LINGUISTICS

F 0 R BEGINNERS
LINGUISTICS
F 0 R 8 E GINNERS

BY W. TERRENCE GORDON
ILLUSTRATIONS BY SUSAN WILLMARTH

-~- ~
FOR BEGINNERS®
an imprint o{Sturforth Prm

Hamw". Ntw Hampshir~

 For Beginners LLC
155 Main Street, Suite 201
Danbury, CT 06810 USA
www.forbeginnersbooks.com

Text: © 2008 W. Terrence Gordon

Illustrations: © 2008 Susan Willmarth
Cover Art: © 2008 Susan Willmarth

All rights reserved. No part of this publication may be reproduced, stored

in a retrieval system, or transmitted in any form or by any means,
electronic, mechanical, photocopying, recording, or otherwise, without prior
permission of the publisher.

A For Beginners® Documentary Comic Book

Copyright © 2008

Cataloging-in-Publication information is available from the Library of Congress.

eISBN: 978-1-939994-14-1

For Beginners® and Beginners Documentary Comic Books® are published

by For Beginners LLC.

v3.1
 Linlut'stlc&
b~eo""~~ ;11 IV~~ eerier ar~a.,
feel llke IW. •~ C)z.,
I
Tv~ try;"'1 to tel( it lik~ it ~'·
-OJ4~ N1a~~
Wh$t ~Yo" ~ ..t', A l"fttftt'"Sf~tl Pt.l•t"~
O>r A ~.ti~ ..L-+ery rafk! (r .. 'tZ)

If you have already heard a"out linguistics or read something a"out

the su,ject, and you share the feeling that Ogden Nash descri"es
a"ove, you've come to the right "ook. If you know nothing at all
a"out the topic "ut have picked up this "ook out of sheer curiosity,
so much the "etter. Read on, and take your first steps toward
learning what linguists do for a living.
HOW LANGUAGES WO/lK
Lessons One and Two
let's start with theee f1aslce:
J.ng~~~~~p ¥ II tool;
1/ngu/lltlerJ ¥ tM IINI~ of J.ngu,_
Why say that
language ls a tool?
Because llke any of the
things that we recognize as
tools, from hammers to
computers, lt lets us do things
that would otherwlee l7e
impo661l71e or a lot harder to
do (try to Imagine driving
nalls without a hammer
or complllng a city phone
directory without
the help of a
computer).
Language is a tool for
getting thoughts out of our
l7ralns and Into our mouths and
into other l7rains.

[3
 How else would we communicate? Sure, you can just let out a yell to
warn of danger, or a groan to expre5s pain, strain, or 17oredom, and a
map or a sketch can give a lot of information. 5ut try sketching this:
f/1 niWtlf' forgtlt
htlf' lllugh.

Apart from "laugh,"

the elements of this
sentence are too
al7stract for a pic-
ture. They are ideas
and concepts,
expressil71e only when
organized by and into
a complex system:
language.

Unlike most other tools, language can 17e used

on itself, and that is exactly what happens in
the study called linguistics. It is analysis of
language, it is language al7out language.
Here's another way to think al7out it: lin-
guistie5 is to language what a mechanic's
manual is to a car. A linguist working on a lan-
guage with analytical tools is not much differ-
ent from a mechanic working on an engine with
his socket wrenche5. The shop manual is not a
driver education handl7ook. and a 17ook on lin-
guistie5 doe5 not teach you how to speak. It's
possil71e to 17e a competent mechanic without
knowing how to drive a car and just as possil71e
to 17e a linguist without 17eing fluent in the lan-
guage you are analyzing. (More al7out this
17elow, where we meet three guys named
4]
Chomsky, Mithridate5, and Fazah.)
 Wh11t? No Wore¥?
So far, we haven't said a
word a11out words, and we're
not defining linguistics as
words a11out words. Why?
Because llngulfJtlc analyeiiJ lfJ
not limited to worde.
linguistics goes 11elow and al1ove the word. It takes words apart
( hopelessly=hope+le5s+/y) and
examines how the parts go
together ( hope+le5s+/y 11ut not
/e!!js+hope+/y). It also looks at
how words form groups (He is
hope/e!!jsly lost 11ut not Lost is
hopele5s/y he).
When a sentence makes sense,
its words are linked like pearls
on a string. What keeps the
words together is a pat-
tern. Many different
sets of pearls could
11e put together on
the same piece of
string, and many
different sets of
words can hang together on the same pattern. The
study of patterns for sentences
Is called syntax t7y linguists
and grammar t7y the rest
of the world. Let's go 11ack
to our example: He is
hopelessly lost.
[5
 Thi5 6entence ha6 the 6ame pattern (we could al6o 5ay the 5ame
model or the 5ame 5tructure. and we will 6ee later that 5tructure
86pecially 16 a favorite word among lingul6t6) a6 the following:

He i6 prol:lal:lly lo6t.
She wa6 very cold.
They will l:le somewhat annoyed.
Linguf5t5 are a lot 1866 inter86t-
ed in word6 than they are In how
word6 coml:line with each other
and in how l:llt6 and piece5 com-
bine to make up a word. The l:llt6
and piece5 of 6poken language
6] tum out to l:le more lntere5tlng SHE WAS VERY COLO
than those of written
language, 17ecause
there is more regularity, more
system, more pattern, more

®
structure (there's that lin-
guist's word or choice again) in speech
than in writing. Think of it this way:
you can line up ten people who all
look very different from each JLOSf
other, 17ut if you line up their x- (
rays, their skeletons will look
very similar. Linguists pay more
t:7
attention to skeletons than to skin; they J
@)\5
spend more time studying the sounds of ~
spe ech and sound systems than words on
the page.

Put It In Writing
But let's not imagine that learning al7out writing Is completely out-
side the important fundamentals of linguistics. We write English In
what is called a phonetic alphabet. This means that the letters of
the alphal7et stand for sounds, the most 17aslc elements of lan-
guage that linguists study.
[1
 A phonetic alpha"et is a medium, in the sense of an extension of our
"odies. It turns the sounds of language that we produce with our
lungs and tongues and teeth and lips into visual marks, whereas the
sounds of language are an extension of the thoughts in our minds.
Here we are "ack at the "asic idea we started with: language as a tool.
A phonetic alpha"et is a tool or medium not only "ecause it extends
our "odies "ut in the even more "asic sense of something that goes
"etween and "rings together. What does a phonetic alpha"et go
"etween and "ring together? Meaning and sound.

ei

If we compare, say, Chinese characters, with a

phonetic alpha"et, we find no "go-"etween" in
Chinese. The writing gives meanings, "ut it
doesn't show how to pronounce what is written.

If you are having trou-

"le understanding this,
think a"out sym"ols like +
or $or %. There is nothing
in the shapes of these sym-
"ols to show how they are
pronounced, "ut there is in
the letters of plus, dollar,
and percent. Imagine if
every word in English was a
sym"ol like + instead of
plus, $ instead of dollar, or
%instead of percent, and
then you've got an idea of
how Chinese works and
how It is different from a
phonetic alpha"et.

g]
 Whitt Do. • UngultJt Do
from Nln11 to Rv11?
Some llngule;te; etudy one lan-
guage and how lte; sounde;
vary In different plaCB6 In a
eound-group (p. for example.
ie not pronounced In exactly
the e;ame way at the 11eglnnlng
of a word In English-pot. and
at the end-top). Some may
examine the etreet elang of
their own neigh11orhoode;, 11ut
othere may race to a far cor-
ner of the world to record con-
vere;atione; among the lae;t few
speakere; of a dying language.
Thie ie; all modern llnguietlce;, twentieth
century linguietice;, ae; it hae; 11een prac-
ticed eince the time of Swiss echolar
Ferdinand de Saussure (1851-lqJS) and
11ecause of hie influence. So far-reaching
hae thie Influence 11een that Sauseure ie;
often called the father of modern lln-
guietice;. But what al1out earlier?
le it not poe;ei"'e that people
have 11een thinking al1out lan-
guage for almoe;t as long ae;
they have 11een using It? In
fact. we find the fln:;t flow-
ering of llnguietlc thought
twenty-two centuriee;
11efore Saue;e;ure.

[q
 The French word llngul5tique had already
17een in U6e for at lea6t 24 year6 when
Sau66Ure wa6 17om; it6 Engli6h cou5in
llngul5tlc appeared fir6t in 1837 in the
writing6 of the Briti6h 6cholar William
Whewell. who
defined it a6
the5cience
of language.
Under the
influence of
American
5cholar6
6UCh a6
Noah Wel76ter and Dwight Whitney. llngui5tlc
wa6 tran5formed into llngui5tiC5.
By compari5on with llngul5tic(5), llngul5t
ha6 a much longer hi6tory. having 17een
U6ed fir6t f,y Shakee;peare in rsqr in rwo
Gentlemen of Verona to mean "one who i6
5killed in the U6e of
language." Language 6cholar6 were known a6
philologi5t5 17efore llngul5t came along. and the
two term6 continued in U6e along5ide each
other through the year6 of Sau55ure'6 lifetime.
What doee; a lingui6t not do at any time of
day? The jol7 of the lingui6t i6 to de6cril7e lan-
guage. to record it. analyze it. explain how it
work6. theorize a17out how we learn it. and much
more. But not to dictate how you 6hould U6e
your
language. ihat would 17e
just as inappropriate as
a geneticist giving you
5ugge5tions al7out
who you 6hould mate
with. But of cour5e
there are case6
where there ha6
17een meddling and
mending ...
10]
POUQES AND POUCE. Pf(OTOCOL AND POYnC5:
Th• Ungul•tlc Mix
I) 5au55ure taught that language change come5 at1out 5pontaneou5-
ly and cannot t1e impo5ed, t1ut thi5 has not kept any numt1er of people
from trying, and their effort5 are 5ometlme5 called linguistic engi-
neering. At Its wor5t, the phenomenon can Induce collective t1rain-
wa5hlng (Newspeak in George Orwelr5 (qg4); at lt5 t1e5t it elfminate5
prejudice and t1ia5, a5 in the introduction of Oown'5 syndrome to
replace mongolism, engender to replace father (pseudogeneric vert1 ),
or founder to replace father (pseudogeneric noun).
2) In Canada, Quet1ed5 Commls5lon de Ia protection de Ia langue
fran~l5e (Comml55lon for Protection of the French Language) 15
commonly known in Engll5h a5 the language pollee. It i5 the t1u5lne55
of thf5 t1ody to enforce Quet1ed5 parochial and mi5guided legl51atlon
for en5uring the 5Urvlval of the French language In the predominant-
ly French-5peaking province (word5 on put11ic slgn5 in language5
other than French mu5t t1e half the 5lze of French words). The lan-
guage pollee have t1een vigilant enough to 5pot unlllngual English
matzo meal packaging. And. a Montreal grave5tone maker ha5 t1een
required to down5ize the Het1rew lettering on a fJfty year old 5lgn
over hi5 t1u5ine55 premi5e5.

[II
 3) Modem 5ociety i5 making progres5 toward eliminating language
that i5 prejudicial again5t per5on5 17ecau5e of their race, 5ex, age,
5exual orientation, di5al7ility, ethnic origin, or 17ellef 5y5tem. But lin-
gui5t5 and non-lingui5t5 alike di5agree on exactly where to draw the
line on what i5 deemed to 17e 17ia5ed language. 15 fellowship ol7jec-
tional71e? Should a new term 17e invented for a female holder of a
fellow5hip? It i5 true that the primary meaning of the root word fel-
low given in mo5t dictionaries i5 that of man or boy, 17ut hi5torically
there are 5everal other meaning5 5Uch a5 associate, companion,
trustee, etc., and the Old Engli5h origin of the term i5 a gender free
word for business partner. Of cour5e, it i5 po55il71e to argue that in
context5 where the meaning5 of associate, companion, etc., are to
17e expres5ed, one 5hould opt for one of these term5. And it i5 prol7-
al71y true that even if Old Engli5h feolaga (business partner) wa5
technically gender-free, there were prol7a171y few if any female "u5i-
nes5 partner5 to "e found a thou5and year5 agol

12]
CHOMK5Y, FAZAH, AND MITHRIDATE5 (NOT A LAW RRM)
If you have read anything
a17out linguistics, you may
have already discovered
that the name of Noam
Chom5ky has dominated
the field, particularly in the
United States, for fifty
years. Chomsky is credited
with recharting the course
of linguistics when, with the
ink still fresh on his Ph.O.
from the University of
Pennsylvania, he pul71ished
Syntactic Structure5 in
1q51, It was a very slim
17ook 17ut enough of a fire-
cracker to start linguists
arguing al7out how they
should approach the analy-
sis of language. They are
still at it. Linguistics has
gone through many phases
of development as a direct
result of Chomsky's work
(as have his own 17asic ideas
on the su17ject).

Not so, says the

Guinnes5 8ook of World
Record5, where the top ~
contender for the title is
one Ziad Fazah of
Brazil, who speaks and .
writes sg language5.
f
[13
 Hi5tory record5 that one of the fir5t per5on5 known for hi5 multilin-
gual 5kill5 wa5 King Mithrldate5 of Pontu5 (132.-63 B.C.E.), who wa5
fluent in 2.2. language5. '-..
The Guinnes5 8ook U5e5 linguist in
the 5en5e we 5aw earlier, a5 per
Shake5peare: one who know5
many language5. Apparently,
in that re5pect, Or. Chom5ky
can't hold a candle to King
Mithridate5, much le55 to
Mr. Fazah.
UngultJtiCIJ Th1111
11ndNow
It didn't take the invention of the
term5 linguist and linguistics for
the analytic 5tudy of language to f,egin.
The link f,etween logic and language goe5 f,ack to ancient time5 in
Arl5totle'5 work and the categorie5 he 5et up mark the f,eginning5 of
what would eventually f,e
called linguistics.
The logic/language connec-
tion i5 5till important to
lingui5t5 today. When
computer5 came on the
5cene, after the Second
World War, 5omef,ody 5oon
got the idea to try U5ing
them for tran5lating. To
do thi5, it wa5 neee55ary
to give the computer
information af,out the lan-
guage5 it would tran51ate
in term5 of very f,a5ic
logic. So, logical and math-
ematical model5 of lan-
guage f,egan to appear,
and they have dominated
lingui5tic5 ever 5ince.

14]
 Input ;o! Output

The story goes that the first experiment in translation f1y computer
(usually called machine translation, even though the computer is not
a machine in the usual sense of having mechanical parts) was not a
success. Supposedly an international team of linguists and transla-
tors had worked long and hard and thought that they had every-
thing ready to get their computer to translate from English to
Japanese. They gave it the sentence "The spirit Is willing. "ut the
flesh is weak." The translation did not take long, "ut unfortunately it
came out meaning "The drink Is all right, "ut the meat is lousy." So
it was "ack to the drawing "oard.
............
It's easy enough to get a computer to recognize a sentence pattern
"ut very difficult to give it all the details a"out the limits of the
pattern. All human languages have patterns, "ecause humans have
pattern-making minds, "ut language patterns are incomplete, imper-
fect, irregular in all kinds of ways:
t~lng-tN!Ing-*lung and rfng-r~~ng-rung
t~lnk-tN!Ink-*lunk "utnot thlnk-thllnk-thunk
and definitely not plnk-p6nk-punk
horror-horrid-horrify Is a complete pattern
tsrror-ttlrrffy. C6ndor-t:~~ndld are Incomplete patterns
These Incomplete patterns are no pro,lem for us; we learn what and
where the quirks are. It's a "ig pro,lem for a computer to "under-
stand" that you can have a 11moke or have 11 drink "ut you can't
[15
 "'have an eat. (In linguistics. an asterisk precedes a word or phrase
that is not found in standard use. This once prompted a linguist to
come up with the rallying cry: Llngui5t5 of the world unite; you have
nothing but your a5teri5k.) So "'terrld, "'candlfy.

Go!ln't
HAVE AN
EAT
Mors
About
Writing It
Down
Let's come uack to three points weve
already touched on and tie them
together: I) Linguists are especially
interested in the l7its and pieces of
language; 2) the average linguist
is more likely to pay attention to
the stream of speech coming
out of someuody's mouth
than to chunks of language
flae;h-frozen in written
words; 3) We write
English in an alphal7et
that shows us how to
pronounce our words.

16]
Think at1out this last statement. It is only partly true. The first letter
in knight, gnome. and p6yche. for example, does not tell us what the
first sound in the word is. In the long history of English, the written
language has not kept up with the changes that have taken place in
speech. So, we have words where different letter5 5tand for the 5ame
5ound (way, weigh. whey) and word5 where the 5ame letter 5tand5 for
different sounds (the o in on, once. onion, only). And then there are
comt1ination5 of letter5 that represent only one 5ound ( th. 6h, for
example) and tho5e 5ilent letter5 in knight, gnome. p6ychic, etc.

R8cslvttd Pronundtltlon
Talking Posh
This phra5e refer5 to the pronunc;iation
An English professor com-
of Received Standard Engli5h, the mo5t
plained to the owner of the
prestigiou5 dialect of Briti5h Engli5h-a
pet shop where he had
c;la55 dialect rather than a loc;al dialect,
recently bought a parrot:
though it i5 a5soc;iated primarily with
"He uses improper language."
the 5outhem counties. "RP," a5 it i5 1

commonly known, i5 u5ed l7y highly edu- I "Well, I am surprised," replied

c;ated Briton5 and memt1er5 of the
E5tat71ishment. It i5 also known a5 the
88C accent, the public 6chool accent,
talking proper, and talking p06h.
the shopkeeper, "I never
taught him to swear."
"Oh, it's not that," explained
'the professor, "but he keeps
splitting his infinitives."

(17
 And Still Mors About Writing It Down
English is not the
only language with
inconsistent spelling
_,,. ~ .J,~
or silent letters. In
:.' _.; French, to take just
)

one example, the

!. verl1 avoir, meaning
to have, has many
forms, including
aie5. Here the four
letters stand for
just one sound. The
word is pronounced
like the vowel e in
English pet.
The muddled up spellings of languages like English or French forced
linguists to invent a new way of writing down speech to show it accu-
rately and consistently. It is called the International Phonetic
Alphabet (IPA). Here are some of the words we have already used as
examples transposed into the IPA:

Ill]
 I
I
I
I
I
I r----------•
1 The column on
; the right shows ;
1

hi,~tt I I [!1-Q.it] 1 the way that I

II Henry H"1gg~ns
. 1I
I I 1 recorded Eliza 1
I
I I ; Doolittle's way of 1
1 speaking Engliah I
I I ; in hia note"ook in ;
I I 1 the opening acene I
; of My Fair Lady. ;
I I ·----------·
Kni5ht I : (_... Qi-t] Those square
I brackets indicate
I a phon•tlc
I transcription - a
I written record of
somebody talking.
I If rm speaking very
I quickly and use
the word "only," I
I I
'lnome. I (}tou~ may pronounce it
1 Coni], and if a
linguist is taking
I NOAM
I down my words,
I I she will transcribe
Coni] not Conli] to
JI$1Cit.ie show exactly what
I said.

I I [wei]
way
on I I c~n]
once 1 [wAne]
onion
I [AnjAn]
only I I [ounll]
(French) aiBfJ
I I [S]
I I (Jq
I I
I I
Of course. in spite of my little slip. I know the usual pronunciation of
the word. and if this is what the linguist wants to show. forward
slashes are used around the transcription instead of square brack-
ets: I onli/. Now we have an example of what is called phonsmlt: or
phonologlt:l!ll transcription-not one person's words but what people
usually say. Here we are moving up from the particulars of speech
(phonstlt:4~) to the patterns and system of sounds of language
(phonology).
If all this talk about phonstiQI and phonology has inspired you already
to become a modem-day Henry (or Henrietta) Higgins. you can start l:7y
going to the section further on here titled "More About Phoneticg."

Introducing Sammy Msnsflsld

Most textbooks of linguistics present a chapter on phont~tlt::~~ and
phonology with a diagram of a cut-away head to show the organs of
speech and the place of articulation of speech sounds. The diagram
is a (presumably) male head and always shown from the left side.
This fellow is called Sammy Mansfield. and his name is a link. via the
initials. to what the diagram represents: Speech Mechanism. (The
corresponding phrase in French is appareil de production 5onore or
organe5 articulatoire5, but efforts to standardize the name of
Sammy's French counterpart as either Albert Philippe Simard or
Odette Ar5enault-the latter a particular admirable effort to count-
er the male bias too often encountered in the old bastions of Gallic
culture-have met with only limited succe5s.) Here fs Sammy... or
Alphonse...or Odette.
 Incidentally, most versions of this diagram show the talking head
(you've never seen one like this on TV) with nostrils 171ocked. which
would make it tough (to say the least) to produce nasal sounds. In
the interest of accuracy. and to
make life easier for Sammy.
we show him un171ocked.
QUIJntum wp #I -
8r8111dng ths Sound B.rri.-
Every language i5 a 5elf-contained 5y5tem.
(By 5y5tem we mean a 5et of element5 and
rule5 for their u5e.) 5ound5 that make a
difference in meaning in one language may
not work the 5ame way in another. We can
5tart with Engli5h veal and wheel to lllu5-
trate thi5 point. Word5 like the5e are called
minimal pair5. They are the conjoined twine of
language. Jw:;t one 5ound in each i5 different
from one 5ound in the other. If we compare
Ukrainian (a Slavic language), we do not find
any minimal pair5 5imilar to veal/wheel.
Minimal pair5 provide lingui5t5 with the evidence of the unit5 they
call phon11mu - the 5ound5 of language who6e jo1:1 It 16 to allow dif-
ferent meaning6. The phont~mu that make up the Engli6h 60und 6y6-
tem (Engli6h phonology) include /vi and lwh in Ukrainian there are
many word5 that 1:1egin with /vi 1:1ut none that 1:1egin with /w/. There
are al5o Ukrainian word6 that end In lw/ 1:1ut none that end in /v/. 5o
two 6ound6 that are 6eparate phontlmtltl in Engl16h 1:1elong to the
5ame phont1mt1 in Ukrainian, where there are no pair6 of word6 con-
tra6ting in meaning comparal:11e to Engli6h veal/ wheel.

[23
 We learn the phtHr.,.. of our Rr5t
language without knowing what a
phon•,. is or how it work5. (You
can learn to drive a car without
17eing a mechanic or even knowing the
principle of the coml1ustlon engine.)
We carry over the ha17its we acquired
in mastering our Rrst language when
we approach a new one and can
aoon 11ump up against a few prol1-
lems with phtHI.,.. that refu5e to
11ehave like the one5 we are familiar
with. This is why a Ukrainian speaker
learning English will 5aY [vel] for well.
[ vik] for week etc.

24]
 QUIIntum Lap #2 -
flult!Jlng ths Cmnbsrry Tst~t
At this point, most 17ooks on lin-
guistics march you straight from
phonology to long lessons on mm<-
~ from a language's system
of sounds (which mean nothing in
themselves) to the way they work
to communicate meaning. Here
we'll limit ourselves to introducing
,~ f7y way of some facts
at7out the phrase cranberry lin-
gui5t. But first a word at7out our
new word~
~comes from the Greek noun meaning form, the same root
word that we find in metamorpho515 (change of form), morphogene515
(structural changes in the evolution and development of an organ-
ism), and even Morpheu5 (in Greek mythology, the god of sleep, so
called t7ecause of the form5 or 5hape5 that he calls up in dreams).
Linguists are not the only group of specialists to
use the term~ For scholars in other
disciplines, it refers to the study of form and
structure of plants, animals, human 17ones, conti-
nents, etc. In linguistics, It means the 5tudy of the
smalle5t forms that carry meaning. These forms
are called DUHjJ/lulu?&
They do not have to t7e words; they can t7e part of
a word, as long as that part gives some meaning.
In is a HUHj.Jhen~ and a word, direct is a~
and a word, indirect is two n~ t7ut only one
word. The in- at the t7eginning of indirect, carrying
the meaning of not, is not the ~word in
(opposite of out). Think af7out the difference
t7etween in direct communication and indirect
communication.
We can think of nw#~ as the part of linguistics that links
phonology (the study of the functional sounds of language) and
syntax (the study of arrangements of meaningful forms) f7y iden-
tifying those meaningful forms.

[25
 Now for that cranberry llngui5t. This is actually a
term of contempt among linguists for one of
their numt7er who studies nothing more impor-
tant than the nw'(Jiwlo.gy/ of cranberry and the question
of where the cran- in cranberry comes from. (It is a vari-
ation on crane, one explanation t7eing that crant7erries
typically grow in marshy land of the kind cranes favor for
nesting. Not all lingulsts-crant7erry and non-crant7er-
ry alike-favor this hypothesis: it has also t7een
suggested that the Pilgrims called the fruit
craneberry t7ecause its arching t71ossoms suggest-
ed the shape of the t7ird. So dis-
agreement is over the motivation for
crane, rather than the
cranelcran- alternation, which
is typical of a sut7stantial numt7er of pairs of
related words: 5ane/5anity, urbane/urbanity,
profane/profanity, etc. Before cranberry was
imported into British English from America,
t7oth plant and fruit were known in
England as mar5h-whort, fen-whort, fen-
berry, mar5hberry, and mo5sberry. In
German, Swedish, Danish, and other con-
tinental languages, the term for cranber-
ry is a compound word and a direct trans-
lation of craneberry.)

26]
QUimtum Lap #S - aut~tiJrinp bsyond ths Cmnbsrrl1111
We move on now from~ to syntax. This term come5 from
Greek syn (together) + tassein (to arrange).
In phtJntJiagy, phon•mu are arranged together,

and the arrangements create the simple units

of meaning called ~ze&;

CAT
in syntax,~ are arranged together,

and the arrangements create the complex units

of meaning called phrases and sentences.

TABBY
CAT

[2.7
Phonology, nw'f.Jhol~_, and syntax are different levels of analysis.
Linguists separate elements of language at these levels for purpos-
es of analysis, uut in practice no such separation occurs.
(You can't drive your car if the carburetor is sitting on the
workbench in your garage.}
Linguistics manuals often give a diagram for an overview of the suu-
ject with mrnj.Jhoi'?!IY' stacked on top of phonology, syntax stacked
on top of mmplzolqg!f• and semantics (meaning) at the very top of
the pile, as if you don't get to meaning till you have worked your way
"up" through the other levels.
Not so! Sure, we can just make a list of the phon•m• of a
language, uut even in doing that we are indicating the
sounds that work together to create meaning. When
language functions, phonology is inseparaule from
semantics, mmplwlo,.gy is inseparaule from
phonology, syntax is inseparaule from
mmplzo/qp;4 etc.

Semantics is the messiest level

• of linguistics, uecause the
human mind is good at making
and multiplying meanings and
uecause the resources of lan-
guage for doing this are very
complex. So uefore we tackle
semantics, let's get syntax
(where things are a little tidier-
at least at the entry level) out
of the way.
 When computers came into general use after World
War II, their potential for endless applications soon
became apparent, and these included trans-
lation. This requires supplying the computer
with enough information to be al71e to recog-
nize how sentences can change form without
changing meaning (Example: The
fireman re5cued the cat; The cat
wa5 re5cued by the fireman).
Oescril7ing the relation of
different forms of a sen-
tence to each other is known
as tran5formational syntax
or tran5formational grammar,
and here we have to rememl7er
that grammar means not just rules
for sentence construction, and e5pe-
cially not rules for constructing "good" sentences, since
linguists a17stain from such value judgments,
17ut simply a description of how lan-
guage works. Though transforma-
tional syntax was first devel-
oped for use in machine
translation, it soon took
on a life of its own and
changed the face of
modern linguistic theory.
Transformational syn-
tax is also known as
generative syntax or
generative grammar. It
is a theoretical frame-
work for viewing lan-
guage as an infinite
set of potential sentences which may 17e generated from a finite set
of al7stract rules and principles. In other words, when we know a lan-
guage, we know how to put together an unlimited numl7er of new
sentences from a limited set of rules. We acquire this amazing skill
as children 17y listening to adults talking at us and figuring out what
the rules are that will let us talk 17ack to them in such a way as to
make ourselves understood.
 Ungul.tlc Fun:
/Orymlng Slang
Cockney in origin, thi5 form
of 5lang replace5 a word l7y
a phra5e that rhyme5 with it
f1ut 5uppre55e5 the rhyming
part. Thu5, "kid," rhyming
with "God forf1id," tum5
"How are the kid5?" into
"How are the godfor5?"
"Leg5" rhyme5 with "f1acon
and egg5," and 50 "She'5 go
5ma5hing leg5" 11ecome5
"She'5 got 5ma5hing
11acon5." It take5 a while to
get the cra5h-11oom of it.

30]
 SsntsncetJ
Grow on Tr886

At lea5t in lingui5tic5 they do. What'5 more, the tree5 grow up5ide
down. Tree, in thi5 ca5e, i5 the name given to a diagram for the 5tep5
and 5tage5 in the production of a 5entence. In keeping with the image
of a tree, we could al5o call the5e 5tep5 and 5tage5 the growth or
the development of a 5entence. The term derivation, in thi5 5en5e, i5
particularly favored ~ lingui5t5. So the lingui5t'5 tree i5 a vi5ual help
for 5howing the particular5 of a 5entence. Branch ~ f:lranch.
Naturally we are going to give only 5imple example5 here,
50 don't expect the tree diagram5 to look very much
like an inverted ver5ion of that magnificent
oak out5ide your window. For one thing, it
doe5n't have any leave5.
Our tree here grow5
into a fully-formed
5tructure, f:lranching from
a 5eed that already con-
tain5 the whole 5entence,
f:lut the growth i5 limited to
the f:lranching it5elf.
The only "f71o550m5" you will find on
the5e tree diagram5 are word5 that can take their place at the final
5tage of it5 growth. Before we get to that 5tage, we find f:lranching
point5 laf:leled with initial5 that 5tand for the type of phra5e that
occur5 there: NP (noun phra5e), VP ( verf:l phra5e), PP (prepo5itional
phra5e), etc.• all 5temming from S (5entence).
(31
 Here is a derivational tree for the
sentence Thomas ate the fruit In
the kitchen.
5
A
NP VP

VP
A NP

NP
A PP
You may wonder why the diagram always shows
phrases. implying groups of words. when some
of the elements of our sample sentence are sin-
gle words: the NP sut1ject of the sentence is
just Thomas and the VP is the simple vero form
ate. Well. it's t1ecause our tree needs to t1e gen-
eral enough to accommodate sentences that
follow the same pattern as Thomas ate the
fruit in the kitchen. and many of these will tum
out to have complex elements. (These would t1e
shown in more detail in an expanded diagram.)
The same tree lets us diagram the sentence
Charles's favorite uncle eventually gobbled up all
the apples. oranges, and bananas In the room
adjacent to the dining room. Our page isn't t7ig
enough to show the full tree for this sentence.
t1ut the t1asic tree for it is the same as the one
we have already.
We can see that two or more sentences (in fact an unlimited numt7er
of sentences) will have the same tree structure. But one sentence can
also have two or more tree diagrams associated with it. In fact. that
is the case for Thomas ate the fruit In the kitchen. In setting out the
diagram for this sentence. we assumed something at1out its meaning.
namely that the fruit was In the kitchen t1efore Thomas ate it.

32]
 We don't really know from the
sentence itself if he was in the
kitchen when he ate it, 17ut this ie
one possi17ility; there is a strong
suggestion for the same inter-
pretation of the sentence that
there was fruit in another room
that Thomas could have eaten
instead. But if we know for sure
that Thomas did the eating in
the kitchen, we need a different
tree diagram to show this:
5
~
NP VP PP

A
VP NP
The 11asic difference 11etween our two trees is that the PP
(Prepositional phrase) in the second is independent of the second
NP (the fruit), whereas it was directly linked to it in the first case.
Now it occurs in the first 17ranching of the tree, and this puts it in
direct relation to the flr5t NP, indicating in this way that Thomas
was in the kitchen when he ate the fruit. (This time we don't know
for sure if he found the fruit there or got it somewhere else and
was just eating it in the kitchen so that the people in the other
room wouldn't see what a pig he was making of himself.)
In a more detailed analysis important differences emerge from such
features and contrasts as rlght-bnlnahlng t!ltll7ttll7a.:
Thi5 i5 the cat that caught the rat that ate the
chee5e that lay in the trap...
and /tift-/?mnahlng t!ltll7ttii7C811:
The racing car'5 driver's 5ide door'5 number deca/'5
paint job'5 color'5•••

[33
 Even though we are ju5t at a very elementary 5tage of learning al7out
generative syntax. we can 5ee already that 5entence 5tructure and
meaning interact. Thi5 17ecomes even clearer when we move on from
the derivation of a 5entence to it5 tran5formation. For our 5ample.
the two meaning5 have to 17e di5tingui5hed when the 5entence i5
turned into a pa55ive con5truction:
The fruit in the kitchen wa5 eaten by Charles (tree #I)
V5.

The fruit wa5 eaten in the kitchen by Charles (tree #2)

or
The fruit wa5 eaten by Charle5 in the kitchen (tree #2)
If you under5tood all thi5 with no prol71em.
it'5 17ecau5e you are coming to it from
in5ide the world'5 mo5t efficient and
complex communication 5y5tem. even
though it i5 made up mo5tly of 5alty
water-the human f7ody with it5
17uilt-in mind. Computer5 can't
under5tand anything unles5 it i5
laid out for them 5tep f7y 5tep and
they can't make the choice5 we
make in under5tanding and U5ing
language without in5truction5 to
make them one at a time.

34]
Tree diagram5 and everything el5e that t7elong5 to the apparatu5 of
generative syntax i5 the legacy of the original 5cheme to U5e com-
puter5 to tran51ate human language5. We can 5ee at
...
a glance the difference tJetween the tree diagram5
,,_,~.....__
~

:--\ for the two interpretation5 of Thoma5 ate the

fruit in the kitchen; a computer can't 5ee any-
thing and can only get the difference tJy
tJeavering away at a 5erie5 of two-way choic-
e5-the tJranching point5 of

~ ~~
our tree diagram5.
# ... •
Before we let Thoma5 alone to '\ '
dlge5t hi5 fruit, let'5 notice another point in / ·-
I ~ ,'
... '
thi5 example that i5 of intere5t to lingui5t5. •' '
... '
..
We have pretty well nailed down the ambiguity \ .. ._ "
( more than ~ne po55it71e meaning ) of the 5en- \ " ~ ' ~..~ ~ ~~
tence. the d1fferent 5entence 5tructure5 a550- • •· 4
•

elated with tho5e po55it7ilitie5, and the contra5t5 '- " ' : \
that 5how up when the different interpretation5 are
paraphra5ed a5 pa55ive5. But there i5 5till a lot we do not know and
cannot know from the 5entence, from it5 5tructure, or from it5 word5.
If we compare the 5entence atJout
Thoma5 with the one atJout Charle5'5
favorite uncle, there 5eem to tJe 5imi-
laritie5 tJeyond the t7a5ic one that let5
U5 diagram tJoth with the 5ame tree. Ate i5
a le55 colorful and le55 5pecific ver5ion of even-
tually gobbled up all. the fruit corre5pond5 to apple5, orange5, and
banana5 a5 category to memt7er5, tJut after that thing5 get a little
fuzzier. Can we tJe 5ure that the kitchen refer5 to the room adjacent
to the dining room? It doe5n't have to. 15 Thoma5 Charle5'5 favorite
uncle? There i5 certainly no rea5on to a55Ume 50 (though I will tell
you confidentially that he i5).
[35
 The5e que5tion5 and other5 related to them are f:1eyond the 5cope of
syntax. f:1eyond the 5cope of anything that we can 5y5-
tematize or 5tandardize a5 part5 of lingui5tic5 or
level5 of analy5i5 that lingui5tic5 can handle.
Of:1viou51y the5e que5tion5 are related to meaning.
f:1ut they are f:1eyond the feature5 and dimen5ion5
' of meaning that are integrated into lingui5tic5
under the heading of
semantics. Such que5tion5
are examined in lingui5tiC5 under the
heading of pragmatics-knowledge
af:1out the relation f:1etween what i5 5aid
and what in the world it refer5 to. But we
are getting ahead of our 5tory.

A Punctul!ltlon Pl!lth In the

5Bml!lntlc ForiJIIt of 5yntllctlc TriJBII
What a difference the dot5 make to the right connection5. Here are
two dear John letter5 with identical word5 f:1ut world5 apart. (From
Game5 Magazine. Jqg4)
Dear John.
I want a man who know5 what love i5 all af:1out. You
are generou5, kind. thoughtful. People who are not
like you admit to f:1eing u5ele55 and inferior. You have
ruined me for other men. I yearn for you. I have no
feeling5 what5oever when we are apart. I can f:1e for-
ever happy. Will you let me f:1e your5? Gloria.
Dear John.
I want a man who know5 what love i5. All af:1out you
are generou5, kind, thoughtful people, who are not
like you. Admit to being U5ele55 and inferior. You
have ruined me. For other men, I yearn. For you. I have
no feeling5 what5oever. When we are apart. I can be
forever happy. Will you let me be? Your5, Gloria.

36]
A Middle Ground
For purpo5e5 of de5crii:Jing language fully and accurately, lingui5tic5
i5olate5 functional unit5 of language at the level5 of analy5i5 we
have identified 50 far. phonology, mmplwl~. syntax, and seman-
tics. It i5 important not to over5harpen the di5tinction among
the5e, important to under5tand that when language i5 in action,
unit5 from different level5 come into play together. Here are exam-
ple5 of (I) he and him playing tag a5 a new 5entence pattern devel-
op5 in Engli5h (mo'f'llo~~ and syntax interact); (2) two 5en5ele55
5entence5, where one 5how5 it5elf to I:Je 5Uperior (syntax and
semantics interact).
(f) "Him wa5 given a book:" How Now, Noam Chom5ky?
Mo5t people would agree that there i5 5omething wrong with 5aying
"Him wa5 given a I:Jook." We 5hould 5ay "He wa5 given a I:Jook." ln5tead
of making thi5 kind of judgment and impo5ing a rule of U5age, the lin-
gui5t confine5 her5elf to de5crii:Jing language. A full lingui5tic de5crip-
tion of Engli5h include5 the change5 in the language that have taken
place 5ince the day5 when 5peaker5 did 5ay "Him wa5 given a I:Jook."
Clearly our 5entence mean5 that the book wa5 given to him, and not
that he did 5omething (it'5 the I:Ja5ic joi:J of 5UI:Jject pronoun5 5Uch
a5 he to indicate who performs an action). "He wa5 given a I:Jook'' i5
le55 logical than "Him wa5 given a I:Jook," with it5 u5e of an object
pronoun (it'5 the I:Ja5ic joi:J of an oi:Jject pronoun to
indicate who i5 on the receiving end of an action,
whether it I:Je a punch in the no5e or the gift of a
I:Jook). But we don't much like "him" up front in the
5entence, I:Jecau5e it doe5n't follow the pattern of
5entence5 we u5e much more often 5tarting with
"he wa5" (followed f:Jy any veri:J). The "He
wa5..•" pattern created a pre55Ure that
cau5ed 5entence5 of the type "Him wa5
given a I:Jook'' to di5appear.
The5e are the fact5 of the ca5e
de5crii:Jed from a lingui5t'5 point of view,
and they 5how why lingui5tic5 avoid5
pa5sing judgment on language U5e: what
was "wrong" once upon a time ("He wa5
given a book'') i5 now "right" and what
was at that time "right'' ("Him wa5 given a
I:Jook'') i5 now "wrong."
[37
 \ J
(2) And Chomsky in his Green Period
'
One of the 11est known and often
quoted examples discussed in the
early writings of Chomsky is the sen- --
tence "Colorless green ideas sleep
furiously." His point in discussing this
sentence was to show that what
makes a sentence grammatically well
formed is not predicta111e on seman-
tic grounds alone. "Colorless green
ideas sleep furiously'' has pretty
good syntax, even though semanti-
cally it is a mess. It shares a tree
structure with "Serious university
students work diligently," and that
makes it superior syntactically to
"Furiously sleep ideas green color-
less," even though !1oth are meaningless. So this example may seem to
give a reason for keeping semantics out of syntax, 17ut we have seen
enough other examples already to understand that syntax and
semantics are fundamentally (and functionally) insepara111e.

" => ~L( L

\; ' ~( \~. t ·Yh
~?, ~ ~ Sf'"'
., .1
\ lf\.?'5

rr ·,~"l R't'
~ I x-~
lr-

- 7 ) s
- ·-, <

'-.(_

)
38 ]
How Many J.anguagss Can Rt on 111 P1111nst?
Let'5 take a little 17reak from looking at
how lingui5t5 do their 5tudie5 and look
in5tead at what they 5tudy: lan-
guage5 worldwide. Nol7ody know5
for 5Ure how many language5
there are, though we do know
that their numl7er i5 dimini5h-
ing. (Since 1q28, lingui5t5 have
17een holding an international
congre55 every five year5, and
the theme of the 1qq2 congre55
in Quel7ec city, Canada, wa5
endangered language5). E5timate5
range from under 3000 to over 5000.
r------------------------------, It i5 rare
for a language to have no
known relative5, f7ut Ba5que,
5poken in the Pyrenee5
Mountain5 5traddling France
and Spain, i5 perhap5 the
17e5t-known ca5e.
rL-----------------------------,
I
I
I
~-----------------------------JI
I
Other5 include Ainu (Japan), I
I
Buru5ha5ki (India), and Ket I
I
( Sil7eria ). I
I

r------------------------------1 I

,·f4r~:(;~1 L_ ;;;;.;;;;;;~;;~-;;;,-;;;~~~ -,-;,-;;;,=---"

· \ . , ,~ 3 lie5 that range dramatically in
5ize. The Romance language5,
derived from Latin, number 13
today (the le55er known mem-
ber5 of the family include
Galician, Roman5ch, Ladin,
Friulian, and Aromunian), but
the Bantu group, 5poken in central and 5outhern Africa, run5 to over
five hundred language5 and dialect5. Va5t a5 it i5, the Bantu family
belong5 to an even larger group known a5 the Niger-Gongo language5.
[3q
 Au6tronesian i6 a group of no les6 than 1,000 languages who6e
6peakers are to be found from Madaga6car to New Zealand. So
there 16 no 6hortage of work for lingui6t6 who are prepared to trek
off and do field work recording the Incredibly rich variety of lan-
guages In our world.
There are more than twenty families of native languages In the
Amerlca6, known collectively a6 the Amerlnd languages. But language
families 6pill over geographic boundaries: the Afroa5iatic languages,
as their name Indicates, 6pan Africa and A6ia, and likewi6e the mem-
bers of the va6t Indo-European group stretch from the northern-
mo5t region5 of Europe to the Indian 6Ubcontinent.
If there i6 no single an6wer to the question we 6tarted with (How
many languages can fit on a planet?), we do at lea5t have an idea of
what the 5queeze-rate is here on earth: in ?ctpua, New Guinea, with
an area only 51ightly larger than California and a population of ju6t

~ l/1/ 1 L(I~Clffi®] I

qoo
over three million, there are nearly 6eparate languages.
Wflve been talking here about natural languages, but lingui5ts start-
ed inventing artificial languages long ago, in the hope of making
international communication ea6ier. Esperanto i6 one of the best
known. It blend6 element6 from exi6ting widely 6poken language in a
noble and democratic effort to allow the peoples of the world to
communicate in a common language. But this approach ha6 not
alway5 been the 5tarting point for llngui6tic internationali5m...

4-0]
Nudllllr EnglltJh/8tJtJ/c EnglltJh
Nuclear Engli5h is not English for the atomic age "ut a proposed core
language derived from full English (compare 8a5ic Engli5h) and intend-
ed as an international medium of communication. What a difference a
capital letter makes. Basic English is "asic, "ut it is not simply "asic
English. A carefully worked out system, consisting of 850 words and
the rules for their use, Basic English was developed f:1y British scholar
Charles Kay Ogden (Jggq-Jq57) as an international auxiliary language.
Ogden "elieved it would help promote world peace. Basic is an acronym
for British, American, Scientific, International, Commercial. The champi-
ons of Basic English included such masters of language as Winston
Churchill, Lawrence Durrell, and Ezra Pound. Critics dismissed it, charg-
ing that it turned a sentence from standard English such as "The offi-
cer led his soldiers against the enemy, "ut the enemy stood firm'' into
a clumsy clunker: "The person in military authority was the guide of his
men in the army against the nation at war, "ut the not-friends stood
solidly upright." This version, conveniently awkward for detractors, vio-
lates the rules of the Basic system; if they are respected, it comes
out rather well: "The lieutenant went in front of his men to the attack,
"ut the other side did not give way."
We'll "e looking at more of the languages of the world a little later on.
In the meantime, it's "ack to some "asics a"out how language worlc5.

WhtJt'tJ th11 Buzz?

What do "ees, computers, and traffic signals have in common?
:1\1 They get messages across, in ways that are "oth similar
~~ ~ and different from each other. And especially differ-
h4' ent from the human languages that linguists
1111 \ study and analyze.

ll ~
~Jo~
l
.
A surge of electrical current through "ul"s
under red, green, and am,er glass tells us to
stop, go, or "e careful.

[41
 The lights give their messages to us. t7ut they do not communicate
with us. (They don't much care if we get the message. they don't
expect us to reply. and they wouldn't understand us if we did.)
Bees give us a definite message if they choose to sting us. t7ut for
the most part they exchange information at7out how far away a
source of nectar is and how to get there. Pretty sophisticated stuff
compared to stop and go. and it is expressed in a code of micro-
metrically controlled gestures (often referred to as a "dance") that
leaves the traffic lights 11ack at the comer. Even so. the point of the
11ees' t7ulletin 17oard 17ody language is not much different from that
of the traffic signals-posting instructions.
Now computers seem to 11eat the
17ees for sophistication. 17ut that is
an illusion created t7y the power of
the computer to display data in an
infinite variety of forms. No matter
how complex the data our pentium
pet may 11e handling. it all reduces
to yes or no. plus or minus. one or
zero. galaxies of ones and zeros.
down in those

solid-state gizzards that we

don't see and don't usually think
at7out. And the software "lan-
guages" that run those circuits
at our pleasure are strings of
commands. 11asically no differ-
ent from the stop and go of
traffic lights.

42]
What makes human lan-
guages 5o fundamentally
different from computer
languages. from the DOS
of the t1ee drive. or from that
tricolored winker at the inter5ec-
tion. i5 meaning. in all the complexity
I
that our mind5 can give it. From the ,
feature5 of language that we have r
already had a look at and the part5 (~ '
r
of lingui5tic5 that we
have defined 50 far.

a definition of lan-
guage that will lead
~
we can put together ~~·'"'"C ~f. u
~~~A
~ ~
;
/ -~
~~
1~
'""'_.
u]
U5 to the next level of
analy5i5:

There are a few thing5 we

would need to add to thi5
definition to make it a5
complete a5 po55it11e. for
example that the only place
a 5pecific language exi5t5 in
it5 fullest form i5 in the
memory of memt1er5 of a
5peech community. And of
cour5e while the idea5 17ehind
word5 like 5et and 5ound and
rules are clear enough from
our general knowledge. the meaning of meaning i5 pret-
ty 51ippery. A5 it happen5, The Meaning of Meaning i5 the name of a
t7ook that ha5 t1een in print continuou51y 5ince fqzs. and it will t1e our
5tarting point for looking at semantics. c43
 Semantics (also called "Meaning")
The Meaning of Meaning is f7y two dead white males who gave the
world this, their first 1:7ook, when they were roughly the same tender
age as the Chomsky of Syntactic Structures. (Linguistics has
always 17een a field for young
persons.) Charles Kay Ogden
(u~gq-fq57) and lvor
Armstrong Richards (fgqg-
fqJq ), who met as students
at Caml7ridge University, were
appalled at what a hash linguists
and philosophers alike were making
of the study of meaning.
fhey set al:7out cleaning up the mess f7y
placing themselves at the crossroads
where linguistics meets psychology, philos-
ophy, and anthropology (linguistics as a
whole displays this interdisciplinary
nature more than ever today). For good
measure, they also chose the meeting
point of linguistics
and 5emiotic5.
This is the
general
study of
r-::--- - signs, i.e.,
of anything
that stands for some-
thing other than itself, a
44]
 study extending t7eyond the scope of vert7allanguage. Ogden and
Richards anchored much of their t7ook in the work of the amazing
American thinker and founder of modern semiotics. Charles Sanders
Peirce (lggq-fql4 ).
For Ogden and Richards, the shat7t7y analysis of language in their day
redeemed itself through Peirce, whose work on semiotics first t7ecame
widely availat71e through an appendix they put71ished in 1he Meaning of
Meaning. In the definition of the
sign at the t7ase of Peirce's
work one discovers the
most far-reaching
influence on Ogden
and Richards. Ogden
later pointed out
that logic for Peirce
h /7 was the equivalent of the general
111
11 t(l , ~ ~~ theory of signs. In this respect

. 0 there is also a parallel with 1he Meaning of

'i, SIG Meaning. t7ecause the rules the authors propose for
· keeping meaning under control are a reworked version
of philosophy's traditional rules of logic.
Ogden and Richards are critical of Saussure for reject-
ing the term 6ymbol to designate the linguistic sign.
Saussure reserves 6ymbol for a limited group of signs
l: ~~ that he defines very carefully, t7ut this is
.............~ ..·--·-..:.. ............................~·~ not good .----------~t----1
L Q G JC enough for
Ogden and
Richards. They see Saussure's dis-
tinction t7etween sign and symt7ol
as irrelevant and sut7stitute their
own: symt7ols communicate t7y put-
ting signs into sign-situations. On
this view, all symt7ols are signs;
not all signs are symt7ols. So
Ogden & Richards set up a rela-
tionship t7etween sign and symt7ol
on the t7asis of function, whereas
Saussure keeps them apart on the
t7asis of their relative qualities.
,,
[45
 The Meaning of Meaning is much too long for us to give a full sum-
mary of it here, so we will look at only two more points af,out it.

WordMIIglc
Ogden and Richards use this phrase
to descrif,e thought under the control
of language instead of language under
the control of thought. Much of
Ogden's later work, especially the gsa-
word system of 6aslc English he devel-
oped (see f,elow under Univer5al
Language) were part of a program for
eradicating word magic. Word magic Is
at the heart of taf,oos on the use of
language (such as saying go/dam to
avoid saying God damn) and f,eliefs
af,out the power of language revealed in
cliches such as 5peak of the devil, with its
suggestion that the word devil, and words
in general, are so powerful that they can
conjure up the things they stand for.

Ths Phonstlc 5ubtsrfugs

The Meaning of Meaning is a rich and
erudite f,oolc. It is also written in
smart-alecky. stay-with-me-if-you-are-
smart-enough style and pokes fun at
cumbersome terminology in linguistics
(the kind we are trying hard to keep out
of this book). This is what the authors are doing when they come up
with a series of phrases that starts with phon•tlc 5ubterfuge. If we
Imagine that a word follows the pattern of meaning
of other words just
f,ecause it sounds similar,
Ogden and Richards declare,
we are guilty of the phon•t-
lc 5ubterfuge. An example?
(Ours, not theirs:) "Leasaf,le''
means availaf,le for lease;
"loanable'' means available for '
loan; "lovaf,le'' does not mean
availaf,le for love.

46]
Some of Ogden's and Richards's whimsy has f:laffled even the most
erudite of readers, including polymath Douglas Hofstadter, the
author of Metamagica/ Themas, Le ton beau de Marot, and Godel.
E5cher, Bach. Nevertheless, the f:look inspired Hofstadter to pen
these lines:
Two experts, to explicate Meaning,
Penned a text called "The Meaning of Meaning,"
But the world was perplexed,
So three experts penned next
"The Meaning of Meaning of Meaning."

Not to Put too Rns s Point on It

But even those who walk away from
The Meaning of Meaning scratching
their heads rememf:ler its three
main points:

TIIOUGHT OR REFERENCE

.···················································
Stands for REFERENT
(an imputed relation)
TRUE

\
... (WELL YOU
KNOW)''

Few scholars have appreciated

that, far from f:leing a mere visual
convenience to readers, the triangle
unifies the ideas developed in The Meaning
of Meaning. (It is also another link from Ogden
and Richards to Peirce.)
[47
 But why a triangle diagram? Because its three points are the
indispensa,le ingredients of meaning. Four are not required an
d two are not enough.

Symbol= word, phrase,

world the sym,ol stands for ( · .--
senl (~_);referent= what in the
); thought= I) what referent the
sym,ol make5 us think of when we hear or read a message; 2) what
sym,ol the referent make5 us think of when we want to tell the
world a"out it ( _, .· fiS·If we take away any one of these,
there is no
\'
communication.

~~,, · -~:;.: \ ·, There are many ways to define meaning and many
~~~~~-·
~ '~/ - ''"! ;;
4 ' . ,~.
.~. approache5 to the study of meaning. If we focus
~-" 1 .) ·; : on words, we can look at them individually
~ ~
- \ /
1 (e5pecially those that have complex
1
1
/ ' meanings), in pairs (e5pecially those
that have similar or opposite mean-
ings), or in groups (e5pecially those
that share meanings). Or we can study
how the5e various state5 of meaning
came a"out.

Multlpls M1111nlng•
Many words have more than one meaning. This is normal. The more a word
is used, the more it occurs in new contexts, and the more chance it has of
acquiring new meaning(s) from those contexts. An extreme case in English
is the word set, with 12g meanings listed in the Oxford English Oictionary.

48]
Homonym5...

NOAM

•••are
word5
that 5ound the same (though they are not nece5sarlly spelled the
same way) "ut have different meanings. Homonyms can lead to con-
fusion, even among persons who know a lot a"out their language.
Htlrt1 '- 11n tlXIImpl~~:
There are versions of the Cinderella story in many languages around
the world. The one commonly told in English came via France, where
Charles Perrault (162g-f703) recorded in writing the
stories told to his children f;7y their Basque
nurse. The gla55 5/ipper figures in Perrault's
version, and corresponds to the crystal 5lip-
1/ /
per of the Scots-Irish and other versions
of Cinderella. Well and good, "ut not good --.
enough two centuries later for two distin-
/
guished French literati, Honore de Balzac
and Emile Llttre, who assumed that a con- /;
fusion had arisen "ecause the French words
for gla55 ( verre) and for
5C[Uirrel fur ( vair) are pro-
nounced the same. Surely
a fur slipper makes
more sense than a glass slipper.
May,e, "ut this is a story. And
it's a crystal or glass
slipper even In languages
where the words for gla55
and fur are not pronounced the
same way and where no confusion could
have occurred. How now, messieurs Balzac
et Littre?

[44
 In Old Engli!Sh, sam, the equivalent of Latin semi-, meant "half." Sand-
blind, atte5ted in the fifteenth century, i5 likely a variant of
samblind-"half-blind." The prefix wa5 already
well on it5 way to being a55ociated only
with sand when Shakespeare wrote in The
Merchant of Venice: "fhi5 i5 my true
begotten father, who being more than
5and-blinde, high gravel blinde, know5
me not." And that pillar of erudition,
Samuel John5on, defined sand-blind in A
Oictionary of the English Language (1755)
a5 "having a defect in the eye5, by
which !Small particles appear to fly
before them." Our cited phra5e
come5 from the poetry of Walter
de Ia Mare: "Hope... Led !Sand-blind
Despair To a clear babbling weii-
!Spring And laved hi5 eye5 there."

...are word!S that !Share meaning. It i!S

a lot le55 common to find two or more
word!S with exactly the !Same meaning.
And the que5tion of !Style come5 into the
picture. A poem titled "The Death of Roget" by
George Hatch, Jr., give5 a whim!Sical account of the fate of Peter
Mark Roget, the man who gave the world it!S be!St-known dictionary
of 5ynonym5:
50]
 Said Roget:
"What's the purpose? Tell me why.
•~i:iiililillii:;; What's the reason? Specify!
The other was silent.
speechless. mum.
close-mouthed, firm-tongued.
tight-lipped and dumb.
word-bound. curt. concise and brief:
He drew a gun. a gat. a rod.
and waved it in Roget's favade.
his face. his mug.
his map. his lug.
his kisser.
"Now wait a minute!" cried Roget.
"Wait a second! Stop! Delay-!"
That's all he said. the other fired.
shot him once; Roget expired.
pegged out. conked out.
kicked the bucket.
croaked and piped. With
Roget's luck it
means he's dead. defunct. passed on.
deceased. demised. lamented. gone.
They wrapped him up in polished oak.
a coffin. casket. wooden cloak.
and on his grave these words bespoke:
"Lie in peace; to God bequest...
R.I.P. and all the rest ..."

[51
5emantlc Accidents
In our last four sections. we isolated
and defined features of language.
illustrating them with
various words. word
pairs. and word groups.
These same features of
language can come togeth-
er in a single example. as
in the case of the phrase
piggy bank. This happens by
chance and produces some
interesting results.
Once upon a time, there was a good old
English word pygg. meaning a type of material much like earthenware.
Containers made out of pygg were called pygg6. So far. no surprises.
After all. a container made out of glass is called a gla66, one made
out of tin is called a tin. etc. A pygg used for storing coins was called
a penny-pig. pence-pig. etc. When pygg went the way of ceramic ham-
mers. speakers of English soon forgot what the pygg part of penny-
pig meant and made a new connection instead with pig. As a result.
coin banks began to be fashioned in the shape of the animal. So the
chain of events here takes us from the development of a new meaning
for pygg (container) through loss of meaning (no more pygg as mate-
rial) and homonyms (pygglpig) to the modem piggy bank. At this last
stage. there seems to be an obvious reason for the thing to be called
piggy bank. but if we know the history of the word, this turns out to
be pure fiction.

A Rurry of Word• for 5now

Even without having read one word
about linguistics before. you probably
have a vague idea that there are "a
lot of Eskimo words for snow."
When we look closely at the
data. we discover that
one part misinforma-
tion has been mixed
. .,. .,= :: -----------
~ll{_~:::::::lliiif' ~ - _,:; with ten parts of
~ -~:.---- exaggeration to
'"' >¥- ~ - ~ produce some boot-
~ ~ ;~::_ leg linguistics here.
'J< ~~ ~ - Arealsnowjob.
 It all started quite
innocently when the
grandfather of
American linguistics,
anthropologist Franz
Boas (rgsg-Jq42),
pointed out that just
as different words in
English express different
forms of water (lake,
river, brook, etc.), so
too "Eskimo" [we're
going to explain the
reason for the quotation
marks two paragraphs
from now, so we'll stop
using them in the

i
meantime] has a variety
-*- of terms: aput 'snow on
-J{ the ground', gana ~
# ~ ;){- 'falling snow', piqsirpoq / "'- }
-*" 1\J 'd;ifting snow',~nd <J0 / .
~ ,- ~ q1muq5uq 'a • <"-
~ k v~. +- snow , t,'/ /
-¥ ~ * ,.- . drift'. J' ~
¥* . ~~/

_ AI f~ar;~-~~Q/Mu~~v~~
____..--- \ ,\i\111 · \1 I I \ ) ' ~
'~ ~ ·' ,~\~ ~
~ ~~ ·j,~'~,;~ :l !l:f . ~
.,

__::_---- \1 \'~~, j ' ,,, ---

~" ~\JI .

[53
 f3oa5'5 point wa5 that Engli5h get5 the5e idea5 acro55 with phra5e5
u5ing the 5ingle term snow. though It could have evolved a variety of
form5, a5 it did for lake. river, brook. etc. In other word5, when we
look 1:1ack to the original comment l:1y f3oa5, we di5cover that he i5
pointing out not a contra5t 1:1ut a 5imllarity 1:1etween how language
can and doe5 work in the ca5e of Engli5h and E5kfmo.
f3oa5'5 df5cU55ion made ft5
way into the wrlting5 of lin-
guf5t Benjamin Lee Whorf
(JgCf]-Jq4J), who turned the
four E5kfmo term5 cited l:1y
f3oa5 (Did he know of other5?
If 50, he didn't 5ay.) into 5even
phra5e5 al:1out 5now in Englf5h,
implying that there could 1:1e any
numl:1er of additional one5 in E5kfmo/ The 1:1ea5t wa5
loo5e. In the meantime. there have 1:1een "report5" of
up to 400 E5kimo word5 for 5nowl
If f3oa5 were alive today, he might well a5k what are the sq6 that
my5teriou5ly got added to hi5 example5. He would
certainly 5hake hi5 head at the 5hal:11:1y 1:1rand of
lingui5tic5 that
refer5 to a non-
exi5tent, generic
"E5kimo" lan-
guage and at the
failure to make
the di5tinction
1:1etween words
a5 5uch and
roots from
which any num-
1:1er of word5 (al:1out 5now or ice or ju5t al:1out anything el5e) can 1:1e
formed in language5 5uch a5 La1:1radorian Inuit and We5t
Greenlandic. Here are 5ome example5 and their English equivalents:

54]
 Labradorian Inuit
*****************
pukak - granular snow
masak - soft snow
mauja - soft, deep snow
mangokpak - watery snow
massalerauvok - snow filled with water

.~fl{l~W~ A total of 4q words for types of snow and

'?~~~: z~ 1:! ice in West Greenlandic is given on the website
cl www.urbanlegends.com.

How Many Linguists Ooes it Take to Change a Lightbulb?

Eight. A semanticist to write a paper on the meaning of" a lightbulb,"
a semiologist to write a paper on the meaning of a lightbulb, a jour-
nal editor to point out that, following Saussure, not everyone would
agree that a semiologist is a kind of linguist, a Chomskyan to set
the switch to on or off, a Neogrammarian to reconstruct the dead
bulb from all the burnt out ones, a proponent of inter-language to
analyze the complex transfer phenomena before attempting a
changeover, a discourse analyst to record, transcribe, and analyze
the work of the other linguists, as they argue in the dark, and a
pragmaticist to find the right words to persuade the janitor to
change the bulb right away.
[55
 Ths 8SI!Jt Mlll!lnlng I#J ths Lsll#Jt Mlll!lnlng
Thies estatement may eseem like an odd way to end our look at seman-
tics, but it i5 nothing more than a le55on on the importance of get-
ting meaning from context. It i5 a way of 5aying that if we do not
know the exact meaning of a word when we hear or read it in a 5pe-
cific context, it i5 esafe5t to choo5e among the poes5ible meanfng5 by
taking the one that make5 the word lea5t different from the mean-
ing already eset up by that context. fo illu5trate the point, let'5 take
aes an example a esentence where we do know the meaning of the final
word, where that meaning i5 50 obviou5 that we would not normally
bother to check for the other po55ible meaning5:
The rider dug in his spurs to urge on his mount.
fhere is enough coherent information in the sentence
before we get to mount to tell us that it is being U5ed
in the 5en5e of horse. We eliminate the meanings i)
object used to diesplay another, ii) object U5ed
for 5upport; iii) hill or mountain. We even
eliminate meanings more closely related
to the context (act of mounting, style
of mounting) virtually without
thinking about them.
fhe be5t-meaning-a5-
least-meaning rule 5imply
reminds U5 that the way we
process meanings more or le5s
automatically when confronted
with known word5 is the 5ame
way we eshould proceed when
a word puzzle5 ues.

56]
 6y W•y of llJwltJW (but no quiz)
Look at the two sentences following. What is special a"out them?
What have you learned a"out linguistics so far that can help you
de5cri"e them?
.>:" · ~~ . ,.

_.:.:..:.:~::.~\?:~ He's none too wise, replie5 the old man, muttering.
'\~/£~.;4 He's known two wise replie5, the old man muttering.
~ • ~ 1/ /
Answer (we're not even going to "other {1-~ ~
.
~ vr-~\\
~
\/11
,·-::?'
=turning it upside down at the "ott om
'
)
~: '\
~~ ~ ~
of the page : \·~ :,\"- ~~·
A. ~~- ; :1J ·'"'1 _
I) The sentence5 are different t1y
only one phontlmtl (the vowel In
none and known);
~ .- i\!l•~ ·
~~ 1
111

1
JLJ
.
2) They contain the homonyms too and two; '. ~i
g) They have very different meanings "ecause of \
I) and 2) and "ecause they have different tree
structure5.
4) The different meanings for the two sentences can "e recognized
in part t1y the contrast in intonation (rise and fall of the voice)
"etween them.

A FEW RNAL LE550N5 ON HOW LANGUAGES WORK

From Tones and Whistle5 to Clicks and Kllngon
and the Case of Case
In our warp-speed survey of linguistics
we have had to leave out a lot of lnter-
£:..
e5ting facts a"out how different lan- ~ :-<>' . _ ::od
guage5 work and the challenge5 that f-r· •
linguists face in analyzing them. Here ·
are a few to help us fill in the picture.

(1) N•ms thllt Tons

Cantone5e, Hausa (Nigeria/Cameroun),
Mandarin, Margi (Nigeria), Thai, and
Yoru"a (Nigeria/6enin). Like more than
50% of the world's language5, each of
the5e is what is known as a tonal language, where the tone5
(change5 in pitch) make a difference in meaning. Say "Ma" with a
high and level tone in English and you are calling your mother; say It
with a high and falling tone and you are letting her know you are
exasperated with her, "ut she's still your mother. In Mandarin, make
[57
 the same switch and the
meaning of the sylla,le
"ma" goes from "mother''
to "scold."
All tonal languages
have at least two
tones, and there
are some with
eight or more, "ut
these are rare.
Differences in
meaning "etween
the same sylla-
"le in two dif-
ferent tones
are not always
as radical as in our Mandarin example. In Nigeria's Kanuri language,
for example, tone distinguishes among the tenses of a vertJ.
In writing, tones can only "e shown tJy sym,ols that differ from each
other, even though the sounds they represent are the same, as hap-
pens with our two mas in Mandarin (and two more that mean
"hemp" and "horse"). But they can make it onto the airwaves tJy
other means than the human voice, as in the case of African "talking
drums" that reproduce the rhythms and tones of speech.

5fl]
(2) Colons/ Bogsy Would H.vs Lovsd It

they normally velong to, can "e whistled

formed
f,y
The5e same rhythms and tones, detached from the syllavle5 that

f,y speakers. The patterns

the rhythms and tones are distinctive enough even with-
out specific speech sounds to make communication possivle. The

in the missing sounds. A5 long ..

f,yy
whistled me5sage is interpreted ~fi:•":in~g:..-~~-::~----
1
as you and the person you are
conversing with stay in key,
there will ve no confusion over
""all" and "small", etc.

Mln/-f(sv/sw
Oo ball and 5mall make a minimal
pair? Look vack to our section on
J
"ecause they differ f,y
phonology. The answer is "no,"
more than
one sound. What a"out ball and
mall or ball and all? "Ye6" to
voth, vecause the difference in
meaning in each pair depends on
a difference of just one sound.
There are no whistled language5,
no communication systems
consisting entirely of whistle5,
only whistled versions of tonal
language5. Whistling is an
op'tion here.

[sq
 (s) Click Here, Click There
Clicking is not an option in the Khoisan languages of southern
Africa. Practically the only clicking sound we make or recognize as
speakers of English is the one usually
written tsk, tsk-a signal of disap- ~ ~ L-
proval. It is a full-fledged click, , S
.J..' k -v 5
1 -----~
f\
produced by pulling the
tongue away from the upper
teeth. A distinctive sound
( V
=
results when the mouth is
suddenly unblocked by the
movement of the tongue and
air enters.
Variations on this process are at work
in all the languages of the Khoisan family, where the difference in
pronunciation between various clicks makes just as much difference
to meaning as lpl, It/, and lml do in pan, tan, and man. Clicks can
occur at just about every point of articulation in the mouth and can
even accompany other sounds produced in the throat or the nasal
cavity, making for a huge number of possibilities. The !Xu language
has 4g different varieties of clicks-in addition to 47 other non-click
consonants! We are a long way from tsk, tsk, but not as far away as
the speakers of Klingon .•.

(4) No Clicks In Kllngon

...those aliens in Star Trek. Paramount Pictures wanted a harsh and
guttural language for them, and Washington linguist Marc Okrand
provided it. Though there is more of cult than culture surrounding
Klingon, the non-profit Klingon Language Institute describes its
chief aim as being the facilitation of scholarly exploration of the
language and its culture. Written Klingon (a translation of Hamlet is
available) uses the English alphabet but breaks with its conventions
by using capitals wherever required in a word as reminders that the
pronunciation of a letter differs from that of English. A beginner's
guide warns that confusing q and Q in Klingon is as serious as con-
fusing If/ and lg/ in English.

60]
 ~ ~ ~ 1> r ;z q

~
I
a
,
b

j
ch

l.
1
D

(t
II n
e gh

ng
H

~ 1 I "'? ..t ~ ~
p q Q r s t tlh

J-. 1-4 f' q

u v • y

-
0
r
1
(
2
(:
3

"(
7 •
8
t
9

Kllngon'6 gh i6 not the gh

of rough, through, or any
other Engll6h word "ut a
gargled 6ound 6omething
like a French lrl. The
a"6ence of cllck6 pro,a-
"ly doe6 not make
Klingon any ea6ier to
pronounce than IXu. q, for
example, mu6t "e pro-
nounced a6 far "ack in
the mouth a6 po66i,le with the "ack of the tongue touching the
uvula. It i6 related to Q, the difference "eing that the fir6t 6ound6 a
little like you are choking and the 6econd 6ound6 a lot like you are
choking. If you don't find th16 too Intimidating or di6couraging, you
could 6oon join the legion6 having fun with Klingon or practice the
lingui6tlc 6kill6 you are acquiring here l:1y analy6ing the language.

[61
 (5) CIJII8
This has to do with the way in which the connections among words
are expressed. In English we make many such connections t7y using
words "elonging to the category known as prepogitiong (of, by, with,
in, from, etc.). In other languages (called inflected languages), it is
etxeaz - t7y means of the house
etxearekin - with the house
etxean - in the house
etxetik- from/out of the house

ETXEAREN

ETXIET~K

62]
 ~ LANGUAGES IN CONTACT

'-..{_
lt'5 not a joke. Tok
Pi5in i5 the name
of a language. :
•rt •••••••••••••r:••:
( .'t/ We Talk Tok Pi5in Here, They Talk Tolaki There...

Tok-Fisin.com
The name come5 • • • • • • • •
from talk pidgin. • • •••••••••••
(When the 5peaker5 of two or more language5 in
contact work out a common mean5 of communica-
tion with a limited num1:1er of word5 from their lan-
guage5 and very 5imple grammatical rule5, the product i5
called pidgin.) Tok Pi5in, 1:1a5ed on Engli5h and in U5e
5ince the end of the nineteenth century, i5 5poken 1:1y
over two million people a5 a 5econd language in
Papua, New Guinea. The older name for Tok
Pi5in i5 Melane5ian Pidgin Engli5h. Here
are 5ome 5ample5 of Tok Pi5in.
Let'5 5tart with one word:
yumitupela
It wa5 four word5 in Engli5h:
you+me+two+people. Thi5 i5 the pro-
noun we, when we refer5 only to the
per5on 5peaking and the per5on 1:1eing
5poken to.
Now full 5entence5:
f3andarap em i kukim (Bandarap cooked it).

Meri hilans i karamapim het wantaim bilas

(Women from the highland5 cover their head5 with decoration5).
You can check out the daily new5 in Tok Pi5in from the Radio
Au5tralia home page.

TOK PISIN ·tf//ff// I

[63
 Map a de Pid!Pns y Criollos del Pacifico
BasaiooseJtel~

- Basalias ea el eQaiial

- Bas alias ell el ~

Basalias ea el ~

- Banoloos eJt autro,..;o:U:o

O.iano Ptlclfu;o

Criollo Nodiolk
\,

rolaki. an Austronesian language. is the mother tongue of af1out

125.000 speakers in south-eastern Sulawesi. Here's a small start on
learning the very distinctive names of some of the languages on
space-ship Earth:
Bof1o Fing - a Niger-Congo language of Burkina Faso and Mali
Li.i - Known f1y three other names. LU is variety of the Tai Yuan lan-
guage of Northern Thailan. It is a tonal language (see f1elow) spoken
in Yunnan.
Gutnius
Jon I raitim
Bipo .po tru, laina oa.ta
I 6 Wupela . . . i lralnap, Delli biloq
1 SUMiq i blup yet, Tot i
DO
Tok i aap wantaim God u Tot em
en Joa. God i biD aaliln C111 I baa. 7 Em
ltap.
i bill biloal audal tok. Em i autiiD tok
yet i Oocf. J Bipo bipo tna Tot i Up bUoaa dispela IM, I* ollela ........-1 I
waDUiila Ood. f Loaa dilpela Tot taiOI kea lwim tok blloat eaa, na ol i kea
Ood I ...tbD bmap olida -dlta· H. bilip. I Sal )'eC all i 110 dilpela a.it.
I DO pi waapela .......... i biDap ~ Nopt. Ea i baa biJolla auaila tok tuol
urapela rot. Nopt. 011eta IMIIIilll • biJoaa dispda lak. 'DiiDda Wt em I lait
kuup, a11 Tok yet i metiiD lluulp. tru Da i save f.ivi• 1d loa& oltda
• LaiP i ltAP loQiea~, na clispda laiD aa IIWIJDCri, a11 i laik i tam lout araua.
i a.k biJoaa ol ......m. ' Diapela lait i
lll'le lail i .., loal tlldak. .. tuciU i
ao bill ......._ em.
64]
Zaza - Thi5 i5 an Iranian language, quite di5tinct from Kurdi5h, 5po-
ken t1y perhap5 two million per5on5 in Iran and Turkey.
Bi5hnupriya
Manipuru - It i5 a
little ea5ier to
rememt:1er tJy it5
other name: Mayang.
Thi5 i5 a form of
Bengali (an Indo-
Aryan language with
UW million 5peaker5
today in Banglade5h
and India). Mayang
i5 5poken in three
region5 of
Banglade5h.

Fox - a memt:1er of the huge Algonquian family

of at:1original language5 of North America
Kic;kapoo - a memt:1er of the huge
Algonquian family of at:1original lan-
guage5 of North America

Hyperoorean - The word referred

in c;la55ical Greek to a mythical
people who lived "1:1eyond the
North Wind" ( thi5 i5 the direct
tran51ation of Hyperborean). In
lingui5tic;5, thi5 i5 the older name
for Palaeo5iberian (meaning "old
Sit:1erian")- minority language5
of Siberia that are unrelated to
other language5.

(65
 Loka of Mustang - also known as Kagate. a TI17etan
dialect spoken f7y the Sherpa and their kin in Nepal
Romt71omanon - an Austronesian language with 200,000
speakers in the Romblon and Sit7uyan Islands
Sre - an Austroasiatic:; language
Here are a few more names of languages, some with
ec:;hoes of those at7ove, t7ut one purely fictitious. Guess
whic:;h?
5ot7o Wule, Li, Hre, Roratongan, Zezuru, Kimt7undu, Coeur
d'Alene, Lokko, Iron, Huihui, 5inkoka, Yi.

Oid you guess the one that seems to stic:;k out the most,
Coeur d'Alene (Frenc:;h for "Alene's heart")? This is the
name of a language in the Salish family of North
Americ:;an at7original languages. (It's also the name of a
town in the great Americ:;an state of Idaho.) All the oth-
ers are genuine too, exc:;ept 5inkoka, though it has an
ec:;ho suggestive of EJingkokak, an Austronesian language.
66]
Crt10lllt!J
A creole i5 a language de5cended from a pidgin (look
t7ack to our Tok Pi5in 5ection). In 5ome ca5e5. a pidgin
t1ecome5 50 well e5tat1115hed in a multilingual 5ociety
that a generation of 5peaker5 grow5 up U51ng only the
pidgin to communicate among them5elve5. When thl5
happen5. the vocat7ulary of the pidgin expand5 and a
full grammatical 5y5tem develop5. promoting the for-
mer pidgin to the 5tatu5 of a complete language 5y5-
tem. Thi5 type of new language. when it develop5 5pon-
taneou5ly. 15 called a creole. The tran5formatlon of a
pidgin into a creole i5 called creolization. (Engli5h and
African tongue5 t71ended to form the Gullah creole
5poken on a narrow coa5tal 5trip of South Carolina.
Georgia. and northea5tem Florida.)

[67
 Ungua fnlnCI!I
Thi5 i5 the name given to a language
u5ed over a large area where people
5peak a variety of language5. The
term originally meant 'Franki5h lan-
guage.' i.e., the language of the Frank5;
the ancient Germanic trif:1e that lived on
the f1ank5 of the Rhine in the early
Chri5tian era. (The Salian Frank5 gave
their name to France when they con-
quered Gaul in the 5th century.) During
the late medieval period, the lingua
franca u5ed for commerce in the
ea5tem Mediterranean wa5 an Italian
dialect mixed with French, Spani5h,
Greek, Araf:1ic and rurlci5h. The term i5 U5ed to
refer to any language adopted a5 a mean5 of communi-
cation f:1y 5peaker5 of different language5 who inhaf:1it the
5ame region. In it5 f:1roade5t 5en5e, lingua franca refer5 to pidgin§
and/or creole5, though mo5t
lingui5t5 would reject thi5
u5age a5 too loo5e to f:1e a
reflection of lingui5tic fact5 a5
we have de5crif:1ed them here.
Prof:1af:11y the mo5t generally
accepted u5e of lingua franca
among lingui5t5 at pre5ent
would f:1e a5 the term for a well-e5taf:11i5hed language who5e native
5peaker5 have the 5tatu5 of a pre5tige group, a5 a con5equence of
which the language come5 into u5e for communication among
group5 of 5peaker5 of variou5 other language5. Thi5 i5 the ca5e for
Swahili in Ea5t Africa, Hau5a in We5t Africa, and Engli5h in
Singapore and el5ewhere.

LangUllglll!l /Jill Ufs mfts

rwo tran51ator5 are talking 5hop af:1oard a 5hip when it 5uddenly
f:1egin5 to 5ink:
"Can you 5Wim?" a5k5 one.
"No," replie5 the other, "f:1ut I can 5hout for help in nine language5."

M]
HOW HUMANS LEARN THEIR LANGUAGES
Ths Ungulst Within
An idea that Noam Chomsky and other lin-
guists have promoted In recent years is
that we are 17om hard-wired to learn lan-
guage. The jury is still out on this question,
f7ut we know for sure that children acquire
the complexities of language with amazing
speed and facility. This happens In a way
that is marked f7y such a high degree of
organization, that it gives weight to
Chomsky's claim. So what does the syntax
of an eighteen-month old linguist look like?

Basically, it's a set of

rules that let tiny tykes
put words together to
express relationships-
~--
the same relationships
that show up later when
the little gaffers get full
control of language-
such as topic and modifi-
i er, verb and of7ject, su/:7-
ject and predicate. And just like full-17/own syntax, the toddler's
scaled-down model excludes certain comf7inations. This has nothing to
do with "f7ad grammar'' (who could understand such a concept when
they are still in diapers?) f7ut
everything to do with Impos-
ing limits and restrictions
that make the childs
emerging grasp of lan-
guage (and the world
that language is
af7out) coherent
and systematic.
 Single word5 are u5ed a5 full 5en-
tence5 when infant5 f1egin to 5peak
(the word infant come5 to U5 from
Greek and mean5 "not 5peaking").
Our lingui5t taking fir5t 5tep5
acro55 the living room
floor and into the wide
and wonderful world of
language may know
f1oth "f1ye-f1ye"
and "daddy"
f1ut will not u5e the word5 together.
When, at around rg month5 of age, the child
doe5 launch into two-word phra5e5, her
entire vocaf1ulary fall5 into two group5. The
larger one con5i5t5 of word5 that can 5tand
f1y them5elve5 to form a me55age (5hoe, hot, milk), f1ut word5 from
the 5maller 5econd group are never U5ed alone (my, green). ln5tead,
they occur in comf1ination with a word from the fir5t group. The full
5et of rule5 for a generative syntax of the 5entence5 of eigh-
teenth-month old 5peaker5 can f1e 5ummarized a5 follow5, where 5
5tand5 for a 5en- -
tence, I for potentially Cjree ~"- ( Ml LK
independent word5, R ~'~"'( v
: ki DAD 0 1
forword5 re5tricted _ -~~~ · ~ ~
go-ne(A,_-
~~~-~
~
'-::~.-·. _--...-_· ~,r
'~
1
to comf1ination5, and H0
the lingui5t'5 arrow ~
ha5 the 5ame function \~~~o ~ .
a5 the mathemati- ( ~;-,~~ :_-::·~'·)
cian'5 equal5ign (=). ~ L2_1~

s~l

s~ 1+1
s~I+K

s~K+I

70]
So, of all 5entence type5 that could occur, two do not, namely 5 -
> R and 5 -> R + R. fhi5 i5 not a theory. Thi5 i5 auout what young
5peaker5 do and do not 5ay: "Milk hot," "Green uall," "Daddy gone,"
"My 5ock'' uut never "Green my'' or "My green" or "My'' or "Green".
Mo5t children learn to make 3-word 5entence5 uy their 5econd uirth-
day. The5e 5entence5 too 5how pattem5 that prohibit certain com-
uination5. Generative syntax, a5 Chom5ky pointed out long ago, i5
very powerful, too powerful even at the .5imple 5entence 5tage,
uecau5e not all po55iule 3-word comuination5 are u5ed. The rule5
that generate even the fir5t 3-word 5entence5 found in child 5peech
are ua5ically the 5ame in form a5 tho5e u5ed in the de5cription of
adult language in all it5 complexity. It 5eem5 that children uegin to
get a handle on lan-
guage i7y recogniz-
ing different rela-
tion5hip5 among
word5. When thi5
knowledge i5 ma5-
tered, they proceed
to learn tran5for-
mational rule5 and
procedure5 for cre-
ating the fully-
formed 5tructure5
of their language
community in all
acceptaule vari-
etie5.

[71
72]
 THE BIRTH OF LANGUAGE,
THE GROWTH ANO OEATH OF LANGUAGES
WordOns
Human5 are naturally curiou5 al:1out how language 1:1egan. We might
a5k: what wa5 the fir5t word ever 5poken? lt'5 an intere5ting que5-
tion, 1:1ut from the lingui5t'5 point of view, not a very good one,
1:1ecau5e it make5 an a55ump-
tion that i5 not true. If there
i5 a word, it ha5 to 1:1elong to
a language 5y5tem. The fir5t
utterance could not have 1:1een
a word, 1:1ecau5e there wa5 no
5y5tem in place yet for it
1:1elong to. Language df!Neloped
over a very long time 1:1efore
anyl:1ody ever wrote down any-
thing al:1out anything, much
le55 al:1out language it5elf, 50
the an5wer to our que5tion5
al:1out the origin5 of language are lo5t forever in the mi5t5 of time.
Thi5 ha5 not 5topped 5peculation, and at lea5t five main theorie5
have emerged. We give them here under the popular name5 j,y which
they are known.
Ths Bow-Wow ThtJOry
Spoken language 1:1egan with
human 1:1eing5 imitating 5ound5
in the natural environment, par-
ticularly animal cal/5.

Ths Pooh-Pooh ThtJOry

Spoken language 5tarted with
human5 producing 5ound5 in5tinc-
tively in re5pon5e to pain, anger,
and other 1:1odily troul:11e5.
[73
 Ths Ding-Dong Thsory
The root6 of language are to
"e found in 6pontaneou5
human reaction to 6timuli in
the environment, taking the
form of 5ound6 which mirror or
harmonize with that environ-
ment. (Thi6 i6 the Bow-Wow
Theory making the move from
the kennel to the co5mo5.)

Ths Yo-Hslws-Ho Thsory

The origin6 of 6peech can "e
traced to communal. phy5ical
la,our. the exertion required
producing rhythmical grunt6
that developed fir6t into
chant6 and eventually into
full-fledged language.

Ths 1.8-I..IJ Thsory

Thi5 i6 the contri,ution of the romantically inclined to explaining the
"eginning5 of human language: it all 5temmed from 5ound6 a55oci-
ated with love. the irrepre561,1e joy of the human 6pirit. the expre5-
6ion of poetic feeling. the 5U,Iimity of the 5ong of the heart that
mu5t out.

74]
LllngiHigs Ch11ngs
Change in language is inevita,le
with the passing of time.
Languages change constantly.
Words are added to the voca,u-
lary. some disappear. and the
meanings of others shift and
grow. Pronunciations. grammatical
forms. and sentence structures
change too. A5 long as a language
remains in use. change cannot "e
avoided.
The why of language change is
complex and varied: new inventions.
new concepts. and new activities "ring in new words; words may
acquire or lose social prestige according to circumstances and
"ecome more widespread or rarer as a result; grammatical struc-
tures may "e simplified as a result of high frequency of use; and
contact among speakers of different languages may lead to the
"orrowing of forms and usages from one linguistic community or
the other.
The result of never-ending language change is a steady increase in
the divergence "etween a language at any stage in its development
and the form of the language from which it derived. And to compli-
cate the picture. a language spoken over a wide area undergoes
changes at different rates in different places. The result. over a
long stretch of time. is a "reak up. At first. distinct regional
dialect5 emerge; finally. divergence "ecomes so great that different
languages come into "eing. and a language family exists where origi-
nally there was just one language.

[75
GEOGRAPHICAL UNGUI517C5

Now our topic i5 geographical lingui5tic5-language in relation to

5pace. (Diachronic lingui5tie5 i5 Sau55ure'5 term for language in
relation to time and it will come in again here.) Sau55ure 5ay5 that
the diver5ity of language5 in 5pace i5 ea5ier to think aj,out than the
diver5ity of language5 in time, j,ut you can j,e the judge of that for
your5elf. He 5tart5 hi5 le55on5 on geographical llngui5tic5 j,y talking
aj,out tendencie5: the tendency to think of language a5 a local cu5-
tom, the tendency to a55ume the 5Uperiority of our own language,
the tendency to look for 5imilaritie5 in language5 once differenee5
have j,een e5taj,li5hed. He al5o talk5 aj,out an impo55ij,ility: the
impo55ij,illty of ever finding out whether all language5 came from
one. There ha5 j,een too much language change to get an an5wer to
thi5 que5tion. But thi5 doe5n't keep the lingui5t from comparing lan-
guage5-tho5e that are very different from each other a5 well a5
tho5e that are 5imilar.

[77
 Three thing5 can happen when different
language5 come together in the 5ame
territory:

They get mixed together like the

ingredient5 of a cake or a 5tew.
Thi5 i5 what happened to Engli5h
after the Norman Conque5t (1066
and all that). Engli5h a"5orbed
~Jt;liii~liii~~Sj~~ element5 of the language of the
· conquering French.

5 WI S S G-ERMAN
SWISS FRENCH
SPoKEN HE Rf
They each get U5ed in SPOKEN HERE
their own 5eparate ~ /
piece of the territory.
which i5 the ca5e for
French and German in
5au55ure'5 "ithplace-
Switzerland.

They co-exi5t without getting mixed

together. In the Ba5que region of
the Pyrenee5 mountain5 "etween
France and Spain, French. 5pani5h
and Ba5que are 5poken. At the time
of the Roman Empire, language5
co-exi5ted thi5 way all around the
Mediterranean.

7i]
When languages are related. how they differ from each other can be
observed and traced back to what unites them. 5aussure's big
interest here is getting at the cause of the differences. He points
out that distance alone does not create differences. Time is the
determining factor. If half the speakers of a language were packed
off to a new location on the other side of the world, the day they
got there they would still be speaking the same way as the people
back home. But after decades and centuries the language would
have undergone certain changes in the new location ANO other
changes among the speakers back home. So the unity of related lan-
guages is to be found in time alone.
Now what happens in a country where there is just one language and
a stable population? fo lead into the answer to this question,
5aussure reminds us of one of his most basic lessons on linguis-
tics: the sign is changeable. Change in language is inevitable and
going on all the time, as we noted in our section above devoted to
this topic. And this change
cannot be uniform over the
territory where the language
is spoken. How do we know
this for a fact? Because
there is no record anywhere
of any language ever having
changed in exactly the same way everywhere that it is spoken. Here
it is visually:
Not: But:

fhis state of affairs only raises more questions. What is the start-
ing point for the language changes that end up producing different
dialects (local speech)? What pattern of changes is involved? To
answer these questions. we have to keep two basic facts in mind:
 language change takes the form of specif-
ic innovations that follow each other one
tJy one AND each of these innovations cov-
ers a definite and well marked-out area. If
the area happens to 17e the whole territory
where the language is spoken, no dialect
differences will 17e
created; if the area covered tJy the innova-
tion is only part of the territory, dialect dif-
ferences will 17egin to emerge.
Now a shift in pronunciation from ~ to Jb for
example, can occur on the same territory as
a shift in pronunciation from ~ to b 17ut in
different ~ of that territory. There is no way of predicting what
direction dialect changes will take when they spread, so if they are
all shown together on a map, they make very complicated patterns.
What is the result of this kind of differentiation over time? One lan-
guage may 17e spoken throughout a vast area, 17ut as centuries go
17y and it undergoes local changes everywhere, speakers from the
extreme regions will likely not 17e al71e to understand each other,
whereas speakers in neighl7oring regions will. Travelling from one
region to another, you would notice only limited differences 17etween
dialects, 17ut the sum of those differences grows till it finally distin-
guishes one language from another.

8arking Up the Wrong Syntactic Tree

A dog walked into the Western Union telegraph office, took a blank
farm and wrote: "woof, woof, woof, woof, woof, woof, woof, woof,
woof." The clerk looked over the form and pointed out that there
were only nine words, and that for the same price the dog could
send an extra "woof." The dog frowned and replied, "that would make
no sense at all."
=-- __-__:_--;;.:,.::_-::::_==-----:::-~~--::..:::....::. __=-:: -~-...==-..:.J

Saussure 17roke with nineteenth

century historical linguistics. He •• I' '.. •
a : c •• \ '
also break5 with the approach to
j--:-., .,.· f' \ __
dialect studies as they had 17een
done up to his time. He rejects
f-·-·--{
. d
·~
,.·'
~
.~ J.
, .1----<•. g ''r
the idea of dialect areas as per-
fectly defined and sitting neatly h
'

.
• ' 4/) •

.·.
•

L.\.
side 17y side, as they were usually
shown on linguistic maps:
_./ e\1 "'
The natural development of dialect5 doe5 not give the5e type5 of
re5ult5, 5au55ure 5ay5. In fact, he 5crap5 the idea of natural
dialect5 altogether and prefer5 to 5ay only that there are natural
dialect feature5. There are a5 many dialect5 a5 there are 5peech
communitie5. (If you rememl:1er Profe55or Henry Higgin5 In My Fair
Lady claiming that he could tell where people lived to within a few
5treet5 1:1y how they 5peak, you've got the Idea 5au55ure i5 trying to
get acro55 here.)
What to 5UI:15titute then for that artificially tidy dialect map?
5au55Ure 5ay5 there are two choiee5. A dialect could 1:1e defined l:1y
one of it5 feature5 and the map could 5how the 5pread of thi5 fea-
ture. But thi5 i5 not a great idea, 5ince the map will wind up 5howfng
a 1:1oundary of an i5olated feature in5tead of any dialect a5 it 15
actually 5poken. A 1:1etter pro5pect i5 defining a dialect l:1y the 5um
of it5 feature5 and choo5ing a community of 5peaker5. Now when we
choo5e a 5econd point for a 5econd community of 5peaker5 what
happen5? ln5tead of the tidy 5ingle line5 5eparating them, a numl:1er
of line5 5eparate them-one for each di5tingui5hlng feature of their
dialect5. We wind up with a map which look5 like thi5:

·A
;..,-·-;.::.--,:,:-·
-----;::~&:·--.,. ,. ,:. . .--·-·-·-.
______-: ......... . . ._ . ... . . . . . ..---
_...-...___...•·· • B
____..... '~...
. ...................
,.,:~-:,_:·····-
. . . ____ >::.•.•••· "·--·-

Now if you think thi5 doe5 not look much like a map, rememl:1er the
5ul:1ject i5 dialect geography and that for now only two 5peech com-
munitie5 are 5hown on the map. When other communltle5 are added
to make a more complete map, they will 1:1e 5eparated from each
other l:1y 1:1oundarle5 that look 5omething like thi5:

A completely complete dialect map would have many 5Uch zone5

marked on it. Not 5urprl5ingly, 5aU55Ure call5 the 1:1oundarle5 defin-
ing dlalect5 INNOVATING WAYE5, 1:1ecau5e of their 5hape, and
1:1ecau5e they mark the point5 where innovation5 occur In the form
(fll
 of the language underlying different dialects. Notice that the waves
can almost coincide or overlap. (You can see this clearly on the map
where only A and B are shown.) Saussure says that a dialect is
defined f,y a sufficient accumulation of such overlaps.
If this is the method for
defining dialects, how are
languages defined?
Saussure admits that it is
hard to say exactly what
separates a dialect from a lan-
guage. Sometimes a dialect takes on
the status of a language l7ecause a
f7ody of literature is written in it. This is
what happened for Portuguese and
Dutch. A more important factor is
understanding: If people do not under-
stand each other's speech, it is agreed
that they speak different languages. (If
they do understand each other and
there are only a few differences in
their speech, they speak related
dialects.) But languages are like
dialects in that innovating waves define
them, it's just that the waves show up over
a much larger territory for languages than for dialects. And tidy
dividing lines l7etween languages are ol7scured f,y areas of transi-
tion, just as they are for dialects.
BUT! When intermediate dialects disappear, groups of dialects that
make up a language do NOT continue merging into each other.
Instead, they collide. This happened for Germanic and Slavic lan-
guages and many others. What is today the national standard for
literary French used to l7e only the local lan-
guage of the lie de France (a former
province of north central France, includ-
ing Paris and the surrounding region).
It now extends to all of France's l7or-
ders, including the one with Italy,
where it comes up against standard
Italian, itself originally a dialect spo-
ken only in the region of Tuscany.

&2]
 How do linguistic waves spread? (Important question for Saussure.)
Wouldn't you know it? After giving us ying/yang style oppositions all
the way through the ~ (language/speech. signifier/signified. etc.)
Saussure comes up with two opposing forces to
answer our latest question. He calls them INTER-
COURSE and PROVINCIALISM. t7ut that may not do
much for you. so let's call them INTERACTION and
INERTIA Inertia keeps us in our t7eds. in our hous-
es. in our towns. It also keeps us
speaking the way we have always
spoken. If not7ody ever moved from place to place.
the result of inertia. according to Saussure. would
t7e an infinite numt7er of peculiarities in speech. But
interaction keeps this from happening. It spreads
language and gives it unity.

-_- :- :--::-.:·.- ~-=--- . . . . .;;

This unity through interaction comes at7out in two ways. Saussure's

explanation for this. as usual. puts the emphasis on forces that
t7alance each other. there is a negative force that wipes out innova-
tion and keeps dialects from splintering; there is a positive force
that promotes unity when an innovation is accepted and spreads.
(We are not going to get into a full-scale review of all of Saussure's
ideas here. t7ut this one is linked to the lessons on the LINGUISTIC
SIGN: t7ecause it is art7itrary. it cannot t7e changed t7y design;
t7ecause it is art7itrary. it does change t7y chance. There is a connec-
tion t7etween what Saussure is saying at7out the dout71e effect of
ARBITRARINESS and what he is saying at7out the dout71e effect of
INTERACTION on the unity of language.)
It's t7ecause of the spread of a new feature of language permitted
t7y t7oth the ARBITRARINESS of the sign and the INTERACTION prin-
ciple that Saussure talks at7out INNOVATING WAVES. The line that
descrit7es the 17oundary of a geographical fact a17out a dialect is like
the leading edge of a wave. (We are not going to get into a full-scale
review of all of Saussure's ideas here-honestl-t7ut you might want
[ll3
 to check out the fact that he used the idea of a wave in his section
on synchronic linguistics. Waves are like the coming together of
thought and sound in the LINGUISTIC SIGN.) The waves of language
change start at a single point in one dialect and spread out gradu-
ally. How these waves spread is different from how the change got
started in the first place. At the single point that is its source, the
change is taking shape in time alone, but once it starts to spread,
time and space both come into play.
It's easy to tell the forces of INTERACTION and INERTIA apart when
dealing with a single geographical point such as one village. Any fact
about the speech of the village
depends on one or the other.

~
Once the linguist looks at a
larger territory, where two or .~f
' \._-c< \
\
more dialects are spoken, it is impos-
'
sible to say which force is responsible
for which feature. Inertia keeps the
speakers of this territory from imitating the speech of
another territory, but interaction within the territory is also at work
and keeps it unified. The interesting little twist that Saussure gives
at the end of this lesson is the observation that linguistic change
can be studied without taking INERTIA into account. INTER/A is the
negative dimension of INTERACTION; it's the INTERACTION within a
region of speech communities.
Saussure liked the wave theory of language origins and language
groupings, because it helped to correct some mistakes that lin-
guists before him had made in their speculation about language ori-
gins and language groupings. The wave theory also helps to under-
stand the causes of language differentiation and the conditions
that determine which languages are related. He had more to say
about this and about
another subject he called
RETROSPECTIVE LINGUIS-
TICS, but this is definitely
not FOR BEGINNERS!

The Linguietic Graveyard

Latin is a language,
Dead as dead can be,
First it killed the Romans, \(I
Now it's killing me.
84]
 This was a popular jingle in the
days (not so long ago) when
schools still taught Latin. Yes,
Latin is dead, except in the
Vatican, where resident scholars
still coin new words to add to the
language, if the Pope's latest
encyclical so requires. Over the
ages Latin has t1een joined in the
linguistic graveyard t7y many oth-
ers. Here is the story of Akkadian,
an extinct language of Iraq.
Akkadian t7elonged to the vast
family of languages spoken in the
Near East, including Arat7ic,
Het7rew, and Aramaic, a family
known as Semitic, t7ecause the
speakers of these languages were descended from Noah's son Shem
(or Sem). Akkadian was a ruling language in Iraq and Syria in the
second millennium B.C.E. Clay tat71ets inscrit7ed with Akkadian
cuneiform script (cuneiform comes from the Latin word cuneus,
meaning "wedge-shaped") preserved some of the world's most
ancient literature, including the Epic of Gilgamesh (a tale of a semi-
divine hero and his quest for the secret of eternal life) and myths
of the Creation and the Flood with close counterparts in Het7rew
Bit71ical texts.
When Aramaic and Greek
replaced Akkadian, it fell into
disuse and may have t7een for-
gotten forever, if inscriptions
in its exotic, spiky script on
clay tat71ets had not t1een dis-
covered t7y travelling adventur-
ers in the 17th century. It was
only then that scholars took
up the daunting challenge of
deciphering this ancient lan-
guage. The texts of those
tablets that survived two
thousand years - literature,
chronicles, t7usiness records, school exercises, and even recipes -
have given us a rich record of Akkadian culture. [85
 Ce c. Cu a( tC iC uC
'. - r-----..,- r- ---,------,

.n ~ j t:E ~ < !
--- ---~-- ----~
: 1
---7------:----~
I i

k ~: <jg lli !
-----~
:I
: 1
!
i
9 ~ -~&_ --- 1:~-! £I i # !~1ol'J i
~ .rgr .~! :
*
q
-- - ----------- .-L-- --:·-----..1. ---- r

p
- ----~--------c ~ :
i ~r-- : ::q
b ~>-T : ~ !
~ I I
t --- - - ,. - - - -I

d3TT ------~----
~T ~ I I I ~
t:a
d ~ ; ~ . . ___<T~~
----~---
I
--oot------,- ______ ,. -- -~----

m S T-- <tt: ~ I ~ i :4* : ~

--- _.;. _____ -L-- --- -'---+------ ... -- --- -:--- -~---j
0 -:'T : ~~ ! ~ , ~ i -T- ! o{l :Ef:m:: till i

t* i
1 _ _ _ _ _ _ j,._ I _.. l

s ' ::m= ~
-----·------ -
I
-~-----1
::
: ..n :
z: ~~.. ~ ~ i ~'H i
H: : .....,tjf
-....- --1 I
I

~ :-: ~if ! ~T ! ~+T ;i . I

-- -- i---,- ----:- --'------+---------:-----

' :1m ~ <To- : ~ : j «< ~ ~
: a= : I :

h ff< 4 · +f\~:.ffi-i
---- - -- --------t-------1- ------ _____
i : ~rr~T
..s... - - - ·

~ ~ JST : ~···-!·: ~ : ~ : 4 :
-------r- -- ~ ~ -r ,--- -~

, E?;:fT ~T ~ '<T--TT<T . ~ : lT3 .

-------------r--- --l------,----. ----,-------
.......t-fl I I I I
y : ~~ 1 I

•
,
"'+ r
--,----

"'~
1
t
--

-------------~----1
.. _ -

I
r
I
:
----t-

- __
I
:
.

_! _ _ _ _ _ _
:
1
1

:
1
_!. ______• __ _

We know too that Akkadian achieved the etatue of one of the firet
international languagee in the domaine of !1oth commerce and diplo-
macy in the Middle Eaet, 11ecauee Akkadian texte from ae early ae
the eecond millennium B.C.£ have 11een found in Egypt, for example,
well away from the areae where the language wae in everyday uee.
Pity the 11rave linguiete who eet al1out cracking the code of the
Akkadian writing eyetem, containing over a thoueand eym11ole. Some
of theee etood for eylla11ee (groupe of eounde), eome for a whole
word or for a concept. Many of the eym11ole developed from othere
ueed in the writing eyeteme of unrelated languagee.
g6]
 UNGUI5TIC5 AND PHILOSOPHY
Another Kind of Semantics
Semantics, a5 we have 5een, i5 the tlranch of lingui5tic5 that ana-
lyze5 meaning, tlut general semantics i5 not part of lingui5tic5 in
the conventional 5en5e of analy5i5 of language. Or at lea5t it i5 not
limited to that. General semantics wa5
the invention of a Poli5h-t1om notlleman,
Count Alfred l<orzyt75ki (lfl7q-Jqso ), who
emigrated to the United State5. While
l<orzyt75ki wa5 intere5ted in language and
prot11em5 of communication, he developed
general semantics in the ICJ305 a5 an
applied form of philo5ophy, intended to
explain the working5 of the nervou5 5y5-
tem and to train people in the more effi-
cient u5e of it. In Jq43, l<orzyt75ki founded
the ln5titute of General Semantics and
the International Society for General
Semantics. The Society'5 journal, Et cetera, reflect5 little of
l<orzyt75ki'5 work today, tlut it i5 a lively putllication and retain5 a
focu5 on communication and education.

Bst:Jsv/1/BtJ by II 81J1Jtl11
Like many other 5Ut1ject5 that
eventually 5pun off on their own,
lingui5tiC5 had its first home in
the traditions of inve5tigation
and detlate that are the domain
of philosophers, and we men-
tioned Aristotle in "LinguistiC5
Then and Now" atlove. Let's fast
forward to a key figure in the evo-
lution of the twentieth century
philosophical movement called
philo5ophy of mind, the intellectu-
al giant whose work i5 credited
with tlringing to philosophy a new
direction called the lingui5tic tum.
Thi5 wa5 Ludwig Wittgen5tein
(lggq-Jq5J ).

[&7
His argument against dualism (the philosophical position that
accepts a 1:1asic division of the universe into mind and matter)
involved a colorful analogy with an imaginary beetle box. Here is what
Witters (Wittgenstein's nickname among the really savvy) asked his
readers to imagine: everyl:1ody has a 1:1ox containing something called
a "1:1eetle." (It turns out later that the key word here is not "1:1eetle"
1:1ut "something.") No1:1ody is allowed to look inside anyone else's 1:1ox.
1:1ut everyone claims to know what a 1:1eetle is just from
what is in their own 1:1ox. (It turns out later that the key
r word here is not "1:1eetle" 1:1ut "what.") Now if the word

••
"1:1eetle" is used among 1:1eetle 1:1ox owners, it cannot
1:1e for the purpose of naming a thing, 1:1ecause
nol:1ody knows what is in anyone else's 1:1ox. (The
I "something" in the 1:1ox may even 1:1e nothing.)
Use of a word and naming of a thing have to 1:1e separate,
and that's the catch for dualists, if they concede that words
such as "pain," for example, have a use. The term
is shared l:1y language users, 1:1ut direct knowledge
of it is not. It's as inaccessil:11e as all those 1:1ee-
tles nol:1ody is allowed to peek at. In use, "1:1eetle''
is not (the name of) something, 1:1ut it is not ?•
nothing either (even if there is nothing in the 1:1ox).
Using the word makes it something 1:1ut does not
make it the name of something. The story goes
that 1:1eetle 1:1oxes 1:1egan springing up on people's mantels in
Caml:1ridge and Oxford to indicate that they knew al:1out
Wittgenstein's posthumously pul:11ished Philo5ophicallnve5tigation5.
A Ungul~tlc 8tltltklry
(I) Witter'5 Lion
Among the well-known quotations from Wittgensteln's writings Is
the passage that says: "If a lion could talk, we could not under-
stand him." Perhaps
our philosopher was
thinking a"out the
fact that "iological
differences exist
"etween the percep-
MARK
tual mechanisms of
humans and ani-
mals. This i5 a fact
and needs to "e
taken account of In
de5crl,ing the dif-
ference5 "etween human ways of communicating and the
exchanges that take place among "irds, ,ee5, and "easts.
Perhaps he had something more su,tle In mind.
A lot of Ink has flowed In the discussion over Wlttgensteln's
o"servation, and a lot of complex lssue5 are involved, "ut there
are some valua,le le5sons even for "eginners In thinking a"out
what the old "oy meant, and especially in examining "oth his
suppositions and the lesson he may have intended to Illustrate
with his memora,le example. Let's put it in the form of que5tions
that connect with "asics in linguistics. And remem,er that here
the questions are more important than the answers:
i) If a lion could talk, why could we not record what he says, just
as linguists do in their field work with unknown language5, and
work out the analysis of lion language "it l:1y "It, phon•m•f:,y
phon•m•, mmjJ/zenw f:,y morpheme, structure l:1y structure?
ii) Did Wittgenstein "elieve that the interplay of speech (what Is
said) and language (what can "e said), a characteristic of all nat-
ural human languages, and deeply Ingrained in our predispositon
to social meaning-making might not exist for that chatty lion?

(gq
 iii) All the languages in the world of human5 are tran51atal:11e. even if
only roughly in 5ome ca5es. into all other language5, 50 would it ue
impo55il:11e to tran51ate lionese into human language?

Wittgen5tein uegan with an if. and we have to add another one here:
if the an5wer to our la5t two question5 i5 yes. then our philo5opher
may have intended to offer a les5on auout a univer5al feature of
human language that di5tingui5hes it from other po55il:11e form5 of
language. Philo5opher Steven Burn5 put5 it in the5e term5: if we go
uack to the context where Wittgem;tein made hi5 ol:15ervation. we
5ee that it requires an account of perception (in particular. of
oujectivity and 5ul:1jectivity in the ca5e of seeing-as) and of meaning
in relation to perception.
And thi5 1:7ring5 u5 uack to the phenomenon of 5ocial meaning-mak-
ing. preci5ely what could keep U5 from under5tanding the lion. If he
i5 not a 5ocial animal like our5elves. he may not even ue 5peaking.
even if he could trick U5 into 5Uppo5ing that he i5 5peaking to U5
(here we are uack to Wittgen5tein'5 if).

qo]
(2) Parrots Are No Good with Paraphraset5 or Paradigms
A linguistic lesson 17eyond Wittgenstein's is 17rought home when we
look closely at the not-so-amazing al7ility of parrots to say just
al7out anything that their humans will teach them to say. Parrots
are mimics, 17ut they have no linguistic skills other than 17eing a171e to
repeat fragments of language with fine accuracy. Once your parrot
learns "Polly wants a cracker," that phrase will always come out in
the same form. Saying "I would sure like to have a soda 17iscuit'' is
not an option for the parrot.
The a17ility of humans to master a language
involves paraphrase skills (re-expressing
a message in other terms), and
this, in tum, involves ~--~--
acquiring the al7ility to
select among items in
a language system that
are related in form and
meaning (paradigms) and
then to coml7ine them prop-
erly (look 17ack at the sec-
tions on mmp/zolflPF and syn-
tax) to make sentences that
make sense. Our parrot, of
course, does not know his syn-
tax from his semantics.
('if
----~~~~~~~-:.-~-~.-~: .. --~.-.~ =---~~c-c·. ..-:.----~' -~---: .-__ .-c·-_c.---~~ ~..-.-. .-.-.:=••=•=-~=-=•==-~-----=-,1

r--7178 parrot'5 phra5e, paraphra5e, and a pared down phra5e \

[1 II
1: A missionary visiting a tribal community gives a very long sermon. One :1
1
! of his anecdotes goes on for nearly half an hour. Then the interpreter Ili
1:

1
!i stands up, says four words, and all the listeners roar with laughter. \i
11 The missionary is baffled. How is it possible that his complicated story \1
1\ can be translated into four words? What kind of amazing language do \
1, these people speak? He asks the interpreter how he managed to pack 11

!I the whole story into four words. The interpreter replies, "Story too il

II long~-l~ay. -~ ~--~~~j~~~~-~g~~·~-"~~------·"~--c-~c:=-=--==~===-~=J
ANTHROPOLOGICAL UNGU/5TIC5-

••. focu5es on the link 17etween language and culture. Cultural anthro-
pologi5t5 (a5 oppo5ed to phy5ical anthropologi5t5, who confine
them5elves to looking at 5tone5 and 17ones) learn the language5 of
the people they 5tudy for the practical purpo5e of communicating
with them, 17ut they al5o examine the language for clue5 to the cul-
ture expres5ed through it. In the USA, it wa5 Franz 13oa5 (go "ack
and have a look at "A Flurry of Word5 for Snow'') who pioneered and
empha5ized the 5tudy of native American culture5.

[qg
His student Edward Sapir
entrenched this approach and set
the direction that American linguis-
tics would take for decades "efore
Noam Chomsky came along to upset
the anthropological applecart.
In Britain, it was the Polish-,om
anthropologist Bronislaw Malinowski
who gave the study of language its
anthropological thrust in the early
twentieth century. Nearly eighty
years after its first pu,lication, his
essay on meaning in relation to cul-
ture continues to "e reprinted as a
supplement to Ogden's & Richards's
The Meaning of Meaning. Malinowski
ended his career in America, "ut his ideas exerted a "ig influence on
John Rupert Firth (Jgqo-Jq6o ), acknowledged as the father of lin-
guistics in Britain. New areas of linguistics have opened up on Its
"orderlands with sociology, psychology, and other disciplines, "ut
anthropological linguistics remains
an active field of research. The cul-
tural anthropologists who work in
linguistics were "Y no means left
in the dust "Y the stampede to
theoretical linguistics. Lively
pu,lications continue to
appear, for example, on
the ever-popular top-
ics of kinship terms
and color names.
Klnt~hlp 18rm11
Studying the expres5ion of term5 for mother, father, brother, 5i5ter,
(etc.) in your language of choice ha5 at lea5t one t7ig advantage over
other area5 of the vocat7ulary. Even if a culture ha5 a very complex
5y5tem of names for cou5in5, let U5 5ay (5eparate word5 for father'5
t7rother'5 5on, father'5 5i5ter'5 5on, mother'5 5i5ter'5 daughter, etc.),
there are clearly defined limit5 on which
word5 need to go into the lingui5t'5 analy-
5i5 to give a complete picture. Starting
with domain5 like kin5hip makes it ea5ier
for the lingui5t to get a handle on a lan-
guage and it5 culture. Here i5 an example
of kin term5 from Swedi5h:

FARMOR

FARFAR

MORFAR
farfar- grandfather (=father's father)
farmor- grandfather (=mother's father)
mormor- grandmother (=mother's mother)
morfar- grandmother (=father's mother)

MORMOR
These words also give us an example of what linguists call nzo#w-
~motivation (look f,ack to our section on nzmplwl~). This
means that the parts of the word "add up" to the meaning of the
whole (it is a fact and clear to the speakers of Swedish that farfar
means the far of your far; compare English. where grandfather is
not necessarily a father who is grand).
Kinship terms vary greatly from one language to another.
Comparisons with English show f,oth simpler and more complex
organization. Parallel to the example of the Swedish grandparents
af,ove. we discover that many languages distinguish f,etween
'father's f,rother' and 'mother's f,rother.' whereas English has only
uncle. But whereas English makes a distinction f,etween niece and
nephew, some languages make do with a single word for f,oth. leav-
ing the masculine/feminine contrast unexpressed. In some cases.
one word covers not only niece and nephew f,ut also grand5on and
granddaughter. And whereas f,oth men and women have 5i5ter5 in
English. in Basque a man has an arreba. while a woman has an ahiz-
pa. In Seneca. an age distinction is made explicit in the kin terms.
giving different words for older 5i5ter and younger 5i5ter.
Recent puf,lications relate kinship terminology to other social phe-
nomena. as well as touching on the role of language in cognitive
development. dialectology. loanwords. language change. universal
features. These are rich additions to the accurate interpretation of
culture. which was. and remains. one of the primary purposes of
studying kin terms.
What is the chief interest of kinship for linguistics? If. as is often
said. every linguist is at heart a collector. nowhere do the f,utter-
flie5 of language showcase in a more 5pectacular way than in kin
terms. Except. perhap5. for...
ColorNllmtlll
Like word5 for kin, tho5e for color appeal
to llngul5t5 a5 an area of 5tudy "ecau5e
it appear5 inherently tidy. After decade5
of 5tudy, thing5 look a little more com-
plicated here than they did at flr5t,
"ut color i5 5till a lot le55 fuzzy a
domain than, 5ay, the expre55lon of
"eauty, art, democracy, fa5hion, etc;.,
though "rave lingul5t5 have tackled
the5e 5u,ject5 too.
The 5tudy of color
term5 ha5 a long hi5-
tory within the
anthropological
approach to lingui5-
tic5, "ut it wa5
tran5formed l:1y
the pu,lication in
Jq6q of a "ook
l:1y Brent Berlin
& Faul Kay called
8a5icColor
Term5. They di5-
covered that
each color name
ha5 a con5i5tent focu5 and thu5 removed varia,le category "ound-
arie5 for color name5 among language5 a5 an o"5tacle to a theory
of 5emantlc univer5al5 in thi5 domain. Berlin & Kay claimed the5e
unlver5al5 take the form of a 5et of 5equentlally applica,le rule5:

[qr
 Rule J: All languages have term5 for /:11ack and white.

Rule 2: If a language ha5 only

three names for color5, the
third will /:1e a term for the
color red.

Rule 3: If the language ha5 only four

names for color5, the fourth will /:1e a
term for green or yellow.

qg]
In other worde, only in the caee of a language more complex than
one deecri"ed l7y rule 3 will the dietinction appear among red,
orange, yellow, or among green, "lue, and violet.
Berlin & Kay found "oth eupportere and detractore. Both campe
offer irrefuta,le empirical evidence for and againet the Berlin & Kay
theeie. Some reeearchere have deeigned experimente to compeneate
for what they eee ae deficienciee of method in Berlin & Kay. Some
offer improved proceduree for inveetigating color terme, In thie caee
providing eupport for Berlin & Kay. Ten yeare after the pu,llcation of
hie original etudy with Berlin, Kay, working with another partner,
concluded that the etructure of the color lexicon ie a phyelologically
determined phenomenon, there"y etrengthening the claim for
eemantic univereale in thie domain. Oe,ate over the Berlin & Kay
theeie continuee into ite fourth decade.
With the euggeetion that univereale of perception exiet in epite of
dlfferencee among languagee, the Berlin and Kay rulee for color
terme challenge the eo-called Sapir-Whorf hypotheeie of
linguietic relativity. American linguiet Edward Sapir
(Jgg4-Jq3q) epoke of the "thought-groovee"
that he viewed ae ineepara,le from the
language of the
thinker, and hie
etudent
Benjamin Lee
Whorf (JgCf7-
Jq4J) pureued
thie view.

[qq
 He 511arpened 11i5 teacher'5 metapf1or (if we may t1e permitted to mix
our own) to tf1e point that encouraged a 5omewf1at over5tated in-a-
nut5hell ver5ion of 5apir-Wf1orf: tf1e 5tructure of our language
determine5 tf1e way we perceive the world. Notice carefully 11ow tf1i5
5tatement differ5 from Wittgen5tein'5 ot15ervation on the connec-
tion t1etween language and world view in A Anal Word t1elow.
Tl1e lingui5tic relativity
hypothe5i5 did not t1egin witf1
5apir-WI1orf. Centurie5 earli-
er. 5cf1olar5 t1egan 5peculat-
ing on the connection5 among
language. mind. perception.
and culture. A notat11e exam-
ple wa5 the eighteentf1 centu-
ry German who held down at
lea5t tl1ree jot15 a5 5tate5- e struGtu.re.
man. pl1ilo5op11er. and lingui5t
(lingui5t5, we noted earlier.
of Dur /an~lQ,e
were called philologists in deterrrr,~e.t the
•
tf1o5e day5), Will1elm von
Humt1oldt (1767-li35).
w1y we percetve
-the worltL

100]
 UNGU/5TIC5 ANO BEYOND
From 5truaturtJIItJm to PoststruaturtJI/tJm
The approach to linguistics which views language as a structured
system is called structuralism. Up until the twentieth century, lin-
guists dealt with language as a collection of individual elements:
speech sounds, words, grammatical word-endings, etc. Just as the
twentieth century dawned, the Swiss
genius Ferdinand de Saussure, whose 1 1
if,
\ 1,\1 ,,1 1 ., /
work we have already mentioned a few ~~\\\ I "0//
/~
times al:1ove, started linguistics on ~ /
the road to a whole new outlook ~
and methods. His Cour5e in
General Lingui5tic5 ( CGL) made
the case that the 1:1est way to
view language is as a structured
system.
He put the emphasis on how each
element of language is related to
other elements, 1:1ecau5e he
5aw thi5 as a way to correct a
lot of error5 that had 1:1een
made j,y earlier linguist5 who
never looked at the 1:1ig picture.
Saus5ure's approach came to
1:1e known a5 5tructurali5m,
though he did not U5e the
term him5elf, nor i5 the term
5tructure very prominent in
the CGL

[101
 From SaU55Ure'5 teaching5,
5tructurali5m eventually
emerged a5 the dominant
approach to lingui5tle5
among European 5cholar5. A
5tructurali5t orientation to
lingui5tic5 al5o developed in
the USA under Edward Sapir
and leonard Bloomfield, "ut
are recogniza,ly di5tinct
from Sau55Ure'5 original
teaching5.
The legendary U.S. philo5o-
pher, mathematician, and
inventor of the geode5ic
dome, 8uckmin5ter Fuller,
once 5aid that the di5ciple5
of Chri5t took hi5 powerful
teaching5 and turned them
into the language of Humpty-Dumpty. Many "elieve that much the
5ame thing happened when the principle5 5et out f7y the granddaddy
of lingui5tic5, Ferdinand de Sau55ure, were U5ed and interpreted "Y
Jac;que5 Oerrida, who i5 generally acknowledged a5 "eing the founder
of po5t-5triJcturali5m (al5o called decon5truction)-an attempt to
correct the perceived 5hortcoming5 of 5tructurali5m.
Oerrida argues that there are three weak point5 in Sau55Ure'5 teach-
ing5: hi5 ideali5m, hi5 empha5i5 on 5poken language, and hi5 u5e of
paired oppo5ite5 to de5cri"e feature5 of language.
8y ideali5m Oerrida mean5 the view that language doe5 not create
meanings 17ut expre5se5 pre-existing one5. Oerrida disregards
Sau5sure'5 key point that the language-5y5tem mediate5 "etween
thought and 5ound. 8oth thought and 5ound are formle55, Sau55ure
5aid, until they are linked and acquire form through the creation of
tho5e link5-called 5ign5. There are no pre-exi5ting meaning5 in thi5
view, a5 Oerrida 17elieve5.

102]
Oerrida makes the concept of difference all important, as it is for
Saussure, except that he doe5 not keep the complementary term
oppo5ition. But he doe5 follow Saussure in making 5ystem a key
idea. On the /:1asis of 5y5tem and difference, Oerrida develops the
concept of archi-writing. This is supposed to 1:1e a system consisting
of pure differences that underlies speaking AND writing. Saussure
did not recognize such a system, according to Oerrida, 1:1ecause he
held a prejudiced view of writing as nothing more than a way of rep-
resenting speech. To call Saussure's view of writing a prejudice is to
disregard his purpose in developing a new approach to linguistics,
namely avoiding the confusion and errors that marked the work of
earlier linguists, who had always limited themselve5 to written texts.
Oerrida rejects Saussure's use of complementary pairs of terms
such as a55ociative relation5 and 5yntagmatic relation5. This pair,
in particular, is related to the even more fundamental one of
ab5encelpre5ence, which Oerrida refuses.
The whole project of eliminating such pairs is unnecessary when we
recall that Saussure ended his lesson on syntagmatic relations and
associative relations l:1y showing how they interact. They are defined
independently 1:1ut they function interdependently. Since definitions
are particularly su/:1ject to the endle5s play of signifiers (a normal
state of affairs, according to the post-structuralist view), there is
no point in o1:1jecting to their provisionally independent status in
Saussurian linguistics. Saussure moved 1:1eyond all his dualities him-
self. In this sense, he deconstructed structuralism more than half a
century 1:1efore Oerrida.

[103
 ··1tt e.
,__ 1M 1fS
oF oufZ-
t." tJ &"'-' t}fE

ARNALWORO
Wittgen5tein 5aid that the limit5 of our language are the
limit5 of our world. The great Canadian literary critic
Northrop Frye wa5 talking about the 5ame thing when he
5aid that the be5t rea5on to 5tudy a language other than
the one you have 5poken from birth i5 to di5engage your
thought proee55e5 from what he called "the 5waddling
clothe5 of their native syntax." If we can get over that
"ever eerier feeling" that Ogden Na5h 5peak5 of in the quo-
tation we u5ed at the 5tart of thi5 book. we can U5e lin-
gui5tic5 to 5tretch our under5tanding of the world too.

104]
 APPENDIX #I Up for th11 Count

*"e 1- "j; -~r'~- b- fer one 1-sa

+~,~-go seWh 7- fY'IP 2- d \1 ~
htte'3 - sa e•~~~ 8-
hYee 1-r~-e
.r.~ 4-- ~ .. :
.f.v.. /t -1 ·1-..tt't.:l.
lVe ~- tr~r

1 Sar-o-
2_-d \J~al
--tvlo

Let'e; have a look at a common initiation ritual for "eginnere; in lin-

guistics: learning the num,ere from one to ten in different lan-
guagee. Thie; little exercise doee not ehow anything linguistically. "ut
for many teachers of linguietic;e;, It ie a favorite way to introduce
their e;tudente; to the diversity of the world's languagee; and to give
practice in hearing. pronouncing. and trane;cri,ing unfamiliar data.
lt'e; a "it too early for you to have mastered phtm•tle; transcription,
eo the examplee "elow are given In conventional English spelling. even
though thie makee; for a certain loe;e; of accuracy In indicating their
pronunciation.

(107
 Bai (China)
yi. go. sa. shi, ngur. fer. chi. f:Jia. jiu. dser

Balinese (lndone5ia)
sa. dua (or kalih). talu (or tiga). pat. lima, nam, pitu, kutus (or ulu).
sia. dasa

Ukrainian (Ukraine)
odin. dva. tri. chotiri, pyat'. she5ts'. syem, vosyem, devyats'. de5yats'

Bikol (Philippine6)
saro'. duwa. tulo, apat, lima, anom, pito. walo. siyam. sampulo'

Bislama (Vanuata. formerly the New Hef:Jride6)

wan. tu. tri, fo, faev. slkis. seven. et. naen. ten

Welsh (Wa/86)
un. dauldwy. tri/tair. pedwarlpedair. pump. chwech, saith. wyth, naw. deg

Manchu (northeastern China. fewer than 1,000 surviving speakers)

emu. juwe. i/an. duin. sunja. niggun, nadan. jakon, uyun, juwan

lOll]
 APPENDIX #2 Ths Hlt~tory of Our Alph11bst

!JX
Historically, Ala is not for apple 17ut for ox. How can this 17e? Our
alphal7et, the Latin alphal7et, derives from the Greek alpha17et, 17ut
its hoariest ancestor is anc;ient Proto-5inaitic;, where Ala 17egan life
as a stylized drawing of an ox. In 17etween came Phoenic;ian and
He17rew, and in all cases, the alphal7ets of these languages 11egin
with the ox (called aleph in Phoenic;ian and aleph or aluph in He17rew ).
The animal's privileged position reflects its c;entral importanc;e in an
agric;ultural soc;iety and as a sym17ol of strength. The upright horns
on the o>ts head turned east l7y the time the aleph passed into
Greek as alpha (a) and then headed (pun intended) due south in
the transformation Into Latin A

The story of f3 17egins with Proto-5inaitic; bayit (house) and moves

to beth (house) in Phoenic;ian and He17rew 17efore 17ec;oming Greek
beta (~). The straight lines of the original floor plan of the house
gave way to loops and c;urved lines, and, like alpha, beta c;hanged
direc;tion.
[loq
 6ut the much 11lgger change that came along the way from He11rew
to Greek was in the loss of meanings. 6eglnnlng with Greek, the
names of the letters of the alphaf1et refer only to themselves. No
more ox. house, camel, door, or any of the other familiar things from
everyday life that lent their shapes and names to make up the f1its
and pieces of the older alpha11ets.
The pun on alphabet invented f1y that great
linguist James Joyce in Rnnegans Wake is
allforablt.

This was a huge change that gave the new alpha11et (a+~) tremen-
dous power. All the visual reminders of oxen, houses, etc., disap-
peared and the whole focus of the alphaf1et shifted to language
Itself. On the one hand, all the associations of meaning for the
of1jects originally linked to letter names were eliminated-traded in,
as It were. for nothing more than the sound at the f1eglnnlng of the
names of those of1jects (all for a f1it). On the other hand, that left
the new sound-letters free to coml1ine with each other in any num-
f1er of ways to create new meaning (all for a f1it).

110]
 APPENDIX #9 Womsn's 5t;ript

~-?kli

~~ M tfk
~!e ~ tJ
1}-~
ililit {l!
~ - .t;t
~jlt.ffr:
Nushu Text Chinese Tranallteratlon

\
\-__

The women of Jiangyong in 5outhem China have kept an extraordi-

nary 5ecret 5ince the day5 of the ancient Song Oyna5ty. According
to legend, it wa5 an emperor'5 concul1ine who invented nu5hu
(women'5 5cript), allowing her to pour out the forl1idden expre55ion
of her loneline55 in letter5 to her 5i5ter5 without revealing it to
male eye5. Nu5hu i5 related to ancient Chine5e 5cript, predating the
(Ill
 time of Qin Shihuang. who was responslt11e for the unification of
China and t1ecame the first Chinese emperor in 221 B.C. Qin's drive
for unification eliminated all local written languages. and the penalty
for creating new ones was death for the innovator. her family. and
nine related families. But nushu escaped the attention of authori-
ties. t1ecause women and women's activities had no official status.
and the secret script survived. Ironically. ac;c;ess to education and
Increased freedom to work and travel for women in modem China
have all t1ut ensured the extinction of the script that gave them a
powerful t1ond for centuries. (A traditional provert1 in nushu says
"Beside a well one does not thirst; t1eside sisters one does not
despair.") The only three remaining women of Jlangyong who know
the script have taught it to no one. t1ecause no one wishes to learn
it. Even though some dedicated scholar may rise to the challenge of
preserving Its texts. nushu will t1ecome a linguistic fossil.

112.]
 APPENDIX #4 Hlt~torlCIJI Ungult~tlet~

Historical linguistics studies changes in language. It is usually said

to have t7egun in the late eighteenth century, when Sir William Jones
(1746-44), formerly tutor to an anc;estor of the late Princ;ess Diana
of Wales, t7egan to document the previously unnoticed resemt71ance
t7etween Latin and Greek and the ancient Sanskrit language. This
was the first step in defining and descrit71ng the vast family of
[113
 languages known a6 Indo-European, 6tretching acro66 the two
continent6 identified in it5 name, a group in which all mem,er6 are
the off6pring of a lingui6tic Adam. Hi6torical lingui5tic5, with it5
empha6i6 on grouping5 and affinities, cla66ification and compari-
5on, dominated lingui5tiC6 until Ferdinand de 5aU66Ure came along
and propo6ed that the hi6torical approach (he called it diachronic)
"e "alanced l1y the 5tudy of language a6 a 5y5tem (he called thi6
the 5ynchronic approach). ln5tead of achieving the "a lance
5au55ure wa6 looking for, twentieth century lingui5t5 went over-
"oard on 5ynchronic 5tudies. But today the 5earch for principles
underlying lingui6tic change i6 "eing revitalized, and lingui5t5 are
looking in particular at the link "etween that change and llngui5tic
variation (6ee our 5ection on geographicallingui5tic5).

114]
 Apptmdlx #5 Mor11 About Phonllt/Ctl
SOME SOUNDS OF ENGU6H
(6Am 60wndz Av EI}gllf)
You In "racket6 a"ove whic;h of the letter6 in 6tandard 6pelling
5ee
carry over from our heading "Some Sound6 of Engll5h'' to lt6 phonetic
tran6Cription.

Wlurt'• m'-lng?
I) all the vowel6, "ecau6e 6tandard 6pelling doe5 not con6i6tently
6how how to pronounce the twenty of more vowe16 that occur In
mo6t variant6 of Engli5h;
2.) 60me con6onant6, "ecau6e the n in Engli5h i6 not the n of ran "ut
of rang and the 5h In Engli5h 16 not two 6ound6 "ut one.

Wlurt'• dlff611nt?
I) the final -5 of 5ound5 i6 tran6cri"ed z, "ecau6e it 16 pronounced
a6 z;
2.) v and not fin the tran6cription of of, "ecau6B It i6 pronounced a6 v.

Now if we fill in, u5ing the International Phonetic Alpha,et (remem-

"er that the principle of thl5 alpha,et i6 one 5ym,ol for one 6ound
and a different 5ym,ol for every 6ound), we get the full tran6crip-
tion a6 a"ove.

(115
 THE REST OF THE SOUNOS OF ENGU5H
And for tho5e of you who can't get enough phonetic5, here i5 the
complete table of the ba5ic 5ymbol5 for con5onant5 and vowel5 in
the International Phonetic5 Alphabet. Our example5 have been limit-
ed to Engli5h, but remember that thi5 alphabet wa5 de5igned 50
that lingui5t5 could accurately tran5cribe the 5poken ver5ion of any
language in the world. To do thi5, in 5ome ca5e5, you need 5ome
additional 5ymbol5 not 5hown in thi5 table.

116]
 THE INTERNATIONAL PHONETIC ALPHABET (revised to 2005)
CONSONANTS (PULMON IC) © 2005 1PA

l Bilabial Labi odenml Dental IAlveolar [Post alveolar! Retroflex. Palalai Vela r Uvul ar Pharyngeal Gloual

!Plosive p b t d Itct. c j k g q G l ?[
I Nasal m 11) n I 11. J1 I) N
Trill
···----
B
-····----- -·-· ... ... __
, , , ___ r I R
Tap or f-l ap V' f I r
i Fric:.1tive <I> (3 f v 8 5 1 s z I I 3 I~ ~ 9 J X y X B' h ) h fi
Lateml
frica tive
·-
i 5 I-
Approximant u 1 I -t J - lll
1 La!eral
' approximant l ! l li. L
Where symbols appear m p:u rs. the one to the nght represents a vo1ced consonant. Shaded areas denote arttculati011Sjudged am poss1ble.

CONSONANTS (NON-PUL MONIC) VOWELS

i\ ---- rniu
From Cent raJ Back
Clicks
I Voiced implosives Ejecti ves

0 5 F...tamtll cs:
Close
y ~--
Id
Hilabint Bilabial
0
I Dental lkntal/al \'tolnr p' Bilabial

IJ t' ~~ --- Y ~ O
Close-mi d
(Posl)alw:ohu Palatal Dc:Oial/nl\'•·:olar
e (/)
=I= P:!.laroalvcolnr Ig Velar k' Vrl:u

II J\ lv!!~:llar l aternl
I cf l)vulnr s' Al\'«tln r fricat i v~ Open· mid
e rere- 3\G-- A• ;)

OTHER SYMBOLS
Open a CE___i__ a .1 o
M. Voiceless labia1-vclar fric:.uiv~ cz Alveolo-pnlatal frit·at ivcs Where symbols appear in pai~. the one
to the ri ght represents a rounded vowel.
W Voiced labial-w.lar appro;o;imanl 1 Voict-daln.'Oiar l:u.::rnJ flap

q VoiL~d labial ·palatnl approximant fj Simultaneou5 J and X SUPR ASEG MENTA LS

VOILciCSS ~ pi g loual rmt~ t iv.::

Pri mary stress
Affric:ttes and tJoubk articulations
Voiced epiglottal fricntive Secondary stress
t·an be n:pn:scmcd b) two S)'mbols
joi ned by a tie bar if lk:L't'S~ary . ,foun~ 'uf~n
Epiglottal plosivt"
Long e:
DIAC RITICS Diacri tics may be placed above a symbol with a dcscendc r. e.g.
0
I] Half-long e·
<} I P. ~ I" ~"tal ~t ad~-l
Exlra-sho rt e
Vuktkss
I}
§ !
Bre.nhpoi«d

Q g l ., l I Minor (fool) group

h
Vokt-d Cre-aky I.'Oict.-d ApkaJ _ _
II Major (intonation} group
A.'lpir.tted
Syll able break Ji . re kt
MOI\: roundl!d
Linking (absence of a break)

TONES AND WORD ACCENTS

LEVEL CONTOU R

€or l E.\lrn
hi gh eur 11 Ri s in~

e I Hi gh e 'J Falling

X X
e e -1 Mid e 1 High
ri.~in p:
Mid -ccmrnlizcd lb isecl (} = ''oiccd al\·eolar fri cath el
e _.j Low e ~ I. A\ W
rising
Syllabi\!
r "''------~---t~--- - - - - -·-
n, Lowen:<! e f A = voic..:d bilabial appro.\imaiUJ
1' - e J Extm
low e 1 Ri sing-
falling
Non ·s) ll.abic fX AJvanc;."d Ton!:u~ Root e .j.
Downs li!ll ?' Globalri ~

~ Rdr..~et!!d Tongue Rooc

d'- r
"'
Rbocicily p_ Upstep Globnl fa ll
L_--~------~i------~----~ L__ _ _ _ _ _ _ _ _ _ _ _

Cou rt~y of the International Phonetic A5sociation (Department of

Theoretical and Applied Linguistics, School of English, Aristotle University
of fh~saloniki, fhessaloniki 54124, GREECE)
[117
Expl11n11tlon of tsrm11 UIIIJd:
The line ac;ro55 the top of the ta111e refer5 to the organ5 involved in
the production of 5peech. Reading from left to right, we 5ee where
5ound5 are produced, moving from front to 1:7ac;k in term5 of
anatomical part5.

The lefthand column de5crll1e5 how 5ound5 are produced in the

ca5e of the con5onant5 and degree of aloeure of the mouth in the
ca5e of vowel5.

Ploelve (think of explo5ive) refer5 to any con5onant produced 1:7y

temporary 111oc;kage of air coming from the lung5 followed l:7y a
relea5e.

[nq
TO LEARN MORE, PICK UP THESE OTHER
FOR BEGINNERs- 500KS

CHOMSKY
FOR BEGINNERS
Written by David Cogswell • Illustrated by Paul Gordon
ISBN tr78-l-43438q-17-l (l-43438q-17-X) • 160 PAGES
$14,qs (CANADIAN $16.qs)

DECONSTRUCTION
FOR BEGINNERS
Written by Jim Powell • Illustrated by Joe Lee
ISBN tr78-l-43438q-z6-3 (l-43438q-26-4) •168 PAGES
$14,qs (CANADIAN $16.qs)

DERRIDA
FOR BEGINNERS
Written by Jim Powell • Illustrated by Joe Lee
ISBN tr78-l-43438q-z6-3 (l-43438q-26-4) •168 PAGES
$14,qs (CANADIAN $16,qs)

[121
About the Author
W. Terrence Gordon has pu,lished more than twenty "oolc5, including
5aussure For Beginners and McLuhan For Beginners. He is currently
at work on a "ook a"out James Joyce and a "iographical fiction
a"out the legendary linguist Charles Kay Ogden. When he is not "usy
writing or teaching, Gordon photographs the haunting "eauty of
Nova Scotia, Canada, where he has lived since the l'l70s.

About the Illustrator

Susan Willmarth was "om in New Mexico and moved in the early '70's
to New York City. Since graduating from Parsons School of Design,
she has worked as a free-lance editorial Illustrator for Push Pin
Press 6oolc5, Edward 6ooth-GII,,om editions, New York Magazine,
The Open Society, Writers and Readers Pu,llshing, and now For
6eginners LLC. Past work includes Black History For Beginners and
McLuhan For Beginners. She lives in Manhattan with her "lcycle.

[12.3