn 9addd gives an account of the end of habitation in alRufa , added to a long quotation from Ibn al#Adm (ed. A.-M. Terrasse- Edd, 394, tr. 21-2): the Mongols had spared the inhabitants on their march in 658/1260, and after the Mamlk reconquest, a governor was left there until 668/1269-70, when the inhabitants left for Salamiya, 0amt and other places, apparently because of the destruction, which is also archaeologically evident. Since then, the site has been deserted. (C.-P. Haase) 4. In Muslim Spain. Munyat alRufa

, in Spanish Arrizafa, Arruzafa, is the name of the country residence founded by #Abd alRamn I (138-72/756-88 [q.v.]) to the north-west of Cordova and to which he gave the name of the Rufa in Syria (see 3. above) founded by his grandfather Hi9m b. #Abd al-Malik. The first Umayyad amr of al-Andalus purchased lands which had belonged to a Berber chief of ri 's army, Razn alBurnus , and built there a palace ( ar ) and gardens. The Arabic sources class the Cordovan Rufa amongst the three most important constructions of #Abd alRamn I's reign (the other two being the Grand Mosque and the palace of Cordova). The amr enjoyed living there very much and spent most of his time there. In the course of his residence at alRufa , he ordered the execution of three rebels: his nephew alMu9ra b. alWald , Wahb Allh b. | Maymn and #Ay9n b. Sulaymn

alA#rb . Their corpses were dragged as far as Cordova and gibbeted on the banks of the Guadalquivir. In his reign, it became, in some measure, the seat of power, since at the time of his death, his son Hi9m , who happened to be at Mrida, hastened to arrive there before his brother Sulaymn (who was at Toledo and who was disputing with him the right of succession). Moreover, #Abd alRamn I made the gardens at alRufa the first botanical gardens in the history of al-Andalus. He had planted in his grounds exotic plants, mainly brought from Syria, to which he sent envoys to contact his sisters. The most famous of these fruits imported from the East was the so-called safar pomegranate, whose name is connected with Safr b. #Ubayd al|il# , from the 3und of al-Urdunn. The latter is said to have cultivated this variety of pomegranate in the region of Rayya [q.v.], whence it was spread throughout al-Andalus. The origin of the date-palm groves in the Iberian Peninsula is equally attributed, with no real basis in truth, by some Arabic authors to a palm tree at alRufa . #Abd alRamn I's successors continued the tradition of periods of residence at alRufa . It was probably in the reign of #Abd alRamn II (206-38/822-52 [q.v.]) that the poet #Abbs b. Firns

tried, at alRufa , to imitate the flight of birds, dressed in a garment of silk covered with feathers and bearing wings. But above all, it was the amr Muammad (238-73/852-86), known for his zeal as a builder, who enlarged and improved the buildings and the gardens of this residence, where he loved to take rest and where he organised hunting parties. The amr transferred from Cordova to alRufa accompanied by his entourage of chamberlains and eunuchs, and thus surrounded by all the splendour of the Umayyad court. He charged his wazr H9im b. #Abd al#Azz with the construction of a new ma3lis at alRufa and provided him with 10,000 dnrs for this. However, the wazr had the ma3lis built at his own expense. When the work of building was finished, H9im gave back to the amr his 10,000 dnrs and, as a further gesture, prepared for him a sumptuous banquet. The first Umayyad caliph, #Abd alRamn III alNsir (300-50/912-61 [q.v.]), had accompanied, whilst he was still young, his grandfather the amr

#Abd Allh during his pleasure sessions at alRufa . But his preferred country residence was the munyat alN#ra . During his reign, alRufa is mentioned as a residence for important visitors, such as the North African chief Ayyb b. Ab Yazd Ma9lad b. Kaydd al^fran in 335/946. AlNir 's son and successor alMustanir (350-66/962-76) preferred above all the munyat Ar" Ni . Between the residential palace complex and the city of Cordova there developed a suburb ( raba' ), equally called alRufa , and the nisba from this was borne by some Cordovan scholars, such as the father of al0umayd [q.v.]. It was there that alManr Ibn Ab #$mir , alMustanir 's 3ib , had his palace built, which he later abandoned for al-munya al-

#miriyya , whose exact location is controversial. With the arrival of Berber troop contingents during the rule of alManr , the suburb of alRufa became the residence of the Ban Mksan b. Zr and the Ban Zw b. Zr , whose houses were destroyed in the course of the troubles during the first reign of Muammad alMahd . Like the other northern suburbs of Cordova, that of alRufa suffered the consequences of the fitna and its name disappears from the Arabic sources after the 4th/10th century. As for the munyat alRufa , it was first of all despoiled by alMahd in 400/1009 during his second reign. | The caliph used the contents of the palace, like those of the munyat alN#ra and the royal palace in Cordova, in order to pay his troops and to support the costs of the fight against the rival army of the Berbers on which his rival Sulaymn alMusta#n depended. In the following year, and in order to ward off the Berber advance, alRufa was totally destroyed on the orders of W'i , the military chief of Cordova. He even had the trees in the famous gardens cut down, but shortly afterwards he realised the uselessness of his action from the point of view of the defence of Cordova. Only

the name of the Umayyad princes' residence survived. After the Christian conquest, in 633/1236, the land involved was bestowed on the counts of Hornachuelos. Later, a monastery was established on the site. The Rufa of Cordova early became a favoured subject of the court poets. Some very famous verses on its solitary palm tree are attributed to the founder, #Abd alRamn I (according to other version, its author is said to have been #Abd al-Malik b. Bi9r b. #Abd al-Malik b. Marwn ). A poem of #Abbs b. Firns on alRufa , reproduced by Ibn 0ayyn [q.v.], describes at length its buildings, streams, plants, birds, etc. After its destruction, Rufa became, like Cordova, a poetic subject for the expression of nostalgia for departed splendours (texts of Ibn Zaydn and Ibn Burd, given by Ibn Bassm , Ibn 9n and alMaar; this same Ibn Zaydn mentions the existence of a garden of marguerites, raw' aluuwn ). In Almohad times, poets still gathered before the site of alRufa

in order to drink and to recite poetry. The poem of al|sim b. #Abbd alRiy develops the theme of ubi sunt. The second alRufa in al-Andalus was situated at Valencia, between the town and the sea (the place-name is still preserved under the form Ruzafa, a quarter of the modern town). There is no information on the foundation of this Valencian Rufa . E. Lvi-Provenal was the first to suggest the name of the Umayyad prince #Abd Allh alBalans , son of #Abd alRamn I, as its possible founder, at the same time warning of the lack of documentary evidence. This hypothesis has nevertheless been commonly accepted by Spanish and Arab scholars. The Arabic sources stress above all the beauty of the grounds of alRufa , considered as the most attractive pleasure-ground in the vicinity of Valencia (together with the munya of Ibn Ab #$mir ). The poet Muammad b. 9lib alRuf (d. 572/1177) was originally from there and devoted some poems to it. In 480/1087, Castilian troops commanded by Alvar Fez, giving aid to the prince al|dir , installed themselves at alRufa

. It was likewise there that king James I of Aragon encamped with his army, besieged the town and conquered it in 636/1238. Like its Cordovan homonym, this Rufa became a literary subject in the poetical or rhymed prose texts written on the occasion of the loss of Valencia (texts of Ibn alAbbr and Ibn #Amra preserved by al0imyar and alMaar ). alRU-$FA RU-$FA , the name of several places in the Islamic world name of several places in the Islamic world, from Cordova in the west to N9pr in the east (see Yt, Buldn, ed. Beirut, iii, 46-50). Amongst the Rufa settlements of #Ir #Ir were: 1. Rufat Abi 'l-#Abbs (#Abd Allh al-Saff), begun by the first #Abbsid caliph in lower #Ir on the banks of the Euphrates, near al-Anbr [q.v.], and probably identical with that town called al-H9imiyya. 2. al-Rufa, the name of a quarter of the city of Ba9dd [q.v.] founded soon after the caliph al-Manr [q.v.] built his Round City. The quarter of al-Rufa (whose name refers to the paved, embanked causeway across the swampy ground enclosed by the bend of the Tigris within which the quarter was laid out) was, according to the historical accounts, built by al-Manr on the eastern banks of the river, opposite the palace of al-9uld and the Round City, for his son and heir alMahd [q.v.] when the latter returned from Rayy in northern Persia in 9awwl 151/October-November 768. It combined a palace complex, with protective rampart and moat, and an army encampment with a review ground ( maydn [q.v.]) and with various estates granted out as a"i# to members of the #Abbsid family and to the great military commanders (see Ya#b, Buldn, 249, 251, tr. 31-2, 35-6). From this last function as a military centre, it was originally known as #Askar al-Mahd. Al-abar (iii, 365-7, tr. H. Kennedy, Al-Manr and al-Mahd , Albany 1990, 56-9) plausibly explains that the caliph wished to separate his Arab supporting forces by the river which divided the two sides of Ba9dd, so that if one section of the army rebelled, he could call upon the forces on the opposite bank. The building of al-Rufa took seven years, and was not completed till 159/776, by which time al- Mahd had (in 159/775) succeeded to the throne. The new quarter was connected to the western side of Ba9dd by a bridge of boats, al- 3isr , whose obvious strategic importance was such that each end was guarded by a police post of the

9ura [q.v.]. Lassner has suggested that al-Manr began the construction in al-Rufa of a palace complex of such splendour in order to buttress his son's right to succeed to the caliphate against his nephew #^s b. Ms [q.v.], thereby asserting al- Mahd's claims. As caliph, al-Mahd made al-Rufa his official | [VIII:630a] residence, but towards the end of his reign preferred to spend much of his time at a new palace and pleasure ground, that of #^sb9, also on the eastern side of the city but away from al-Rufa. His successors Hrn al-Ra9d and al-Amn [q.vv.] chose, however, to reside at al9uld on the western side; and eventually, al-Mu#taim [q.v.] moved his seat to the new military centre of Smarr [q.v.] some 100 km/60 miles upstream from Ba9dd. The foundation of al-Rufa was the starting- point for the expansion of Ba9dd into such suburbs as 9ammsiyya to its north-east and Mu9arrim to its south; in later times, the tombs of the #Abbsid caliphs were located along the river bank above alRufa. (C.E. Bosworth) Bosworth, C.E. 3. In Syria. This place, distinguished as Rufat Hi9m , Rufat al-9m, is now a ruinous site 30 km/19 miles to the south of the Euphrates in a depression near the 3abal Bi9r, on the ancient desert route from 0im-Salamiya to al-Raa or al- Raba, containing the pilgrimage place of the Arab Saint Sergius, martyred here in the early 4th century, after which it was officially named Sergiupolis in Byzantine times. Archaeological excavations have shown it to be a Roman site, which in the 6th century was embellished with four churches inside the impressive rectangular city walls; among them, the basilica of the Holy Cross, founded in 559 and housing the relics, shows by inscriptions a continuous building tradition until the 12th century. Also inside the city walls, three large cisterns and an ingenious system of supply and distribution of the spring rain water remained famous throughout mediaeval sources. It seems to be referred to in the K. al-#Uyn wa 'l-ad"i: It had been a Byzantine city of ancient foundation with cisterns and a 'water way' (ar li 'l-m") from the margins of the desert (in Fragmenta, 101). Outside the north gate a church or praetorium (?) is connected to the 9assnid prince al- Mun9ir b. al-0ri9 (569-82) by an inscription, and literary sources point to the presence of the 9assanids as well: al-Nu#mn b. al- 0ri9 b. alAyham is mentioned as a governor there, and is said to have repaired the cisterns destroyed by a La9mid and to have constructed a large new one (Yt, ii, 955, 784 (according to the A9br mulk 9assn ); 0amza al-Ifahn, Ta"r9, Berlin 1340/1921-2, 79 (and quoted by Ibn al-#Adm, Bu9ya, i, 114), as well as Abu 'lFid", Mu9taar , Cairo 1325, i, 73, only mention the restoration). Al-Ama# mentions, besides the B. 3afna of 9assn, the B. 0anfa (of Bakr b. W"il, Ibn al-Kalb-Caskel, ii, 156; Kala, |ab"il, i, 312-13) as inhabitants (quoted by al-Bakr, Mu#3am , i, 441); it is he who gives a further name of this, Rufa alZawr" (compare Bakr with Yt, ii, 784, 955; Musil, Palmyrena, 267; the Byzantine name for al-Rufa, |. ml, supposed by Ibn al-#Adm, i, 113, remains unexplained, cf. Ibn 9urrad9bih, 218, B.t. lmiy, 35 mls from it?). Earlier, the

city was within the region of the Tan9 (al- Bal9ur, Fut, 145; I. Shahid, Byzantium and the Arabs in the fourth century, Washington 1984, 405, 465), andafter them?of the Ta9lib (al-abar, i, 2072; Rotter, Die Umayyaden und der zweite Brgerkrieg, Wiesbaden 1982, 131). This may have led to the occasional attribution of al- | [VIII:630b] Rufa to Diyr Mu'ar (e.g. al-abar, iii, 2219). In the #Abbsid period, it came under the control of the B. 9af3a b. #Amr, a branch of the (North Arabian) #Uayl (al-Ama#, in Yt, ii, 284; Caskel, ii, 338; Kala, |ab"il, i, 351); al-Bakr alone connects this Rufa with a verse by al- A9nas b. 9ihb alTa9lib on the extended animal-hunting of the (South Arabian) B. Bahr" (? 9arak un lib un, 56; Yt, ii, 782). Ibn al-#Adm mentions some B. -li of H9im here at the time of Hrn al-Ra9d (Bu9ya, iii, 1467-8, vii, 3446). Administratively, al-Rufa belonged to |innasrn or Aleppo in Umayyad and later #Abbsid times (Ibn al-#Adm, i, 113-14), under Hrn al-Ra9d it was added to the #awimprovince ( Ibn al-Fah , 111), but some transmitters were uncertain about its district (Rufat al-Raa , even al-Rufa in the 3azra , Ibn al- #Adm, v, 2103 with correction; Ibn #Askir, iv, 259; Ibn 9urrad9bih mentions it twice, apparently with its closer neighbours, 74, and together with Blis in the #awim, 75). From Zangid until Mamlk times it seems to have mostly been known as a Christian suburb or in the district of |al#at 3a#bar, which was also called Klnks/ Kallinikosin this way a remark by Barhebraeus, Chronography, ed. Budge, 120, tr. 2111, could be understood: Hi9m died in Rufa of Kallinikos (cf. ibid. tr. 2218; Syriac Chronicle, ad a. 1234, i, 215), and it would fit the reading of an Arabic inscription on the silver goblet from the Rufa treasury suggested by R. Degen, This is what Zayn al-Dr, daughter of ust9Ab Durra, bestowed to the church of the protected |al#at 3a#ba[r], probably meaning al-Rufa (before 1243, in Ulbert, Resafa, iii, 72; for |al#at 3a#bar as a district (a#ml), cf. Ibn al-Dawdr, vii, 283, year 624/1227). Nothing is reported on the Islamic conquest of al- Rufa, and the sources rather convey the impression of its lying in ruins until the building activities of Hi9m (Ibn al-#Adm, i, 113). But it is mentioned as being on the march of the |ays 3af b. 0akm from the 3azra against the B. Ta9lib with their poet al-A9al and their day at 3abal Bi9r in 73/692-3 (al-Bal9ur, Ansb, v, 329; A9n, Bl, xi, 59; Ibn al-#Adm, i, 431 ff.; 9iznat al-adab, Cairo, ix, 4); and also, before 724, a Bishop Abraham of al-Rufa is documented (Degen 70-1, quoting Wright, Cat. of Syriac mss., ii, 796 ff.). The main information on al-Rufa in Muslim sources pertains to the caliph Hi9m (105- 24/724-43), whose residence it became at least in summer and who was buried there. While these sources unanimously locate the residence in or next to this al- Rufa (e.g. Ibn Fa'l Allh al-#Umar, Maslik, Cairo 1342/1924, 332-3), some modern authors have doubted this and identified it with the ruins of |ar al-0ayr al- 9ar [q.v.] (Sauvaget, Remarques sur les monuments omeyyades, in JA [1939], 1- 13). This theory was finally rejected by O. Grabar (City in the desert, 1978, 1-2, 31). It is not very easy to recognise the original tradition within the several additions transmitted by the historians; the news of the death of his predecessor and the regalia were brought to Hi9m at al-Zaytna, where he possessed a small dwelling (duwayra), and he then rode from al-Rufa to Damascus (al-abar, ii, 1467); perhaps because of the unexplained leap, later sources locate either the transmission of the news and the regalia to al-Rufa (al-#Uyn wa 'l-ad"i, 82; Abu 'l-Fid", Mu9taar , Cairo 1325, i, 203) or from al-

Zaytna to al-Rufa (Yt, ii, 784). While O. Grabar still thinks of an identification of al-Zaytna with Kar al-0ayr al- | [VIII:631a] 9ar (City in the desert, 13-14), a hint by Ibn Buln [q.v.] possibly gives the clue to understanding the sequence of residences: Hi9m was fleeing from the mosquitoes on the banks of the Euphrates to al- Rufa (in Yt, ii, 785)this fits the surroundings of the straitened dimensions of his princely residence al-Zaytna, perhaps in the vicinity of his further possessions near al- Raa. Another often-embellished story mentions him avoiding the Syrian cities in favour of al- Rufa because of the plague; like other Umayyads, he fled to the desert (e.g. al-abar, ii, 1737; Ibn al-#Adm, i, 113-14). Hi9m is reported to have reconstructed al- Rufa and erected two castles there (arayn, al-abar, ii, 1738; al-#Uyn wa 'l-ad"i, 101; Ibn Buln, ibid.). Whether their descriptions as possessing a pool and olive yard (al- abar, ii, 1813) or as being luxurious constructions with floral paintings (Ibn al-#Adm, vii, 3044; mani# here evidently not meaning the cisterns, as in ibid., i, 113) go back to eye witnesses or are literary topoi, cannot be decided. Brief archaeological soundings and a survey showed several large Umayyad structures to the south of the city (Otto-Dorn, Ulbert, Sack). The court of Hi9m was magnificent, and must have shown Persian traditions in several respects (cf. the analysis by R. Hamilton, Wald and his friends, Oxford 1988, passim). Whoever was interested in Persian topics among the early #Abbsid caliphs would also refer to accounts of Hi9m's court (e.g. al- Manr, al-abar, iii, 412; al-Mas#d, iv, 47-8, 133-4 = 2234, 2379). Even translations from the Persian seem to have originated from Hi9m's secretaries in al- Rufa, Slim b. #Abd al-Ramn or #Abd Allh and his son 3abala (al-abar, ii, 1750, 1649-50; al-Mas#d, Tanbh, 106, 113; Ibn al-#Adm, ix, 4143; M. Grignaschi, in BEO, xix [1967] 12-13, 24-5, 51- 2). The Arab tradition at his court was upheld by the poets (see, besides the famous competition by al- Farazd, 3arr and al-A9al, descriptions by Ism#l b. Yasr al-Nas", in A9n, iv, 125; 9lid b. -afwn al-Ahtam, in Ibn al-#Adm, vii, 3044; Abu 'lNa3m, in A9n, ix, 78 ff., Ibn al-#Adm, x, 4640). And in one respect he tried to surpass pre-Islamic customs: he himself built the greatest hippodrome (alba) for 3,000 horses here, six bowshots long (A9n, x, 64; al-Mas#d, iv, 41 = 2219; Ibn al#Adm, vi, 2858). Also, the biographies of traditionists at his court contain material on al-Rufa, most famous among them being al-Zuhr (GAS, i, 280-3), Ab Man# #Ubayd Allh b. Ab Ziyd and his grandson Ab Muammad al-0a333 b. Ysuf (Ibn al-#Adm, v, 2100 ff.; al- Sam#n, vi, 135; Ibn #Askir, x, 669-70, iv, 259-60), 9uayf b. #Abd al-Ramn or Ibn Yazd al-0arrn (Ibn al-#Adm, vii, esp. 3265-6). Out of Hi9m's family, his son Sulaymn stayed in al-Rufa until his defeat by Marwn II (al-abar, ii, 1908). Hi9m's tomb and body were desecrated under the first #Abbsid (al-abar, iii, 24989; al-Ya#b, ii, 427-8). The town had an #Abbsid governor in 137/754 (al-abar, iii, 94-5); but apart from occasional visits of a caliph, only one event is mentioned, the sack by the Carmathians at the end of 289/December 902, when the mosque, adjoining the cathedral, was burnt and the #Abbsid defender Sabk al-Daylam killed (al- abar, iii, 2219; Ibn al-#Adm, ii, 946: better, 9ibl al-Daylam; cf. the excavations by D. Sack). Eyewitnesses are few, and are repeatedly quoted in geographical sources: al-Ama#, who mentions merchants, rich and poor, travelling abroad and employing local Bedouins (#arab), a small s with ten shops | [VIII:631b] and textile manufacturing

(Yt, ii, 284-5; still in al-Idrs, ed. Rome, 649, Ibn al- #Adm, i, 113-14, and 033 9alfa, 3ihn-num, 594, without any contemporary observations). Apart from the cisterns and the walls the city is especially famous for its Christian buildings, figuring in the mentioning of its dayr , listed separately by al-Bakr, Yt, and al0imyar. This gave rise to several topoi and anecdotes of nostalgia of the Umayyads and of the monasteries, which do not seem to correspond with reality, as already noticed by Musil (Palmyrena, 268; cf. the story connected to the visit of al-Mutawakkil in 244/858, al-abar, iii, 1436; al-0imyar, Raw' (Beirut 1975), 253; Ibn al-#Adm, i, 114, quoting the K. al-Diyrt by al-9im9; for the clich of the monastery, see L. Conrad, in The quest for understanding, ed. S. Seikaly et alii, Beirut 1991, 271-2). Only the Christian Ibn Buln was interested in the great church, of which he describes the external gold mosaic in 440/1048- 9 (cited in Yt, ii, 785). Judaeo-Arabic inscriptions in one building, dated 1102 and 1127, prove the presence of a Jewish community there (A. Caquot, in Syria, xxxii [1955], 70-4). Ibn 9addd gives an account of the end of habitation in al-Rufa, added to a long quotation from Ibn al-#Adm (ed. A.-M. Terrasse- Edd, 394, tr. 21-2): the Mongols had spared the inhabitants on their march in 658/1260, and after the Mamlk reconquest, a governor was left there until 668/1269-70, when the inhabitants left for Salamiya, 0amt and other places, apparently because of the destruction, which is also archaeologically evident. Since then, the site has been deserted. (C.-P. Haase) Haase, C.P. 4. In Muslim Spain. Munyat al-Rufa, in Spanish Arrizafa, Arruzafa, is the name of the country residence founded by #Abd al-Ramn I (138-72/756-88 [q.v.]) to the north-west of Cordova and to which he gave the name of the Rufa in Syria (see 3. above) founded by his grandfather Hi9m b. #Abd al-Malik. The first Umayyad amr of al-Andalus purchased lands which had belonged to a Berber chief of ri's army, Razn al-Burnus, and built there a palace ( ar ) and gardens. The Arabic sources class the Cordovan Rufa amongst the three most important constructions of #Abd al-Ramn I's reign (the other two being the Grand Mosque and the palace of Cordova). The amr enjoyed living there very much and spent most of his time there. In the course of his residence at al- Rufa, he ordered the execution of three rebels: his nephew al-Mu9ra b. al-Wald, Wahb Allh b. | [VIII:632a] Maymn and #Ay9n b. Sulaymn al-A#rb. Their corpses were dragged as far as Cordova and gibbeted on the banks of the Guadalquivir. In his reign, it became, in some measure, the seat of power, since at the time of his death, his son Hi9m, who happened to be at Mrida, hastened to arrive there before his brother Sulaymn (who was at Toledo and who was disputing with him the right of succession). Moreover, #Abd al-Ramn I made the gardens at al-Rufa the first botanical gardens in the history of al-Andalus. He had planted in his grounds exotic plants, mainly brought from Syria, to which he sent envoys to contact his sisters. The most famous of these fruits imported from the East was the so-called safar pomegranate, whose name is connected with Safr b. #Ubayd al-|il#, from the 3und of al-Urdunn. The latter is

said to have cultivated this variety of pomegranate in the region of Rayya [q.v.], whence it was spread throughout al-Andalus. The origin of the date-palm groves in the Iberian Peninsula is equally attributed, with no real basis in truth, by some Arabic authors to a palm tree at al-Rufa. #Abd al-Ramn I's successors continued the tradition of periods of residence at al-Rufa. It was probably in the reign of #Abd al-Ramn II (206-38/822-52 [q.v.]) that the poet #Abbs b. Firns tried, at al-Rufa, to imitate the flight of birds, dressed in a garment of silk covered with feathers and bearing wings. But above all, it was the amr Muammad (238-73/852-86), known for his zeal as a builder, who enlarged and improved the buildings and the gardens of this residence, where he loved to take rest and where he organised hunting parties. The amr transferred from Cordova to al-Rufa accompanied by his entourage of chamberlains and eunuchs, and thus surrounded by all the splendour of the Umayyad court. He charged his wazrH9im b. #Abd al-#Azz with the construction of a new ma3lis at al- Rufa and provided him with 10,000 dnrs for this. However, the wazrhad the ma3lis built at his own expense. When the work of building was finished, H9im gave back to the amr his 10,000 dnrs and, as a further gesture, prepared for him a sumptuous banquet. The first Umayyad caliph, #Abd al-Ramn III al-Nsir (30050/912-61 [q.v.]), had accompanied, whilst he was still young, his grandfather the amr #Abd Allh during his pleasure sessions at al-Rufa. But his preferred country residence was the munyat al-N#ra. During his reign, al-Rufa is mentioned as a residence for important visitors, such as the North African chief Ayyb b. Ab Yazd Ma9lad b. Kaydd al-^fran in 335/946. Al-Nir's son and successor al-Mustanir (350-66/962-76) preferred above all the munyat Ar" Ni. Between the residential palace complex and the city of Cordova there developed a suburb ( raba' ), equally called al-Rufa, and the nisba from this was borne by some Cordovan scholars, such as the father of al-0umayd [q.v.]. It was there that al-Manr Ibn Ab #$mir, al- Mustanir's 3ib , had his palace built, which he later abandoned for al-munya al- #miriyya, whose exact location is controversial. With the arrival of Berber troop contingents during the rule of al-Manr, the suburb of al-Rufa became the residence of the Ban Mksan b. Zr and the Ban Zw b. Zr, whose houses were destroyed in the course of the troubles during the first reign of Muammad alMahd. Like the other northern suburbs of Cordova, that of al-Rufa suffered the consequences of the fitna and its name disappears from the Arabic sources after the 4th/10th century. As for the munyat al-Rufa , it was first of all despoiled by al-Mahd in 400/1009 during his second reign. | [VIII:632b] The caliph used the contents of the palace, like those of the munyat al-N#raand the royal palace in Cordova, in order to pay his troops and to support the costs of the fight against the rival army of the Berbers on which his rival Sulaymn al-Musta#n depended. In the following year, and in order to ward off the Berber advance, al-Rufa was totally destroyed on the orders of W'i, the military chief of Cordova. He even had the trees in the famous gardens cut down, but shortly afterwards he realised the uselessness of his action from the point of view of the defence of Cordova. Only the name of the Umayyad princes' residence survived. After the Christian conquest, in 633/1236, the land involved was bestowed on the counts of Hornachuelos. Later, a monastery was established on the site. The Rufa of Cordova early became a favoured subject of the court poets. Some very famous verses on its solitary palm tree are attributed to the founder, #Abd al-Ramn I

(according to other version, its author is said to have been #Abd al-Malik b. Bi9r b. #Abd al-Malik b. Marwn). A poem of #Abbs b. Firns on al-Rufa, reproduced by Ibn 0ayyn [q.v.], describes at length its buildings, streams, plants, birds, etc. After its destruction, Rufa became, like Cordova, a poetic subject for the expression of nostalgia for departed splendours (texts of Ibn Zaydn and Ibn Burd, given by Ibn Bassm, Ibn 9n and al-Maar; this same Ibn Zaydn mentions the existence of a garden of marguerites, raw' al-uuwn). In Almohad times, poets still gathered before the site of al- Rufa in order to drink and to recite poetry. The poem of al-|sim b. #Abbd al-Riy develops the theme of ubi sunt. The second al-Rufa in al-Andalus was situated at Valencia, between the town and the sea (the place-name is still preserved under the form Ruzafa, a quarter of the modern town). There is no information on the foundation of this Valencian Rufa. E. LviProvenal was the first to suggest the name of the Umayyad prince #Abd Allh alBalans, son of #Abd al-Ramn I, as its possible founder, at the same time warning of the lack of documentary evidence. This hypothesis has nevertheless been commonly accepted by Spanish and Arab scholars. The Arabic sources stress above all the beauty of the grounds of al-Rufa, considered as the most attractive pleasure-ground in the vicinity of Valencia (together with the munya of Ibn Ab #$mir). The poet Muammad b. 9lib al- Ruf (d. 572/1177) was originally from there and devoted some poems to it. In 480/1087, Castilian troops commanded by Alvar Fez, giving aid to the prince al-|dir, installed themselves at al-Rufa. It was likewise there that king James I of Aragon encamped with his army, besieged the town and conquered it in 636/1238. Like its Cordovan homonym, this Rufa became a literary subject in the poetical or rhymed prose texts written on the occasion of the loss of Valencia (texts of Ibn al-Abbr and Ibn #Amra preserved by al-0imyar and al-Maar).

MAWL$

(A.), pl. mawl , a term of theological, historical and legal usage which had varying meanings in different periods and in different social contexts. Linguistically, it is the noun of place of the verb waliya, with the basic meaning of to be close to, to be connected with someone or something (see LA , xx, 287ff.; TA , x, 398-401), whence acquiring the sense to be close to power, authority > to hold power, govern, be in charge of some office (see Lane, s.v.) and yielding such administrative terms as wl governor, and wilya [q.v.] the function of governor or, in a legal context, sphere of jurisdiction, competence.

I.

IN

THE

|ur"n
AND

TRADITION

Here we find

mawl used in two meanings.

(a) Tutor, trustee, helper.

In this sense, the word is used in the |ur"n , XLVII, 12: God is the mawl of the faithful, the unbelievers have no mawl (cf. III, 143; VI, 62; VII, 41; IX, 51; XXII, 78; LXVI, 2). In the same sense, mawl is used in the 9# tradition, in which Muhammad calls #Al the mawl of those whose mawl he is himself. According to the author of the Lisn , mawl has the sense of wal in this tradition, which is connected with 9adr 9umm (q.v.; cf. C. van Arendonk, De opkomst van het | Zaidietische imamaat, 18, 19). It may be observed that it occurs also in the Musnad of Amad b. 0anbal (i, 84, 118, 119, 152, 330-1; iv, 281, etc.).

(b) Lord.
In the |ur"nit it is in this sense (which is synonymous with that of sayyid ) applied to Allh (II,286; cf. VI,62; X,31), who is often called

Mawln our Lord in Arabic literature, Precisely for this reason, in Tradition the slave is prohibited from calling his lord mawl (alBu9r , 3ihd , bb 165; Muslm , Alf , trad. 15, 16). It is not in contradiction to this prohibition that Tradition frequently uses mawl in the sense of lord of a slave, e.g. in the well-known ad9 Three categories of people will receive a two-fold reward ... and the slave who fulfils his duty in regard to Allh as well as to his lords (alBu9r , #Ilm , bb 31; Muslim, Aymn , trad. 45). Composition of mawl and suffixes are frequently used as titles in several parts of the Muslim world, e.g. mawly (a) (mulay), my lord (especially in North Africa and in connection with saints); mawlaw ( mull [q.v.]), lordship (especially in India and in connection with scholars or saints). The term

mawl is also applied to the former lord ( patron) in his relation to his freedman, e.g. in the tradition Who clings to a [new] patron without the permission of his [legal] mawl , on him rests the curse of Allh (alBu9r , 3izya , bb 17; Muslim, #It , trad. 18, 19). (A.J. Wensinck*)

II.

IN HISTORICAL

AND

LEGAL USAGE

Here the meaning of mawl , a person linked by wal" (proximity) to another person, similarly known as mawl , varies according to the context in which it is found. In pre-Islamic poetry, it usually designates a party to an egalitarian relationship of mutual help, that is, a kinsman ( ibn #amm ), confederate ( alf ), ally or friend, a meaning also attested in the |ur"n (IV, 37; XIX, 5; XLIV, 41) and some later literature (P. Crone, Slaves on horses, Cambridge 1980, appendix VI). In later literature, however, it more commonly designates a party to an unequal relationship of assistance, that is a master, manumitter, benefactor or patron on the one hand, and a freedman, protg or client on the other. This sense too is attested in the |ur"n , where the typical mawl is God (VIII, 41; XXII, 78, and passim; cf. XXXIII, 5, where it means protg). Applied to the inferior party in an Islamic context,

mawl almost always means a client of the type recognised in early Islamic law, though its use in the opposite sense was more flexible. The Islamic world has of course known many other types of client, but not by this name. The client recognised in early law was a non-Arab freedman, convert or other newcomer in Muslim society. Since non-Arabs could only enter this society as clients, mawl came to be synonymous with non-Arab Muslim, and the secondary literature usually employs the word in this sense (though the lexicographers fail to list it, cf. LA and TA , s.v. w-l-y). It is also with non-Arab Muslims that this article will be concerned.

1. Pre-IslamicArabia.
The Islamic institution of wal" is generally assumed to be of Arabian origin (cf. Goldziher, Muh. Stud., i, ch. 3; J. Juda, Die sozialen und wirtschaftlichen Aspekte der Mawl in frhislamischer Zeit, Tbingen 1983), but this is scarcely correct. Leaving aside | foreign merchants and colonists under imperial protection, the non-Arab population of pre-Islamic Arabia consisted of Jews, slaves and freedmen of African and Middle Eastern extraction, half-bred descendants of colonists, and presumably also ethnic and occupational pariah groups of the type attested in modern times ( |awwila , Baydr , -ulubbs , etc.). There is no reason to doubt that all were known as mawl in the sense of kinsmen, in so far as they were free and came under Arab protection (cf. the modern use of the word a9 brother), but the question is, what this implied. Are we to take it that all non-Arabs were individually assigned to Arab patrons and acquired partial membership of Arab tribes through them, having no social organisation of their own? Or did they form social groups of their own, so that they were collectively placed under the protection of Arab tribes in which they acquired no membership at all, merely becoming their satellites? The first solution is that enshrined in Islamic wal"

, but it is the second which is attested for Arabia. Thus it is well known that the Jews of Arabia formed tribal groups of their own. In fact, Jewish tribes were sometimes strong enough to escape Arab protection altogether (and thus also the status of mawl ). But this was hardly the common pattern. The Jews of Fadak, for example, paid protection money to Kalb (M. J. Kister, On the wife of the goldsmith from Fadak and her progeny, in Muson, xcii [1979], 321); the Jews of Wd 'l|ur similarly paid what would nowadays be known as 9uwwa to Arab overlords (alBakr , Mu#3am m ista#3am , ed. F. Wstenfeld, Gttingen 1867-7, i, 30); and those of Ya9rib were reduced to client status by the Aws and 9azra3 [q.vv.] some time before the rise of Islam (J. Wellhausen, Skizzen und Vorarbeiten, iv, Berlin 1889, 7ff.). Naturally, client status weakened the tribal organisation of the Jews; the same is true of modern pariah groups. But the Jewish tribes were not dissolved, nor were the Jews assigned to individual patrons: clientage was a relationship between groups. Similarly, the Arabised descendants of the Persian workmen and prostitutes of Ha3ar clearly formed a quasi-tribal group of their own under #Abd protection, for all that they adopted the nisba of their protectors (alabar , i, 986). The question is thus, whether freedmen were treated differently? On this point, the evidence is less conclusive. Continuing relations between manumitter and freedman were clearly common, and there is evidence that the pre-Islamic Arabs practised manumission with what the Greeks called paramon , a clause requiring the freedman to stay with the manumitter for a specified number of years or until the latter died (P. Crone, Roman,

provincial and Islamic law, Princeton 1987). But continuing relations between manumitter and freedman in no way imply that the latter was incorporated in the manumitter's kin; the paramonar freedman only became a member of his household, and then only for a specified time; and one would in general have thought the pre-Islamic Arabs as reluctant to contaminate their agnatic kin with non-Arab freedmen as were their descendants in more recent times. The freedmen (and indeed slaves) of modern Arabia formed lineages of their own, and it was through them, not through their manumitters, that they acquired their rights and duties in respect of marriage, succession and vengeance; and it was to the manumitters' tribe as a political entity that they stood in a relationship of dependence, paying it military assistance and/or 9uwwa (J. L. Burckhardt, Notes on the Bedouins and the | Wahbys, London 1830, i, 181f.; A. Jaussen, Coutumes des Arabes au pays de Moab 2, Paris 1948, 125f.; cf. A. Musil, The manners and customs of the Rwala Bedouins , New York 1928, 276). It should be noted that non-Arabs were not generally affiliated as confederates. 0ilf [q.v.] was a mechanism for the partial or total incorporation of foreigners, individually or as groups, and it is frequently regarded as ancestral to Islamic wal" . But ilf was used only for Arab foreigners, or more precisely for foreigners with full tribal status. The Jews thus qualified on occasion, as did others such as the Abn" [q.v.] but only to the extent that they escaped client status (though there are admittedly ambivalent cases); non-Arab freedmen never qualified (cf. Goldziher, Muh. Stud., i, 106). The alf is thus irrelevant to the question. By the same token, so is most of pre-Islamic poetry. The vast majority of mawl mentioned in this poetry (where the word is exceedingly common) are Arabs from whom help of one sort or another could be expected: real, fictitious or metaphorical kinsmen (cf. ibid., 105 n.; many more examples are given in Crone, Roman, provincial and Islamic law). Confederates are occasionally singled out as mawl 'lyamn , kinsmen by oath, as opposed to mawl

'lwilda or arba , kinsmen by birth/kinship (G. W. Freytag, Hamasae carmina, Bonn 1828, 187; C. Lyall, ed. and tr., The Mufa''alyt , Oxford 1918-21, no. 12:3; Ibn Hi9m , 467; Nbi9a al3a#d , Dwn , ed. and tr. M. Nallino, Rome 1953, no. 12:41). But the distinction is usually quite neutral, and though confederates could obviously find themselves in a subservient position so that the dividing line between them and client groups was blurred (as it is in Nbi9a; cf. Juda, Aspekte, 14-15) they were followers of a type quite different from that of non-Arab clients. It is only when the poets distinguish between amm (or ar ) and mawl that the latter would seem regularly to encompass servile and nonArab elements, and the same is perhaps true when they speak of tails and fins of tribes (cf. Goldziher, Muh. Stud., i, 105-6); but these passages are perfectly compatible with the proposition that freedmen and other non-Arabs should be seen as members of satellite groups rather than as persons assimilated to the tribe by affiliation (Goldziher).

2. The
R9idn calips and the Umayyads. With the conquests, the Arabs found themselves in charge of a huge non-Arab population. Given that it was non-Muslim, this population could be awarded a status similar to that of clients in Arabia, retaining its own organisation under Arab control in return for the payment of taxes [see 9imma

]. But converts posed a novel problem in that, on the one hand they had to be incorporated, not merely accommodated, within Arab society; and on the other hand, they had forgotten their genealogies, suffered defeat and frequently also enslavement, so that they did not make acceptable alf s; the only non-Arabs to be affiliated as such were the 0amr" and Aswira , Persian soldiers who deserted to the Arabs during the wars of conquest in return for privileged status (alBal9ur , Fut , 280, 373). It was in response to this novel problem that Islamic wal" was evolved, presumably by the authorities and at an early stage, though nothing can be said with certainty about its emergence.

(a) Early Islamic

wal" . What follows is based on a collation of information in classical law (below, section 5), ad9 and historical sources. All non-Arabs who aspired to membership of Arab society had to | procure a patron (an upper mawl in the terminology of the lawyers). Freedmen automatically acquired a patron in their manumitter, unless the latter renounced the tie. Free persons and those freed without wal" had to acquire one by agreement. Contractual clientage was known as wal" of muwlt (inclination, attachment, the term generally used in 0anaf literature), tib#a (following, alabar , ii, 1853), 9idma

(service, A9n , xii, 48) or islm , conversion: whoever converted at the hands of another became a client of the other according to a famous ad9 (man aslama #al yad 9ayrih fa-huwa mawlhu ). Wal" was a solution to the problem of affiliating non-tribesmen to a tribal society, and though most such non-tribesmen were clearly converts, conversion was not necessary for the legal validity of the tie. A fair number of non-Muslim clients, both freeborn and freed, are attested (Crone, Slaves, n. 358; alabar , i, 3185). From the point of view of the authorities, the main role of the patron was to provide the client with an #ila [q.v.]. The patron and his agnates were required to pay compensation ( #al , diya [q.v.]) for bodily harm inflicted by the client, to the extent that the latter had no agnates of his own. Refusal to pay seems to have been a common problem (alKind , The governors and judges of Egypt , ed. R. Guest, Leiden and London 1912, 333f., 335f.). Conversely, if the client was killed they were entitled to blood-money in compensation for him (cf. alKind , op. cit., 333f.). In return for his obligations, the patron acquired a title to the client's estate, though not an indefeasible one. The classical rules of exclusion are not, on the whole, favourable to the patron, but it is not known whether they applied in pre-classical law. The role of the client, on the other hand, was purely passive. He did not contribute to blood-money payable for damage inflicted by the patron, nor did he share in the receipt of blood-money if the patron was killed or acquire a title to his estate. He was not formally obliged to render obsequium. The patron by manumission could make over the patronateby manumission could make over the patronate to a third party by sale, gift or bequest, and the parties to the contractual relationship could terminate theirs (in which case the client would

need a new patron). If not, it would pass to the descendants of the two parties in perpetuity, though it would lose legal (but not necessarily social) importance as the client acquired agnates in Islam, Muslim clients could and frequently did have clients of their own. From the point of view of the client, the main role of the patron was to provide him with access to a privileged society, and in practice the patron's rewards were far greater than those provided by the authorities. For one thing, the patron might qualify his grant of freedom with stipulations requiring the freedmen to pay regular sums, gifts or labour services to himself or a third party for a specified period, or reserving part or all of the freedman's estate for himself regardless of the presence of heirs (practices condemned in early ad9 ; cf. Crone, Roman, provincial and Islamic law). For another thing, freedmen were notoriously loyal. They would stay by their manumitters in danger (alabar , i, 3001f., ii, 1959; cf. Ch. Pellat (tr.), The life and works of Ji , London 1969, 215, 260), share their sorrows (cf. alabar , ii, 384), assist them in need (though for one who refused, see A9ni 3 , xvi, 107), attend to them in death (alabar , i, 3046, ii, 1751) and seek to avenge them (ibid., ii, 1049; alBal9ur , Ansb , v, 338). Concerning the services provided by contractual clients, we are less well informed. Like freedmen, they clearly went to swell their patrons' retinues, both military and | civilian (M. J. Kister, The Battle of the 0arra , in Studies in memory of Gaston Wiet, Jerusalem 1977, 44; Crone, Slaves, 53f.), and they must have performed other types of 9idma too. But patrons preferred freedmen to mawl tib#a (alabar , ii, 1852f.).

(b) The

mawl in Umayyad society. No formal disabilities seem to have attached to the status of client in public law. In principle, clients were in a dependent position only vis-vis their patrons, enjoying the same rights and duties as other Muslims in society at large (lahum m lan wa#alayhim m #alayn ). In practice, of course, there was massive prejudice against them. The Arabs generally equated them with slaves, partly because they were unwarlike agriculturalists ( 0assn b. 9bit , Dwn , ed. and tr. H. Hirschfeld, Leiden and London 1910, no. 189:8; cf. 9u 'l-Rumma, Dwn , ed. C. H. H. Macartney, Cambridge 1919, no. 29:48), partly because they had suffered spectacular military defeat (O men, do you not see how Persia has been ruined and its inhabitants humiliated? They have become slaves who pasture your sheep, as if their kingdom was a dream, Nbi9a al3a#d , no. 8:12f.), and, finally, because the majority of clients were freedmen. Christian and Muslim sources are agreed that the Arabs took enormous numbers of prisoners-of-war during the wars of conquest. He killed and took prisoners is the standard expression for the activities of a conqueror in the Muslim ones ( 9alfa b. 9ayy , Ta"r9 , ed. S. Zakkr , Damascus 1967-8, i, 127, 163, 168, 171, 178, 237, 242; Sebeos, Histoire d'Hraclius, tr. F. Macler, Paris 1904, 100f., 110, 146; Bar Penkaye in A. Mingana, ed. and tr., Sources syriaques, Leipzig [1907], *147 = *175; Michael the Syrian, Chronique, ed. and tr. J.-B. Chabot,

Paris 1899-1910, iv, 417 = ii, 422); and the usual fate of prisoners-ofwar was enslavement. Moreover, many localities were required by treaty to supply a specified number of slaves to the Arabs, such as 30,000 or 100,000 once and for all or a smaller number annually (alabar , ii, 1238, 1245, 1246, 1321, 1329, 1667; alBal9ur , Fut , 208; C. E. Bosworth, Sstn under the Arabs, Rome 1968, 17; M. Hinds and H. Sakkout, A letter from the governor of Egypt to the King of Nubia and Muqurra concerning Egyptian-Nubian relations in 141/758, in Studia Arabica et Islamica I. #Abbs , ed. W. alQ' , Beirut 1981; Juda, Aspekte, 64-5). Victims of war and their descendants thus outnumbered freeborn clients, and slaves was the standard term of abuse for a client of any kind (alabar , ii, 1120, 1431; alBal9ur , Ansb , iva, 247, v, 356; A9ni 3 , xvi, 107). Naturally, such men were subject to numerous disabilities in the society of their conquerors. Slaves and clients were vile ( TA , s.v. h-w-y), and clients of clients were the most miserable persons to walk on earth (alBal9ur , Ansb , ivb, 10; cf. al-Farazdak in LA and TA , s.v. w-l-y). Thus a mawl who married an Arab woman risked both penalties and the dissolution of his marriage ( A9ni 3 , xvi, 106f.; cf. also Goldziher, Muh. Stud., i, 128ff.), though such marriages are unlikely to have been officially prohibited (below, (d)). A mawl 's life was felt to be worth less than that of an Arab, so that an Arab should not be killed in retaliation for a client (alBal9ur , Ansb

, iva, 220), while conversely, retaliation inflicted upon a client failed to compensate for harm suffered by an Arab (alabar , ii, 1849). A mawl was not worth avenging (though |utayba b. Muslim [q.v.] invoked a moral obligation to do so in an unusual situation where the client was a Transoxanian prince, alabar , ii, 1249); it was by way of insult to their victim that Arab avengers would claim to have killed so-and-so for a mere mawl or slave (al- | Azd , Ta"r9 alMawil , ed. #A . 0abba , Cairo 1967, 62; alDnawar , alA9br aliwl , ed. V. Guirgass, Leiden 1888, 350; alabar , i, 3276, ii, 710). Above all, mawl were felt to be unsuitable for positions of authority, such as that of prayer-leader, judge or governor (Goldziher, Muh. Stud., i, 109 n., 116; . A. al#Al , alTanmt ali3tim#iyya wa 'litidiyya fi 'lBara 2 , Beirut 1969, 96f.; Ibn |udma , al-

Mu9n , ed. T. M. alZayn and others, Cairo 1968-70, vii, 33; Juda, Aspekte, 182-4); as late as 133/750-1 the inhabitants of Mawil objected to the appointment of a mawl as governor (alAzd , Mawil , 146). Confronted with this prejudice, non-Arab Muslims initially made their careers mainly in the service of their patrons, and the tie between patron and client remained important throughout the Umayyad period; but their education, skills and sheer number was such that they rapidly achieved positions of influence in their own right.

Civilian careers.
Many non-Arabs had worked as labourers, craftsmen, traders and shopkeepers while still slaves, and many continued in such occupations on their manumission (cf. Juda, Aspekte, 109ff.) But we hear more about those who remained members of their patrons' households, especially those of governors, who would employ them as messengers, spies, executioners and other agents of various kinds (alabar , i, 2138, ii, 40, 246f., 268, 1276, 1649). Governors and caliphs alike also employed their own freedmen as 3ib s [q.v.], see the information at the end of each reign in 9alfa , Ta"r9 , and alMas#d , Tanbh , 284ff.; alBal9ur , Ansb , v, 172; alabar , ii, 1650). But Mawl

played a more important role as administrators. Some administered their patron's estate (alBal9ur , Fut , 8; alabar , ii, 1734; Juda, Aspekte, 119-20); a great many administered the empire. Thus a secretary ( ktib [q.v.]) was usually a non-Arab, sometimes a non-Muslim (alMas#d , Tanbh , 302, 307, 312; alBal9ur , Ansb , xi (= Anonyme arabische Chronik, ed. W. Ahlwardt, Greifswald 1883, 343), but more commonly a convert (cf. al3ah9iyr , Kitb alwuzar" wa 'lkuttb , ed. M. alSa" and others, Cairo 1938, 61) or a freedman; being appreciated for their skills rather than their personal loyalties, such men were employed not only by their own patrons (e.g. al3ah9iyr , 54, 64; alabar , ii, 837ff.) but also by others (e.g. alMas#d , Tanbh , 302, 312, 316, 317, etc.; alabar , loc. cit.; al3ah9iyr , 66). The various sections of the dwn [q.v.] in a particular province, or indeed the entire dwn

, were commonly headed by mawl (alabar , ii, 837ff., 1649, 1650; al3ah9iyr , 42, 69; alBal9ur , Ansb , ivb, 83, 123). Moreover, mawl soon came to be appointed as fiscal governors, sometimes on behalf of top Arab governors, but frequently in their own right. We hear of such appointments in Mecca and Medina (Ibn 0abb , alMuabbar , ed. I. Lichtenstdter, Hyderabad 1942, 379). Transoxania (alabar , ii, 1253, 1421, 1509), and #Ir , where non-Arab fiscal governors played a major role in Arab politics (ibid., 1282f., 1648; al3ah9iyr , 42f., 49, 63; Ibn #I9r , i, 39; alBal3ur , Ansb , ivb, 123), as well as Egypt, where three non-Arabs rose to the position of effective governor thanks to their control of the taxes, that is Usma b. Zayd alSal ,a mawl of Mu#wiya (al3ah9iyr , 51) under Sulaymn , #Ubayd (or #Abd )

Allh b. al0abb [q.v.], a mawl of Sall under Hi9m , and #Abd al-Malik b. Marwn b. Ms b. Nuayr ,a mawl of La9m (or the Umayyads) under Marwn II (alKind , Governors, 93). Lesser administrative jobs were also in the hands | of mawl (alabar , ii, 1845; in general, see also Juda, Aspekte, 119-20). Outside the administration, non-Arab Muslims rapidly came to dominate the world of scholarship Mawl , mainly descendants of captives, played a crucial role in the formation of the Islamic faith [see ALasan
AL-

bar ], Islamic law [see ab anfa

, ,
AL-

awz# ws ],

|ur"nic studies [see ab #ubayda ] and the Prophet's biography [see IBN is ], as well as in the collection of pre-Islamic poetry [see ammd
AL-

rwiya ]. They also produced some notable poets [see ba99r B. BURD ]. Contemporaries were well aware of the preponderance of mawl in scholarship (Goldziher, Muh. Stud., i, 114f.), and in the second half of the Umayyad period it was usually mawl who were accorded the role of tutor to the caliph's children (Ibn 0abb , Muabbar , 476ff.; alabar , ii, 1741). In the same period they also began to receive appointment as judges (al#Al , Tanmt , 96 n).

Careers in the army.

Leaving aside the Persian soldiers enrolled during the wars of conquest, non-Arabs initially entered the army only as private servants of their patrons. Every soldier had a number of slaves and freedmen registered under him in the dwn ; some governors acquired sizeable bodyguards of slaves and freedmen; and towards the end of the period it was common for governors and generals to have semi-private retinues of freedmen, contractual clients and other protgs (Crone, Slaves, 38, 53, 198f.). The aras , or palace-guard, of the caliphs and their governors also seems usually to have been composed of and headed by mawl , though not necessarily mawl of the employer (cf.

9alfa , Ta"r9 , at the end of each reign; alabar , ii, 1384, 1499, 1569, 1650; alBal9ur , Ansb , v, 172f.). But already Mu#wiya placed a mawl in command of troops in an expedition against Byzantium ( 9alfa , Ta"r9 , i, 198, cf. 102), an example followed by #Abd al-Malik (alabar , ii, 1487; alBal3ur , Fut , 160f.); and the Second Civil War decisively undermined the Arab monopoly of military power. Non-Arabs being available, everybody made use of them: thus Mu9tr [q.v.]; adherents of Ibn al-Zabayr (Kister, Battle of the 0arra , 44f.); and the Umayyads themselves (Crone, Slaves, 198; 9alfa , Ta"r9 , i, 335; alBal9ur , Ansb , v, 356ff.), with the result that a mawl became military governor of Medina for #Abd al-Malik (alabar , ii, 834, 852, 854). Thereafter, non-Arabs were regularly admitted as soldiers in their own right (cf. the rich documentation in Juda, Aspekte, 120ff.). Some were placed in special corps for mawl with native skills of their own, such as the Berber

Waddiyya or the Indian |niyya . Others joined the ordinary army, mawl divisions being set up for their reception. According to a Kitb Mawl ahl Mir cited by Yt (i, 734), there were mawl divisions in the Egyptian army already at the end of the First Civil War, when a freedman from Balhb was made #arf [q.v.] of the mawl of Tu3b . But the #arf in question belongs to the end of the Second Civil War (cf. alKind , Governors, 51), and it is only after the Second Civil War that such divisions are regularly mentioned, be it in Egypt, 9ursn or elsewhere (Crone, Slaves, 38). There were mawl in the Syrian army too, for all that mawl divisions are not mentioned here after #Abd al-Malik (cf. ibid., 274); we are incidentally given to understand that the Syrian troops in Egypt in 125/742-3 included mawl ahl 0im (alKind , Governors, 83), and that those brought to Spain by Bal3 b. Bi9r [q.v.] included mawl

of the Umayyads, clearly among others (Lvi-Provenal, Hist. Esp. mus., i 98; | cf. also Juda, Aspekte, 84-5). Mawl also participated in the revolt against alWald II, but whether as regular soldiers or private retainers is not clear (alabar , ii, 1800, 1806f., 1809). The proportion of non-Arabs to Arabs in the late Umayyad armies cannot be estimated. In 96/714-15 the mawl of 9ursn were numerically on a par with Bakr b. W"il [q.v.], numbering 7,000 out of a total of 54,000 (alabar , ii, 1290f.). They must have become more numerous thereafter. Contrary to what is often stated, the Umayyads did not try to keep non-Arabs out of the army (cf. below); and the attempt to show that their numbers decreased after 96, at least in 9ursn , rests on a misreading of the sources (M. A. Shaban, The #Abbsid revolution, Cambridge 1970, 113, 115 and passim: the figure of 1,600 given for the rearguard in the Battle of the Pss scarcely refers to mawl , and at all events, not to their total number; we are explicitly told that the governor had dispersed his troops before the battle and that two mawl commanders, one i charge of 10,000, were among those who had gone elsewhere, cf. alabar , ii, 1532f., 1538, 1549, 1551). In fact, mawl must have been particularly numerous in 9ursn , where they are constantly mentioned and their participation in all military activities is taken for granted, be it in campaigns (ibid., 1023, 1080, 1225, 1447, 1478, 1485, 1516, 1518, 1538, 1630f., cf. 1184f.) feuds (ibid., 1856), or revolts and their suppression (ibid., 1582, 1589, 1605, 1920f., 1926, 1933, cf. 1163, 1867, 1918f., 1922). Once admitted to the army, mawl

began to receive both military and fiscal governorship. Thus the governor and general of |innasrn in 75/694-5 was a mawl of #Abd al-Malik (alBal9ur , Fut , 188), and we incidentally learn that the governor of Ba#labakk in 126/743-4 was also a client of the Umayyads (alabar , ii, 1790). The 3azra received its first non-Arab governor under #Umar II and/or Yazd II in Maymn b. Mihrn ,a mawl of B. Nar (or Azd) who had been tutor to #Umar II's children and who was later to command the Syrian army against Byzantium (alAzd , Mawil , 37; Ibn 0abb , Muabbar , 478; alabar , ii, 1487). Mawl appear as sub-governors in Transoxania from the time of |utayba onwards (alabar , ii, 1206, 1448, 1694f.), and they regularly ruled North Africa. Abu 'lMuh3ir

, who administered North Africa for ten years under Mu#wiya , was appointed by his own patron, the governor of Egypt (alabar , ii, 94; alBal9ur , Fut , 228). Similarly, Ms b. Nuayr [q.v.], a mawl of disputed origin who enjoyed the protection of #Abd al#Azz b. Marwn [q.v.], #Abd al-Malik's governor of Egypt, was appointed by his protector. He in turn appointed a mawl of his own, ri b. Ziyd [q.v.] as commander in the conquest of Spain (alBal9ur , Fut , 230f.; Ibn #I9r , i, 39f., 43). Thereafter, a succession of mawl were appointed by the caliphs themselves. Sulaymn chose Muammad (or #Abd Allh ) b. Yazd ,a mawl of the Umayyads ( Ibn

#I9r , i, 47). #Umar II appointed Ism#l b. #Abd Allh b. Abi 'lMuh3ir ,a mawl of Ma9zm who had worked as tutor to his children (alBal9ur , Fut , 231; Ibn 0abb , Muabbar , 476). Ism#l was followed by Yazd b. Ab Muslim, a freeborn mawl of al0a333 's [q.v.] appointed by Yazd II (al3ah9iyr , 42; alBal9ur , Fut , 231); and when he in turn was murdered for his harsh policies, a mawl of the Anr by the name of Muammad b. Yazd

was elevated from the troops by popular choice and, subsequently, caliphal appointment (alabar , ii, 1435; al3ah9iyr , 57; according to alabar , | he was identical with the previous governor of that name). Hishm appointed #Ubayd Allh b. al0abb , mawl of Sall , to Egypt, North Africa and Spain, and #Ubayd Allh neatly reversed the pattern by appointing his own patron to Spain ( Ibn #I9r , i, 51ff.; alBal9ur , Fut , 231).

Local influence.
From the Second Civil War onwards, mawl begin to appear in Muslim society and politics as men of local importance. Thus 0umrn b. Abn , a captive from #Ayn al-Tamr [q.v.] and former secretary of #U9mn 's, joined the pro-Umayyad 3ufriyya in Bara

in the Second Civil War and briefly achieved the position of governor there (Ch. Pellat, Le milieu barien et la formation de @i , Paris 1953, 270, 278; cf. 268, probably a doublet). Similarly, the wealthy family of #Abd Allh b. Hurmuz, mawl of the Umayyads and directors of the Baran dwn al3und from the time of al0a333 onwards, are said to have been very influential in this city (alBal9ur , Ansb , ivb, 123). In 9ursn , the B. -uhayb , mawl of the B. 3adar , enjoyed a position of eminence among the Rab#a and intervened in the local feuds during the Second Civil War (alabar , ii, 491ff.). In the Third Civil War, a certain Murib b. Ms , mawl of the B. Ya9kur , emerged as #am aladr in Frs

, where he took to expelling Umayyad governors (ibid., 1976f.). And of 0ar9 ,a mawl of 9uz#a who joined the #Abbsid revolution, we are told that he was #am ahl Nas (alDnawar , A9br , 341). All in all, the mawl must thus be said to have penetrated Arab society extremely fast. They played a predominant role in most activities outside the world of politics, controlled the civil administration almost from the start and made their presence felt in military politics within a generation of the conquests. Certainly, the Arabs retained their control of military politics until the end of the Umayyad period, most governorship and other politically influential posts being allocated to them; but the popular image of mawl as an excluded people passively exposed to Arab whim and prejudice is quite wrong. Given the cultural and numerical discrepancy between the conquerors and their subjects, it is not really surprising that the latter acquired influence so fast: the conquerors simply could not govern without non-Arab help, as later 9u#bs were to point out; indeed, they needed their advice even in matters of food and drink (al-l , Adab alkuttb , ed. M. B. alA9ar , Cairo 1341, 193). What is surprising is that the Arab integument of Muslim society withstood the pressure.

(c) Fiscal status.

The secondary literature generally associates

mawl with fiscal disabilities. Thus all the Umayyads other than #Umar II are said wrongfully to have collected poll tax ( 3izya [q.v.]) from converts and to have refused them registration in the army, being assisted in this by the leaders of the non-Muslim communities who had an interest in penalising conversion (J. Wellhausen, The Arab kingdom and its fall, Calcutta 1927, ch. 5; D. C. Dennett, Conversion and poll-tax in early Islam , Cambridge Mass. 1950; H. A. R. Gibb, The fiscal rescript of #Umar II, in Arabica ii/1 [1955], 1-16). But this view is in need of modification. On the one hand, the vast majority of mawl were freedmen and descendants of freedmen who had never paid any poll-tax at all; and free converts who acquired a respectable patron also escaped fiscal disabilities (Crone, Slaves, 52). The conventional picture applies only to a special type of convert, the fugitive peasant. On the other hand, the Umayyad treatment of such mawl should not be seen as a violation of the law; the law on this question was what the | Umayyads themselves decreed. The fact that the classical rules of taxation have been attributed to #Umar I does not mean that they in fact existed so early (cf. K. Morimoto, The fiscal administration of Egypt in the early Islamic period, Kyoto 1981); and they would not have helped the Umayyads even if they had. Thus the classical rules lay down that the convert should be freed of his poll-tax, but not of his land-tax ( 9ar3 [q.v.]). In the Umayyad period, however, no villager converted without leaving his villages and thus also such land as he might possess: the distinction between a 9imm poll-tax and a religiously neutral land-tax was quite irrelevant. Converts invariably left their land because the attraction of conversion lay in its promise of access to the ranks of the conquerors: converting without joining these ranks would have been pointless and, locally, extremely unpleasant. To convert was thus to make a hi3ra [q.v.] from the land of unbelief to the land of Islam, that is, the garrison cities, as #Umar II explained (Ibn #Abd al0akam , Srat

#Umar b. #Abd al#Azz , ed. A. #Ubayd , Beirut 1967, 93f.); and the problem confronting the Umayyads was not whether converts should be freed of this or that part of their fiscal burden, but whether they should be allowed to make their hi3ra and thus escape their fiscal burden altogether. Naturally, Umayyad policies varied. #Umar II accepted such converts (his problem was thus the fate of their land, cf. ibid.). But most Umayyads adopted a harsh policy vis--vis fugitives regardless of whether they claimed conversion or not (cf. Morimoto, Fiscal administration , 120ff.; Crone, Slaves, 52), resettling them in their villages or at best allowing them to stay where they were in return for continuing fiscal liability. Three points follow from this. First, the fugitives in question were required to pay all their customary taxes, whatever these might be, not merely a religiously neutral land-tax: having been denied access to the conquest society, they were not Muslims at all in the eyes of the authorities; and all their taxes, not merely the poll-tax, were regarded as 9imm taxes at this stage. Secondly, such converts were not eligible for membership of the army. Naturally, when #Umar II decided to admit them to Muslim society, he admitted them to the army as well; but the fact that others refused them entry to the army does not mean that mawl as such were kept out of the dwn . The numerous mawl who fought in the army without pay were runaway peasants who were still being held to their fiscal obligations and who fought as volunteers in the hope of being picked up by a patron, as is clear from the story of Ynus b. #Abd Rabbih, who acquired a patron in Nasr b. Sayyr (Crone, Slaves, 52f.). Thirdly, it was such converts, not converts in general, who were open to penalisation by the leaders of the Christian, Jewish and Zoroastrian communities; had their

hi3ra been accepted, they would of course have been placed under Muslim administration (cf. alabar , ii, 1688; Dennett, Conversion, 124ff.). It should be clear that the entire problem of the fiscal treatment of converts was a problem for 9imm s, not for mawl . It was 9imm s who were frustrated by the closure of the gate of the hi3ra (cf. Ibn #Abd al0akam , Sra , 94); a mawl , by contrast, was somebody already admitted. It was accordingly also 9imm s not mawl , who would enrol in the service of anyone who promised them taxrelief. Whoever converts is freed from khar3 is a slogan on a par with that addressed to slaves, whoever joins us is free (cf. Crone, Slaves, nn. 399-400, 647). Both are addressed, usually by rebels, to malcontents outside | free Muslim society, not to oppressed elements within in; it was only on responding to such slogans that the non-Muslims and slaves in question acquired client status.

(d) The issue of assimilation.

The Umayyads are generally credited with an active policy of discrimination against non-Arab Muslims (cf. most recently, M. A. Shaban, Islamic history, i, Cambridge 1971). This impression arises largely from their treatment of fugitive peasants. But though they discouraged flight from the land and no doubt shared the common prejudice against non-Arabs, they do not seem to have had an actual policy of discrimination against accredited members of Muslim society. Practically every Umayyad caliph is known to have appointed a

mawl governor. Al0a333 , a man notorious for his harsh treatment of runaway peasants, appointed the first non-Arab judges in #Ir (al#Al , Tanmat , 96 ); he also appointed a mawl to his 9ura (alYab , ii, 328), an unusual step (for a later example, see alKind , Governors, 70). #Umar II, a caliph famous for his encouragement of conversion, is said to have disapproved of intermarriage between Arabs and mawl (al#Al , Tanmt , 96 ). But no prohibition of such unions has been recorded, and mawl are known to have married female relatives of other Umayyads (alBal9ur , Ansb , iva, 247; alabar , ii, 1420); the right to repudiate or endorse such unions presumably rested with the guardians of the bride (cf. also Juda, Aspekte, 178ff.). The fact that mawl formed quasi-tribal groups of their own in the army reflects the tribal organisation of this army, not a policy of segregation; and it was the Umayyads themselves, not Ab Muslim, who abolished this organisation in 9ursn (Cron, Slaves, 38). The belief that

mawl were relegated to the infantry rests on a failure to distinguish mawl inside Muslim society from runaway peasants trying to enter it: governors and generals such as ri b. #Amr , Dnr b. Dnr , Ms b. Nuayr , ri b. Ziyd or Ibrhm b. Bassm were scarcely disqualified from riding horses. (See also alabar , ii, 1118f., 1599; Lvi-Provenal, Hist. Esp. mus., i, 98.) Conversely, the enemies of the Umayyads do not seem to have regarded assimilation as an issue of political opposition. No rebel of the Umayyad period mentioned the treatment of non-Arab Muslims in his proclamations, and the belief that the #Abbsids [q.v.] regarded assimilation as their prime objective is gratuitous. Obviously, assimilation accelerated on the fall of the conquest society, but scarcely as a result of official encouragement. The legitimacy of favouring Arabs over non-Arab Muslims in matters of appointment, vengeance and marriage clearly did become an object of debate in the Umayyad period, as did the refusal to accept runaway converts as Muslims, and in principle the question could have been taken up by politicans. In practice, however, the debate remained divorced from politics, and it continued long after the Umayyads had fallen.

(e)
Mawl grievances.

As the prominence of non-Arab Muslims in Muslim society increased, so did their contribution to revolts. It is customary to interpret their participation in rebellious and/or heterodox movements as an expression of protest against a social inequality which ultimately led to the fall of the Umayyad dynasty. But for one thing, it is by no means clear that mawl were disproportionately represented in movements of protest. For another, they scarcely clamoured for social equality. Not one revolt of the Umayyad period was conducted exclusively by mawl in the name of concerns | exclusive to them; and the only two revolts in which such concerns came to the forefront, revealed somewhat different aims. The first is that of Mu9tr , an Arab opportunist whose non-Arab followers are described as slaves and freedmen ( #abd wamawl ) in the Muslim sources, and as prisoners-of-war in the contemporary Christian work of Bar Penkaye (Mingana, Sources syriaques, *156ff. = *183ff.). They were thus captives trying to escape their Arab masters, not converts seeking equality with them (indeed, the extent to which they were Muslims is disputable); and though Mu9tr was of course forced to extend Arab privileges to them in order to gain their co-operation, neither he nor they would seem to have had any views on the position of non-Arab Muslims in general. The second is that of the Berbers, recruited into Khri3ism [see 9ri3ites ] by Baran missionaries; and Berber Khri3ism did not of course express a desire for social equality, but rather for political independence in Islam. Once more, the conventional picture rests on a failure to distinguish between 9imm s denied recognition as Muslims on the one hand and mawl within Muslim society on the other. The former did indeed clamour for Arab privileges, such privileges being denied them altogether; but the latter clamoured for such privileges (social, cultural or political) as were appropriate to the social group in which they happened to find themselves. The fact that all non-Arabs were exposed to insult and

prejudice does not mean that they responded by forming a tradeunion. A non-Arab peasant in search of a patron such as Ynus b. #Abd Rabbih had little in common with non-Arabs who had long been members of Muslim society; and of these, a general such as Ibrhm b. Bassm had little in common with mawl working as secretaries, scholars or businessmen, let alone as domestic servants. To attribute the fall of the Umayyads to mawl discontent is accordingly meaningless; what grievances did 0ammd alRwiya , an #Ir collector of Arabic poetry at home at the Umayyad court, share with an uncouth Berber general such as ri b. Ziyd , and what sympathy did either feel for the miserable peasants rounded up by Umayyad governors (Arab and non-Arab) throughout the caliphate? Non-Arab Muslims simply did not form a single social group. The fact that numerous mawl participated in the #Abbsid revolution is accordingly also meaningless unless it is specified what kind of mawl they were. In fact, they were of three quite different kinds: longstanding members of Muslim society such as the family of the abovementioned Ibrhm b. Bassm (alabar , ii, 1996-7, iii, 17f., 21, 37, 48, 75ff.; for their origin, see alBal9ur , Fut , 393); freedmen and other clients who automatically followed their (Arab or non-Arab) patrons (e.g. al-

abar , ii, 1954); and 9imm villagers for whom joining the rebel armies constituted both conversion and admission to Muslim society, as it did for Sunb9 [q.v.] and other recruits of Ab Muslim's who were later to opt out of this society as followers of prophets of their own. The causes of the revolution are clearly to be sought in the first type of mawl and his Arab counterpart (from whom he is frequently indistinguishable), the long-standing member of 9ursn society. Such men were subject neither to fiscal disabilities nor to exclusion from the army, the cavalry, high office or general respect. The identification of their aims depends on whether one regards them as coming from inside or outside the local army, the evidence suggesting the former (on this question see Shaban, #Abbsid revolution; M. Sharon, Black banners from the East, Jerusalem 1983; cf. also E. L. | Daniel, The political and social history of Khurasan under Abbasid rule, Minneapolis and Chicago 1979).

3. The
#Abbsids . The #Abbsid revolution deprived the Arabs of such social and political privileges as they retained. Access to political office, influence and wealth now rested overwhelmingly on membership of an army recruited mainly in 9ursn and a bureaucracy recruited mainly in #Ir , as well as of the ruler's household. Non-Arab Muslims reached top positions through all three institutions (Crone, Slaves, appendix 5; cf. barmika ), while at the same time the majority of Arabs and mawl found equality as ordinary subjects. Since Muslims society was no longer constituted by Arab privilege, non-Arab Muslims ceased to require a patron for membership of it. Freedmen continued to become clients of their manumitters, but most of the classical schools rejected the patronate over converts as offensive (below, section V), and the

careers of free converts and the descendants of freedman ceased to be shaped by wal" . Yet Arab superiority on the one hand and the institution of wal" on the other were still to be of major importance in other ways.

(a) The
9u#biyya . In cultural and religious terms, the Arabs continued to be regarded as a superior people, a fact which underlay the so-called 9u#b movement, the movement of the gentiles. 9u#b sentiments had undoubtedly been common already in the Umayyad period, but it was only in the early #Abbsid period that they came into the open, clearly because the mawl were now in a position to get a hearing for their case: the exponents of 9u#bism were drawn primarily from among members of the caliphal bureaucracy and court. Purely literary in manifestation (cf. Goldziher, Muh. Stud., i, chs. 4-5), the movement campaigned for cultural rather than social or political objectives, its ultimate aim being to break the nexus between Islam and Arabism, partly because this nexus stood in the way of non-Arab self-esteem and more particularly because it obstructed the reception of non-Arab culture. Ultimately, the issue behind the controversy was the cultural orientation of Islam (cf. H. A. R. Gibb, Studies on the civilization of Islam , Princeton 1962, ch. 4). The controversy only petered out in the 6th/12th century (cf. R. P. Mottahedeh, The Shu# byah controversy and the social history of early Islamic Iran , in IJMES , vii [1976]), and the issue was never properly resolved, though in practice the 9u#bs lost (cf. P. Crone and M. Cook, Hagarism, Cambridge 1977, 102 f.). For further details, see 9u#biyya .

(b)

Wal" . Having lost its social importance, the institution of wal" acquired a new political significance. Unlike the Umayyads, the #Abbasids trusted their freedmen and other private servants better than the public servants of the state. Thus alManr [q.v.] is said to have esteemed mawl (in the sense of clients, not non-Arab Muslims) for their loyalty and to have accumulated more of them than any caliph before him, recommending them to his son (alabar , iii, 414, 444, 448). Clients of the caliphal household appear as a separate group at court soon after the revolution, and both alManr and alMahd [q.v.] selected a fair number of governors from their ranks (alabar , iii, 429, 545, 1027; Crone, Slaves, appendix Vb). AlMahd , who similarly expressed a preference for mawl (alabar , iii, 531), turned them into an army of their own (ibid., 459). Mawl of domestic origin continued to form troops of their own side by side with Turks and others far into the #Abbsid period, as well as in lnid Egypt | (ibid., 1400, 1501; alYa#b , ii, 606, 624; cf. A9n , xii, 52). Already alManr , however, recruited non-Muslims for military use, attaching them to the #Abbsid house by contractual wal" , and this example was followed by Hrn

[q.v.] (Crone, Slaves, 74). And from the time of alMu#taim [q.v.] onwards, the core of the caliph's armies typically consisted of men who were both slaves and non-Muslims by origin [(cf. 3ay9 and 9ulm ]; D. Ayalon, Preliminary remarks on the Mamluk military institution in Islam , in V. J. Parry and M. E. Yapp, eds., War, technology and society in the Middle East, Oxford 1975; D. Pipes, Slave soldiers and Islam , New Haven 1981; Crone, Slaves, ch. 10). As political power came to rest on private ties with the caliph, the title mawl amr almu"minn became a common honorific. First attested under alManr , it was bestowed on governors and other dignitaries of non-Arab origin regardless of whether they were clients of the caliph in a legal sense. From the time of alMu#taim onwards, it was regularly granted to Turkish generals and other favourites. It was also the title usually held by non-Arab rulers of successor states (Crone, Slaves, 75, appendix Vb, note 610; alBal9ur , Fut , 134, 330; alYa#b , ii, 597; Hill al-bi" , Rusm dr al9ilfa , ed. M. #Awwd , Ba9dd 1964, 122 f.).

4. Muslim Spain.
The relationship between Arab and non-Arab Muslims in Spain differed from that of the east in three major respects. First, wal" played virtually no role in it. On the one hand, many of the conquerors were Berbers, and such ties of wal" as they had with Arab patrons lost all significance when they acquired the status of conquerors themselves. On the other hand, the conquerors settled all over the land, not merely in garrison cities. Muslim Spain thus lacked not only the purely Arab conquest lite characteristic of the east, but also the privileged amr which elsewhere attracted 9imm immigrants and caused the Muslims to exclude from their ranks all converts without a patron. Conversion in Spain did not normally involve either hi3ra or wal" , the converts adopting Islam wherever they happened to be. Indeed, they were not normally known as mawl at all, but rather, in the first generation, as muslima and thereafter as muwalladn (originally meaning home-born slaves; Lvi-Provenal, Hist. Esp. mus., i, 75). Having adopted Islam in their own homes, the non-Arab Muslims of Spain failed to penetrate Arab society. Naturally Spain had its freedmen who, here as elsewhere, entered Arab society as clients. But whereas freedmen of the most diverse origin were exceedingly numerous in the cosmopolitan East, they were relatively few in Spain. Spanish society thus came to be characterised by the coexistence of three quite distinct ethnic groups, Arabs, Berbers and muwallad s, rather than by relationship of dependence between Arabs and ethnically heterogeneous clients. Furthermore, since Spain escaped 9ursn conquest, these groups were able to retain political importance right down to #Abd alRamn III [q.v.].

Secondly, Spain saw armed conflicts between Arabs and indigenous Muslims. Throughout most of the Umayyad period, the mawl of the East were ethnically too diverse and socially too dispersed in Arab society to rebel as mawl against Arabs, while at the same time non-Arabs who had stayed together had also failed to adopt Islam. Only shortly before and after the #Abbsid revolution, when on the one hand whole localities adopted Islam together, while on the other hand government was still identified as Arab, | did non-Arab Muslims rebel against Arab rule. They did not, however, rebel as mawl , but rather as heretics (as in North Africa) or even non-Muslims, rejecting the Arabs and Islam together (as in both North Africa and Iran). In Spain, where Arabs and muwallad s coexisted as distinct groups, such revolts could in principle have erupted any time. In practice, they only came in the 3rd/9th century, perhaps provoked by the growth of the Umayyad state (Arab and Berber leaders also rebelled, and the upshot was the centralised state of #Abd alRamn III); and here for once the rebels took action as mawl , explicitly invoking the cause of the non-Arab Muslims ( da#wat almuwalladn wa 'l#a3am , Ibn 0ayyn , alMutabis , ed. M. M. Antua, Paris 1937, 24) under the leadership of men such as #Umar b. 0afn [q.v.]. Being short of traditions of their own, partly because they were natives of provincial Spain and partly because they were Muslims of long standing, they had no alternative to Cordovan Islam. Accordingly, they did not deny the legitimacy of the Cordovan state as heretics; and though Ibn 0afn

did in the end reject Islam for Christianity, few muwallad s followed suit (cf. Lvi-Provenal, Hist. Esp. mus., i, 295 ff.). Thirdly, there were practically no 9u#bs in Spain. The fact that the muwallad s did not have much of a cultural legacy to vindicate would hardly in itself have prevented them from adopting 9u#b arguments in response to Arab prejudice: the one 9u#b author attested for Spain, a Slav secretary equally lacking in cultural traditions of his own, simply adopted the arguments of eastern writers (cf. J. T. Monroe, The Shu#biyya in al-Andalus, Berkeley and Los Angeles 1970). But having avoided enslavement and migration, the muwallad s had also failed to acquire culture and positions of influence in the society of their conquerors. Where eastern mawl had spokesmen among bureaucrats and courtiers, the leaders of the muwallad s were country squires more noted for their virt than for their polish; indeed, the smarts and insults suffered by such rural lords at the court of Cordova played a role in the outbreak of several muwallad revolts. The muwallad s thus lacked both the education and the influence required for a literary onslaught on Arab superiority. Instead, however, they were in a position to take up arms in their castles, as they did until #Abd alRamn III reduced both them and their opponents to docile subjects. In political terms, however, the institution of wal" played much the same role in Spain as it did in the #Abbsid East. Thus #Abd alRamn I [q.v.], who relied considerably on freedmen and clients of the Umayyads for the conquest of Spain, is said by some to have recruited an army among non-Arab Muslims; al0akam [q.v.] expanded this army and created the palatial guard of 9urs

(mute ones), i.e. foreign slaves and freedmen as well as local Christians (Lvi-Provenal, Hist. Esp. mus., i, 129 f., iii, 71 ff.). The 3und [q.v.], however, survived much longer in Spain than it did elsewhere, being abolished only by Ibn Ab #$mr [see ALmanr ]; and thanks to the geopolitical position of Spain, it was Berber mercenaries rather than Turkish slaves and freedmen who replaced it.

5.
Wal" in classical law.

(a) Sources of
wal" . All schools accept manumission as a source of wal" . Only the 0anafs , Imms and |sim Zayds , however, accept contractual agreement as such, and then in different ways. According to the 0anafs and Imms , contractual clientage ( wal" almuwlt or, in the terminology of the Imms , alta'ammun bi 'l3arra

) arises from a | contractual agreement distinct from the act of conversion at the hands of another; conversion does not in itself give rise to the tie. But according to the |sims , it arises from conversion at the hands of another; mere agreements cannot create wal" .

(b)
Wal" al#it . All the schools are agreed that wal" arises automatically on manumission, but they disagree about the invariability with which it does so. According to the Imms , it only arises when the manumission is gratuitous, i.e. not expiatory, not made in fulfilment of a vow or other legal requirement, and not made by kitba [see #abd ]; and both the Imms and other schools enable the manumitter to exempt himself from wal" by declaring the freedmen s"iba , though the Sunns disapprove of the practice. In 0anaf and 9fi# law, however, manumission invariably gives rise to wal" , whatever the circumstances or the inclinations of the manumitter. If the freedman is a non-Muslim, the tie is deprived of most of its legal effects. The manumitter acquires responsibility for the payment of diya on behalf of the freedman and qualifies for the role of marriage guardian

to his freedwoman or freedman's daughter. In return, he is granted a title to the freedman's estate in all schools except that of the Ib's (not that of the hirs , as stated in #abd ). In Sunn and |sim Zayd law, he inherits as the remotest agnate of the freedman; he is thus excluded by the freedman's own agnates (e.g. a son), but inherits together with |ur"nic heirs (e.g. a daughter) and himself excludes remoter relatives ( 9awu 'larm , e.g. a sister's son). In Imm , Ism#l and Nir Zayd law he is excluded by any blood relation of the freedman, though he inherits together with the spouse. The freedman does not, on the whole, acquire any corresponding rights and duties. Only the Ism#ls call him to succession, and only in default of all other heirs. The Mliks do hold him responsible for the payment of diya on behalf of the manumitter if the latter has no agnates, ahl aldwn or patrons of his own; but a similar opinion transmitted from al9fi# failed to become school doctrine, and all other schools exempt him. The possibility that he might act as marriage guardian is not considered. The relationship survives the death of both parties, passing to their heirs in perpetuity, though it loses practical importance as the client acquires agnates of his own. It also extends to the freedmen of the

freedman and their freedmen in perpetuity, again with decreasing practical significance.

(c)
Wal" almuwlt . The prospective patron must be a free, male and adult Muslim. The prospective client, according to the 0anafs and Imms , must be a free and adult non-Muslim of either sex who has no agnates or patrons in Islam, that is a 9imm , convert, foundling or (in Imm ) law) a freedman without wal" ; the |sims , however, require him to be a arb convert: conversion of a 9imm does not give rise to wal" . The patron agrees to pay blood-money on behalf of the client in return for a title to the latter's estate; the parties may stipulate mutual succession. Either way, the heir by contractual wal" is excluded by any blood-relation of the deceased. Whether the contractual patron may act as marriage guardian is disputed. Unlike wal" al#it , the contractual relationship may be terminated as long as the patron has not had occasion to pay, but it becomes permanent thereafter.

(d) The nature of

wal" . Practically every lawbook states that

wal" should be regarded as a fictitious kinship tie (alwal" luma kalumat al-nasab , as | a famous maxim has it), and this view underlies a number of subsidiary rules generally accepted by Sunns and nonSunns alike. Thus wal" cannot be alienated by sale, gift or bequest in classical law, though such transactions were permitted in pre-classical law; one cannot sell or give away nasab , as various authorities point out in ad9 . Equally, wal" cannot be inherited in the strict sense of the word; the devolution of the rights and duties vested in the tie follows special rules ensuring that the relationship functions like an agnatic tie (cf. R. Brunschvig, Un systme peu connu de succession agnatique dans le droit musulman, in his Etudes d'Islamologie, Paris 1976). Pace Brunschvig, however, this view of wal" is not an archaic survival, but on the contrary a juristic interpretation of the late Umayyad and early #Abbsid periods. It was adopted with particular forcefulness and consistency by the Sunns , to whom the essence of wal" lies in ta#b , agnatisation. In fact, however, the legal nature of wal" is quite different from that of an agnatic tie even in classical law. For one thing, it is only in Sunn law that the patron inherits as an agnate, and then only if he is a manumitter, not a contractual patron. For another thing, the relationship lacks reciprocity. The client is a purely passive member of the patron's agnatic kin. Indeed, for some purposes he is not a member of it at all. Thus Sunn

lawyers do not usually consider clients of |uray9 [q.v.] eligible for the caliphate; and the question whether clients of the H9imites were excluded from receipt of zakt [q.v.] on a par with their patrons remained controversial; as Ibn alA9r pointed out, the maxim mawl 'lawm minhum (which originated in this very context) could be interpreted in a purely metaphorical vein (alNihya , Cairo 1963, v, 228). In legal terms, wal" is a tie of dependence which derives its efficacy from the fact that the client is detached from his natal group without acquiring full membership of another. The tie undoubtedly owes its existence primarily to administrative convenience, though the administrators may well have been influenced by the legal institutions of the preconquest Near East (see further Crone, Roman provincial and Islamic law, with full references).

(e)
Mawl and kaf"a [q.v.]. Classical law does not in general attach any legal significance to servile and/or non-Arab origin outside the private relationship between patron and client, but there is one major exception. NonArabs and freedmen cannot marry Arab women, according to the 0anafs and the 9fi#s . The same view prevails among the 0anbals , while contradictory views are found in the Zayd schools. The Mliks

, who see no harm in such unions, nonetheless allow an Arab woman to have her marriage dissolved if she marries a freedman in the belief that he is an Arab (as opposed to merely freeborn ( 9all b. Is , Mu9taar , tr. I. Guidi and D. Santillana, Milan 1919, ii, 37). Only the Ib's , the Imms and, following them, the Ism#ls , consistently refuse to distinguish between Arab and non-Arab, freeborn and freed for purposes of marriage [cf. nik ]. The complete assimilation of Arab and non-Arab Muslims allegedly brought about by the #Abbsids cannot be said ever to have been achieved.