Você está na página 1de 137




1ª edição
rio de janeiro  2016
Conselho editorial  luis claudio dallier, roberto paes e paola gil de almeida

Autora do original  paula bullio

Projeto editorial  roberto paes

Coordenação de produção  paola gil de almeida, paula r. de a. machado e aline

karina rabello

Projeto gráfico  paulo vitor bastos

Diagramação  bfs media

Revisão linguística 

Revisão de conteúdo  rosângela salviano

Imagem de capa  anna bogush | shutterstock.com

Todos os direitos reservados. Nenhuma parte desta obra pode ser reproduzida ou transmitida
por quaisquer meios (eletrônico ou mecânico, incluindo fotocópia e gravação) ou arquivada em
qualquer sistema ou banco de dados sem permissão escrita da Editora. Copyright seses, 2016.

Dados Internacionais de Catalogação na Publicação (cip)

B935o Bullio, Paula

Oficina de tradução I – prosa / Paula Bullio.
Rio de Janeiro: SESES, 2016.
136 p: il.

isbn: 978-85-5548-371-4

1. Literatura. 2. Tradução. 3. Coesão. 4. Coerência. I. SESES. II. Estácio.

cdd 418

Diretoria de Ensino — Fábrica de Conhecimento

Rua do Bispo, 83, bloco F, Campus João Uchôa
Rio Comprido — Rio de Janeiro — rj — cep 20261-063

Prefácio 5

1. The difference between translation and

interpreting 9

1.1  The skill profile of technical translators 10

1.1.1  Interpreter qualifications 11
1.2 Textuality 11
1.2.1  Textuality as a secure translation model 12
1.3  Units of translation 15
1.3.1  Definitions of the unit of translation 16
1.4  Coherence and cohesion 22
1.4.1 Cohesion 22
1.4.2 Coherence 23
1.4.3  Coherence – Semantic property 24
1.4.4  Cohesion – Flow of a text 25
1.5  Connotation and denotation 25

2. History and intertextuality 37

2.1  Types of intertextuality 38

2.1.1  Horizontal or vertical reference 39
2.1.2  Manifest or constitutive reference 39
2.1.3  Socio-cultural objects and socio-textual practice 40
2.1.4  Intertextuality in newspaper headlines 42
2.2  The translation problem and its strategies 43
2.3  A semiotic approach 45
2.4  Translation equivalent 46
2.4.1  Features of translation equivalence 46
2.4.2  Necessity of translation equivalence 48
2.4.3  Different paths 50
3. Formal correspondence and the dynamic
equivalence 61

3.0.1  Nida’s translation theory of dynamic equivalence 62

3.0.2  Formal correspondence 63
3.1  Semantic equivalence and pragmatics 70
3.1.1  The Effect of the Pragmatic Motivation of the Original Message 71
3.1.2  Pragmatic equivalence 72
3.2  The meanings of words and structures 76
3.3  The cooperative principle and its maxims 76
3.4  The context of the utterance 77
3.5  Other items of background knowledge 81
3.6  Grammatical equivalence 81
3.7 Procedure 83

4. Techniques of literary translation 87

4.1  The literary translation and its specificities 89

4.1.1  Literary translation as a career 90
4.2 Authorship 94
4.3  Tips to becoming a better literary translator 107

5. Examples of translations from English to

Portuguese 111

5.1  Examples of translations from Portuguese to English 121

Prezados(as) alunos(as),

In this book about translating prose texts, we are going to observe some tex-
tual aspects, such as textuality, coherence, cohesion, connotation, and denota-
tion. The historical and cultural aspects will also be observed, since they play an
essential role in the making of a good translation.
A considerable part of the book will cover questions about translation and
equivalence, such as formal correspondence, dynamics, semantics, pragmati-
cs, and grammatical and textual equivalence. The last part of the book will co-
ver literary translation. It will show some examples of translation from English
to Portuguese, and from Portuguese to English.
Pondering about translation is almost as old as translation itself. The most
renowned and cunning translators have mulled over the different ways to redu-
ce the chasm between the source and target texts. By looking at written texts,
we can notice that the process of translation can lean towards source or tar-
get text. For instance, according to Hatim (2001: 43), it can move towards the
models which focus on the source text like formal equivalence: Catford, dyna-
mic equivalence: Nida, pragmatic equivalence: Koller, text-based equivalence:
Beaugrande, foreignization and equivalence: Venuti; or it might shift towards
the target text like translation as meta-text: Holmes, translation as re-writing:
Bassnett and Lefevere, transformation: the feminists, deconstructionists, ma-
nipulationists, polysystem, and norms: Toury, skopos: Reiss and Vermeer.
Beyond diverging models, which operate in different fields and modes of
translating, the textual model understands that the text is a set of mutually
relevant communicative functions that hang together and are constructed in
such a way as to respond to a particular context,thus achieving an overall rheto-
rical purpose (Hatim & Mason 1997: 224), and the translator (communicator) is
the one who attempts to communicate concepts from the source to the target.
Considering the source as a text, we believe that the adequacy of translation is
achieved when the target (translation of the source text) is a text which commu-
nicates as efficiently as the source text. Textuality, a linguistic-oriented work on
translation, was primarily initiated by scholars like Beaugrande (1981), Neubert

1 This introduction was based on the article: MIKHCHI, H. Standards of Textuality: Rendering English and
Persian Texts based on a Textual Model. Journal of Universal Language 12-1, 2011

and Shreve (1992), and Hatim and Mason (1997), who have made valuable con-
tributions to textual approach and translation.
To begin with, I would like to pose two questions: What is translation? And
who uses translation services?

What is translation?
Translation is the transmittal of written text from one language into ano-
ther. Although the terms translation  and  interpretation  are often used inter-
changeably, by strict definition, translation refers to the written language,
and interpretation, to the spoken word. Translation is the action of interpre-
ting the meaning of a text, and then producing an equivalent text, also called
a  translation, which communicates the same message in another language.
The text to be translated is called “source text”, and the language to which it is
going to be translated is called “target language”. The final product is usually
called the “target text”.
Translation must take into account constraints such as the context, the ru-
les of grammar of both languages, their writing conventions, and their idioms.
One common misconception is that there is a simple word-to-word correspon-
dence between any two languages, and that translation is a straightforward me-
chanical process. A word-to-word translation does not take into account con-
text, grammar, conventions, or idioms.

Who Uses Translation Services? 

Businesses often seek translation services in an effort to better serve their
customers and keep up with their demands. As the world gets more and more
competitive, it’s imperative that businesses convey their message clearly and
accurately to their customers. Translation mistakes can potentially affect a
company’s reputation, thus resulting in financial loss. 
Generally, most industries require this type of service. The legal field needs
translations of  depositions, petitions, court records, and court proceedings.
Law enforcement may need to have statements translated, and the medical
field often needs medical records and notes translated. Other examples of in-
dustries that benefit from document translation services include insurance
and financial companies, as well as the media.
As more and more companies make their services available
through the Internet, the demand for translations of websites and web

content has also grown. Furthermore,  talk radio, podcasts, surveys, fo-
cus groups, and corporate meetings often require translation services.
Businesses are not the only ones that require document translation. On a per-
sonal level, individuals also employ this type of service. As more and more peo-
ple migrate to foreign countries in pursuit of better life conditions, they find
themselves in environments that are unfamiliar. This includes the language
spoken. As a result, they may need to have legal documents translated, such as
birth certificates, marriage licenses, passports, contracts, and leases or mort-
gage contracts.

Enjoy your studies!

The difference
between translation
and interpreting
1.  The difference between translation and
interpreting 2

Interpreting and translation are two closely related linguistic disciplines.

However, they are rarely performed by the same people. The differences
concerning skills, training, aptitude, and even language knowledge are so
substantial, that few people can do both successfully on a professional level.
On the surface, the only difference between interpreting and translation is
the medium: the interpreter translates orally, while the tratnslator interprets
written text. Both interpreting and translation require a certain degree of
language knowledge, and deep knowledge of more than one language.

1.1  The skill profile of technical translators

The differences in skills are arguably greater than the similarities. The key
skills of the translator are the ability to understand the source language and
the culture of the country where the text originated, then use a good library
of dictionaries and reference materials to render that material clearly and
accurately to the target language. In other words, while linguistic and cultural
skills are still critical, the most important mark of a good translator is the ability
to write well in the target language.
Even bilingual individuals can rarely express themselves in a given subject
equally well in both languages, and many excellent translators are not fully
bilingual to begin with. An interpreter, on the other hand, must be able to
translate in both directions on the spot, without using dictionaries or other
supplementary reference materials. Interpreters must have extraordinary
listening abilities, especially for simultaneous interpreting. Simultaneous
interpreters need to process and memorize the words that the source-language
speaker enounces, while simultaneously outputting in the target language
the translation of the words the speaker said 5-10 seconds ago. Interpreters
must also possess excellent public speaking skills and the intellectual capacity
to instantly transform idioms, colloquialisms and other culture-specific
references in analogous statements the target audience will understand.

2  This part of the chapter was based on the website: <http://www.languagescientific.com/the-difference-

between-translation-and-interpreting/>, which is recommended for further information.

10 • capítulo 1
1.1.1  Interpreter qualifications

Interpreting, just like translation, is fundamentally the art of paraphrasing—

the interpreter listens to a speaker in one language, grasps the content of what
is being said, and then paraphrases his or her understanding of the meaning
using the tools of the target language. However, just as you cannot explain a
thought to someone if you did not fully understand that thought, neither can
you translate or interpret something without mastery of the subject matter
being relayed.

1.2  Textuality

Since texts can only be completely understood in relation to the context

in which they occur and communicate a particular function or purpose,
they are considered a more appropriate unit of analysis for many purposes
like translation. Although phrases, clauses, sentences, and paragraphs are
considered units of translation in some models of translation, we will see that
texts can be regarded as more comprehensive and practical units of translation.
However, the units mentioned have occasionally been considered texts, too.
So, what exactly is a text? Is it just a unit of analysis similar or larger than the
formers? A text can be defined as an actual use of language, as opposed to a
sentence, which is an abstract unit of linguistic analysis. We identify a piece
of language as a text as soon as we recognize that it has been produced for a
communicative purpose (Widdowson 2007: 4).
The communicative purpose is an essential aspect of a text because in the
actual use of language, i.e. text, it is a way of connecting the reader to the text.
The connection consists of transfering the meaning and intention of the text’s
author, who is one of the parties involved, to the reader, who is another party. As a
text is read, the reader is placed in a state created by the communicative purpose.
If a text does not communicate, its reader will not understand the meaning
and intention of that text. Linguistically speaking, a text is a communicative
occurrence which meets seven standards of textuality. If any of these standards
is not considered satisfactory, the text will not be communicative. Again, non-
communicative texts are treated as non-texts.
These standards include: cohesion, coherence, intentionality, acceptability,
informativity, situationality, and intertextuality (Beaugrande & Dressler 1992).
It may be noted that these standards are the constitutive principles which

capítulo 1 • 11
define the communicative purpose of the text. They show us how occurrences
are connected to one another: through syntactic relations on the surface
(cohesion); through conceptual relations in the text (coherence); through
the attitudes of the author and reader towards the text (intentionality and
acceptability); through the transfer of information (informativity); through
the setting (situationality); and through the reciprocal relationship of separate
texts (intertextuality).
On the other hand, to show explicit linguistic expression in a text, textuality
is a property that a complex linguistic object (the text) assumes when it
reflects certain social and communicative constraints. The operation of these
constraints is manifested in recognizable linguistic patterns at the textual
surface. Textuality is induced by the linguistic surface, but it is not confined
to it. The linguistic surface of a text is no more than a pointer to its textuality
(Neubert & Shreve 1992: 70). By investigating the textual surface regarding the
standards of its textuality via textual analysis, one may be able to unravel the
complexity of the linguistic features of the surface, analyze the relationships
between constituents of the text, and ultimately, learn about the meaning and
intention of the text, which are related to social and communicative constraints
comprised by the context.
Assuming that there are linguistic mechanisms that combine the standards
of textuality to create the text, we might perceive that the primary duty of the
translator is to discover how the standards of textuality of source texts hold
the communicative purpose of the texts. Furthermore, discerning linguistic
mechanisms of the standards of textuality of target texts could help the
translator better convey the meaning and intention of the source text to the
target. In this view, translatability relies on the potential for textualization,
and in a broader way, on the potential for communication. Textualization is
the global strategy that makes translation possible (ibid. 133-147). To better
understand the concept of textuality and how it can work in translation, we will
study textuality as a secure translation method.

1.2.1  Textuality as a secure translation model

Translating is looked upon as an act of communication which attempts to

relay, across cultural and linguistic boundaries, another act of communication
(which may have been intended for different purposes and different readers/

12 • capítulo 1
hearers) (Hatim & Mason 1997: 1). With this in mind, one needs, on the one
hand, to thoroughly analyze the source text in order to unravel any complex and
thorny points of the text, and on the other hand, to communicate the extracted
information to the target text. But the question here is how can a translator be
sure that she is properly communicating the right meaning between the texts?
To suggest a solution, textuality is referred by which at the first step, one
analyzes the text based on its basic standards of textuality, and at the second
step, the translator, having previous knowledge of the basic standards of
textuality of the target text, synthesizes the textual features of the source into
the target text. To put it another way, while we assume that every text has a
communicative purpose, the translator (communicator) needs, firstly, to do a
textual analysis of the source text to receive the messages -- the meaning and
intention -- encoded in the communication system of the source text. The
textual analysis of the source text is a way of analyzing the standards of the text
to comprehend the message. Secondly, the translator has to communicate the
messages transmitted in the source text by re-encoding them in the target text.
Indeed, the translation process is successful when the target text is a text
with a communicative purpose likewise the source text. But the problem
here is that sometimes the concerned standards between the texts are less
manageable (linguistically speaking, static), and sometimes they are more
manageable (linguistically speaking, dynamic). In this case Hatim & Mason
(1997) state: At one extreme, there will be those local and global-level textual
occurrences which display maximal cohesion, and consequently maximal
coherence, where intertextuality is least intricate, intentionality least opaque,
situationality least cumbersome, and informativity sparingly used. At the other
extreme, there will be local and global-level textual occurrences where cohesion
is not straightforward, and coherence is problematical to retrieve. In such
cases, values yielded by other factors, such as intentionality and intertextuality,
become slightly less transparent.
Needless to say, when the standards of textuality between the texts conform
to each other, the creativity of the translator is needed at its minimal level;
however, when they are not matched across the texts, the translator’s creativity
is highly required. More specifically, in case of convenient matches, there
are also many open options for the translator to choose between; hence, the
importance of creativity is again remarkable.

capítulo 1 • 13
Globally speaking, as Landers (2001: X) describes it, translation problems
are not like math problems, which have only one, or at most a strictly limited
number of right answers (…) translation is subjective in its essence. It is
important to be aware that analyzing a text based on textuality is essentially
an abstract endeavor. The combination of the standards of textuality makes
a text an actual use of language produced for a communicative purpose.
Furthermore, the interpretation of a text, which depends on proper analysis
based on encoding conventions in order to unravel desirable meaning, is by
and large relative.
As a result, to analyze a text, the translator primarily needs to holistically
contemplate the text as an inextricable phenomenon in which textual features
make sense in close proximity. Meanwhile, she should delve deeper into
the context in order to shed light on convoluted relationships in the text.
Nevertheless, the text analyst has to be cognizant of the delineation of his or
her analysis, not related to the deviation of the main purpose. Bearing all this in
mind, to analyze a text, the only criterion that the text analyst can resort to, is the
text itself. According to Hatim (2001): The parameters for an adequate transfer
are set by texts in communication, yielding not simply one definite meaning,
but rather an array of possible meanings (…) the unit text may be seen in relation
to “rhetorical purpose” and in terms of the way sequences of sentences are
formally organized.
The primary concern of the text analyst is the analysis of textual phenomena
such as sequential relationships, intersentential structure, and text organization
(33-34). In light of this, we suggest the establishment of the “translator’s triad”
(knowledge-ideology-creativity), which is necessary to fulfill the translation process.
The triad is an important instrument which every translator must possess in order
to carry out the process. To begin with, the translator should, as a communicator,
own the knowledge (and skills) of both languages in the process of translation. She
should have semantic knowledge, syntactic knowledge, and pragmatic knowledge.
These must be inherent in a good translator. Furthermore, the translator needs
to have sufficient knowledge of both cultures; knowledge of the rules of the code
governing “usage”, and knowledge of how to “use” the limiting conventions.
Meanwhile, ideology, a set of ideas and attitudes that strongly influence
the way in which a translator operates, is crucial to the whole process by which
most of the choices are dominated in between. It has always been remarked
that translating is not a neutral process. Suffice it here to say that the variety

14 • capítulo 1
and diversity of translated works, if not incorrect or irrelevant, may arise from
the translator’s ideology. Therefore, because of the diversity of ideologies, there
are numerous translations that have been rendered from the same original
text, making it extremely difficult to decide which one should be considered
the most accurate one (problem that will be addressed later). As mentioned
earlier, creativity plays a major role in rendering from the source to the target
text. On one extreme, it helps the translator pick the best option amongst many
estimated choices; on the other, it makes the translatorunique, likewise the
original author, who has created a unique text. Additionally, transferring the
uniqueness of the style of the original text to the target requires lots of creativity
from the translator.
Even more importantly, there is another issue which affects the whole
mechanism: The effect of external factors, such as clients who demand special
translations, organizations who patent the translated works, institutions with
specific tastes, and so on. In this respect, Pym (1998: ix) notes that it is only
through translators and their “ social entourage (clients, patrons, readers)”
that we can try to explain why translations “happened”, i.e., were produced in
a particular time and place. Again, these factors have “direct effects” on the
whole process of translation, which could lead the text towards their direction.
Obviously, the direct effects implicitly divert the standards of textuality of the
target text from the source text. As a result, it is the translator’s job to consider
the effects and, by properly setting out the standards, direct the effects in a
way that the least loss of communication remains. In what follows, we shall
concentrate on putting the issues discussed together in order to establish a
model which may help translators in the act of translating.

1.3  Units of translation3

Both the word and the sentence, the period and the paragraph, must be subject
to analysis and interpretation in context. When it comes time to translating, we
work with an organic semantic whole: The text, which is articulated through
subunits of meaning. Therefore, the unit of translation is the entire text.
Newmark insists that the unit of translation, understood as a segment of the
original text from which the translator can begin his or her reformulation in a

3  This part of the chapter was based on the website: <http://translation-blog.trustedtranslations.com/unit-of-

translation-2009-05-29.html>, which is highly recommended for further studies.

capítulo 1 • 15
different language, is part of a movable scale: “The word, the lexical unit, the
collocation, the group, the clause and the sentence–rarely the paragraph, never
the text”.
This great linguist defends an intermediate posture between the restricted
unit and the laxer unit of the speech analysis theorists, who consider
the unit of translation the entire text. “Unit of translation”, in contrast,
is a phrase  commonly used to make reference to the unit of analysis or
interpretation. Perhaps, it should be reserved to designate a segment of the
dialectic process of the negotiation of the meaning of the source text and its
placement in the target language.

1.3.1  Definitions of the unit of translation4

The UT, which has various interpretations, has been a subject of debate since
it was raised and defined by Vinay and Darbelnet, in 1958 (Nord [1997] 2001:
68). According to them (1958/1995: 21), the UT is “the smallest segment of
the utterance whose signs are linked in such a way that they should not be
translated individually.” Hatim and Munday simply call it “normally the
linguistic unit which the translator uses when translating” (2004: 25). Snell-
Hornby (1988/1995: 16) calls the UT “a cohesive segment lying between the
level of the word and the sentence.” UTs claimed by scholars range from the
culture of the language to the whole text, and also to the morpheme, with
Newmark stating that “all lengths of language can, at different moments and
also simultaneously, be used as units of translation […],” but “operatively, most
translation is done at the level of the smaller units (word and clause)” (1988:
66-67). Guo Jianzhong, based on his own experience, claims that in Chinese-
English translation, “the best UT is the paragraph” (2002: 544). Like other
translation theories, these are scholarly opinions. Since the first definition of
UT was given, fifty years ago, translation scholars have been interpreting and
repeating it, but in-depth analysis of translations is unheard of.
Developing translation theories, especially translation quality assessment
(TQA) models, without knowing the UT is no different than studying medicine
without knowledge of the human cell. Success is possible, but it may be
accidental. Identifying the UT is an attainable task, though not by repeating old

4  This part of the chapter was based on the website: http://www.erudit.org/revue/Meta/2009/v54/n1/029796ar.

html, which is recommended for further studies.

16 • capítulo 1
definitions and citing confusing opinions. Two favorable conditions have long
existed, which are the abundance of translations available in public libraries
and bookstores, and the large number of professional translators throughout
the world. The UT should not have been defined by scholars in the first place,
but it should have been synthesized from practice. A sample of translations
and a survey conducted among the latter could have identified the UT five
decades ago.
Instead of jumping into Vinay and Darbelnet’s over-interpreted definition or
its various confusing versions, all of which are primarily set within the context
of ST, the authors identify the UT through an analysis of sampled translated
texts, excerpts and sentences of over 23,000 pages, and confirm it with an
international survey among 66 professional translators.
As said before, being one of the fundamental concepts always argued
about in the realm of translation, the unit of translation (UT) has been given
various definitions by different theorists. Shuttleworth and Cowie (1997) define
it as: “a term used to refer to the linguistic level at which ST is recodified in
Target Language (TL)” (p. 192). In other words, it’s an element with which the
translator decides to work while translating the ST. Barkhudarov (1993) defines
an UT as “the smallest unit of Second Language (SL) which has an equivalent in
TL” (as quoted in Shuttleworth and Cowie, 1997, p. 192). He states that this unit
of translation, no matter how long, can “have a complex structure” of its own (as
quoted in Shuttleworth and Cowie, 1997, p. 192), although its parts separately
cannot be translated and replaced by any equivalent in the TL. Phonemes,
morphemes, words, phrases, sentences, and entire texts are potential units of
translation for him. What determines the appropriate UT, according to him, is
the wording at a given point in ST.
When a translator commences his work, i.e. translation, in accordance with
the type of ST he’s working on, he decides about the basic segments in ST to
be translated into TT. These segments range from a whole text, as in poetry, to
a single phoneme. The argument about the length of an UT also dates back to
the conflict between free vs. literal translation. Literal translation is pretty much
focused on individual words, or even morphemes. Therefore, in literal translation,
UTs are as short as words. On the contrary, a free translation “aims at capturing
the sense of a longer stretch of language” (Hatim and Munday, 2004, p. 17).
It always chooses the sentence. Of course, with the arising of text linguistics, the

capítulo 1 • 17
focus of free translation has moved from the sentence to the whole text. Once
a translator decides to work on larger segments than necessary to convey the
meaning of ST, free translation is at work. In the same way, when translating
smaller segments than needed, literal translation is under discussion. In Koller’s
terms (1979/1992), translating from a SL which is not that much related to TL
will usually result in choosing larger units, while closeness of SL and TL involves
smaller UTs.
Vinay and Darbelnet (1985/95) totally draw on the concept of word as a basis
for UT. Of course they do not believe in the non-existence of words, especially in
written languages. For them, a translator doesn’t need dictated criteria about
an UT, since the translation process is all done semantically. So, sentencing a
formal segment as a basic UT is not desired at all. Consequently, what should be
identified and distinguished as an unit for a translator, who translates thoughts
and concepts, is an unit of thought. Vinay and Darbelnet consider the three
following terms as being equivalent: “unit of thought”, “lexicological unit”, and
“unit of translation”. What they suggest as a definition for UT is “the smallest
segment of the utterance whose signs are linked in such a way that they should
not be translated literally” (as quoted in Hatim and Munday, 2004, p. 138). The
lexicological units of Vinay and Darbelnet contain “lexical elements grouped
together to from a single element of thought” (as quoted in Hatim and Munday,
2004, p. 138).
Several types of UTs are recognized by them, such as:
1. functional units
2. semantic units
3. dialectic units
4. prosodic units

The last three types are, according to them, counted as UTs, but the
functional units are almost too long to include just one UT. Three other different
categories arise while looking at the relationship between units of translation
and words inside a text:
1 - Simple units: Vinay and Darbelnet correspond this type to a single word.
It›s the simplest, as they state, and at the same time, the most widely used unit.
In this case, the number of units equals the number of words. A replacement of
words will not lead to a change in the sentence structure.

18 • capítulo 1
2 - Diluted units: These units contain several words which, in turn, shape a
lexical unit, since they pursue a single idea.
3 - Fractional units: «A fraction of a word» is what this type of UT consists of.
For Newmark (1988), “a sentence is a natural unit of translation” (p .65). He
then considers some other sub-units of translation in the sentence, the first
of which is the morpheme. Unless placed in special cases, Newmark states,
morphemes shouldn’t be considered seriously. Clause, group, collocation,
and words including idioms and compounds are grammatical and lexical sub-
units of translation proposed by him. Certainly, Newmark’s proposed category
partly relies on a scale formerly established by Michael Halliday in 1985. The
following scale is the one according to which Halliday performs a systematic
analysis of English:


Morpheme definition
A morpheme is a meaningful linguistic unit consisting of a word (such as dog) or a word
element (such as the -s at the end of dogs) that can’t be divided into smaller meaningful parts.
Adjective: morphemic.
Morphemes are the smallest units of meaning in a language. They are commonly classified
as either free morphemes (which can occur as separate words) or bound morphemes (which
can›t stand alone as words).
Many words in English are made up of a single free morpheme. For example, each word
in the following sentence is a distinct morpheme: “I need to go now, but you can stay.” Put
another way, none of the nine words in that sentence can be divided into smaller parts that
are also meaningful.

capítulo 1 • 19
In linguistics, a word is the smallest element that may be uttered in isolation
with semantic or pragmatic content (with literal or practical meaning). This contrasts deeply
with a morpheme, which is the smallest unit of meaning, but will not necessarily stand on its
own. A word may consist of a single morpheme (for example: oh!, rock, red, quick, run, expect),
or several morphemes (rocks, redness, quickly, running, unexpected), whereas a morpheme
cannot stand on its own as a word (in the words just mentioned, these are -s, -ness, -ly, -ing,
un-, -ed). A complex word will typically include a root and one or more affixes (rock-s, red-
ness, quick-ly, run-ning, un-expect-ed), or more than one root in a compound (black-board, rat-
race). Words can be put together to build larger elements of a language, such as phrases (a
red rock), clauses (I threw a rock), and sentences (He threw a rock too, but he missed).
The term word may refer to a spoken word or to a written word, or sometimes to the
abstract concept behind either. Spoken words are made up of units of sound called phonemes,
and written words of symbols called graphemes, such as the letters of the English alphabet.

In grammar, a clause is the smallest grammatical unit that can express a
complete  proposition. A typical clause consists of a subject and a predicate, where the
predicate is typically a verb phrase – a verb put together with objects and other modifiers.
However, the subject is not always expressed. This is often the case in null-subject languages
(for example, Portuguese) if the subject is retrievable from context, but it also occurs in certain
cases in other languages such as English (as in imperative sentences and non-finite clauses).
A simple sentence usually consists of a single finite clause with a finite verb that is
independent. More complex sentences may contain multiple clauses. Main clauses are those
that can stand alone as a sentence. Subordinate clauses are those that would be awkward
or incomplete alone.

A sentence is a linguistic unit consisting of one or more words that are grammatically
linked. A sentence can include words grouped meaningfully to express a statement, question,
exclamation, request, command, or suggestion. A sentence is a set of words that, in principle,
tells a complete thought (although it may make little sense taken in isolation out of context).
Thus, it may be a simple phrase, but it conveys enough meaning to imply a clause, even if it is
not explicit. For example, “Two” as a sentence (as an answer to the question “How many were
there?”) implies the clause “There were two”. Typically, a sentence contains a subject and a
predicate. A sentence can also be defined, purely in orthographic terms, as a group of words

20 • capítulo 1
starting with a capital letter and ending in a full stop. (However, this definition is useless for
unwritten languages, or languages written in a system that does not employ both devices
or precise analogues thereof.) For instance, the opening of Charles Dickens’s novel Bleak
House begins with the following three sentences:
London. Michaelmas term lately over, and the Lord Chancellor sitting in Lincoln’s Inn
Hall. Implacable November weather.
The first sentence involves one word; a proper noun. The second sentence has only
a non-finite verb (although using the definition given above, e.g. «Chancellor sitting in Lincoln’s
Inn Hall.”, would be a sentence by itself). The third sentence is a single nominal group. Only an
orthographic definition encompasses this variation.
In the teaching of writing skills (composition skills), students are generally required to
express (rather than imply) the elements of a sentence, leading to the schoolbook definition
of a sentence as one that must [explicitly] include a subject and a verb. For example, in
second language acquisition, teachers often reject one-word answers that only imply a
clause, commanding the student to “give me a complete sentence”, by which they mean an
explicit one.
As with all language expressions, sentences might contain function and content words,
and properties such as characteristic intonation and timing patterns.
Sentences are generally characterized in most languages by the inclusion of a finite verb,
e.g. “The quick brown fox jumps over the lazy dog”.

Going back…
Newmark considers no priority for each of the lexical or grammatical units, since wherever
they exist, he believes, enough importance should be paid towards them.
Briefly speaking, Newmark (1988) labels paragraphs and texts as higher UTs, and
sentences, groups, clauses, and words as lower UTs. He contends that “the mass of
translation uses a text as a unit only when there are apparently insuperable problems at the
level of the collocations, clause or sentence level” (p. 64). Recent emphasis on communicative
competence and language is what Newmark counts as a factor which made the text as
unit renowned.  In his terms, most of the translation is done at the smaller units, i.e. word
and clause.
Trying to delve more into the details and providing a clearer elaboration on the concept
of UT, Newmark (1988) states that in informative and authoritative texts, the focus is on
the word; in informative texts, on the collocation and the group; and in vocative texts, on the
sentence and the text, as a unit.

capítulo 1 • 21
He concludes in this way: “all lengths of language can, at different moments and also
simultaneously, be used as units of translation in the course of the translation activity(…) to me
the unit of translation is a sliding scale, responding according to other varying factors, and (still)
ultimately a little unsatisfactory” (pp. 66-67).

1.4  Coherence and cohesion

Coherence: The property of unity in a written text that stems from the
relationship between its underlying ideas, and from the logical organisation
and development of these ideas.
Cohesion: The property of flow and connection in a written text that stems
from the linguistic links among its surface elements.
A paragraph has good cohesion when each sentence is clearly linked to
the next. Coherence and Cohesion mean that all of the parts are logically and
linguistically connected to form a whole.

1.4.1  Cohesion5

Cohesion is the first of the seven standards, and has the function of attaching,
syntactically and lexically, the text together in order to create textual unity. It is
a function of syntax in communication that imposes organizational patterns
upon the surface text (the presented configuration of words) (Beaugrande &
Dressler 1992: 48). Hatim and Mason define a cohesive text as:

A text is cohesive in the sense that the various components of the surface text (the
actual words we see) are mutually connected within a sequence of some kind. In terms
of both lexis and grammar, that is, the surface components depend upon each other in
establishing and maintaining text continuity. (1997: 15)

Cohesion is obtained from five main markers of cohesive relationships:

references, substitutions, ellipsis, junctions, and lexical cohesion. To be cohesive,
a text should consist of short-range stretches of the surface structure which are set
up as closely-knit patterns of grammatical dependencies. Long-range stretches,

5  This part of the chapter was based on the article: MIKHCHI, H. Standards of Textuality: Rendering English and
Persian Texts based on a Textual Model. Journal of Universal Language 12-1, 2011.

22 • capítulo 1
in contrast, could be handled by re-utilizing previous elements or patterns,
economizing were possible (Beaugrande & Dressler 1992: 79).
Among the standards of textuality, cohesion is the most probably
linguistic. Cohesion is a property of the linguistic surface of the text. The
linguistic elements that occur in sequences of sentences act together to form
texture, a term which refers to cohesive ties at the level of connected discourse
(as opposed to cohesive ties within individual sentences). Cohesion makes
coherence linguistically evident. The cohesive text is, as a result, the end
product of translation (Neubert & Shreve 1992: 102-103).

1.4.2  Coherence

Coherence, in contrast to cohesion, includes the layout and ordering of the

concepts and relations of the text which are caught on by the surface text.
According to Hörmann (1976), continuity of senses is defined as the foundation
of coherence which results from the configuration of concepts, expressed
relations, and the receivers’ knowledge of the world. He notes: A text makes
sense because there is a continuity of senses among the knowledge activated
by the expressions of the text. A senseless or nonsensical text is one on which
text receivers can discover no such continuity, usually because there is a serious
mismatch between the configuration of concepts and relations expressed and
the receivers’ prior knowledge of the world. We would define this continuity of
senses as the foundation of coherence, being the mutual access and relevance
within a configuration of concepts and relations. (apud Beaugrande & Dressler
1992: 84)
The continuity of senses created by coherence is, in fact, the interpretation
of the text that readers (if the text is written) appreciate and make sense of. The
lack of the continuity of senses disturbs the communication purpose of the text.
On the other hand, Grice’s maxims of manner and relation tell us that there
is order imposed on the information content. This order is a logical structure
which defines the semantic connections between information units in the text.
“Coherence is the connection of individual information elements with a certain
logical structure” mention Neubert & Shreve (1992). They say that coherence is
a property which texts assume when their information contents take on such
a logical structure. Coherence is not an information unit; it is the connection
of individual information elements to create larger, more global structures of
meaning. (Neubert & Shreve 1992: 93-96).

capítulo 1 • 23
Also, Hatim & Mason (1990), after criticizing the views which state that
coherence is not something created by text, but rather an assumption made
by language users that, in accordance with the cooperative principle, texts are
intended to be coherent (p. 194), define coherence as the procedures which
ensure conceptual connectivity, including:
(1) logical relations
(2) organization of events, objects, and situations
(3) continuity in human experience (p. 195)

A coherent text has an underlying logical structure that guides the reader
through the text. This structure helps the reader overcome his ignorance of
specific details. The maintenance of coherence could be established as a
criterion for adequate translation, because in this sense, translatability refers
to the way in which a text makes sense to the readers through the organization
of its content, and the relevance and comprehensibility of its concepts and
ideas. In essence, coherence is not imported from the source text; coherence
is constructed anew in the target text, using the source sense relations as a
template. Re-establishing coherence is an example of how translation is a
creative textual act (Neubert & Shreve 1992: 95-100).

1.4.3  Coherence – Semantic property

Coherence is an aspect of a piece of text that makes it meaningful in the minds

of the readers. When the text begins to make sense on the whole, it is said to be
coherent. If the readers can follow and understand a text easily, it obviously has
coherence. Rather than the text appearing linked together perfectly, it is the
overall impression of the text that appears to be smooth and clear.
Coherence can be achieved through the use of titles, subtitles, paragraphing,
formatting, logical ordering, orthography (spelling, punctuation, capitalisation),
and so forth. For example, this article itself is coherent because it has proper
paragraphs that are logically ordered from one another; it has proper subtitles
to divide the text into coherence and cohesion; and it makes use of bolding and
capitalisation to signal important parts.
Inference is very important in achieving coherence, because sometimes in
a text, the reader may need to have prior knowledge about a subject. So, while
a text may appear coherent to one person, it may appear incoherent to another

24 • capítulo 1
person (possibly because of lack of prior knowledge). However, even in this
case, the text may be cohesive, because the sentences join well together, create
meaning, and flow on from each other. So, you could have an incoherent, but
cohesive text.

1.4.4  Cohesion – Flow of a text

Cohesion can be thought of as the glue sticking different parts of furniture so

that it takes the shape desired by the writer.
Cohesion is the grammatical and lexical link within a text or sentence that
holds a text together and gives it meaning. In short, the links that join different
sentences and make the text meaningful can be thought of as cohesion in
the text. Establishing connections between sentences, sections, and even
paragraphs using synonyms, adverbials, conjunctions etc. is what brings
cohesion to a text.

Consider this cohesive example:

•  John went to the shop and he bought an ice-cream then ate it.
Now, let’s remove the cohesion (flow) of this text and see what happens.

•  John went to the shop. John bought an ice-cream. John ate the ice-cream.
A little bit repetitive and monotonous, right? As it can be seen in the cohesive
example above, three cohesive devices have been used:

1. Ellipsis (‘he’ has been omitted in ‘then [he] ate it’)

2. Conjunctions (‘and’ to join the sentences together)
3. Substitution (John becomes ‘he’, and the ice-cream becomes ‘it’)

1.5  Connotation and denotation6

Connotation and Denotation are the two main methods of describing the
meanings of words. Connotation refers to the wide array of positive and
negative associations that most words naturally carry, whereas denotation is
the precise, literal definition of a word that might be found in a dictionary. To

6  This part of the chapter was based on the article: <https://www.csun.edu/~bashforth/098_PDF/


capítulo 1 • 25
introduce the idea of connotation, let’s define it as the associations that people
make with a word. You can contrast connotation with the denotative value of a
word (its literal meaning), and give an example of a word. Connotation is the
emotional and imaginative association surrounding a word. Denotation is the
strict dictionary meaning of a word.
Connotation and denotation are not two separate things/signs. They are
two aspects/ elements of a sign, and the connotative meanings of a word exist
together with the denotative meanings. Connotation represents the various
social overtones, cultural implications, or emotional meanings associated
with a sign. Denotation represents the explicit or referential meaning of a sign.
Denotation refers to the literal meaning of a word, the ‘dictionary definition.’
For example, the name ‘Hollywood’ connotes things such as glitz, glamour,
tinsel, celebrity, and dreams of stardom. At the same time, the name ‘Hollywood’
denotes an area of Los Angeles, worldwide known as the center of the American
movie industry.
Diction, an element of style, refers to the words writers use to express ideas.
Words convey more than exact, literal meanings, in which case they “connote”
or suggest additional meanings and values not expressed in general dictionary
definitions. Words that “denote” a core meaning are those that are generally
used and understood by the users and the audience to represent an object or
class of objects, an act, a quality, or an idea. However, because of usage over
time, words that denote approximately the same thing may acquire additional
meanings, or connotations, that are either positive (ameliorative) or negative
Consider the changes undergone by these words in the 20th century: liberal,
diversity, team player, right wing, follower, gay, minority, feminist, left wing, abuse,
conservative, motherhood, extremist, rights, relationship, harassment, family,
propaganda, peacekeeper, comrade, drug addict, druggie, drug fiend, substance
abuser, handicapped, crippled, disabled, differently abled, horse, steed, nag,
plug, house, home, abode, domicile, residence, thin, slender, slim, skinny,
lean, beanpole, attractive, pretty, beautiful, handsome, fair, reporter,journalist,
broadcaster, newshound, unattractive, plain, dull, ugly.
Words have both denotations (literal meanings) and connotations
(suggestive meanings). “Fungus” is a scientific term which denotes a certain
kind of natural growth, but the word also has certain connotations of disease
and ugliness. Connotations can be both positive and negative; for example,
“lady” carries a hint of both elegance and subservience. The influence of the

26 • capítulo 1
connotative meaning can also change the denotative meaning, one example
being the thoroughly transformed word “gay”.
Understanding the difference between denotation and connotation is
important to understand definitions and how concepts are used. Unfortunately,
that is complicated by the fact that these terms can be used in two different
ways: grammatical and logical. Even worse, both uses are worth keeping in
mind, and both uses are relevant to logical, critical thinking. In grammar, a
word’s denotation is whatever the word directly refers to, which is roughly
equivalent to its lexical definition. Thus, the word “atheist” denotes a person
who disbelieves in or denies the existence of gods. A word’s connotation refers
to any subtle nuances that might or might not be intended by its use. For
example, one possible connotation for the word “atheist” might be someone
who is immoral and wicked, depending upon who is doing the speaking or
listening. Separating grammatical denotation from connotation is important
because while one might assume that a word’s denotation is fully intended, it
is far more difficult to determine whether a word’s connotations are intended.
Connotations are often emotional in nature, and thus if they are intended,
it may be for the purpose of swaying a person’s emotional reactions rather than
the logical evaluation of an argument. If there are misunderstandings about
how a person is using a word in a particular debate, a primary source of that
misunderstanding might lie in the word’s connotations: people might be
seeing something not intended, or the speaker may be intending something
people don’t see. While constructing your own arguments, it’s a good idea not
to merely look at what your words denote, but also at what they connote.
The relationship between words and meanings is extremely complicated,
and it belongs to the field of semantics. For now though, what you need
to know is that words do not have single, simple meanings. Traditionally,
grammarians have referred to the meanings of words in two parts: Denotation:
A literal meaning of the word; and Connotation: An association (emotional or
otherwise) which the word evokes. For example, both “woman” and “chick”
have the denotation “adult female” in North American society, but “chick”
has somewhat negative connotations, while “woman” is neutral. For another
example of connotations, consider the following:
Negative: There are over 2,000 vagrants in the city.
Neutral: There are over 2,000 people with no fixed address in the city.
Positive: There are over 2,000 homeless people in the city.

capítulo 1 • 27
All three of these expressions refer to exactly the same people, but they invoke
different associations in the reader’s mind: a “vagrant” is a public nuisance,
while a “homeless” person is a worthy object of pity and charity. Presumably,
someone writing an editorial in support of a new shelter would use the positive
form, while someone writing an editorial in support of anti-loitering laws
would use the negative form. In this case, the dry legal expression “with no fixed
address” quite deliberately avoids most of the positive or negative associations
of the two other terms.A legal specialist would try to avoid connotative language
altogether when writing legislation, often resorting to archaic Latin or French
terms which are not part of ordinary spoken English, and thus, are relatively
free of strong emotional associations.
Many of the most obvious changes in the English language over the past
few decades have had to do with the connotations of words which refer to
groups of people. Since the 1950’s, words like “Negro” and “crippled” have
acquired strong negative connotations, and have been replaced either by
words with neutral connotations (i.e. “black,” “handicapped”) or by words
with deliberately positive connotations (i.e. “African-Canadian,” “differently-
abled”). Language meaning is continually shifting, always contextual, and
influenced by historical, cultural, and economic factors. For instance, terms
that were used years ago, such as gangster and thug, denoted (that is, specifically
referred to or explicitly meant) individuals involved in criminal activities, who
were prone to violence, and who had general disregard for laws and social
order. Also, particularly during the Depression era, gangsters and thugs were
associated with male immigrants from Italy, Ireland, and other European
countries. However, today’s gangsters and thugs are associated with African-
American males, and the terms are used to connote (that is, suggest or imply)
that these individuals are concerned with accumulating material wealth, are
hyper-sexual, and are threats to middle-class suburban folks. The terms also
suggest a particular urban ethic and cultural cachet that far transcend the
original suggestion of criminal activity. Just think of the category of “gangster
rap,” a musical genre that practitioners have argued captures the “truth” of
the black, urban male experience. The terms “thug” and “gangster” have
also become prevalent all across youth culture, designating clothing styles,
postures, attitudes, values, etc. and spawning a vast array of related terms.
Much has changed since the 30s, and these changes are reflected in language,
as demonstrated in the example above.

28 • capítulo 1
In denotation7, the referential or lexical meaning of a word denotes a core
meaning of an object, act, or quality that is generally used and understood by the
users; whereas connotation implies the associations that a word may bring to
the hearer’s mind according to his cognition and experience that are additional
to its literal or dictionary meaning. Some words that have approximately the
same denotation may hold different connotations. The words “house” and
“home” have a shared denotation of “a dwelling place”, but “home” has the
additional connotations of “comfort”, “privacy” and “domesticity” that are
absent from the word “house”. The denotation of the word “snake” in the
Advanced English Dictionary is: “a long legless, crawling reptile, some kinds of
which are poisonous”. As for the connotations associated with the word “snake”,
they include “evil” and “danger”, as reflected in the idiomatic expression
“a snake in the grass”. The English word “bus” may connote “low cost” and
“convenience” for some people (especially the poor), but it may be associated
with “discomfort” and “inconvenience” for others who own private cars. It may
have the connotation of “school” for many children who go to school by bus.
The following stanza (McCrimmon, 1963, p.131) shows the significant
connotative differences between some pairs of synonyms (kitten and cat,
mouse and rat, chicken and hen):
Call a woman a kitten, but not a cat.
You can call her a mouse, but you cannot call her a rat.
Call a woman a chicken, but never a hen.

Different scholars have tackled connotative meaning differently. John

Stewart Mill, as early as 1843, related the term ‘connotation’ to the attributes or
properties that a word connotes in opposition to its denotation (cited in Lyons,
1977, p. 175). Pyle and Allgeo (1970, pp.198-200) consider a word’s associations
to be restricted to “the senses of all the words with which it is always used”,
i.e. regardless of its referents. Osgood et al (1971, pp.15-16) focus on the
psychological condition of using a word, and view connotation as the emotive
reactions of the users of the language to it. Nida and Taber (1974, pp. 91-94) view
connotation in terms of the emotional effects of a word on speakers and the
emotional response of hearers. Lyons (1987, pp. 54, 143) states that connotation
as a psychological and social aspect of expressive meaning reflects the feelings

7  This part of the chapter was based on the article: ILYAS, A. The Importance of Connotation in Literary Translation.
AWEJ Special Issue on Literature No.1, 2013.

capítulo 1 • 29
of language users towards a certain issue or thing, and that words like ‘huge’,
‘enormous’, ‘gigantic’, and ‘colossal’ reflect the speakers’ feelings rather than
the things they describe. Palmer (1981, p. 92) views connotation as the emotive
overtones that result from different styles and dialects. Hervey, S. & Higgins,
Ian (1992) classify connotative meaning into six types: Associative, Emotive,
Attitudinal, Reflected, Collocational, and Allusive Connotation. Hatim (1997,
p. 228) defines connotation as the “additional meanings that a lexical item
acquires beyond its primary, referential meaning”. Graddol et al (2005, p. 103)
equate connotation with “the associations that words have for us”. Munday
(2001, p. 154) states that shifting the ST connotations may sometimes produce
a shift in ideology. Many connotations are well-established and constitute part
of the linguistic competence of speakers (Nord, 2005, p. 102). Carter (2004, p.
116) classifies words into core words that are neutral in connotation, such as
the word “thin”, but a core word may have a synonym that is positive (slim) or
negative (skinny).
Connotation is one way in which synonyms may differ (Palmer 1981, p. 89).
They are subject to continuous change. Many of the most obvious changes
in the English language have resulted from changes in word connotations.
Words that have positive connotations may become negative and vice versa. An
example of this is the word “gay”, which was quite positive in the past. The word
“charm”, which has positive associations today, had negative connotation in the
Elizabethan age, as it was associated with sorcery. The expression “nowadays”
had associations of vulgarity and was avoided by the educated people in
Shakespeare’s time. Its use then was restricted to people of low status. The word
“liberal” has a negative connotation in: “He is too liberal”, but has a positive
connotation in: “He is liberal in an area of dictatorship”. The word “bug” can
have a positive or negative connotation in different co-texts and contexts: This
room is full of bugs! (negative connotation.) John is as cute as a bug. (positive
connotation) The connotation of a word is also affected by the context of its use
(setting, occasion, purpose or function, and participants). The word “laser” is
admirable among engineers, but many people have negative feelings about it,
as a result of the medical risks associated with laser technology. That is why
advertisers use the word “scanner” in advertisements, which is an euphemism,
instead of “laser-using equipment”.
Connotation is not restricted only to words. Morphemes, syntax, sounds,
spellings, and even typographical features can all connote certain meanings

30 • capítulo 1
and have specific intended functions. For example, the suffix -ish was neutral
as to connotation in the past, but nowadays, it has acquired unfavourable
connotations in words such as “boyish”. The suffix -ese (in words like
journalese, translationese, officialese) also contains negative connotations.
Some grammatical structures too can have certain connotations. In the two
following examples, the second one has more negative connotations, as the
speaker seems to be annoyed by the person asking him such questions:
1. He always asks me such questions.
2. He is always asking me such questions. (negative connotation).

Some have even attached associative meanings to vowels and consonants,

saying that long vowels are more peaceful and solemn than short ones, as the
latter may express quick motion, agitation, and triviality (Boulton, PP.53-59).
Poets make use of certain sounds to connote certain associative meanings,
as in the deliberate and symbolic use of alliteration and assonance. Users of
language can use spellings to express certain associative meanings, as dealers
do when they use the older spelling forms over the doors of their shops (Gufte
Shoppe instead of Gift Shop) in order to attach a favourable quality of antiquity
to their goods. The romantic poet Keats, in his poem “Ode to a Nightingale”,
changes the spelling of the word “fairy” into its older spelling form “faery” to
create a connotation of antiquity:

“She stood in tears amid the alien corn

The same that oft-times hath
Charmed magic casements, openings on the foam
Of perilious seas, in faery lands forlorn”

As for the associative functions of typographical features, some poets have

made use of such features to support the general theme or message of their
works, as did the poet Malcolm Timperly in his poem: “The Fan” (in which the
poem is presented in the format of a fan); and John Hollander in his poem:
“Swan and Shadow” (in which the poem is presented in the format of a swan
with its reflected shadow). In advertisements, second-hand cars are sometimes
advertised as pre-owned cars, in which the emotional effect is clearly played
out. In the political discourse too, connotations have a significant role and

capítulo 1 • 31
function. For example, a news item once appeared in the British magazine:
Reader’s Digest , (1989, October,135) as : “Why Russia Can’t Feed Itself “. The
word “feed” is mainly used for animals and babies, and this use of the verb
with reference to Russians is based on its negative connotative function, which
reflected Britain’s negative view of Russia then.
A translators’ main concern is trying to reproduce a similar version of
the Source Text’s (Henceforth ST) meaning/function in their Target Text
(Henceforth TT) versions. Beaugrande (1999, P.12) tackles the question of
meaning in terms of discourse, in which meanings cannot be simply handled
at the literal level, since they “mutually constrain each other”. Hickey supports
Fawcett’s view that a translator faces a special difficulty in relation to the
degree “the target audience may need hints as to what is presupposed rather
than explicitly conveyed in the original (1998, P.7). A translator is not only a
bilingualist, but also a biculturalist who does not only consider linguistic and
referential factors in the process of translating, but also takes connotations
into consideration (Jianzhong & Yan, p.180).
In literary translation, where form acquires much importance, connotations
are very important as a tool for enriching meanings and arousing imagination.
Firth relates the difficulties of translation to matters of meaning (1957, P.
32). Some pairs of synonyms that are denotationally and referentially similar
are not interchangeable in certain texts and contexts. For Mc-Guire (1980, P.
15), synonyms such as perfect/ideal include within each of them a set of non-
translatable associations or connotations. Synonymous words that denote
approximately the same thing may convey different connotations, not only
between different languages (such as English and Arabic), but also between
different varieties of the same language (Ilyas, 2001). The translator has to
choose the right TL synonym (out of a set of synonyms embodied with positive,
negative, and neutral associations). In the translation of certain literary texts, it
is often the case that many culturally-charged SL connotations are mistranslated
or lost, either because translators focus on denotation only, are unaware of the
connotative function of the SL items, or cannot find the proper TL connotative
equivalents. When such lost or distorted connotations constitute the gist of the
SL message, the harm done to the original text is beyond measure.
Connotations are indeed very important in literary translation (and may also
apply to some political, religious, and advertising texts), where form and formal
features have important functions and associations. Unfortunately, it is often

32 • capítulo 1
the case that translators focus on denotative and referential meanings when
rendering texts in which connotative meanings play an important functional
and artistic role, producing thus incongruent and awkward renderings. SL
connotations are mistranslated or lost, either because translators focus on
denotation only, are unaware of the connotative function of the SL items, or
cannot find the proper TL connotative equivalents. When such lost or distorted
connotations constitute the gist of the SL message, the harm done to the
original text is beyond measure indeed.


01. Identify the references in the following texts:
a) Every organization, as soon as it gets to any size (perhaps 1,000 people), begins to feel
a need to systematize its management of human assets. Perhaps the pay scales have
gotten way out of line, with apparently similar-level jobs paying very different amounts;
perhaps there is a feeling that there are a lot of neglected skills in the organization that
other departments could utilize if they were aware that they existed. Perhaps individuals
have complained that they don’t know where they stand or what their future is; perhaps
the unions have requested standardized benefits and procedures. Whatever the historical
origins, some kind of central organization, normally named a personnel department, is
formed to put some system into the haphazardry. The systems that they adopt are often
modelled on the world of production, because that is the world with the best potential
for order and system.
b) We all tend to complain about our memories. Despite the elegance of the human memory
system, it is not infallible, and we have to learn to live with its fallibility. It seems to be
socially much more acceptable to complain of a poor memory, and it is somehow much
more acceptable to blame a social lapse on ‘a terrible memory’, than to attribute it to
stupidity or insensitivity. But how much do we know about our own memories? Obviously
we need to remember our memory lapses in order to know just how bad our memories
are. Indeed one of the most amnesic patients I have ever tested was a lady suffering
from Korsakoff’s syndrome, memory loss following chronic alcoholism. The test involved
presenting her with lists of words; after each list she would comment with surprise on her
inability to recall the words, saying: ‘I pride myself on my memory!’ She appeared to have
forgotten just how bad her memory was.

capítulo 1 • 33
Substitution and ellipsis
Identify examples of substitution and ellipsis in this text:
c) The human memory system is remarkably efficient, but it is of course extremely fallible.
That beingso, it makes sense to take full advantage of memory aids to minimize the
disruption caused by such lapses. If external aids are used, it is sensible to use them
consistently and systematically - always put appointments in your diary, always add
wanted items to a shopping list, and so on. If you use internal aids such as mnemonics,
you must be prepared to invest a reasonable amount of time in mastering and practicing
them. Mnemonics are like tools, and cannot be used until forged. Overall, however, as
William James pointed out (the italics are mine): ‘Of two men with the same outward
experiences and the same amount of mere native tenacity, the one who thinks over his
experiences most and weaves them into systematic relations with each other will be the
one with the best memory.’
d) This conflict between tariff reformers and free traders was to lead to the “agreement
to differ” convention in January 1932, and the resignation of the Liberals from the
government in September 1932; but, until they resigned, the National Government was a
genuine coalition in the sense in which that term is used on the continent: a government
comprising independent yet conflicting elements allied together; a government within
which party conflict was not superseded, but rather contained - in short, a power-sharing
government, albeit a seriously unbalanced one.
e) The number of different words relating to ‘camel’ is said to be about six thousand. There
are terms to refer to riding camels, milk camels and slaughter camels; other terms to
indicate the pedigree and geographical origin of the camel; and still others to differentiate
camels in different stages of pregnancy and to specify innumerable other characteristics
important to a people so dependent upon camels in their daily life (Thomas, 1937)
f) There were, broadly, two interrelated reasons for this, the first relating to Britain’s economic
and imperial difficulties, the second to the internal dissension in all three parties.

Identify examples of conjunction in the following texts:
g) These two forms of dissent coalesced in the demand for a stronger approach to the
Tory nostrum of tariff reform. In addition, trouble threatened from the mercurial figure
of Winston Churchill, who had resigned from the Shadow Cabinet in January 1931 in
protest at Baldwin’s acceptance of eventual self-government for India.

34 • capítulo 1
h) These two sets of rules, though distinct, must not be looked upon as two co-ordinate and
independent systems. On the contrary, the rules of Equity are only a sort of supplement or
appendix to the Common Law; they assume its existence but they add something further.

Lexical cohesion
Identify examples of lexical cohesion in the following texts:
i) The clamour of complaint about teaching in higher education and, more especially, about
teaching methods in universities and technical colleges, serves to direct attention away
from the important reorientation which has recently begun. The complaints, of course,
are not unjustified. In dealing piece-meal with problems arising from rapidly developing
subject matter, many teachers have allowed courses to become over-crowded, or too
specialized, or they have presented students with a number of apparently unrelated
courses failing to stress common principles. Many, again, have not developed new
teaching methods to deal adequately with larger numbers of students, and the new
audio-visual techniques tend to remain in the province of relatively few enthusiasts
despite their great potential for class and individual teaching.
j) When we look closely at a human face we are aware of many expressive details - the
lines of the forehead, the wideness of the eyes, the curve of the lips, the jut of the chin.
These elements combine to present us with a total facial expression which we use to
interpret the mood of our companion. But we all know that people can ‘put on a happy
face’ or deliberately adopt a sad face without feeling either happy or sad. Faces can lie,
and sometimes can lie so well that it becomes hard to read the true emotions of their
owners. But there is at least one facial signal that cannot easily be ‘put on’. It is a small
signal, and rather a subtle one, but because it tells the truth it is of special interest. It
comes from the pupils and has to do with their size in relation to the amount of light that
is falling upon them.

Barkhudarov, L. (1993). The problem of the unit of translation. In P. Zelateva (Ed),Translation as
social action: Russian and Bulgarian perspectives (pp. 39-46). London and New York: Routledge.
Halliday, M. A. K. (1985/1994). An introduction to functional grammar. London, Melbourne and
Auckland: Edward Arnold.
Hatim, B. & Munday, J. (2004). Translation: An advanced resource book. New York: Routledge

capítulo 1 • 35
Koller, W. (1979/1992). EinfÜhrang in die Übersetzungswissenschaft. Heidelberg: Quelle and
Newmark, P. (1988). A textbook of translation. Singapore: Prentice hall international (UK) ltd
Shuttleworth, M. & M. Cowie. (1997). Dictionary of translation studies. Manchester: St. Jerome.
Vinay, J. P & Darbelnet, J. (1958/ 1995). A methodology for translation. In L. Venuti (Ed.), The
translation studies reader (pp. 84-93). London: Routledge.

36 • capítulo 1
History and
2.  History and intertextuality 8

Communication is essentially incomplete and inferential — it is impossible

to “say everything about anything at any point in time” (Winter 1994, p.47). To
derive intended meaning from a spoken utterance or text, the hearer or reader
needs to enrich or modify semantic representations of linguistic input (literal
or prototype meanings) by using inferences based on context. This context,
or background, is “the space of possibilities that allows us to listen to both
what is spoken and what is unspoken”; and meaning is created in an active
process whereby “linguistic form triggers interpretation rather than conveying
information” (Winograd and Flores 1986, p.57).
This “space of possibilities” forming the context of a text or utterance is a
subset of the recipient’s entire cognitive environment, selected on the basis of
relevance. A person’s cognitive environment includes information that can be
perceived externally, as well as knowledge stored in memory, and information
deriving from previous utterances or texts (Gutt, 1991, p.26). This latter aspect
of the cognitive environment is referred to as intertextuality. Intertextuality,
which has been described as an “all pervasive textual phenomenon” (Hatim
1997a, p.29), and a “precondition for the intelligibility of texts” (Hatim and
Mason 1997, p.219), is essentially a mechanism through which a text refers
backward (or forward) to previous (or future) texts, by alluding to, adapting,
or otherwise invoking meanings expressed in those other texts. In order to
retrieve the full range of intended meaning in a given text, readers need to be
able to recognise and understand such intertextual references. Failure to do
so will result in partial understanding, or incomplete retrieval of the intended
meaning of the text concerned.

2.1  Types of intertextuality

Intertextuality can operate at “any level of text organisation” (Hatim and

Mason, 1997, p.18), involving phonology, morphology, syntax or semantics
(Hatim 1997b, p.201); and its expression ranges from single words or phrases
that have special cultural significance in a given community at a certain time,
to macro-textual conventions and constraints associated with genre, register,
and discourse. Intertextuality, therefore, encompasses any element (macro or

8  This part of the chapter was based on the article: "ENNIS, T. MA Translation Studies – Translation and Discourse
- Question 5.

38 • capítulo 2
micro) that enables readers to identify and derive meaning from the surface
features of the text in question by reference to other texts or text features
they have previously come across. Intertextual reference is not usually made
casually or for embellishment purposes, but it is nearly always motivated; it is
used deliberately to convey meaning. Several writers have attempted to classify
different types of intertextuality and provide structure to what is a very wide-
ranging and varied phenomenon. These are summarized and expanded upon
in Hatim (1997a).

2.1.1  Horizontal or vertical reference

Quoting the work of Bakhtin, Hatim distinguishes between horizontal and

vertical intertextuality. In the first case, the relation between two texts is
explicit—a text, or extract thereof, written in reply to or development of another
one, for example. This type of intertextuality is a key feature of academic writing,
and has been identified by Hoey (1991, p. 31-34) in terms of “academic oeuvre”
and “text colony”. Vertical intertextuality, on the other hand, is more implicit,
and may relate, for example, to writing conventions.

2.1.2  Manifest or constitutive reference

The second dichotomy discussed by Hatim stems from the work of Fairclough
(e.g. 1992), which sees intertextual reference as manifest (typically expressed
explicitly through surface textual features such as quotations and citations), or
constitutive and hence more opaque. In the latter case, the reader has to activate
the reference by tracing it back to its source.The reference is in the surface
features of the text, but the reader has to make an effort to retrieve it. A reader’s
inclination and/or ability to do so will vary from individual to individual.

Degree of mediation
Each of these classifications of intertextuality is seen by de Beaugrande and
Dressler (e.g. 1981) in terms of the degree to which the author, or translator,
introduces his or her personal assumptions or beliefs into the text in question,
i.e. the extent of mediation. Clearly these three types of classification are
overlapping, with horizontal and manifest reference tending to be least
mediated, whereas vertical and constitutive reference will usually involve
greater author/translator presence.

capítulo 2 • 39
2.1.3  Socio-cultural objects and socio-textual practice

Hatim (1997a) goes on to make a distinction between socio-cultural objects and

socio-textual practices as vehicles of intertextual reference. The first of these
(socio-cultural objects) operate at a micro-level and may be conveyed in a single
word or phrase that has particular significance for a given culture at a given
time. An example provided by Hatim and Mason (1997, p.18) is the biblical
reference to Job in the phrase “the patience of Job”. Another similar example
from the same source would be references to “Jonas” in describing someone
as a traitor. Intertextual references to the Bible and other universal and more
or less timeless literary works (Shakespeare, for instance) are likely to be long-
lasting and retrievable by a wide range of cultures.
Other more ephemeral references, to the political or entertainment domain
for example, would normally be more restricted to their spatiotemporal
accessibility. Examples of this type would include the terms “Thatcherism” and
“Son of Star Wars”, or the use of “Drugs industry” and “Pretoria case” in the
following headline: A: Drugs Industry Humiliated in Pretoria Case (Guardian
Weekly, May 2, 2001). Each of these is likely to conjure up a series of images,
attitudes and meanings in the minds of individual readers, going beyond mere
semantic representation, that can be traced back to previous texts they have read;
although as time passes and the sources become historically more distant and
lose their relevance or fade from public view, the references concerned are likely
to become weaker. Socio-textual practices, on the other hand, are the macro-
constraints and conventions governing register, genre, discourse, and text type,
which make it possible to recognise a given text as a member of a wider universe of
texts. Thus, for example, text A is recognisable as a newspaper headline through
conventional features such as the absence of definers and auxiliary verbs. Fowler
(1991, p.23) also argues that newspaper editorials are likely to display styles and
ideologies that identify them as dependent upon the economic and/or political
interests that finance them. Moon (1994, p.121) points out the role of metaphor
and fixed expressions in establishing intertextuality, in the form of shared
cultural knowledge, and Fowler (1991, pp. 165, 213) argues that the meaning of
metaphor and proverbs is accessible through intertextuality. The same idea is
echoed by McCarthy and Carter (1994, p.115), who see intertextual competence
involved in recognising “oblique references” to shared cultural knowledge and
experience, in terms of sustaining cultural membership and solidarity.

40 • capítulo 2
Analysis of examples
The term “Son of Star Wars”, which refers to the missile defense system
proposed by President George W. Bush, is very rich in intertextual reference,
as analysed below following the approach used in Hatim (1997b, pp.201-2). It
links back to the original strategic defense. It is not usually necessary to know
the exact source of a reference to be able to retrieve its meaning. The term
would be introduced in the 1980s by President Reagan. This was dubbed “Star
Wars” in reference to the highly popular science-fiction film of that time, and in
reference to President Reagan’s Hollywood background as a second-rate movie
actor. The whole scheme was considered by many people to be in the realms of
science-fiction, and therefore, was not taken very seriously. Yet its implications
were extremely serious, since they involved the possibility of conflict in space,
and a further ratcheting up of the Cold War. The new version proposed 20 years
later came at a time when a new version of the Star Wars film was released, and
it was proposed by an American President with right-wing views considered to
be very close to those of Ronald Reagan. George W. Bush was seen by many as
Ronald Reagan’s ideological offspring, so the term “Son of Star Wars” is an apt
way to describe that new generation missile defense policy, which is so close
to his heart. The fact that it was proposed at a time when the Cold War threat
no longer existed made it seem to many people all the more ridiculous and
deplorable. The blunt-sounding “Son of…” rather than “The Son of…” seems
to suggest a person lacking in intelligence, and it made the term “Son of Star
Wars” sound even more derogatory than “Star Wars” as applied to the original
idea in the 1980s.
As a second example, we can consider the following headline from the Daily
Telegraph: B: Holiday High Jinks at 5am Land Euan in More Trouble (Daily
Telegraph, August 18, 2000). In this newspaper headline referring to Euan
Blair, son of the British Prime Minister, the single word “Euan”, especially when
used in conjunction with “trouble”, for many British people in the summer
of 2000 would have conjured up images of a fun-loving teenager constantly
causing embarrassment to his famous father. The readers’ attitudes towards
this would probably depend, among other things, on their political views.
In the ensuing nine months, however, the teenager in question had largely
been out of the news, and the evocative power of his first name had probably
faded considerably. The alliterative phrase “holiday high jinks” supports this
image of having fun, while “land [someone] in more trouble” evokes images

capítulo 2 • 41
of an accident-prone schoolboy, and recalls stories of the “Just William” or
“Boys Own” type. The word “more” reminds the reader that the teenager in
question (Euan Blair) had been in trouble on previous occasions. The reader
is likely to know this is a result of having read previous news reports sufficient
to be familiar with the phrase “the patience of Job” without knowing the Old
Testament story, for example.
In the early 1970s, the following “knock-knock” joke was current among
Conservative supporters, after Harold Wilson’s Labour government had been
defeated in the general election of 1970: — Knock-knock, — Who’s there? —
Harold. — Harold Who? — Have you forgotten already? As Fowler puts it, “a
small reference, powerfully supported outside the text, economically provides
readers with a whole frame of values…” (Fowler 1991, p.118), “cueing in readers’
knowledge and attitudes [and] briefly inviting a ready-made point of view” (ibid,
p.228). This is how intertextual reference can be used to sustain ideologies.

2.1.4  Intertextuality in newspaper headlines

Newspaper headlines make extensive use of intertextual reference, as the

following examples taken from the Guardian Weekly (GW) show: Silence of the
Damned (GW March 1-7, 2001) (reference to the film “Silence of the Lambs” in
the title of an article on humanitarian crises in West Africa and Afghanistan.
The reference here is given additional potency by the fact that the follow-up to
the movie released to the accompaniment of major publicity). The Blame in
Spain (GW, Feb 22-28, 2001), reference to the song “The Rain in Spain”, from
the musical “My Fair Lady”, as the title to an editorial poking fun at frictions in
relations between Britain and Spain.
These two examples provide an interesting contrast, in which the intertextual
reference in the first case appears to be much more seriously motivated than in
the second. The film to which the title refers was particularly horrifying and
gruesome, and the newspaper article describes potential humanitarian crises
involving loss of life on an enormous scale. The editorial about disagreements
between Britain and Spain is written in a far more light-hearted tone, which is
in keeping with the musical to which its title is intertextually related.

42 • capítulo 2
Intertextuality and matching relations
These examples can also be seen in terms of matching relations (Coulthard,
1992; Winter, 1994), where the emphasis is on the single word that is different in
the matched clause: “damned” replacing “lambs”, in the first case, and “blame”
instead of “rain” in the second. It is notable that in both cases the replacement
word is phonologically very similar to the original. Winter (ibid, p.60) makes the
point that incongruous matching relations form the basis for much cartoon-
type humour, where “knowledge of the world”, obtained intertextually, makes
it possible to see the humour. A good example of this is a cartoon (GW March
8-14, 2001) referring to the pardon granted to Mark Rich by Bill Clinton at the
very end of his presidency, which led to allegations of a kickback in return for
donations to the Democratic Party. In the cartoon, Bill Clinton is depicted
holding up a bag full of money, and saying “I never… never had quid with this
pro quo”. This clearly matches his infamous untruthful statement denying
sexual relations with Monica Lewinsky, with “quid” replacing “sex” and “pro
quo” replacing “that woman” in his statement of denial. The fact that Bill
Clinton was at the time reported to have had an affair with Mark Rich’s wife,
adds further power to the intertextual mechanisms employed in this cartoon.

2.2  The translation problem and its strategies

Hatim (1997b, p.200) identifies the intertextual context of a text as “all the other
relevant prior texts which the various textual clues in a given utterance conjure
up for a given language user on a given occasion of use.” These prior texts
need to be “revisited” in order to fully retrieve the meanings associated with
the linguistic item in question. The translation problem, therefore, is basically
a question of the extent to which these intertextual references are accessible
to the targetlanguage readership, or the degree to which such “relevant prior
texts” are known to it. This is mediated by the type of equivalence required
in the translation concerned, which in turn depends on the purpose of the
translation. The translator has firstly to identify any intertextual references,
and then judge the likelihood that the target language readership will be able
to recognise them and cue in to the intended inferences. This tends to be more
difficult the more culturally specific the reference is, and the more distant the
cultures concerned are.

capítulo 2 • 43
Baker (1992, pp.71-77 and 228-243) discusses this in terms of implicatures
generally (i.e. implied meanings that are not explicitly written down in the text),
and offers the following strategies for dealing with them:
a) literal translation
b) cultural substitution
c) elaboration and explication supplied by the translator either within the
text or in a footnote
d) translation by omission

In addition to this last, one might add (e) transliteration by retaining

part(s) of the text in the source language. In a translation to Portuguese,
the first of these strategies would work well in the article title “Silence
of the Damned”: O Silêncio dos Condenados (the Portuguese title of the
movie was “O Silêncio dos Inocentes”). The single-word replacement in the
matching relation (“condenados” for “inocentes”) is phonologically and
morphologically quite similar to the original, and arguably the semantic link
between “inocentes” and “condenados” (evoking the idea of lambs to the
slaughter) is stronger than the link between “lambs” and “damned” in the
English version.
In this case, given that the movie was a hit worldwide, a literal translation
would achieve a very high degree of communicative equivalence and provide
target language readers with very much the same textual experience as
that achieved by the source text. The example of “Son of Star Wars”, which
also partly refers to a worldwide cinema box-office hit, would perhaps be a
candidate for the last strategy (e) mentioned in the list. Although the movie
itself has a perfectly acceptable title in Portuguese (Guerra nas Estrelas),
and the original Reagan missile defense system was popularly referred to in
this way by Portuguese language press, a literal translation “Filho da Guerra
nas Estrelas” is cumbersome. Given that the English title of the movie is
widely accessible, a case could be made for “Filho de Star Wars” as a more
satisfactory translation of “Son of Star Wars”. The blander alternative Star
Wars II basically corresponding to strategy (b) would be more predictable,
but would involve some loss of meaning.

44 • capítulo 2
2.3  A semiotic approach

Hatim and Mason (1990, pp.133-7) take a semiotic approach at strategies

for translating intertextuality, in which the translator’s task is to identify the
specific text elements that act as signs or pointers to intertextual reference
(intertextual signals), and then trace the ways in which these signals link
back to previous texts. Each intertextual sign needs to be evaluated in terms
of informational content, intentionality, and semiotic status (socio-cultural
signification), in order to decide which aspects of the sign can be omitted in
the translation. Hatim and Mason (ibid, p.136) suggest that priority should
always be given to intentionality over informational content, and offer a list of
procedural priorities for translation, which are exemplified below for the sign
“Euan” in the headline about Euan Blair.
1. Preserve semiotic status. This means conveying the socio-cultural
meaning of the word “Euan” as a typical teenager who happens to be the Prime
Minister’s son.
2. Preserve intentionality. This means conveying the sympathetic,
familiar, and affectionate attitude conveyed by first-name use.
3. Preserve linguistic devices that uphold coherence. This would involve
references in the text of the article itself.
4. Preserve informational status. In this case, this would involve retaining
the name “Euan”.
5. Preserve extra-linguistic status, i.e. the subgenre of newspaper

Assuming the translation is for a readership unlikely to be familiar with

the names of the British Prime Minister’s children, and still less their personal
lives, complying with such translation procedures would require some
explication, which might conflict with item 5 on the list. Hatim put this as the
lowest priority, however, so a headline that back-translates as follows would be
a possible outcome of their approach: Fun-loving Teenage Son of British Prime
Minister, Tony Blair, in Trouble Again. This fulfils criteria 1, 2, and 5. Coherence
relations (criterion 3) can be preserved with respect to the main body of the text,
and use of the name “Euan” (criterion 4) can also be made there. Although the
first name has disappeared from the headline, the implied affectionate and
sympathetic attitude is compensated for by “fun-loving”

capítulo 2 • 45
2.4  Translation equivalent9

An expression from a language which has the same meaning as, or can be used
in a similar context to one from another language, can be used to translate it: for
example, English: I don’t understand, French: Je ne comprends pas, Italian: Non
capisco, Modern Greek: Dhen katalaveno, Japanese: Wakarimasen. Achieving such
correspondences involves special bilingual skills to cope with the tendency among
languages to ‘lack of fit’ (technically,  non-isomorphism  or  anisomorphism).
Thus, the source-language expression may be a single word, a phrase, or a
sentence within a text, but its target-language equivalent may have to be rendered
at a different level: for example, the English idiom It’s pouring (with rain) cannot
be translated word-for-word into German, but the meaning can be redistributed
as Es regnet in Strömen (It rains in streams). Most bilingual speakers can supply
examples of such equivalents, and bilingual dictionaries codify them in bulk,
but it is the job especially of the translator and interpreter to decide whether a
particular expression is a fitting match for a particular passage.
A number of complex strategies are needed to find translation equivalents,
ranging from literal procedures, such as direct transfer, substitution, and loan
translation to devices of free translation, such as transposition, adaptation, and
circumlocution (which aim at finding the closest functional equivalent). The
literal approach can work well when language pairs have a similar structure: for
example English and German with mother/Mutter; Mother’s Day/Muttertag. The
free style, however, is demanded even in similar languages whenever anything
close to an idiom occurs: mother-country/Heimat (homeland), necessity is the
mother of invention/Not macht erfinderisch (need makes inventive).

2.4.1  Features of translation equivalence10

Translation equivalence is a major concept in Western translation theory. It is a

constitutive feature and the guiding principle of translation. As Catford points out,
“the central problem of translation-practice is that of finding Target Language
equivalents. A central task of translation theory is that of defining the nature
and conditions of translation equivalence.” (Catford 21: 1965) Actually, since

9 This part of the chapter was based on the website: http://www.encyclopedia.com/doc/1O29-

TRANSLATIONEQUIVALENT.html, which is recommended for further information.
10  This part of the chapter was based on the article: YINHUA, X. Equivalence in Translation: features and necessity.
International Journal of Humanities and Social Science. Vol. 1 no. 10, August, 2011.

46 • capítulo 2
the 1950s, many translation theorists have involved and elaborated translation
equivalence in their respective theories. However, the concept of translation
equivalence is sometimes distorted, and perhaps this is why some people deny
its validity and necessity. To argue for the necessity of translation equivalence,
we should first clarify its features. First of all, it is necessary to understand the
exact meaning of the word “equivalence” itself. According to Mary Snell-Hornby
(17: 1988), for the last 150 years, the word “equivalence” in English has been
used as a technical term in different kinds of exact sciences to refer to a number
of scientific phenomena or processes. For instance, in mathematics, it indicates
a relationship of absolute equality that involves guaranteed reversibility. At
the same time, however, it can also be used as a common word in the general
vocabulary of English, and in this sense, it means “of similar significance”. In
other words, the word “equivalence” is used in the English language both as
a scientific term and as a common word. As a central concept in translation
theory, “equivalence” cannot be interpreted in its scientific sense. It can only
be understood in its common sense as a general word. As J.R. Firth points out
in his writing on translation, it was in the common sense and as an item of the
general language that the word “equivalence” was originally used in English
translation theory (Snell-Hornby: 17). Philosophically speaking, nothing is
absolutely identical. Nida expresses this view as follows: There are no two stones
alike, no flowers the same, and no two people who are identical. Although the
structures of the DNA in the nucleus of their cells may be the same, such persons
nevertheless differ as the result of certain developmental factors. No two sounds
are ever exactly alike, and even the same person pronouncing the same words
will never utter it in an absolutely identical manner. (Nida 1986: 60)
As far as languages are concerned, there are no two absolute synonyms
within one language. Quite naturally, no two words in any two languages are
completely identical in meaning. As translation involves at least two languages,
and since each language has its own peculiarities in phonology, grammar,
vocabulary, ways of denoting experiences, and reflects different cultures, any
translation involves a certain degree of loss or distortion of meaning of the
source text. That is to say, it is impossible to establish absolute identity between
the source text and the target text. Therefore, we can say that equivalence in
translation should not be approached as a search for sameness, but only as
a kind of similarity or approximation, and this naturally indicates that it is
possible to establish equivalence between the source text and the target text
on different linguistic levels and on different degrees. In other words, different

capítulo 2 • 47
types of translation equivalence can be achieved between the source text
and the target text, such as phonetic equivalence, phonological equivalence,
morphological equivalence, lexical equivalence, syntactical equivalence, and
semantic equivalence. (Le Meiyun 1989)

2.4.2  Necessity of translation equivalence

Since translation is a kind of communication, the main task in translation-

practice is to establish equivalence of the original text in the target language. In
other words, any translation involves a kind of equivalence between the source
text and the target text. Without equivalence in certain degrees or certain
aspects, the translated text cannot be regarded as a translation of the original
text. In short, equivalence is of absolute necessity, and a basic requirement of
translation. This can be illustrated in the following aspects:

1. Necessity of equivalence as implied in definition of translation

Translation is so complex a kind of activity that defining it adequately is not
an easy job. So far, various kinds of definitions have been given, some of which
are quoted as follows:

(1) E. Tanke, the Director of the Translation Institute at Siemens, defines translation as
“the process of communication in which the translator is interposed between a transmitter
and a receiver who use different languages to carry out a code conversion between
them.” (Huang Long 1988: 18, and later he improves it to “transfer of a text from a
source language into a text in target language, the objective being a perfect equivalence
of meaning between the two texts.” (Huang Long: 18)
(2) Peter Newmark defines translation as “rendering the meaning of a text into another
language in the way that the author intended the text.” (Newmark 1988: 5)
(3) Nida defines translation as “reproducing in the receptor language the closest natural
equivalent of the source-language message, first in terms of meaning and secondly in
terms of style.” (Nida 1982: 12)
(4) The traditional definition: “the process of transfer of message expressed in a source
language into a message expressed in a target language, with maximization of the
equivalence of one or several levels of content of the message....” (Huang Long: 19)

48 • capítulo 2
As can be easily seen above, no matter how translation is defined, the
concept of equivalence is inseparable and implied in one way or the other. In
a sense, each of the above definitions is constructed around the basic concept
of equivalence, or as Marry Snell-Hornby points out: “definitions of translation
may be regarded as variations of the concept of equivalence”. (Snell-Hornby: 15)
The essentiality of the concept of equivalence in any definition of translation
adequately demonstrates the necessity of equivalence in translation.

2. Necessity of equivalence as required by essence of translation

Just like definitions of translation, there are also various opinions concerning
the nature of translation, such as “Translation is a science”; “Translation is an
art”; “Translation is a language activity” etc. However, translation, in essence, is
basically a kind of communication. In history, translation has always functioned
as a bridge for people who do not know foreign languages to understand the
source text. As a matter of fact, translators and translation theorists worldwide
have long realized the essence of translation as a kind of communication. Nida
has said time and again that translating means communication. Professor
Fan Zhongying has also expressed the same opinion, saying that “translation
is a language activity, the cardinal aim of which is to communicate”. (Fan
Zhongying 1994: 9)
Since translation in essence is a kind of communication, equivalence
between the source text and the target text naturally becomes an essential
requirement. It is generally agreed that the fundamental requirement of
any kind of communication is to guarantee that the message is adequately
transmitted from the source to the receptor. Similarly, in translation, the
translator should try his best to reproduce the closest equivalent message of the
original text in the target text so that the target text reader can understand the
source message adequately. Otherwise, translation as a kind of communication
would end up in failure. Therefore, it might be safe to say that the essence of
translation as a kind of communication calls for the necessity of equivalence
in translation.

3. Necessity of equivalence as demonstrated by limitations of

translatability and difficulty of translation
When we say that something is translatable, in a sense, it means that a
certain degree of equivalence of the source text can be achieved in the target

capítulo 2 • 49
language. Contrarily, when we say that something is untranslatable, it means
that no equivalence of the source text can be achieved in the target language.
In other words, the limitations of translatability are just caused by the necessity
of equivalence in translation. (Catford, 93) If translation were not to seek
equivalence, there would be no limitation of translatability, and any translated
text would be regarded as a correct version of the original text. Therefore, we
can say that the existence of limitations of translatability well demonstrates the
necessity of equivalence in translation. Likewise, the difficulty of translation
sometimes arises from the necessity of equivalence in translation.
It is generally agreed that translation is more difficult than original creation,
and this is mainly due to the requirement of equivalence in translation. In the
original creation, the author is free to say whatever he wants to say, however he
wants to say it. In translation, however, the translator does not have the same
freedom, because he has to say what the author said in the original text in a
way that is as close as possible to what the original author said. Liu Zhongde, a
Chinese professor, argues: “The difficulty in translation just lies in the fact that
both the content and the style are already existent in the original and as a result,
you will have to do your best to reproduce them as they are in quite a different
language.” (Liu Zhongde 1991: 7) The necessity of equivalence in translation
is also suggested in the famous remark made by Yan Fu when he exclaimed:
“it often takes as long as ten days or even a whole month to establish a term in
translation after repeated consideration and hesitation” (Liu Zhongde: 6)
As a matter of fact, it is equivalence that connects the source text and the
target text, and only after the realization of equivalence in some degree or in
some aspects can we say that the target text is the translation of the source
text. Without equivalence in some degree or in some aspects, nothing can be
regarded as a (successful) translation of a certain text. Let’s see the following

2.4.3  Different paths11

Equivalence has been a central notion in discussions about translation across

the ages, whether these discussions are theoretical or practical. In fact, it has
been so central, that translation itself is defined in terms of equivalence, for
11  This part of the chapter was based on the article: BAKER, M. "The Status of equivalence in translation studies:
an appraisal. In: Yang Zijian (ed.) English-Chinese Comparative Study and
Translation, Shanghai: Foreign Languages Education Press, 2004.

50 • capítulo 2
example, in Nida (1959:19), Catford (1965:20), and Wilss (1982:62). Not all early
theorists made the mistake of defining translation itself in terms of equivalence.
However, Jakobson was one of the few early theorists who avoided the issue of
equivalence altogether in his definition of translation (1959:233). This is later
picked up and asserted more clearly in Frawley (1984:160). In assessing the
centrality of the notion of equivalence in translation studies we might note the
•  The notion of equivalence is important because it is used to define
translation itself. This also makes it problematic because it is circular –
translation is defined in terms of equivalence and equivalence is, at the same
time, used for assessing and describing actual translation acts.
•  Equivalence is also central in the study of translation because it is closely
linked to other important theoretical notions in translation studies. In fact,
the assumption of its existence is a prerequisite for the discussion of most
theoretical notions in the discipline. For example, it is central to the notion
of fidelity/faithfulness to an original, which clearly presupposes not only the
possibility, but the desirability for equivalence.

Similarly, the notion of ‘shift’, an important tool of analysis in descriptive

studies, and an important notion in normative approaches, also relies on an
assumption of equivalence. Shifts are changes that occur or may occur in the
process of translating. The notion of shift presupposes the existence of what
is sometimes called an ‘invariant’ (not much different from ‘equivalent’). An
invariant is what remains unchanged. Invariants are elements which are not or
should not be affected by shifts in the process of translation. Whether they are
not or should not be affected depends on whether the approach being adopted
is descriptive or prescriptive. Definitions of equivalence can similarly be
either normative (postulating a specific relation to be achieved) or descriptive
(discovering a relation of equivalence/correspondence between source and
target elements).
•  The idea of an unit of translation – again, the subject of much debate
in translation studies – similarly rests on an assumption of equivalence.
Discussions about an unit of translation centre on what units are to be
considered equivalent (words, clauses, etc.) or what units translators work with
in real life in order to produce an ‘equivalent’ version of the source text.

capítulo 2 • 51
Given that the notion of equivalence has been so central in translation
studies (it is both used to define translation itself, and is taken as a given in
attempts to elaborate other theoretical notions), it is somewhat worrying
to find it discredited in so much of the recent literature on translation. It is
worrying because discarding it involves discarding a whole set of notions that
go with it, like shifts, fidelity, and so on. This would necessarily involve a radical
reassessment of the nature and goals of translation studies as a discipline,
as well as the nature of translation activity itself. There is also the question
of whether we have or need other notions to replace ‘equivalence’ in order to
underpin theoretical work in the discipline.
What is wrong with the notion of equivalence? Let us first look at why the
notion of equivalence has been discredited before we move on to discussing
whether there is anything to be salvaged in this concept. Equivalence has
traditionally been treated as a semantic category. By this I mean that translation
scholars have traditionally stressed equivalence of meaning; of semantic
content. For example, Rabin defines translation as “a process by which a
spoken or written utterance takes place in one language which is intended and
presumed to convey the same meaning as a previously existing utterance in
another language. It thus involves two distinct factors, a ‘meaning’, or reference
to some slice of reality, and the difference between two languages in referring
to that reality” (1958:123; my emphasis).
In ‘same meaning’, read ‘equivalent meaning’. The notion of equivalence
here is similar to that of synonymy, except that one applies to items in two
different languages and the other applies to items in the same language. As
a semantic category, the notion of equivalence is static – it is not dictated by
the requirements of the communicative situation, but purely by the content of
the source text. The semantic view of equivalence draws on a representational
theory of meaning, as can be seen in Rabin’s definition: The idea that reality is
unproblematic, it exists out there, and that representation of reality (whether
in language or any other form) is not only possible, but can provide direct,
unmediated access to this transparent reality (Niranjana 1992:2). The function
of language is to directly represent this reality– this is how meaning is generated.
The function of a translation is to represent the same reality that is represented
in the source text. There is no question of mediation here.
Given the fact that the representational theory of meaning has now been
rejected in most disciplines (very few people still believe that words and texts

52 • capítulo 2
represent reality as such), the treatment of equivalence as a semantic category
soon came to be regarded as untenable in translation studies. One of the first
alternatives to be offered was a definition of equivalence not as a question of
‘how close’ a target text is to the reality portrayed in the source text, but rather
as how close it comes to reproducing the same effect or response in the target
readers that the source produced in the source readers.
This approach originated with Bible translators: Nida (1958, 1964; Nida
& Taber (1969); Larson (1984), Beekman & Callow (1974). The notion of
‘equivalent effect’ brought with it other dynamic notions, such as ‘receptor’ as
opposed to ‘target’ language, and dynamic equivalence as opposed to formal
equivalence. This emphasis on the dynamics of translation (as opposed to
static features of text) became more popular because it allowed us to bring in
the human element, albeit in the form of the reader rather than the translator
or the commissioner, for instance. Other definitions of equivalence stress
other participants in the translation process, for example the client or
commissioner. The idea of equivalent effect, though more attractive than
the semantic alternative, helped undermine the notion of equivalence even
further. It was soon pointed out by various people that there is no reliable
way of measuring effect in readers – not only is it impossible to know how
two people are going to respond to a given text, but it is also true that even the
same reader will respond differently to the same text on different occasions.
‘Equivalent effect’ is simply a shorthand way of saying that the translator has
to imagine how a reader or group of readers might respond to both the ST
and the TT – subjectively. Alternatively, we can limit the notion of equivalent
effect to ‘similarity’ in a very global and limited sense. For instance, Hervey
& Higgins (1992:23) suggest that the translator of a portion of a ST which
makes the source reader laugh can attempt to produce a TT which makes the
target reader laugh. As they rightly explain, this is “a gross reduction of the
effects of a text to a single effect”.
The notion of equivalent effect is also linked to the idea of reproducing
the ‘intention’ of the source author, i.e. emphasizing the equivalence of
intended meaning. This is also highly problematic because it assumes that
the translator ‘understands’ rather than ‘interprets’ the source text – that
somehow he or she has direct access to the communicative intentions of
the original author. But translators cannot know with any certainty what the
source author intended to convey, especially when there is a large temporal

capítulo 2 • 53
gap between the source and target texts. All they can do is try to interpret
it, which makes any theory or model based on some notion of equivalence
of intention impossible to verify. Another alternative which gained much
ground in the 1970s and 1980s was speaking of equivalence of functions,
rather than effect or intention.
Some scholars try to define the possible functions or purposes of
communication in order to suggest ways in which “equivalence of message”
may be achieved in relation to the function which is most in focus. Roberts
(1985), for instance, suggests that there are three main functions:
•  Expressive (primary focus on source);
•  Informative (primary focus on object: subject matter);
•  Imperative (primary focus on intended receptor).

If the function is imperative, the translator must make sure that the target
reader reacts to the message in the same way as the source reader (as in the
case of advertisements). Irrespective of changes in the translated version,
equivalence is achieved if the target reader reacts to the message in the way
intended by the source writer. Apart from the obvious problems of defining
a single function for a text (as well as the earlier problem of identifying the
intention of the source writer and the effect on the target reader), this approach
has rightly been criticized and divorced from the realities of translation in that
it assumes that the function of the target text is determined by (and therefore
has to be equivalent to) that of the source text. This is not at all the case in many
situations. If, as it often happens, a client gives a translator an advertisement and
asks him or her to produce a rough translation of it for informative purposes,
it would be perverse of the translator to insist on producing a target text which
can function as an advertisement in its own right. In response to this challenge,
new approaches emerged in the 1980s, particularly in Germany, which
pointed out that the reasons for commissioning or initiating a translation are
independent of the reasons for creating the source text. What matters therefore
is the function of the translated text, not that of the source text. Equivalence
here, it is suggested, becomes a function of what is sometimes called the
skopos or commission accompanying a request for translation. Sager (1994)
similarly suggests that equivalence is a function of the specification that comes
with a translation. Scholars like Vermeer (1989) therefore talk of ‘adequacy’

54 • capítulo 2
with regards to the skopos, rather than equivalence, as the standard for judging
translations. Nord (1991) takes this even further by suggesting that it is not the
text itself that has a function – rather a text acquires its function in the situation
in which it is received.
There are other reasons why the notion of equivalence is no longer as
appealing as it used to be – reasons not connected to the lack of rigour in
defining it, nor to whether or not it is valid in its own right. One of these reasons,
for instance, is that the notion of equivalence suggests a denial of originality
in translation (Pym 1992:39, footnote). If equivalence (in whatever form –
semantic, functional, etc.) is to be aspired to, then what we are really saying
is that we do not expect translations to be original or creative: we expect them
to adhere as closely as possible to some aspect of the source text, as specified
by a participant other than the translator. This means that translating is not
a creative, original process, and therefore, the translator cannot be treated as
an author. This assumption underlies the legal attitude towards translators,
the fact that copyright law in most countries places strict limitations on the
translator’s control of the translated text, and that authors continue to enjoy an
exclusive right to translations of their works (Venuti 1995).
Translators are typically paid a flat fee to translate a literary text, but no
percentage of the royalties or subsidiary rights sales. Similarly, according
to the allotment of loan rights under the Public Lending Right, the author
receives 70% and the translator, 30% of the royalties accrued each time a book
is borrowed from a public library. Finally, equivalence, however defined, means
‘sameness’. Irrespective of the types of things which may form the two poles
of the equivalence relation (meaning, effect, function, etc.), the assumption is
that if they are equivalent, their relationship is one of sameness or identity. Any
kind of theoretical notion in the humanities that starts from an assumption
of identity or sameness (whether this assumption is descriptive or prescriptive
in nature) is unlikely to be popular in the present intellectual climate. The
emphasis on various branches of the humanities is increasingly on highlighting
difference rather than sameness or similarity, especially when people and
cultures are concerned. Niranjana (1992) – like many other scholars – tends to
regard the impulse to see ‘sameness’ or homogeneity as a colonial enterprise
and to favour approaches which “emphasize the need to reinvent oppositional

capítulo 2 • 55
cultures in nonessentializing ways” (ibid:46). She suggests that translation
studies ignore the fact that languages are not ‘equal’ and that translation is a
tool of colonialism (ibid.:58).

The notion of equivalence assumes a fairly close relationship between
a specific original source text and a translated version of it. The weaker the
relationship between the two, the more elusive the notion of equivalence
becomes. It is, and probably will always be, more precise as a theoretical construct
when it is treated as a semantic category (as equivalence of form or meaning).
Moving away from a semantic view of equivalence by necessity renders the
notion of equivalence itself less useful, although it opens up the door to other
notions that are likely to be more useful. Equivalence as a semantic category may
offer rigour, but it does not often relate to what goes on in real life. What goes on
in real life includes forms of translation which do not lend themselves easily to
being described or assessed in terms of notions of equivalence. I am referring,
for example, to phenomena such as adaptations and pseudo-translations. These
suggest that we have to look for other notions to supplement (if not necessarily
replace) equivalence. And we might have to focus on relationships other than
those that may exist between source and target texts.
There is also another problem which is rarely addressed in discussions
about equivalence: the fact that many source texts are badly written, and that
the professional tasks of the translators become extensively stretched to include
such things as copywriting, linguistic consultation, writing summaries, editing,
etc. Equivalence to a badly written source text is the last thing any client wants.
One thing that has come out of the debate so far is that, rather than seeing
equivalence as an a-priori relationship which exists between discrete, static
systems or texts, we should perhaps see it as a textual (not systemic) relationship
which emerges from situations in contact and is shaped by a variety of dynamic
factors, including the translator’s interpretation of the source text, the
requirements of the commissioner, the context of translation, and of course the
translator’s own ideological make-up. Pym (1992:43) suggests that the notion of
equivalence is similar to (or should be treated like) that of value in economics:
a coat may be equivalent to 20 yards of linen this week and 15 yards next week.
It would seem then that we still cannot throw ‘equivalence’ out of the
window. Even the idea of producing a target text that addresses a specific reader

56 • capítulo 2
– rather than one that is faithful to the original – still implies transferring some
part of the source text that is considered to be of value in the particular exchange
situation. The concept of equivalence is thus likely to be with us for a long time
to come.

01. Translate these oral extracts from Portuguese to English:
a) Bom, numa sala de aula basicamente não tem muita coisa, né? Tem geralmente um
quadro, que é onde os professores escrevem e dão as suas aulas, apagador, giz ou
caneta, que alguns quadros são brancos e têm caneta, tem as carteiras também dos
alunos. Eu como estudei num colégio militar tinha uma bandeira, tinham várias coisas
sobre militarismo, essas coisas na minha sala de aula, mas geralmente não tem. Na
faculdade só tem isso mesmo: carteira, livros às vezes algumas faculdades têm, um lugar
pra botar os livrinhos, às vezes uma televisão, um vídeo pra mostrar alguma coisa, um
retroprojetor, mas fora isso não tem muita coisa numa sala de aula, não.
b) Bom, essa cor aqui, várias pessoas falam de vários jeitos: abóbora, laranja; eu falo que é
laranja. Aqui, preto, que é uma cor básica. Esse aqui é rosa ou pink, que na verdade as
pessoas chamam de pink porque é um rosa mais forte. Esse aqui é branco, né? Cor que
todo mundo usa... Azul, aqui dá pra ver, né? O azul. Aqui um rosa mais clarinho, e acho
que basicamente ainda tem o vermelho que dá pra ver, não sei se dá pra ver mais aqui,
bem pequenininho o vermelho, que também é uma cor importante. Aqui a cor do cabelo
castanho também, mais pro preto, pro castanho, e aqui marrom, também é uma cor.
c) Bom, em português nós chamamos as quatro operações básicas da matemática de
adição, subtração, multiplicação e divisão. Bom, o verbo para adição seria somar, “eu
posso somar coisas” ou “adicionar coisas” – são conceitos universais, e eu acho que o
uso em português e no Brasil seria mais ou menos como nos outros lugares, nos outros
países do mundo, pelo menos no mundo ocidental. Com relação à subtração, nós temos
o verbo subtrair ou diminuir; retirar também estaria no mesmo campo semântico da ideia
de subtração. Divisão, nós temos o verbo dividir, sei lá, outros verbos do mesmo campo
semântico poderiam ser espalhar ou... Enfim, seria essa a ideia básica de dividir; o verbo
básico seria dividir. Para multiplicação, nós temos multiplicar... Acho que também seria
esse o verbo básico pra ideia de multiplicação em português. Por exemplo, com relação à
adição, o sinal gráfico... Nós temos a palavra “mais” nas operações matemáticas. Poderia
dizer “três mais três é igual a seis”, né? Com relação à subtração, nós temos o sinal de
«menos», então eu digo «cinco menos dois é igual a três.» São coisas que são faladas no

capítulo 2 • 57
dia-a-dia, a todo momento, pelos falantes do português no Brasil. A multiplicação: o sinal
do “xis” seria... A gente diz oralmente “vezes”, então “dois vezes dois igual a quatro.” E a
divisão, nós temos “dividido por”, ou seja, “nove dividido por três igual a três” ou “nove por
três igual a três.” Seriam exemplos básicos.
d) Eh, meu pai era um contador e, eh, ele trabalhou em, como, fazendo a contabilidade de
um supermercado, eh, durante muitos anos. Depois, agora que ele tá mais velho, ele
começou a fazer, eh, bijuteria. Ele resolveu fazer bijuteria, e é muito engraçado porque
ele é o tipo do macho, assim né, e resolveu fazer bijuteria e tá fazendo bem. E a minha
mãe era professora de português, depois ela fez outro curso e começou a fazer, eh,
orientação estudantil, então ela gostava muito de trabalhar com educação e gosta... Eh,
depois, agora ela tá aposentada, ela escreveu um livro de educação e, pra fazer uma...
Assim, eh, pra ajudar os alunos a interagirem com os professores e com os pais.
e) A minha mãe é uma mulher muito alta, tem o cabelo cacheado, escuro, castanho, meio
avermelhado, é mais ou menos magra, assim, uh, tamanho normal. A minha irmã mais
nova também é muito alta, mais alta do que eu. Ela já é um pouco mais gordinha, mas
tem o cabelo louro, sardas e uma pele bem clara. A minha irmã mais velha já é um
pouco mais baixa, mais escura, tem o rosto bem redondo, mas um sorriso muito bonito.
Ela também já é um pouco mais cheinha. Eu sou alta, tenho o cabelo naturalmente
cacheado, muito enrolado, mas hoje ele está liso, olhos castanhos. Ninguém na minha
família tem olhos claros, olhos azuis, mas eu também tenho uma pele mais escura. O
meu pai também é muito alto, tem o cabelo super encaracolado, super cacheado, cabelo
curto, escuro, olhos castanhos também e uma pele bem escura. Os meus avós já são um
pouco diferentes, minha avó já é mais branca, mais europeia, tem o cabelo mais louro,
uh, castanho-claro e uma pele também super clara.

Baker, M. 1992. In Other Words: A Coursebook on Translation, London: Routledge.
Beaugrande, R. de and Dressler, W. 1981. Introduction to Text Linguistics, London: Longman.
Beekman, John and John Callow ((1974/1986) Translating the Word of God, Dallas: Summer
Institute of Linguistics.
Catford, J. C. (1965) A Linguistic Theory of Translation. An Essay in Applied Linguistics, London:
Oxford University Press.
Catford, J. C.. A Linguistic Theory of Translation. London: Oxford University Press, 1965.
Coulthard, M. 1992. “On the importance of matching relations in the analysis and translation of
literary texts” Ilha do Desterro 27, Florianoplis, Brazil: Federal University of Santa Catarina, pp.33-34.

58 • capítulo 2
Fairclough, N. 1992. Discourse and Social Change, Cambridge: Polity Press. MA Translation Studies
– Translation and Discourse - Question 5 Tim Ennis 12 Fowler, R. 1991. Language in the News,
London: Routledge.
Frawley, William (1984) ‘Prolegomenon to a Theory of Translation’, in William Frawley (ed)
Translation: Literary, Linguistic, and Philosophical Perspectives, London & Toronto: Associated
University Presses.
Gutt, E-A. 1991. Translation and Relevance, Oxford: Blackwell.
Hatim, B. 1997a. “Intertextual intrusions: towards a framework for harnessing the power of
the absent text in translation” in Simms, K. (ed.) Translating Sensitive texts: Linguistic Aspects,
Amsterdam, Rodopi.
Hatim, B. 1997b. Communication Across Cultures: Translation Theory and Contrastive Text
Linguistics, Exeter: University of Exeter Press.
Hatim, B. and Mason, I. 1990. Discourse and the Translator, London: Longman.
Hatim, B. and Mason, I. 1997. The Translator as Communicator, London: Routledge.
Hervey, Sándor and Higgins, Ian (1992) Thinking Translation: A Course in Translation Method: French
to English, London & New York: Routledge.
Hoey, M. 1991. Patterns of Lexis in Text, Oxford: Oxford University Press.
Jakobson (1959) ‘On Linguistic Aspects of Translation’, in R. Brower (ed) On Translation,
Cambridge M.A.: Harvard University Press.
Larson, Mildred (1984) Meaning-based Translation: A Guide to Cross-language Equivalence,
Lanham, New York & London: University Press of America.
Liu Zhongde. Ten Lectures on Literary Translation. Beijing: China Translation & Publishing
Corporation, 1991.
McCarthy, M. and Carter, R. 1994. Language as Discourse, London: Longman.
Moon, R. 1994. ‘The analysis of fixed expressions in text’ in Coulthard, M. (ed.) Advances in Written
Text Analysis, London: Routledge, pp.117-135.
Newmark, Peter. A Text Book of Translation. Hemel Hempstead: Prentice Hall International Ltd.,
Nida, Eugene A. (1964) Toward a Science of Translating, with Special Reference to Principles
and Procedures Involved in Bible Translating, Leiden: E. J. Brill.
Nida, Eugene A. and Charles R. Taber. The Theory and Practice of Translation. Leiden: E. J. Brill,
Nida, Eugene A. and Charles Taber (1969) The Theory and Practice of Translation, Leiden: E. J.
Nida, Eugene A. and Jan de Waard. From One Language to Another: Functional Equivalence in Bible
Translating. Nashville: Thomas Nelson Publishers, 1986.

capítulo 2 • 59
Nida, Eugene, A. (1959/1975) ‘Principles of Translating as Exemplified by Bible Translating’, in
Anwar S. Dil (ed) Language Structure and Thought. Essays by Eugene A. Nida, Stanford: Stanford
University Press.
Niranjana, Tesjaswini (1992) Siting Translation: History, Post-structuralism, and the Colonial Context,
Berkeley, Los Angeles & Oxford: University of California Press.
Nord, Christiane (1991) Text Analysis in Translation, Amsterdam & Atlanta: Rodopi.
Pym, Anthony (1992) Translation and Text Transfer, Frankfurt and Main: Peter Lang. Rabin, C.
(1958) ‘The Linguistics of Translation’, in Aspects of Translation (# Studies in Communication 2, The
Communication Research Centre, University College London), London: Secker and Warburg.
Roberts, Roda (1985) ‘Translation and Communication’, NUCLEO 1, Caracas, 139-76.
Sager, Juan C. (1994) Language Engineering and Translation – Consequences of Automation,
Amsterdam: Philadelphia.
Snell-Hornby, Mary. Translation Studies: An Integrated Approach. Philadelphia: John Benjamin’s
Publishing Company, 1988.
Venuti, Lawrence (1995) ‘Translation, Authorship, Copyright’, The Translator 1(1): 1- 7 25.
Vermeer, Hans (‘Skopos and commission in translational action’, trans. Andrew Chesterman, in Andrew
Chesterman (ed.) Readings in Translation Theory, Helsinki: Oy Finn Lectura Ab, 173-87.
Wilss, Wolfram (1982) The Science of Translation: Problems and Methods, Tübingen: Gunter Narr.
Winograd, T. and Flores, F. 1986. Understanding Computers and Cognition, Reading MA: Addison-
Winter, E. 1994 “Clause relations as information structure: two basic text structures in English”, in
Coulthard, R.M. (ed) Advances in Written Text Analysis, London: Routledge, Ch. 4, pp.46-68.

60 • capítulo 2
and the dynamic
3.  Formal correspondence and the dynamic

E.A. Nida establishes four priorities as guiding principles for translating and
bases for judgment: Contextual consistency; dynamic equivalence over formal
correspondence; the aural (heard) form of language over the written form;
forms that are used by and acceptable to the audience for which a translation
is intended over forms that may be traditionally more prestigious (Nida and
Taber, 1969: 14). Among them, dynamic equivalence is of primal concern, and it
is the core or essence of Nida’s entire theoretical system. Besides, the question
of equivalence, as agreed by many translation theorists home and abroad, is a
decisive factor in disclosing the nature of translation, and a criterion by which
the quality of the translation is judged. Basically, the diversity of translation
theories is attributed to the varied views on equivalence.

3.0.1  Nida’s translation theory of dynamic equivalence12

Nida’s dynamic equivalence theory studies translation from a totally new

perspective, deviating from the traditional source text-centered theories,
shaking off the straitjacket of sticking to some specific linguistic problems,
and shifting the focus to the function of translation: Making sure the receptor
understands accurately the message carried by the source text. In this sense, it
is a big step forward in translation studies. Nida bases his dynamic equivalence
theory on some linguistic achievements made by Jakobson and Chomsky, who
claims that a dynamic dimension can be added to language structure through
the use of transformation. Nida thus categorizes the kernel sentences of a
language into seven types. In other words, the surface structure of any language
is but the logic organization of those kernel sentences, which justifies the
possibility of dynamic equivalence between different ways of expression within
one language (戴灿宇1987: 61). He therefore concludes that all languages have
the same capability of expression by stating: “Anything that can be said in one
language can certainly be said in another language…”, with reasonable accuracy
by establishing equivalent points of reference in the receptor’s culture and

12  This part of the chapter was based on the article: DAYAN. L. Dynamic Equivalence and Formal Correspondence
in Translation between Chinese and English. International Journal of Humanities and Social Science Vol. 2 No. 12
[Special Issue - June 2012]

62 • capítulo 3
matching his cognitive framework by restructuring the constitutive elements
of the message (Nida, 1984: 13).
A dynamic equivalence, as defined by Nida, is to reproduce “in the
receptor language the closest natural equivalence of the source-language
message…”(Nida and Taber, 1969: 12). The key words are “closest”, “natural”
and “equivalence”. By “closest”, he indicates that, owing to the impossibility
of absolute equivalence, the “closest” equivalence is the most ideal one. Nida
(1964: 167) particularly stresses that “a natural rendering must fit the receptor
language and culture as a whole; the context of the particular message; and
the receptor-language audience”. To put it plain, either the meaning or the
form should not sound “foreign”. The essence of dynamic equivalence is the
receptor’s response, in Nida’s own terms, “the degree to which the receptors
of the message in the receptor language respond to it in substantially the same
manner as the receptors in the source language” (Nida and Taber, 1969: 68). The
reaction or response is based on the comprehensive reception of the message,
not only understanding the meaning or content, but also feeling the way the
original readers do. By laying stress on the receptor’s response, he underlines
the improvement to the source text by the receptor’s subjectivity and aesthetic

3.0.2  Formal correspondence

Nida puts forward dynamic equivalence in opposition to formal correspondence.

In speaking of naturalness, he is strongly against translationese - formal
fidelity, with resulting unfaithfulness to the content and impact of the
message. Basically, a formal equivalence translation, as Nida (1964, 165) states,
is source-oriented, which is designated to reveal as much as possible the form
and content of the original message, that is, to match as closely as possible the
formal elements, like grammatical units, consistency in word usage, meanings
in terms of the source context, just to name some. David Crystal, J.R.Firth,
Catford, and other linguists and translation theorists agree upon the six levels
of formal equivalence, namely: Phonetic, phonological, morphological, lexical,
syntactical and semantic equivalence (乐眉云, 1989: 38). A formal equivalence
translation, strictly speaking, is impossible because of the differences between
linguistic structures and socio-cultures. As Saussure points out, there is no
essential link between the signified and the signifier, and such arbitrariness

capítulo 3 • 63
determines that languages, particularly those of different language families,
differ greatly in form.
In many instances, certain formal elements of the source language, as
mentioned by Nida, cannot be reproduced, like puns, chiasmic orders of words,
instances of assonance, or acrostic features of line-initial sounds. Crystal, too,
agrees that it is impossible to achieve equivalence on all formal levels - the absolute
formal equivalence - and, on usual occasions, semantic equivalence should be given
priority, while other levels, especially phonetic, lexical, morphological, syntactical
equivalence, etc., are given attention only to achieve a special translation effect.

Formal correspondence and translation equivalence

The two concepts featured in the title belong to two different, though (as
will be shown) by no means unrelated, activities. Formal correspondence is
a term used in contrastive analysis, while translation equivalence belongs to
the metalanguage of translation. In principle, perhaps, the two terms could be
discussed separately in their two disciplines. It is indeed possible to imagine
a theory of translation which would operate with the concept of equivalence
defined without reference to formal correspondence, just as it is possible to
imagine contrastive analysis which would rely on the concept of correspondence
established without the use of translation. In practice, however, both terms have
been found necessary by students of translation and by contrastive analysis.
Issues that are raised in connection with formal correspondence and
translation equivalence are certainly more than just terminological: A discussion
of formal correspondence in translation concerns the role of linguistic units
in translation and the place of linguistics in translation theory, whereas a
discussion of translation equivalence in contrastive analysis concerns the role
of translation in contrastive work. The relationship between them has been
discussed by Catford (1965) from the point of view of translation theory, and
by Marton (1968), Ivir (1969, 1970), Krzesowski (1971, 1972), and Raabe (1972)
from the point of view of contrastive analysis.
Our understanding of the concept of translation equivalence will depend
on the view we take of translation itself. Looking at translation as a result or
product, faced with two texts, one of which being a translation of the other, we
might be tempted to conclude that translation is “the replacement of textual
material in one language (SL) by equivalent textual material in another language
(TI.)”, or more generally, that it is “the rendition of a text from one language to

64 • capítulo 3
another” (Bolinger, 1966: 130). Equivalence would then exist between texts, i.e.,
it would hold together chunks of textual material or linguistic units (texts being
simple linguistic units of a higher order than the smaller units which compose
them). This is a static view both of translation and of equivalence: Pushed to its
extreme, it forces in the conclusion that, for any linguistic unit (text of portion of
a text) in the source language, there is an equivalent unit in the target language,
and it is the translator’s job to find that unit. Hence the search for different
textual types and their characteristics in different languages.
Another picture of translation and translation equivalence is obtained
when a dynamic view is taken, and translation is regarded as a process rather
than a result. One then speaks about substituting messages in one language for
messages in some other language (Jakobson, 1959: 235); about “reproducing
in the receptor language the closest natural equivalent of the message of the
source language” (Nida, 1969: 495); or about “the nature of dynamic equivalence
in translating”. (Nida, 1977).
This latter view of translation is the communicative view, and it sees translation
equivalence not as a static relationship between pairs of texts in different
languages, but rather as a product of the dynamic process of communication
between the sender of the original message and the ultimate receivers of the
translated message via the translator, who is the receiver of the original message
and the sender of the translated message. Messages are configurations of
extra-linguistic features communicated in the given situation. The original
sender starts from these features, and relying on the resources of his language,
on his command of that language, and on his assessment of the nature of the
sociolinguistic relationship between him and his (actual or potential) receivers,
codes them to produce the source text. The coded message (source text) reaches
the translator through the (spatiotemporal) channel of communication. He
decodes it and receives the original sender’s message, which he then proceeds
to code again in the target language, relying on the resources of that language,
on his command of that language, and on his assessment of his relation to the
ultimate receivers.
Under this view, what is held constant (i.e., equivalent) are not texts but
rather messages, and it is messages that the participants return to at every
step in the process of communication. The translator, in particular, does not
proceed directly from the source text to the target text. Rather, he goes from
the source text back to that configuration of extra-linguistic features with which

capítulo 3 • 65
the original sender tried to communicate his message, and when arriving there,
he codes that message again, in a new and different communicative situation,
producing a text in the target language for the benefit of the ultimate receivers.
Several points must be made in connection with the view of translation and
equivalence presented here. First, the nature of the translator’s job in receiving
the original sender’s message does not essentially differ from the job of other
source-language receivers of that message, and his job in recoding the received
message in the target language is not unlike the task performed by the original
sender (only the communicative situation is different, that is, the translator
is a different “linguistic person” than the original sender; he uses a different
language and codes the message for different receivers than those of the
original sender).
Secondly, messages are not communicated absolutely. The original
message undergoes modifications in the process of coding (depending on the
potential of the language, on the sender’s command of that language, and on
the intended audience), in the process of transmission (owing to the “noise
line the channel”), and in the process of decoding (depending on the receiver’s
command of the language and his ability, coming from the shared experiential
background, to grasp the sender’s message).
Clearly, such modifications also take place when the translator receives
the message, when he recodes it in the target language, when he transmits
the coded message through the channel of communication which links
him to his receivers, and when the ultimate receivers decode the translated
message. This relativity of communication – any communication, not just that
involving translation – places the concept of equivalence in translation on a
new perspective: equivalence holds between messages (communicated by the
original sender, received and translated by the translator, and received by the
ultimate receivers) which change as little as possible and as much as necessary
to ensure communication. Thus, true translation is by no means limited to
communicative situations involving two languages. An act of translation
takes place every time a text is produced as a coded expression of a particular
configuration of extra-linguistic features, and is decoded to enable the receiver
to receive the message (cf. Steiner, 1975: 47).
The third point that can be made about translation equivalence follows
from what has just been said: Equivalence is a matter of relational dynamics
in a communicative act – it is realized in that act and has no separate existence
outside it. It can thus be compared to abstract units of the linguistic system,

66 • capítulo 3
such as phonemes, which do not exist physically outside the speech act in which
they are realized, and whose realization in speech is somewhat different, and
yet produced and received as the “same” phoneme. Or, it could be compared
to a person’s signature. There is no “ideal” signature of a given person, and in
each act of signing, it comes out a little bit different visually; yet, it is recognized
as “equivalence” along with any other of its realizations, allowing for the fact
that different realizations take place in different communicative situations.
Since translation equivalence is the translator’s aim, and since it is
established at the level of messages, in the communicative act, and not at the
level of linguistic units, it may appear that there is no need for the concept
of format correspondence in the model of translation presented here. I will
argue further below that this is not so, and that there is a sense in which formal
correspondence holds together the source and target texts. But in order to
demonstrate this, a modification of some of the available definitions of formal
correspondence will be needed.
Catford has defined formal correspondence as identity of function of
correspondent items in two linguistic systems: To him, a formal correspondent
is “any TL /target language/ category which may be said to occupy, as nearly
as possible, the ‘same’ place in the economy of the TL as the given SL/source
language/ category occupies in the SL” (Catford, 1965: 32). Marton (1968)
and Krzeszowski (1971, 1972) postulated an ever closer relationship between
linguistic expressions in the source and target languages – one of congruence,
which is characterized by the presence, in both languages, of the same number
of equivalent formatives arranged in the same order. Realizing that by relying
on a concept defined in this way, would prevent the contrastive analyst
from working with real language (and thus make his results useless for any
conceivable pedagogic purposes). Krzeszowski later (1972: 80) went back to
the concept of equivalence. However, he applied it to sentences possessing
identical deep structures (i.e., semantic representations of meaning), rather
than those which were translations of each other. At the level of deep structures,
equivalent sentences were also regarded as congruent, with their congruence
disappearing in later derivational stages leading to the surface structure.
Both Catford’s “formal correspondence” and Marton-Krzeszowski’s
“congruence/equivalence” represent attempts to bring linguistic units of
the source and target languages into some kind of relationship for purposes
of contrasting, with the necessary tertium comparationis being provided by
the identity of function or meaning. Without a tertium comparationis, no

capítulo 3 • 67
comparison or contrasting of linguistic units is possible. But the question
is what can serve as the tertium comparationis. One possibility would be
an independently described semantic system whose categories would be
held constant, while their linguistic expressions in pairs of languages under
examination would be contrasted. oweverHowever, such a system has not yet
been proposed, and we do not know what its categories might be.
Another possibility might be a common metalanguage in terms of which both
the source and the target language would be described to the same degree of
exhaustiveness. This metalanguage would supply categories in terms of which the
appropriate parts of the two systems could be contrasted. Since the descriptions
would be matchable, their contrasting would consist in simply mapping one
description upon the other to establish the degree of fit. Again, there are no two
languages whose descriptions meet this requirement. Formal correspondence,
as defined by Catford, can hardly be said to exist: Even in pairs of closely related
languages, it is practically impossible to find categories which would perform the
“same” functions in their respective systems, and this probability decreases with
typological and genetic distance. Marton-Krzeszowski’s concepts of “congruence/
equivalence” in fact make use of the metalanguage of the transformational-
generative grammar, in particular of the notion of deep structure, to avoid relying
on the postulate of translational equivalence. But the postulate of deep structure
and transformation are no easier to work with: The status of deep structures is far
from clear, as is also the meaning-preserving nature of transformations.
Therefore, one falls back on the concept of translation equivalence in their
search for a suitable tertium comparationis for contrastive purposes. (One feels
all the more justified in doing this when observing actual contrastive practice:
No matter what they otherwise profess, contrastive analysts begin with sentences
which are obviously translational pairs, and proceed to demonstrate the bilingual
person’s – which is the analyst – intuition of their equivalence.) However, we
must remember that translation equivalence holds together communicated
messages, and not linguistic units used to communicate them, and that we
must go beyond equivalence to find the necessary tertium comparationis which
holds linguistic units together. It has been suggested (Ivir, 1969: 18) that a good
candidate for the job would be formal correspondence – but not as defined
with reference to linguistic systems (as Catford would have it), but rather with
reference to translationally equivalent texts. Formal correspondents – to modify
Catford’s definition given above – would be all those isolable elements of

68 • capítulo 3
linguistic form which occupy identical positions (i.e., serve as formal carriers of
identical units of meaning) in their respective (translationally equivalent) texts.
The difference between language-based and text-based (or system-based
and equivalence-based) formal correspondence is seen in the fact that, while
the former type of correspondents stand in a one-to-one relationship, the
relationship in the latter type is one-to-many. Typically, a given formal element
of the source language, when used in different texts produced in different
communicative situations, will have several target-language formal elements
which will correspond to it in translated target texts. But one should be aware
that, precisely for that reason, the formal elements which are correspondent
in translationally equivalent texts are never matched in totality, as they would
be if parts of the systems of the two languages were contrasted. Rather, they
are matched in those of their meanings with which they participate in the
particular source and target texts.
A procedure is needed that will enable the contrastive analyst to isolate
formal correspondents in translationally equivalent texts. The recommended
procedure is that of back-translation (Spalatin, 1967), which is intended to serve
as a check on the semantic content. Because of its function, back-translation,
unlike proper translation, does not deal with messages, but with formal
linguistic elements isolated from the target text, which are then translated
back to the source language to give the corresponding linguistic element of
that language. Back-translation can thus be defined as one-to-one structural
replacement. This means that an element of form isolated from the target
language as a likely candidate for a formal correspondent of an element in the
source text, is translated literally and only once back to the source language to
see if it will yield exactly that element whose correspondent it is thought to be.
It was said above that the process of translation is characterized by repeated
recursions to the extra-linguistic content of messages. However, the process of
translation is also a linguistic process, and a strict separation of the translation
would look as follows:

Extra-linguistic message

source text target text

Figura 3.1  – 

capítulo 3 • 69
The contrastive pair of formal correspondence links forms the base of
the triangle of communication by translation, and serves as a basis for the
establishment of translation equivalence. The translator begins his search
for translation equivalence from formal correspondence. It is only when the
identical-meaning formal correspondent is either not available, or not able to
ensure equivalence, that he resorts to formal correspondents with not-quite-
identical meaning, or to structural and semantic shifts which destroy formal
correspondence altogether. But even in the latter case, he makes use of formal
correspondence as a check on meaning – to know what he is doing, so to speak.
A realistic theory of translation would have to account for the communicative
and linguistic (in the narrow sense) aspects of the translator’s work. The
linguistic aspects are contrastive in nature. Equivalence appears as a product
of the contrast between textually realized formal correspondents in the source
and the target language, and the communicative realization of the extra-
linguistic content of the original sender’s message in the target language. Both
components are present in the process of translation, and together ensure
dynamic equivalence, which avoids both literalness and paraphrasing.

3.1  Semantic equivalence and pragmatics

The role of semantic, syntactic, and pragmatic relations13

Semiotics (the science that investigates the general properties of sign
systems) distinguishes the following types of relations: Semantic (sign to
object), syntactic (sign to sign) and pragmatic (sign to man). One of the most
essential requirements imposed on translation, is that both texts (the original
and its translation) should be semantically equivalent. The goal of translation
is to produce a text, bearing the same relation to the extra-linguistic situation
as the original. Semantic equivalence of messages does not necessarily
imply the semantic identity of each linguistic sign. Semantically equivalent
utterances include not only those made up of the semantically identical signs
(for instance,  He lives in Paris  – Ele mora em Paris), but also of utterances
comprising different sets of signs which, in their totality, add up to the same
type of relationship to the extra-linguistic world and denote the same extra-
linguistic situation (e.g., Wet paint – Tinta Fresca.). Semantic relations affect
13  This part of the chapter was based on the website: <https://www.academia.edu/10453537/Lectures_on_the_

70 • capítulo 3
translation both in the initial stage of analysis, and in the production of the
target-language text.
As distinct from semantic relations, syntactic relations are important only
at the stage of analysis, since relations between linguistic signs are essential for
their semantic interpretation (cf. Bill hit John and John hit Bill). But although
they may occasionally be preserved in translation, the translator does not set
himself this goal. Very often, syntactically non-equivalent utterances prove to
be semantically equivalent. Pragmatic relations are superimposed on semantic
relations and play an equally important role in analyzing the original text and
producing an equivalent text in the target language. Semantically equivalent
messages do not necessarily mean the same thing to the source and target-
language receptors, and therefore are not necessarily pragmatically equivalent.
The phrases “He made a fifteen-yard end run” and “Ele correu quinze milhas”
are semantically equivalent for they denote the same situation, but the American
reader, familiar with American football, will extract far more information from
it than his Brazilian counterpart, who would neither understand the aim of the
maneuver nor appreciate the football-player’s performance.
The pragmatic problems involved in translation arise from three types of
pragmatic relations:
1. The relation of the source-language sender to the original message
2. The relation of the target-language receptor to the target-language
3. The relation of the translator to both messages

3.1.1  The Effect of the Pragmatic Motivation of the Original Message

The first type of relations amounts to  the sender’s communicative intent or
the pragmatic motivation of the original message. The translator, in other
words, should be aware whether the message is a statement of fact, a request,
a command, an entreaty, or a joke. Very often, the speaker’s communicative
intent differs from what the message seems to say. “I don’t know” may be not
only a statement of fact, in which case it would be translated as “Eu não sei”,
but also an expression of hesitation, which would be translated as “Será?”. “Is
Mr. Brown there, please” is not a question, but a disguised request “Sr. Brown
está, por favor?”.
The last type of equivalence we are going to deal with is pragmatic

capítulo 3 • 71
3.1.2  Pragmatic equivalence

In order to contrast languages on the pragmatic level, one has to decide

what the equivalence of contrasted structures on the pragmatic level means.
One definition could be that there is pragmatic equivalence between two
expressions in the SL and the TL if they can be used to perform the same speech
act (directive, commissive, perlocutionary etc.) All we have to do then is look at
the strategies used in both languages to perform these acts.
For this purpose, we will consider two aspects above all: coherence
and implicature

While cohesion concerns the surface relations that organize and create
a text, coherence is the network of conceptual relations which underlie the
surface text as perceived by the language users. The mere presence of cohesive
markers, such as linkers or lexical chains, is not sufficient to create a coherent
Imagine we said:
(a) I’m terribly tired because (b) bananas are yellow.
(c) I’m very tired at the end of the week. (d) On Wednesdays, I usually see
my sister.

Although a conjunction such as because suggests a cause/effect

relationship between the two clauses or parts of the sentence, it would be very
difficult to find a logical reason why (b) should cause (a). And although week
and Wednesday belong to the same lexical chain, it is hard to find a connection
between (c) and (d). Thus, what really gives texture to a stretch of language is
the possibility to recognize in it underlying semantic relations that establish
continuity of sense.
Coherence is mostly receiver-centred. It relies on the ability of the hearer/
reader to interpret a stretch of language based on his/her expectations and

14  This part of the chapter was based on the website: <https://www.google.com.br/url?sa=t&rct=j&q=

72 • capítulo 3
experience of the world, which, in turn, are influenced by the society he/she
lives in.
For example, in order to attribute sense to a stretch like:

He looked like Frodo coming down the mountain. The hobbit was walking slowly and
singing to himself.

One has to know that “Frodo” and “the hobbit” are the same person, which
means that he/she must have read the novel Lord of the Rings, seen the movie
drawn from it, or at least heard about one of the two. If a translation of the
sentence were addressed to a public who is not likely to have done any of the
above mentioned things, the translator would probably have to intervene and
modify it, for instance in the following way:

He looked like Frodo, the hobbit, coming down the mountain. He was walking slowly
and singing to himself.

Thus, coherence is not really a property of the text, but of the event/situation
and of the people and things involved in it. A dialogue can be coherent to
one observer or participant, and not coherent to another. For example, from
the patient’s point of view, a medical interview can appear incoherent. The
physician may be considering two or three diseases as potentially being the
patient’s condition and the patient may not know what diseases are being
considered and why certain questions are being asked. However, another
physician listening to the dialogue might understand the doctor’s intentions
and regard the interview as coherent.
From the point of view of translation, this means that the difficulties
encountered will not so much depend on the grammatical and semantic aspects
of the source text itself, but on the audience the translation is addressed to. Like
a writer, a translator has to consider the range of knowledge available to his/her
target readers, their expectations, their views on how the world is organized and
on the structure of social relations, and the conventions of particular text types
in the target language.

capítulo 3 • 73
These are all elements that affect the coherence of a text, because we can
only make sense of new information in terms of our own knowledge, beliefs,
and previous linguistic and non-linguistic experience.

Grice proposed a well-known distinction between what is said and what
is implied, distinguishing truth-conditional aspects of meaning as what is
said, and conventional and conversational implicatures as what is implied.
In this distinction, semantics and pragmatics overlap: There is no clear-cut
boundary. Conventional implicatures, such as the meaning of contrast in ‘but’,
the conclusion to premises in ‘therefore’, or the idea of overcoming difficulty
in ‘manage’, are part of word meaning, but do not contribute to the truth-
conditional content of sentences.
Generalized conversational implicatures, such as an enrichment from ‘three’
to ‘exactly three’, do not require context for their occurrence. They are regarded
by some as semantic, and by others, as either semantic or pragmatic, unlike
particularized conversational implicatures, which are context-dependent and
certainly achieved through pragmatic processes of inference. Bearing this in
mind, we might test the following hypotheses:
(A1) Semantic equivalence is the equivalence of what is said.
(A2) Pragmatic equivalence is the equivalence of what is
implicitly communicated.

Implicature must not be confused with idiomatic meaning. Idiomatic

meaning is conventional, and its interpretation depends on a good mastery
of the linguistic system rather than on interpretation. For instance, in the
following exchange:
A. Shall we go for a walk?
B. Could I take a rain-check on that?

The interpretation depends on knowing the meaning of the expression

“take a rain check” in American English, which is: “If you don’t mind, we’ll do it
another time”. While in the case of:
A. Shall we go for a walk?
B. It’s raining.

74 • capítulo 3
The answer could be interpreted as: “No, thanks, I don’t want to get wet”,
or “Okay, but let’s take an umbrella”, etc. According to Grice, who is mainly
concerned with spoken language, a speaker can signal an implied meaning
conventionally or non-conventionally.
In the first case, he/she uses textual resources, such as conjunctions or
grammatical structures. Implied meaning which is not conventionally signalled
derives from the Cooperative Principle and its maxims of Quantity, Quality,
Relevance and Manner:

•  Make your contribution as informative as required;
•  Do not make it more informative than required.

•  Do not say what you believe is false;
•  Do not say that for which you lack adequate evidence.

•  Make your contribution relevant to the current exchange.

•  Avoid obscurity;
•  Avoid ambiguity;
•  Avoid prolixity;
•  Be orderly.

The maxims of the Cooperative Principle can be flouted for various reasons,
but generally speaking, we assume that an utterance that follows a question
provides an answer to that question. Therefore, we try to find an interpretation
that suits the question. Again, the inferences we draw will depend on our
knowledge of the world, of the participants in the discourse, and of the situation.
Implicatures, then, are pragmatic inferences which allow us to understand
a stretch of language beyond its literal meaning by taking into account the
Cooperative Principle.

capítulo 3 • 75
This complicates the task of a translator, who might knowingly or
unknowingly eliminate certain possible interpretations of the original text.
Grice suggests a number of factors which can contribute to our success or
failure to interpret implicatures, and they are:
a) The conventional meanings of the words and structures used (mastery
of the language) and the recognition of any references involved.
b) The Cooperative Principle and its maxims.
c) The context of the utterance.
d) Other elements of background knowledge.
e) The fact that the above mentioned elements are available to
both participants.

3.2  The meanings of words and structures

This point is quite obvious. If we do not know the meaning of the words and
structures used in the text, we cannot understand its implied meanings. A
mistranslation can provoke the total loss of an implicature. On the other hand,
in English, for example, the use of rhetorical questions often conveys irony, but
it may not have the same function in other languages.
The ability to identify references is also essential for drawing inferences
and maintaining the coherence of a text. For instance, mentioning a type of
food or a fictional character that is unknown to the reader, can disrupt the
continuity of a text.
One of the strategies a translator can use to overcome this problem is cultural
substitution. He/she may substitute a reference to Camilleri’s “Commissario
Montalbano” for a reference to a similar fictional character who is more familiar
to his/her target public.

3.3  The cooperative principle and its maxims

According to Grice, the maxims of the Cooperative Principle are not arbitrary, but
a feature of rational behaviour. However, some linguists do not agree with him
and consider the possibility that the Principle and its maxims are not universal.
There are contexts in which the use of a particular implicature differs
from one language community to another. And even in the same cultural and

76 • capítulo 3
linguistic community, there are special contexts in which one or more of the
maxims do not apply (e.g., quality and quantity in court questioning).
There is also the question whether Grice’s list is exhaustive and whether
the maxims have the same virtue in different cultures. Grice himself admitted
that a politeness maxim could be added to them, and in some cultures, the
Politeness Principle could overwhelm all the other maxims. Politeness is
a relativistic notion, and different cultures have different norms of polite
behaviour and different taboos. In some translation contexts, being polite can
be more important than being accurate. A translator might decide to omit or
replace whole stretches of language which violate the reader’s expectations of
how taboo subjects should be dealt with.
Another important factor that seems to override Grice’s maxims, above
all those concerning relevance and brevity, and support the idea that they are
both language and culture-specific, relates to norms of discourse organization
and rhetorical functions in different languages and text types (e.g., use of
digressions and repetition).
A further question that has been raised is how Grice’s notion of relevance
relates to a participant’s level of interest in a topic. This problem is particularly
important in any translation activity which involves some form of rewriting,
such as editing or summarizing. It raises the question of how well the maxims
transfer from speech to written discourse, from a context which involves a
single receptor and one that involves a range of receptors.
In fact, Grice’s maxims seem to reflect notions which are valued in the
English-speaking world, such as sincerity, brevity and relevance, but which do
not necessarily have the same value in other cultures.
A more plausible theory would be that all discourse, in any language, is
basically cooperative, and the phenomenon of implicature is universal. In other
words, the interpretation of a maxim may differ from one cultural-linguistic
community to another, but the process of conveying meaning by exploiting
whatever maxims are operating in that community will be the same.

3.4  The context of the utterance

The context (participants and situation), co-text and linguistic conventions of a

community in which an utterance occurs determine the range of implicatures
that may be derived from it. The meaning of a word, utterance or gesture does

capítulo 3 • 77
not hinge so much upon an universal, abstract and fixed semantic system, but is
strictly connected to the context. Due to this context dependence, the meaning
of a word, utterance or gesture varies each time, being subject to the conditions
of a certain situation. For this reason, in different contexts, the same message
may receive different interpretations.
Furthermore, there may be a basic ambiguity between a given communicative
intention by the speaker and the ascription of another intention to him/her
by the recipient. In particular, the meaning of any utterance, word or gesture
cannot be expressed and grasped in a totally explicit way, but it always entails
some implicit aspects. It is practically impossible to enunciate all the features
of any meaning, because we take many things for granted in speaking and
interpreting the utterances of others. Such things are given as “presupposed”,
and they are often entailed by the appropriateness condition of the sentence
itself. For instance, if I say that Peter has stopped drinking, I presuppose that
Peter used to drink, since the verb to stop implies a specific behaviour in a
previous time.
On this subject, Searle introduced the notion of background assumptions
as the totality of things taken for granted by the interlocutors in a given
communicative exchange.

For example:

Suppose I go into a restaurant and order a meal. Suppose I say: Bring me a steak with
fried potatoes. I take it for granted that they will not deliver the meal to my house, or to
my place of work. I take it for granted that the steak will not be stuffed into my pockets
or spread over my head. But none of these assumptions was made explicit in the literal

Though not explicitly said, those assumptions contribute in a basic way to

determining the meaning of the utterance. Besides background assumptions,
circumstantial assumptions concern the specific conditions of a given context,
the communicative intention of the speaker, as well as the intention ascribed
by the recipient. People’s interpretations of their own and of other people’s
expressions are not necessarily stable or constant over a period of time, but may
change as the context changes.

78 • capítulo 3
Searle identifies two types of meaning in speech: sentence meaning or
word meaning, and speaker meaning. Sentence meaning is the conventional
meaning of the words as they are usually employed in a lexical sense. Thus, in
the sentence below there is a clear meaning. A woman observes a couple leaving
a party and comments to her partner:

Jim and his wife are leaving the party.

The conventional or sentence meaning here is as follows: two people, a man

(Jim) and his wife are leaving a social function (as opposed to a political party).
The speaker meaning, as Searle puts it, differs from the sentence meaning or
conventional meaning in terms of the speaker’s intentions:
To return to the sentence above, the table below illustrates the sentence
meaning (SEM1), the speaker meaning (SPM1) and the listener meaning
(LIM1). A woman, on seeing a couple leave the party, comments to her partner:

Jim and his wife are leaving the party.

•  Sentence meaning (SEM1)
•  Two people are leaving the party (social function).
•  Speaker meaning (SPM1)
•  Two people are leaving the party (social function).
•  Listener meaning (LIM1)
•  Two people are leaving the party (social function).

But it is possible that these same words could have a different meaning
imposed upon them. For example, the same speaker and her partner may
have become concerned that they are staying too long at social functions. They
decide that in the future they will leave well before the end. The table below
shows the same conventional sentence in a similar context, but with a different
speaker meaning and listener meaning. On this occasion, the lexical approach
does not reveal the speaker’s meaning.
•  Jim and his wife are leaving the party. (SEM2)
•  Two people are leaving the party. (SPM2)
•  We should be leaving now. (LM2)

capítulo 3 • 79
She utters the sentence, ‘Jim and his wife are leaving the party’ with
this meaning (‘We should be leaving now’) on the basis that, in a previous
conversation about their social habits, the speaker and her partner have decided
not to be among the last group to leave social functions. The immediate physical
context is the same, but the socio-linguistic context has been considerably
altered by a previous conversation.
In addition to the realities of a situation, the context includes certain
strategies that people generally use to impose some kind of structure upon the
world around them. When someone describes something, retells an event or
makes a list, he/she normally follows a certain sequence. For example, when
retelling a series of events, one generally follows a temporal order. This order
can be modified or even reversed, but still represents a preferred ordering
strategy according to Grice’s sub-maxim of Manner: “Be orderly”.
However, as we said before about brevity, sincerity and relevance, Grice’s
maxims seem to reflect notions which are valued in the English-speaking
world, but do not necessarily have the same value in other cultures.
It is impossible to determine what the “natural order” is in different
cultures and languages, also because the ordering of events may be adapted to
maintain point of view or thematic progression. In a translation, the order can
be modified to fulfil the expectations of the readers of a different culture. For
example, it is normal to expect entities that are closer to one’s own environment
to be mentioned first in a list.
Another point which can be included in the category of “context” is the
language user’s sense of what is socially (politeness) or textually (genre)
appropriate. The first aspect not only covers the use of personal pronouns,
but also includes the use of appropriate personal and occupational titles, first
names and surnames, nicknames and terms of affection. In translation, these
peculiarities have to be brought back to the conventions of the TL.

Yesterday Mr. Blair left for the United States.
Ontem (o primeiro ministro) Blair foi para os Estados Unidos.

80 • capítulo 3
3.5  Other items of background knowledge

A text may confirm, contradict or extend what we know about the world, as long
as it relates to it in some way. Whether a translator decides to explain a referen-
ce, or to recur to cultural substitution, depends on how much he/she assumes
the reader is familiar with it, and on his/her freedom of intervention.
Of course, a translator may not grasp an implicature himself or herself. In
many cases, it is a good rule to do some research work in order to access the
relevant background knowledgeand sometimes include it, either in the text
itself or in a footnote.

The Availability of All Relevant Items

In order to convey an intended meaning, the speaker/writer must assume
that the hearer/reader has access to all the necessary background and can work
out any intended implicatures. Apart from filling gaps in the reader’s knowledge,
a translator can sometimes adjust the text to the reader’s expectations in order
to avoid conveying the wrong implicatures. But he/she does not necessarily have
to do it. We are normally prepared to accept deviations from usual linguistic
behaviour provided that they are justified, for instance, on the basis of poetic
creativity or humour, and interpretable.

3.6  Grammatical equivalence15

Grammar is the set of rules which determine the way in which units such as
words and phrases can be combined in a language. Grammar has two main
dimensions: morphology and syntax. Morphology concerns the structure of
single words; the way in which their form varies to indicate specific contrast
in the grammatical system (example: singular/plural, number, present/
past). Syntax concerns the grammatical structure of groups of words (clauses
or sentences); the linear sequence of classes of words (noun, verb, adverb,
adjective, etc).

15  This part of the chapter was based on the website: https://www.google.com.br/url?sa=t&rct=j&q=&esrc=

capítulo 3 • 81
Different grammatical structures in the SL and TL may cause remarkable
changes in the way the information or message is carried across. These changes
may induce the translator either to add or to omit information in the TT because
of the lack of particular grammatical devices in the TL itself, which might cause
problems in translation.
As far as translation is concerned, the most important difference between
grammatical and lexical choices is that the former are generally obligatory,
while the latter are largely optional. In the process of translation, such difference
between the source language and the target language often implies some change
in the information content. When the source language has a grammatical
category that the target language lacks, this change can take the form of adding
information to the target text. On the other hand, if it is the source language
that lacks a category, the change can take the form of omission. Grammatical
rules may vary across languages, and this may pose some problems in terms of
finding a direct correspondence in the TL.
•  Number
Number is the inflection of nouns, pronouns, verbs, adjectives and
determiners to show singular, dual, or plural forms. There are three classes of
number: singular (‘one’), dual (‘two’), and plural (‘more than two’). The idea of
accountability is probably universal, but not all languages have the grammatical
category of number, even if they might make distinctions at the lexical meaning.

•  Gender
The term gender, usually attributed to Protagoras in Shery Simon, is derived
from a term meaning class or kind, and refers to the division of Greek nouns
into masculine, feminine and neuter. Gender is the grammatical category
according to which a noun or pronoun is classified as masculine or feminine.

•  Person
Morphological category of the verb used to mark the singular and plural
finite verb forms as ‘speakers’ (first person), ‘addressees’ (second person), or a
‘person, state or thing’ referred to in the utterance (third person). Subgroups of
pronouns refer to people as speakers (I, we), addressees (you), or other people/
things (he, she, it) (inclusive vs. exclusive). The category of person relates to the
notion of participant roles.

82 • capítulo 3
•  Tense and aspect
English has two way tense systems, so that instead of the past tense form
“was”, we can use the corresponding present tense form “is”. “Aspect” is a term
used to describe the duration of the activity described by a verb, and whether
the activity is on-going or completed. In languages which carry these categories,
the form of the verb usually provides two types of information: time relations
and aspectual differences.

•  Voice
Voice is the grammatical category that defines the relationship between
subject and verb. It’s called active clause if the subject is responsible for
performing the action, and it’s called passive clause if the subject is the affected

3.7  Procedure

Translation procedures or translation shifts are defined as “the smallest

linguistic changes occurring in translation of ST (source text) to TT (target
text)”.  Translation is a field of various procedures. Translation procedures are
used to obtain equivalence between source language and target language in the
translation process. There are many kinds of translation procedures, but the
writer wants to explore some procedures that must be used by the translator to
conform to the stylistic demands and grammatical conventions of the target
language. These possibilities are expanded below.

Additional information
The change can take the form of an addition to the target text of information
which is not expressed in the source language, if the target language has the
grammatical category which the source language lacks. Information which is
not present in the source language text may be added to the target language text.

Deletion of information
Baker refers to deletion as the “omission of a lexical item due to grammatical
or semantic patterns of the receptor language”. In the process of translating,
the change of information content of the message can be done in the form
of omitting information in the source language, if the target language lacks a
grammatical category.

capítulo 3 • 83
Structural Adjustment
Structural adjustment is another important strategy for obtaining equivalence
between source language and target language. Structural adjustment can also be
called shift, transposition, or alteration.  Newmark states that “A ‘shift’ (Catford’s
term) or ‘transposition’ (Vinay and Darbelnet) is a translation procedure which
involves a change in the grammar from SL to TL”.

01. What is dynamic equivalence?

02. What is the difference between “Formal correspondence” and “Translation equivalence”?

03. What is the difference between language-based and text-based in formal correspondence?

04. What is Semiotics? What does it mean to have semantic equivalence?

05. What is coherence and cohesion?

BOLINGER, d., 1966. “Transformulation: Structural Translation,” Acta Linguistics. Hafniensia IX, 130-144,
CATFORD, J.C., 1965. A Linguistic Theory of Translation (Oxford UP)
FILIPO VI, R., 1971. “The Yugoslav Serbo-Croatian – English Project,” in: G. Nickler, ed., Papers in
Contrastive Linguistics (Cambridge UP), 107-114.
IVIR, V., 1969. “Contrasting via Translation: Formal Correspondence vs. Translation Equivalence,”
Yugoslav Serbo-Croatian – English Contrastive Project, Studies 1, 13-25.
IVIR, V., 1970 “Remarks on Contrastive Analysis and Translation”, Yugoslav Serbo-Croatian English
Contrastive Project, Studies 2, 14-24
JAKOBSON, R., “On Linguistic Aspects of Translation,” in: R.A.Brower, ed., On Transation (Cambridge,
Mass,: Harvard UP), 232-239.
KRZESZOWSKI, T.P., 1971. “Equivalence, Congruence and Deep Structure,” in: G. Nickel, ed., Papers
in Contrastive Linguistics (Cambridge UP), 37-48.
KRZESZOWSKI, T.P., 1972. “Kontrastive Generative Grammatik, “ in: G. Nickel, ed., Reader zur
kontrastiven Linguistik (Athenäum Fischer Verlag: Frankfurt), 75-84.

84 • capítulo 3
MARTON, W.,1968.“Equivalence and Congruence in Transformational Conrastive Studies,” Studia
Anglica Posnaniensia I, 53-62.
NIDA, E.A., 1969. “Science of Translation,” Language 45, 483-498. 1977 “The Nature of Dynamic
Equivalence in Translating,” Babel XXIII, 99-103.
RAABE, H., 1972. “Zum Verhältnis von kontrastiver Grammatik und Űbersetzung,” in: G. Nickel, ed.,
Reader zur kontrastiven linguistik (Athenäum Fischer Verlag: Frankfurt), 59-74.
SPALATIN, l., 1967. “Contrastive Methods,” Studia Romancia et Anglica Zagrabiensia 23, 29-45.
STEINER, G., 1975. After Babel: Aspects of Language and Translation (Olxford UP)

capítulo 3 • 85
86 • capítulo 3
Techniques of
literary translation
4.  Techniques of literary translation 16

In her book “Translation and Translation Studies: Introduction to Translation”

(2001), Professor Amparo Hurtado Albir, a leading translation specialist,
defines five literary translation techniques as presented below:
1. Adaptation
Albir describes adaptation as a “technique whereby one cultural element is
replaced by another which is typical of the receiving culture. This technique
is very useful when translating advertisements, slogans, etc., which employ a
number of different linguistic processes. In these cases, the most important
thing is the actual meaning of the message rather than the words making it up.”

2. Linguistic amplification
According to Albir, “this translation technique adds new linguistic elements in
the target text. It is the opposite of the linguistic compression technique.” This
is usually about using a paraphrase to explain a word that has no equivalent in
the target language.

3. Compensation
Compensation, on the other hand, is a “translation technique whereby a piece
of information or stylistic device is moved to another location in the text,
because it does not have the same effect if maintained in the same place as in
the original text”. This process is intended to compensate for the losses that
a text suffers when it is translated. The technique is especially useful when it
comes to wordplay: If the translator cannot directly adapt a pun, for instance,
which tends to happen quite often, then they will try to create another play on
words in another part of the text.

4. Elision
The fourth technique of literary translation described by Albir is elision.
Elision is a process that “involves removing items of information in the original
language text so that they do not appear in the target text As with the linguistic
compression technique, elision is the opposite of the amplification process.”
It is frequently the case that the literary translator is obliged to condense the
information contained in certain passages being translated. To do this, some

16  This part of the chapter was based on the website: <http://culturesconnection.com/5-techniques-of-literary-

88 • capítulo 4
items which are not considered essential must be removed, as their elision will
improve the stylistic quality of the translated work.

5. Borrowing
Borrowing is a technique frequently used in literary translation, but which
can also be applied in medical  and business translations, for instance. For
Albir, this translation technique involves “using a word or an expression in the
original text and placing it as it is, with no modification, in the target text.” This
can be an expression taken from a third language (e.g., Latin), an expression
which is familiar to speakers of the target language, or even an untranslatable
expression which is not worth explaining.

4.1  The literary translation and its specificities17

What is Literary Translation? Traditionally, Literary Translation is one of four

broad categories of translation, the other three being Interpreting, Scientific
and Technical Translation, and Commercial/Business Translation. There are
also a number of special fields, such as Legal Translation. Literary Translation
is not confined to the translation of great literature. When the Copyright Act
refers to ‘literary works’, it places no limitations on their style or quality. All
kinds of books, plays, poems, short stories and other writings are covered,
including items such as a collection of jokes, the script of a documentary, a
travel guide, a science textbook and an opera libretto.
What qualifications are required? Is it possible to be a translator without
holding any formal qualifications? Some translators develop their skills
naturally, due to having a bilingual family background or having lived abroad
for long periods. However, there is no doubt that a formal university education
in Modern Languages is helpful, especially if it includes classes in Translation.
There are courses and workshops designed to improve translation techniques.
These may be useful, even though there is no guarantee that after attending
such a course the translator will actually obtain commissions. People with
qualifications in addition to Language diplomas often find themselves in
great demand. For example, a publisher who wants someone to translate a
book on Genetic Engineering may be keen to commission a translator with

17  This part of the chapter was based on the article: The Society of Authors, 84 Drayton Gardens, London SW10
9SB, Quick Guide to LiteraryTranslation, 1988.

capítulo 4 • 89
a degree in Biology. Information about translation courses may be obtained
from Educational bodies and cultural institutes of individual foreign countries.
Although they should not be regarded as free advice services, the Institute of
Linguists and the Institute of Translation and Interpreting may also be willing
to help. Since the members of the Translators Association are accomplished
literary translators who have already achieved publication, the Association itself
does not usually offer ‘training’ courses, although it holds talks and seminars
that are helpful for working translators.
Qualities rather than qualifications. When experienced members of the
Translators Association were asked to produce a profile of a literary translator,
they listed the following points:
1. The translator needs to have a feeling for language and a fascination
with it.
2. The translator must have an intimate knowledge of the source language
and of the regional culture and literature, as well as a reasonable knowledge of
any special subject that is dealt with in the work that is being translated.
3. The translator should be familiar with the original author’s other work.
4. The translator must be a skilled and creative writer in the target
language (i.e., the language to which the translation is made), and nearly always
he/she will be a native speaker of it.
5. The translator should be capable of shifting from one style to another
in the language when translating different works.
6. The aim of the translator should be to convey the meaning of the
original work, as opposed to producing a mere accurate rendering of the words.
7. The translator should be able to produce a text that reads well, while
echoing the style and tone of the original – as if the original author were
writing in the target language. As evidenced in this description, the flair, skill
and experience required in a good literary translator resemble the qualities
that are needed in an ‘original’ writer, and it is not surprising that writing and
translating often go hand in hand.

4.1.1  Literary translation as a career

Almost without exception, translators of books, plays etc. work on a freelance

basis. In most cases, they do not translate the whole of a foreign language
work ‘on spec’: They only go ahead with the translation after a publisher or

90 • capítulo 4
production company has undertaken to issue/perform the translation and has
signed an agreement commissioning the work and specifying payment. As in
all freelance occupations, it is not easy for the beginner to ensure a constant
flow of commissions. Only a few people can earn the equivalent of a full salary
from literary translation alone. Literary translators may have other sources
of income, such as language teaching or an academic post. They may also
combine translating with running a home; they may write books themselves,
as well as translating other authors’ work; or they may be registered with a
translation agency and accept shorter (and possibly more lucrative) commercial
items between longer stretches of literary translation. Through membership
of the Translators Association, translators who work alone in a wide variety
of circumstances can obtain support and advice concerning all the business
aspects of Literary Translation.

The ground rules of the trade

As a trade, Literary Translation operates within the framework of copyright
law. An article about the UK Copyright Act can be found in The Writers’ and
Artists’ Yearbook, published annually by A & C Black. Under the Act, a translation
is an adaptation of the original foreign language work, so the translator must
ensure that the owner’s permission has been obtained before starting work.
It is a golden rule that a translation must be faithful to the original work, but
translating is still a creative process. There is a popular misconception that
translating a text from one language to another is a mechanical exercise – a
matter of straight conversion or even copying. Yet, if two translators are given
the same source text, the result may be two quite different, but equally valid
versions of the text in the target language. In theatre, we are also used to seeing
a succession of new translations of classic plays, proving that each translator
creates something original that is specially made to ‘speak’ to a particular
audience. The law recognises this ‘original’ nature of a translation and affords
copyright protection to the translation, separate from the copyright protection
to which the original foreign work is entitled and also separate from the
protection of someone else’s translation of the same work.
The translator’s ‘moral rights’ are also protected: A translation cannot
be used in a derogatory way, and if the translator wishes, it must carry the
translator’s name when it is published. In European countries, copyright
protection generally lasts until the end of the seventieth year after the death of
the translator.

capítulo 4 • 91
The translator’s income
When a translation is made in the course of employment, the employer is
the owner of the copyright. A self-employed translator, on the other hand, is not
simply ‘doing a job’. According to copyright law, he/she is creating an original
literary work in which he/she is the owner of the copyright. Anyone who wants
to publish the translation or reproduce it in any way must obtain a license from
the translator for the relevant rights, in return for suitable remuneration. Like
the original author, the translator should expect his/her remuneration to reflect
the amount of use that is made of the translation. This general principle was
confirmed in the 1976 Nairobi Recommendations of UNESCO. If the payment
is in the form of a fee, i.e., a lump sum, it should not be ‘for the translation’,
but for a specified use of the translator’s work, e.g., for the right to print 5,000
copies for sale in the UK.
Such an arrangement makes fair allowance for additional fees to be paid,
for example, if further copies are sold or if the license is extended to include
America. For the translation of a book, the Model Contract issued by the
Translators Association recommends that there should be an advance payment
on account of a royalty on each copy sold and a share of the proceeds from
uses such as serialisation. In the case of a play, the translator should receive
a percentage of the gross box office receipts. The translator should be able to
obtain additional payment if he/she is asked to edit a literary work as well as
translate it, and there should be an additional fee for the preparation of an
index for the translated edition.
When translations are borrowed from public libraries, the translator
receives a 30% share of the full Public Lending Right payment. For the
beginner, the usual method of obtaining commissions is by sending a letter
and a short sample translation to selected publishers, letting them know that
the translator is seeking commissions. Details of British publishers are to
be found in regularly updated publications such as The Writer’s Handbook
(published by Macmillan) and The Writers’ and Artists’ Yearbook (A & C Black).
Only a fraction of all the works published in their original language have their
translation subsequently published, but a translator who knows what sells in
their own market may occasionally spot a suitable candidate for translation.
The first step then is to contact the owner of the translation rights, who may be
the original author, or in many cases, the foreign publisher of the original work.

92 • capítulo 4
The translator needs to ascertain whether the translation rights are available,
and whether the owner is willing to authorise him/her to approach potential
publishers. It is not an easy task to obtain commissions by either method. It
may be a matter of luck whether the translator is ‘in the right place at the right
time’, but eventually, the commissions should begin to come along, and a list
of contacts is created. Opportunities are likely to arise at or shortly after some of
the large book fairs, such as the Frankfurt Book Fair and the Bologna Children’s
Book Fair, where publishers and literary agents are busily engaged in buying
and selling translation rights.

The translator’s relationship with the author

Usually, the contractual arrangements are made in two stages. The
publisher/ theatre/broadcasting organisation decides that it wants to publish/
produce a translation and signs a contract with the owner of the original work.
Then, they sign a contract with the translator. The translator should bear in
mind that usually the author/rights owner accepts a lower advance payment
and lower royalties on the translated edition than on the original edition of the
work in his/her own country. Often, the author receives a 7.5% royalty on sales of
a translated book in the UK, compared to 10% or more on sales of the work in its
original language. On a theatre production, the original author might receive
6% instead of 10%. The author may also have forgone part of his/her share of
secondary rights, e.g., of the proceeds from the sale of American rights.
This means that some or all of the payment received by the translator is
money that might otherwise have been paid to the original author, i.e., to that
extent, the author is the person who, indirectly, bears the cost of the translation.
Clearly it would be unfair if the translator did not receive from the publisher
the share of the royalties that the author/rights owner believed to have been
allocated towards the translation costs. On rare occasions, the foreign author
of a work is so keen to see it translated, that he/she will offer to pay the
translation costs directly. In these cases, the author and the translator should
be conscious that translating the text does not guarantee that the translation
will be published. There should be a written contract between the author and
the translator, clarifying the rights of each party and how the proceeds are to be
divided if, subsequently, a publisher is willing to issue the translation.

capítulo 4 • 93
4.2  Authorship

Authorship attribution is the following issue: For a given text, to determine

the author of said text among a list of candidate authors. Determining authorship
is difficult, and a host of methods have been proposed: As of 1998, Rudman
estimated the number of metrics used in such methods to be at least 1,000
(Rudman, 1997). For comprehensive recent surveys, see (Juola, 2006; Koppel
et al., 2008; Stamatatos, 2009). The process of authorship attribution consists
on selecting markers (features that provide an indication of the author), and
classifying a text by assigning it to an author using some appropriate machine
learning technique.

Attribution of translated texts

In contrast to the general authorship attribution issue, the specific problem
of attributing translated texts to their original author has received little attention.
Conceivably, this is due to the common intuition that the impact of the translator
may add so much noise, that proper attribution to the original author will be very
difficult. For example, in (Arun et al., 2009), it was found that the imprint of the
translator was significantly greater than that of the original author. The volume
of resources for natural language processing in English appears to be much
larger than in any other language, and it is thus, conceivably convenient to use
the resources at hand for a translated version of the text, rather than the original.
To appreciate the difficulty of purely lexical or syntactic characterization of
authors based on translation, consider the following excerpts from three different
translations of the first few paragraphs of Turgenev’s: Liza “A nest of nobles”.

Translated by W. R. Shedden Ralston:

A beautiful spring day was drawing to a close. High aloft in the clear sky floated small
rosy clouds, which seemed never to drift past, but to be slowly absorbed into the blue
depths beyond. At an open window, in a handsome mansion situated in one of the
outlying streets of O., the chief town of the government of that name–it was in the
year 1842–there were sitting two ladies, the one about fifty years old, the other an old
woman of seventy.

94 • capítulo 4
A Nobleman’s Nest Translated by I. F. Hapgood

The brilliant, spring day was inclining toward the 65 evening, tiny rose-tinted cloudlets
hung high in the heavens, and seemed not to be floating past, but retreating into the
very depths of the azure. In front of the open window of a handsome house, in one of
the outlying streets of O * * * the capital of a Government, sat two women; one fifty
years of age, the other seventy years old, and already aged.

A House of Gentlefolk Translated by C. Garnett

A bright spring day was fading into evening. High overhead in the clear heavens small
rosy clouds seemed hardly to move across the sky but to be sinking into its depths of
blue. In a handsome house in one of the outlying streets of the government town of O
- (it was in the year 1842) two women were sitting at an open window; one was about
fifty, the other an old lady of seventy.

As translators express the same semantic content in different ways, the

syntax and style of different translations of the same text will differ greatly due
to the footprint of the translators. This footprint may affect the classification
process in different ways, depending on the features. For markers based on
language structure, such as grammar or function words, it is to be expected
that the footprint of the translator has such a high impact on the resulting text,
that attribution to the author may not be possible. However, it is possible that
a specific author/translator combination has its own unique footprint, which
is discernible from other author/translator combinations: A specific translator
may often translate frequently used phrases in the same way. Ideally, the
footprint of the author is (more or less) unaffected by the process of translation,
for example, if the languages are very similar or the marker is not based solely
on lexical or syntactic features. In contrast to purely lexical or syntactic features,
the semantic content is expected to be roughly the same in translations and
originals. This leads us to hypothesize that a marker based on semantic frames,
such as found in the FrameNet database (Ruppenhofer et al., 2006), will be
largely unaffected by translations, whereas traditional lexical markers will be
severely impacted by the footprint of the translator.

capítulo 4 • 95
The FrameNet project is a database of annotated exemplar frames, their
relations to other frames, and obligatory as well as optional frame elements for
each frame. FrameNet currently numbers approximately 1,000 different frames
annotated with natural language examples. In this paper, we combine the data
from FrameNet with the LTH semantic parser (Johansson and Nugues, 2007),
which until very recently (Das et al., 2010) was the semantic parser with best
experimental performance (note that the performance of LTH on our corpora
is unknown and may differ from the numbers reported in (Johansson and
Nugues, 2007).

Related work
The research on authorship attribution is too voluminous to include. See
the excellent surveys (Juola, 2006; Koppel et al., 2008; Stamatatos, 2009) for an
overview of the plethora of lexical and syntactic markers used. The literature on
the use of semantic markers is much scarcer: Gamon (Gamon, 2004) developed
a tool for producing semantic dependency graphs and using the resulting
information in conjunction with lexical and syntactic markers to improve the
accuracy of classification. McCarthy et al. (McCarthy et al., 2006) employed
WordNet and latent semantic analysis to lexical features with the purpose of
finding semantic similarities between words. It is not clear whether the use of
semantic features improved the classification. Argamon et al. (Argamon, 2007)
used systemic functional grammars to define a feature set associating single
words or phrases with semantic information (an approach reminiscent of
frames). Experiments of authorship identification on a corpus of English novels
of the 19th century showed that the features could improve the classification
results when combined with traditional function word features. Apart from a
few studies (Arun et al., 2009; Holmes, 1992; Archer et al., 1997), the problem
of attributing translated texts appears to be fairly untouched. As pointed out in
(Luyckx and Daelemans, 2010), the size of data set and number of authors may
crucially affect the efficiency of author attribution methods, and the evaluation
of the method on some standard corpus is essential (Stamatatos, 2009). Closest
to a standard corpus for author attribution is The Federalist Papers (Juola,
2006), originally used by Mosteller and Wallace (Mosteller and Wallace, 1964).
We employ the subset of this 66 corpus consisting of the 71 undisputed single-
author documents as our Corpus I.

96 • capítulo 4
For translated texts, a mix of authors and translators across authors is
needed to ensure that the attribution methods do not attribute to the translator
instead of the author. However, there does not appear to be a large corpus of
texts publicly available that satisfy this demand. Based on this, we elected to
compile a fresh corpus of translated texts: Our Corpus II consists of English
translations of 19th century Russian romantic literature chosen from Project
Gutenberg, for which a number of different versions, from different translators
existed. The corpus consists primarily of novels, but it is slightly polluted by a
few collections of short stories and two nonfictional works by Tolstoy, due to the
necessity of including a reasonable mix of authors and translators. The corpus
consists of 30 texts by 4 different authors and 12 different translators, some of
which having translated several different authors. The texts range in size from
200 (Turgenev: The Rendezvous) to 33,000 (Tolstoy: War and Peace) sentences.
The option of splitting the corpus into an artificially larger corpus by sampling
sentences for each author and collating these into a large number of new
documents was discarded. We deemed that the sampling could inadvertently
both smooth differences between the original texts and smooth differences in
the translators’ footprints. This could have resulted in an inaccurate positive
bias in the evaluation results.
Some translation theorists18 claim that the translator’s authorship
deserves recognition, or better recognition. The claim is made with respect to
literary translation (cf. Zeller 2000), political metaphors of translator-author
relationships (cf. Chamberlain 1988/2000), copyright regimes (cf. Venuti 1995:
12), and more recently, a “creative turn” in Translation Studies, associated
with postmodern theory, that works to blur all boundaries, to recognize the
translator’s subjectivity, and above all to keep turning (cf. Perteghella and
Loffredo eds 2006). Lawrence Venuti has probably been the most consistent
and vociferous proponent of the translator’s authorship. He initially opposed
the idea to a “Romantic conception of authorship” that would accord all
creativity to the author of the source text, thereby relegating the translator’s
work to “derivative” status (cf. Venuti 1992).
This critique is largely in tune with mainstream literary ideology: The
“death of the author” was identified by Foucault and Barthes a long time ago;
theories of intertextuality took off from Kristeva’s reading of Bakhtin. The idea

18  This part of the chapter was based on the website: <http://usuaris.tinet.cat/apym/on-line/translation/2010_

capítulo 4 • 97
that all creativity is translational is now a keystone of postmodern thought.
From this perspective, saying that the translator has authorship is also saying
that all authors work translationally. And if that means that translators, like all
authors, transform texts, bring novelty into the world, have complex productive
cognition processes churning within them as they work, and are all different,
then I have no qualms about the proposition at all. Translators are indeed
subjective in their minds and creative in their writing, as any piece of empirical
research should be able to show. “Authorship”, however, can be understood
in several senses. It concerns not only creativity or individuality, but also
ethical responsibility, a point that has been overlooked by many of the literary
Here, I want to focus on authorship, precisely and exclusively in the sense
of responsibility within communication acts, that is, from the perspective of
formal pragmatics. This is not to suggest that the other senses of authorship are
somehow wrong. They merely sidestep an entire domain that is nevertheless
crucial, not only for contemporary social understandings of what translators do,
but also, through Habermas, for an ethics of liberal humanism. Bear with me—
this might be more important than translation and literature combined. The
Canadian micro-sociologist Erving Goffman provided one of the clearest and
most provocative analyses of authorship. Although his universalist categories
are certainly not beyond dispute, his terms will here provide no more than
convenient hooks on which to hang my arguments. Goffman most interestingly
uses the term “author” to denote the role of the “principal”, the person who
takes responsibility for an utterance: “someone whose position is established
by the words that are spoken, someone whose beliefs have been told, someone
who is committed to what the words say” (1981: 146).
We will soon attempt to unpack each of those three descriptors. Goffman
thus distinguishes the role of “author” from the actual creation of a text, which
is the point where he diverges from the literary ideologies. Smart politicians, for
instance, do not write their own speeches: They merely “authorize” them and
thereby take moral responsibility for them. This distinction dispatches with
the whole question of Romantic authorship and “creative turns” – creativity
is not the issue, ethical responsibility is. Goffman also usefully distinguishes
authorship from the role of the “animator”, a person who says the words
but might be doing so on behalf of someone else, perhaps by quoting, using
indirect reported speech, parodying, acting, or indeed translating. From this

98 • capítulo 4
perspective, to say that a translator has authorship is to posit that the translator
is more than an animator: The translator would somehow have their position
established by the words, or be someone whose beliefs have been told, or even
be someone committed to what the words say. This is the proposition at stake.
Seen from this perspective, translators are not usually authors (despite many
possible nuances – see Torikai 2009 for an analysis of interpreters’ interview
data in precisely these terms). To bluntly state that translators have (or should
have) authorship is thus to overlook quite an important dimension.
Here, I will try to produce evidence of why translators are not authors. This is
in the spirit of testing what seems obvious, in the same way as first-year science
classes ask students to demonstrate that the Earth is a sphere (if I remember
correctly, you are supposed to make hypotheses about ships disappearing over
the horizon, and things like that). Culture is not physics, and the Flat Earth
Society still exists (“Deprogramming the masses since 1547”, says their website),
but the search for empirical proof might yet be instructive. So where shall we
look? 1. An author is someone whose position is established by the words. In
all languages that have personal pronouns, linguistic subjects are positioned
in relation to the I-here-now of an utterance (that is a matter of definition; it
requires no proof). The linguistic subject that is translating, however, cannot
occupy an I-here-now while they are actually translating. Whenever they say “I”,
that position is ostensibly occupied by someone else, the author of a previous
text. This is an essential feature of what I call the “translation form” operative
in our immediate cultures (see Pym 2004).
It is something that children have to learn when they start experimenting
with interlingual communication—the movement is mostly from reported
speech (“Mummy says that…”) to translation (what I am saying is an un-
annotated report of what Mummy said…). I currently have a five-year old son
who makes occasional experiments with this transition, but only when no
personal pronouns are involved. It is a hard trick to learn, as indeed are personal
pronouns themselves at a younger age. Now, the upshot of these lessons that we
all had to learn (and then we forgot, apparently) is that translating translators
have no “I”, or are condemned to use what is called “the alien I”. That could
perhaps be an universal feature of what we would want to call translations, but
I have no proof of any such universality and I have not tested the cultural or
historical frontiers involved.

capítulo 4 • 99
This is just what I see in the language usage around me, and I warmly
recommend that you have some children in order to see how hard and complex
the use of such linguistic features can be. So, here is an adaptation of Goffman’s
first description of authorship: Translators, when translating, are not authors
because their pronominal position is not established by the words said. Or more
precisely, their position in the discourse is only established as a non-author.
You can say as much as you like on covers, titles and introductions about the
identity and creativity of the translator, gather reams of process analysis and
post-hoc interview data, but while they are translating, beyond the paratexts
and processes, translators do not have authorship in this pronominal sense.
Is this a necessary distinction? It is quite possible to imagine a community in
which the interdicted first person is not needed.
All translated utterances would be marked at all times as reported speech;
interpreters in court would be sworn in as witnesses rather than operate
as officers of the court; there would be no use of the alien “I”; imperial laws
would carry the signatures of those who redacted each linguistic form, and
each language version would be a different law; translators might even get a
meatier deal out of copyright conventions, albeit not as translators. When we
try to imagine a community without the “alien I” translation form, however, we
quickly come across a prime reason for the very existence of such a form: simple
efficiency. If we had to note explicit discursive positioning for each utterance
rendered from another language, or even rewritten within the one language,
communication would become even more unwieldy and more prone to doubt
than it already is. In the interests of efficiency, in the first place, translators seem
excluded from the positions established by the words they produce. Efficiency,
however, merely posits one condition for the development of a translation form.
It says very little about the way the form is actually used. Once the translator’s
non-manifest “I” is understood, is the rest just a case of convenient forgetting?
2. An author is someone whose beliefs have been told. In Habermas’s formal
pragmatics (for example, Habermas 1990), we find three “universal validity
claims” that are supposed to be made as a part of communicative competence:
1. A claim to the truth of what is said or presupposed
2. A claim to the normative rightness
3. A claim to the truthfulness or sincerity of the speaker

100 • capítulo 4
These claims map with some difficulty onto Goffman’s three conditions
of authorship, constituting a problem that I do not wish to pursue here
(although “truthfulness” comes close to “commitment”, and Habermas simply
assumes authorship and therefore positioning). The three claims are also quite
timorous whistles when it comes to cross-cultural communication, since there
is considerable empirical evidence of non-shared competencies. This, too, will
be left for another day. More interesting in the present context, I suggest, is the
task of identifying which of these claims are inherent in translatorship.
It seems to me that the translator is not normally required to claim
anything about the truth, rightness or truthfulness of the source text or author.
Translators are, however, regularly required to make such claims about the
way the translation represents a source text – we claim the translation is a
true representation, in accordance with the communicative norms of the
(translation) situation, and that the translator believes in the truth of the
representational act. Such claims are thus rather like the person who certifies
that a photocopy is a true representation of a document, even though the
document photocopied may be an outright forgery. Furthermore, all users of
the translation will understand and share in these implicit claims. From this
perspective, the communicative competence of the translator would be based
on no more than that mimetic or representational function.
It would not encroach upon the specific claims of authorship. At this
point, there should be cries of protest from translators and theorists alike.
No, a translation is not a photocopy! Heaven forbid! Of course, the complaint
is entirely correct – translators can only claim to represent some aspects of
their sources, and considerable creative transformation is often required to
achieve the illusory representation of those aspects. Along with the claim to
representation, there must also be implicit claims to transformational work,
perhaps in the name of “rightness” to the new communicative situation
(Habermas’s second claim). So let us politely withdraw the photocopy simile
for a while. The main point is not the extent of what translatorship entails
here, but what it does not comprise. Whatever their truth claims with respect
to representation, translators are usually not required to attest to the truth of
their source texts. It is this limitation that distinguishes translatorship from
authorship, rather than the problematics of representation as such. How can
this limitation be tested?

capítulo 4 • 101
The fundamental restriction of validity claims can be seen at work in pseudo-
translations, understood as texts presented as translations, but which have no
corresponding source text – there are indeed authors who choose to write as
translators. Pseudo-translations abound in periods where new knowledge is
institutionally repressed, since translational representation becomes a way of
getting around the censorship. Referring to Latin proto-science of the twelfth
and thirteenth centuries, Thorndike argues that many of the texts presented as
translations may well have been pseudo-translations: “The number is suspiciously
large of works of which the lost originals were supposedly by Greek or Arabian
authors but which are extant only in later Latin ‘translations’” (1923: 2.26-27).
Adelardus de Bada, for instance, claimed to have disguised many of his
personal opinions in a way quite compatible with pseudo-translation: “For I
am aware what misfortunes pursue the professors of truth among the common
crowd. Therefore it is the cause of the Arabs that I plead, not my own” (cit.
Thorndike 1923: 2.25). By presenting knowledge as translational, authors could:
1. Limit their own public responsibility and thus liability to persecution
2. Present the knowledge as being operational in another culture, and
thus of a threatening or rival status (“we need this knowledge because our com-
petitors already have it”)
3. Give the knowledge the authority of age, if not of associated auctorial
prestige (in cases where the author was supposed to be Aristotle, for example).

Not all three logics are operative in all situations, and there can certainly be
further reasons for pseudo-translations. None of this, however, would be possible
in the absence of a translation form based on a distinction between authorship
and translatorship. In other words, if translators were authors in this sense of
making truth claims about the original, there would probably be no pseudo-
translations, or not as many. Goffman’s general descriptor of this point might
thus be crudely adapted as follows: “Don’t shoot me – I’m only the translator!”
And yet, many translators have indeed been persecuted because of their assumed
authorship: Translators of the Bible (William Tyndale, Jan Hus), of oppositional
philosophy (Étienne Dolet) or of more contemporary heresies (Hitoshi Igarashi,
Ettore Capriolo, Aziz Nesin, all translators of Rushdie). And a very long et cetera.
The translation form is obviously not always a successful defense. Yet, it is
there. The real problem is that our contemporary theorists, who would make
translators into authors in the name of recognizing creativity, would perhaps
thereby be attributing to translators rather more validity claims than is their due.

102 • capítulo 4
Would the theorists willfully ratify the burnings and shootings and stabbings to
which translators have been subjected, all in the name of validity claims? No, this
is not a question of mere efficiency. An author is someone committed to what the
words say. Translators may be sincere in their belief that they are representing a
source text, but should this commitment be extended to belief in the validity of
the source text itself? In some cases, such extensions are quite normal.
Translators of the Bible, for example, traditionally express their collective
belief that the text is the Word of God, and this commitment somehow
remains unshaken by all the ensuing debates over how to render the text. More
normally, though, the professionalism of the translator requires detachment
rather than commitment. The translator of pharmaceutical instructions, for
example, would certainly have to trust in the validity claims of the source text,
but that trust would be essentially the same as that of any user of the actual
pharmaceutical products. If the translation misrepresents the source, the
translator is responsible; if the source instructions are badly written or the drug
has unforeseen sideeffects, the translator is not responsible.
There are institutional authors that are legally liable in such cases. An
interesting test case of commitment is a speech delivered in English on
November 10, 1991 in Weinheim, Germany, by the American Fred Leuchter. In
this speech, Leuchter argued, among other things, that it was impossible that
so many people could have been killed in Auschwitz. Such questioning is illegal
in Germany. The real problem, however, is that the speech was rendered into
German by a translator named Günter Deckert, and it was the German translation
that became the object of a series of court cases (there were two retrials). Could
the translator be condemned for the validity claims of a discourse he merely
translated? The question in this case concerned neither the position of the
translator as non-author, nor the exclusive ownership of the beliefs recounted,
but the translator’s commitment to those beliefs and to their effects (see the
discussion in Pym 1997). Indeed, at the 1995 retrial, Deckert was found guilty
of gefährliche politische Brandstiftung (literally, “dangerous political arson”),
since he had lit the fire that enabled Leuchter’s speech to burn in German.
Now, a professional translator would not normally be expected to share that
guilt, and as such, should have a reasonable defense. Deckert, however, was the
head of a neo-Nazi party, and it seems he had invited the American to deliver
the speech in the first place. Just as pseudo-translations can be used to import
incendiary ideas, here, a kind of pseudo-original was being used to evade
German law. Is it possible to condemn the translator Deckert, and yet retain

capítulo 4 • 103
the professional detachment of other translators? I think it is. One merely
has to insist that the elements that suggest commitment (head of a political
party, organizer of the speech) belong to a subject quite different from the one
positioned in the translational discourse. As a politician and organizer, Deckert
quite probably shared the commitment, and certainly sought the incendiary
effects; as a translator, however, he had no need to do so.
The translation form should thus survive this case, and Deckert’s supposed
innocence should not. Some nevertheless claim that translators should indeed
only translate messages to which they are committed. Baker (2009) strangely
posits that you should only translate people with whom you have mutual
respect, so she, for example, would never do a translation for the CIA, Walmart
or the State of Israel. This position is as confusing as it is honorable. It is one
thing to work with clients with whom you have mutual trust – this would be
good advice for a commercial relationship of any kind. But this relationship has
nothing to do with the status of the messages involved. All clients want to know
about what their rivals or enemies are saying; all translators have been called
on to render words with which they do not agree, for a great many noble and
ignoble reasons. Baker’s position, however, starts from the assumption that
narrative is a basic way we construe knowledge about the world, so the function
of translation is to control the narratives that circulate in a culture.
Thus, it makes sense to edit out the bad narratives through non-translation,
and to tell stories about your enemy rather than talk to them – a point at which
there is no essential difference between translatory or auctorial narration, since
the fascistic binding of the culture apparently has priority over all else. This
view willfully excludes dialogue as another basic means by which we construe
knowledge about the world. If you only believe in narrative, you can require
commitment to all texts in a culture, and ultimately a control of that collective
commitment – Baker constantly uses her own narratives to reduce the complex
subjectivity of the other (“he is not the sort of person who…”). If you believe in
the acquirement of knowledge through dialogue, however, you will ultimately
require mediation with your enemy, on one level or another, at which point you
will need someone prepared to do more than just expound their own beliefs.
Let me be careful here, before some postmodern theorist concocts a host of
hasty extrapolations. To say that the current translation form does not require
commitment to content does not imply, in any way, that translators are neutral,
subjectless or without personal involvement in their utterances.

104 • capítulo 4
On the contrary, countless text comparisons, process studies and
psychological attention after war-crimes trials show that translators intervene
in their productions, whether they want to or not, and thus that neutrality is
a profoundly ideological professional construct (see Pym 1992/2010: 167ff.).
It is something quite different to insist that translators do or should always
believe in the content of their translations. In general, then, there is no need for
translators to claim (or be attributed with) any commitment to the content of
what they are translating. To that extent, they are translators, not authors, and
they have no obligation to sign up to this week’s good causes. To summarize our
adaptations of Goffman’s descriptors: I claim that, with respect to discursive
positioning, beliefs and commitment, translators are not authors. Such, at least,
is the conclusion invited by the translation form in our immediate cultures.

A suggestion from process studies

Most of the discussions about authorship and translatorship are based on
products; texts that have been authored and translated. One might nevertheless
hope to find some evidence from studies of how translators actually work when
translating – process studies should also have their word to say. Here, we review
the little evidence that has become available. What is the difference between
translating and just regular writing? A naïve answer would be that the translator has
a complete text as a point of departure, whereas other forms of writing only have an
incomplete text and a frightening white space to fill. That answer would be naïve
because, one might argue, all writing works from previous texts or text models,
either written out or in the mind. All writing is to some extent re-writing, and
translation need not be fundamentally different in this regard. But is it? Another
facile conception would be that translators just work at phrase level, replacing text
fragments with text fragments, without the wider macro-textual or communicative
frame deemed operative in other modes of writing. Some studies contradict this
view. Englund Dimitrova (2005: 14) summarizes the research as follows: In several
process-oriented studies on translation, professional translators have been shown
to have a high degree of consciousness regarding textual features, global strategies
and the communicative purpose of the translated text.
Thus, professional translators verbalize, in their think-aloud protocols,
global strategies, translation principles and personal theories of translation
to a greater extent than students and non-professionals would. This level of
awareness is surely indicative of authorship, at least in the “creativity” sense of
the word, if not as an indicator of responsibility for the direction and success of

capítulo 4 • 105
the discourse. Enhanced consciousness on this level also seems to be acquired
as one becomes a translator, and could thus be included in catalogues of
professional translation competences. Englund Dimitrova continues (2005: 14-
15): Furthermore, professional translators have, in comparison with students, a
higher degree of TL [target-language] pragmatic and stylistic awareness, as well
as an awareness of the purpose of the translation. This is all grist to the mill of
translational authorship.
Working in parallel with the theorists of products, the researchers of
processes seek and find evidence that translators are actually thinking as
authors, albeit without that term. There could, however, be a few snakes in
the woodheap, or flies in the ointment. Immonen (2006) finds that translators
produce texts quite differently from monolingual writers: They pause fewer
times in the first draft; they revise for longer; their pauses are longer at clause
level and below and shorter at above clause level. What might this mean? One
can imagine the monolingual writer stopping to think about relatively large
stretches of text (“Where the hell am I going? Where the hell have I come
from?”). The translator, however, seems more concerned with relatively small
stretches (“What the hell does this mean?”), and then compensates for that
narrow vision by putting more effort into post-draft revision. So much for all
writing being the same... Second snake or fly: Astrid Jensen (2001) compared
expert translators to “young professional translators” with respect to a classical
psychological account of writing (from Bereiter and Scardamalia 1987).
The basic opposition is between text production as “knowledge
transforming”, which is when the actual writing process identifies and
transforms ideas (hence we find writing harder work than speaking), and writing
as “knowledge telling”, which would be more linear and less problematic.
Perhaps surprisingly, Jensen finds that expert translators use the “knowledge
telling” model more consistently, since they “engage in less problem-solving,
goal-setting and re-analyzing behavior vis-à-vis young professional translators”
(Jensen 2001: 177). That is, they deal with the text as it comes, with limited
involvement and responsibility. Indeed, we might surmise that, when becoming
experts, translators learn to become rather less like authors. Of course, those
findings could be no more than small stories based on smaller experiments.
Then again, these different kinds of cognitive processes regularly explain why
many writers start to translate when they are psychologically blocked in their
monolingual work. Translating is not just ethically different from authoring; it
can be a psychologically different kind of writing.

106 • capítulo 4
4.3  Tips to becoming a better literary translator

It is a well-known fact that a good literary translator has to be a skillful writer

at the same time. So when it comes to translating a book, or more generally,
literary translation, one must use its creativity at the maximum. It is like writing
the book all over again, trying to make it compatible with another language and
a different culture. This is indeed a challenging job, and requires full attention
and dedication.
Before deciding to translate a book, you should make sure that you have vast
knowledge in the field on which the book focuses. Thus, if philosophy is not
your favourite subject, you will definitely have a hard time trying to translate a
philosophical book.

Diversify the topics that you read

In order to become a great book translator, it is advisable to read as much
as possible in all the languages you master. A good idea is to diversify the topics
that you read, even if they are out of your comfort zone.
This is really helpful, particularly in terms of understanding the styles
and subtleties of foreign literature. This will help you adapt them to your own
culture and achieve a professional translation.

Work on developing your writing skills

In addition to reading a lot, it is a very good idea to work on developing your
writing skills. This can be achieved by putting your ideas on paper or building
your own blog, where you can post regularly. This way, you can receive feedback
and improve the areas where you don’t sound so professional. Writing regularly
also helps you create your own style, which will give a touch of originality to your
work. Only with a lot of practice and hard work can professionalism be achieved!

Research about the writer’s style in advance

Another good advice for book translators is to  research about the writer of
the work in advance. Before interpreting a vast book, you must know the style of
the author in advance.
Try to read other publications of his in order to understand his personality
better. Afterwards, check the period when the book was written in order to
contextualize it accurately. Attention to details is a must if your goal is to reach
perfection in your work.

capítulo 4 • 107
To conclude, translating a book is the type of work that requires a lot of
effort and attention to details. But in the end, it can be extremely rewarding.
After all, being a part of promoting a valuable work that brings so many benefits
to its readers is a real privilege.

01. What are the five literary translation techniques?

02. How can we understand authorship from the intertextuality point of view?

03. How do you understand the difference between author and translator?

ACM. Efstathios Stamatatos. 2009. A survey of modern authorship attribution methods. Journal of the
American Society for Information Science and Technology, 60(3):538–556.
Andrew I. Schein, Johnnie F. Caver, Randale J. Honaker, and Craig H. Martell. 2010. Author attribution
evaluation with novel topic cross-validation. In Proceedings of the 2010 International Conference
on Knowledge Discovery and Information Retrieval (KDIR ’10).
Baker, Mona. 2009. “Ethics of renarration”, an interview with Andrew Chesterman, in Cultus 1(1):
Benno Stein, Moshe Koppel, and Efstathios Stamatatos, editors. 2007. Proceedings of the SIGIR
2007 International Workshop on Plagiarism Analysis, Authorship Identification, and Near-
Duplicate Detection, PAN 2007, Amsterdam, Netherlands, July 27, 2007, volume 276 of CEUR
Workshop Proceedings. CEURWS.org. Frank Yates. 1934. Contingency tables involving small
numbers and the – 2 test. Supplement to the Journal of the Royal Statistical Society, 1(2):pp.
217–235. 70
Bereiter, Carl and Marlene Scardamalia. 1987. The Psychology of Written Composition, Hillsdale,
NJ: Lawrence Erlbaum.
Chamberlain, Lori. 1988/2000. “Gender and the Metaphorics of Translation”, in Venuti, Lawrence (ed.)
The Translation Studies Reader, London and New York: Routledge: 314-29
David I. Holmes. 1992. A stylometric analysis of mormon scripture and related texts. Journal of the
Royal Statistical Society, Series A, 155(1):91–120.

108 • capítulo 4
Dipanjan Das, Nathan Schneider, Desai Chen, and Noah A. Smith. 2010. Probabilistic frame-semantic
parsing. In Proceedings of the North American Chapter of the Association for Compututional
Linguistics Human Language Technologies Conference (NAACL HLT ’10).
Englund Dimitrova, Birgitta. 2005. Expertise and Explicitation in the Translation Process,
Amsterdam and Philadelphia: John Benjamins.
Frederick Mosteller and David L. Wallace. 1964. Inference and Disputed Authorship: The Federalist.
Springer-Verlag, New York. 2nd Edition appeared in 1984 and was called Applied Bayesian and
Classical Inference.
Geoffrey Leech, Paul Rayson, and Andrew Wilson. 2001. Word Frequencies in Written and Spoken
English: Based on the British National Corpus. Longman, London.
Goffman, Erving. 1981. Forms of talk, Philadelphia: University of Pennsylvania Press. Habermas,
Jürgen. 1990. Moral consciousness and communicative action, trans. C. Lenhardt & S. Weber
Nicholson, Cambridge, MA: MIT Press.
Immonen, Sini. 2006. “Translation as a writing process. Pauses in translation versus
monolingual text production” in Target 18(2): 313-335. Jensen, Astrid. 2001. The effects of time
on cognitive processes and strategies in translation, Copenhagen: Copenhagen Business School,
Faculty of Modern Languages. Unpublished doctoral dissertation.
John B. Archer, John L. Hilton, and G. Bruce Schaalje. 1997. Comparative power of three author-
attribution techniques for differentiating authors. Journal of Book of Mormon Studies, 6(1):47–63.
Joseph Rudman. 1997. The state of authorship attribution studies: Some problems and solutions.
Computers and the Humanities, 31(4):351–365.
Joseph Ruppenhofer, Michael Ellsworth, Miriam R. L. Petruck, Christopher R. Johnson, and Jan
Scheffczyk. 2006. FrameNet II: Extended Theory and Practice. The Framenet Project.
Juola. 2006. Authorship attribution. Found. Trends Inf. Retr., 1(3):233–334.
Kai-Bo Duan and S. Sathiya Keerthi. 2005. Which is the best multiclass svm method? an empirical
study. In Proceedings of the Sixth International Workshop on Multiple Classifier Systems, pages
Kim Luyckx and Walter Daelemans. 2010. The effect of author set size and data size in authorship
attribution. Literary and Linguistic Computing. To appear.
Mark D. Smucker, James Allan, and Ben Carterette. 2007. A comparison of statistical significance
tests for information retrieval evaluation. In Proceedings of the sixteenth ACM conference on
Conference on information and knowledge management, CIKM ’07, pages 623–632, New York,
Michael Gamon. 2004. Linguistic correlates of style: Authorship classification with deep linguistic
analysis features. In Proceedings of the 20th International Conference on Computational Linguistics
(COLING ’04), pages 611–617.

capítulo 4 • 109
Moshe Koppel, Jonathan Schler, and Shlomo Argamon. 2008. Computational methods for
authorship attribution. Journal of the American Society for Information Sciences and
Technology, 60(1):9–25.
Moshe Koppel, Jonathan Schler, and Shlomo Argamon. 2010. Authorship attribution in the wild.
Language Resources and Evaluation, pages 1–12. 10.1007/s10579-009-9111-2. 69
Perteghella, Manuela and Eugenia Loffredo (eds). 2006. Translation and Creativity: Perspectives on
Creative Writing and Translation Studies, London and New York: Continuum. Pym, Anthony. 1997. Pour
une éthique du traducteur, Arras: Artois Presses Université.
Philip M. McCarthy, Gwyneth A. Lewis, David F. Dufty, and Danielle S. McNamara. 2006. Analyzing
writing styles with coh-metrix. In Proceedings of the International Conference of the Florida Artificial
Intelligence Research Society, pages 764–769.
Pym, Anthony. 1992/2010. Translation and Text Transfer. An Essay on the Principles of
Intercultural Communication, revised edition, Tarragona: Intercultural Studies Group. http://www.
tinet.cat/~apym/publications/publications.html Thorndike, Lynn. 1923-58. A History of Magic and
Experimental Science during the First Thirteen Centuries of our Era, 8 vols. London: Macmillan.
Pym, Anthony. 2004. “Propositions on Cross-Cultural Communication and Translation” in Target
16(1): 1-28.
Quinn McNemar. 1947. Note on the sampling error of the difference between correlated proportions or
percentages. Psychometrika, 12:153–157.
R. Arun, V. Suresh, and C. E. Veni Madhaven. 2009. Stopword graphs and authorship attribution in text
corpora. In Proceedings of the 3rd IEEE International Conference on Semantic Computing (ICSC
2009), pages 192–196, Berkeley, CA, USA, sep. IEEE Computer Society Press.
Richard Johansson and Pierre Nugues. 2007. Semantic structure extraction using nonprojective
dependency trees. In Proceedings of SemEval-2007, Prague, Czech Republic, June 23-24. Patrick
Shlomo Argamon. 2007. Interpreting Burrows’ Delta: Geometric and probabilistic foundations.
Literary and Linguistic Computing, 23(2):131–147.
Sindhu Raghavan, Adriana Kovashka, and Raymond Mooney. 2010. Authorship attribution using
probabilistic context-free grammars. In Proceedings of the ACL 2010 Conference Short Papers,
pages 38–42. Association for Computational Linguistics.
Torikai, Kumiko. 2009. Voices of the Invisible Presence. Diplomatic Interpreters in Post-World War II
Japan, Amsterdam and Philadelphia: John Benjamins.
Venuti, Lawrence. 1992. “Introduction” in Venuti, Lawrence (ed.) Rethinking Translation. Discourse,
Subjectivity, Ideology, London and New York: Routledge: 1-17. Venuti, Lawrence. The Translator’s
Invisibility, London and New York: Routledge.
Zeller, Beatriz. 2000. “On Translation and Authorship” in Meta 45(1): 134- 139

110 • capítulo 4
Examples of
from English to
5.  Examples of translations from English to

Sentences by Nelson Mandela

In this first example, we have a classic sentence by Nelson Mandela and its
translation. We can observe that, in this case, due to the fact that he uses formal
vocabulary, it becomes quite easy to have a good translation to Portuguese:

“I have fought against white domination, and I have fought against black domination. I
have cherished the ideal of a democratic and free society in which all persons will live
together in harmony and with equal opportunities. It is an ideal for which I hope to live
for and to see realized. But, if it needs to be, it is an ideal for which I am prepared to die.”

In Portuguese, using the translation offered by the website, we have:

“Eu lutei contra a dominação branca, e lutei contra a dominação negra. Eu tenho
prezado pelo ideal de uma sociedade democrática e livre, na qual todas as pessoas
possam viver juntas em harmonia e com iguais oportunidades. É um ideal pelo qual eu
espero viver e que eu espero alcançar. Mas, caso seja necessário, é um ideal pelo qual
eu estou pronto para morrer.”

We can observe that the biggest problem in this translation is the verb tense:
They translate the present perfect using the past in the first sentence. However,
I would dare saying that it would be much more suitable to use the present,
considering that Nelson Mandela was still fighting for humans rights at the
moment of this speech. Some adaptations are also possible of being observed.

112 • capítulo 5
Harry Potter and the Prisoner of Azkaban:

The owl post

Harry Potter was a highly unusual boy in many ways. For one thing, he hated the
summer holidays more than any other time of year. For another, he really wanted
to do his homework but was forced to do it in secret, in the dead of night. And he
also happened to be a wizard.
It was nearly midnight, and he was lying on his stomach in bed, the blankets drawn
right over his head like a tent, a flashlight in one hand and a large leather-bound
book (A History of Magic by Bathilda Bagshot) propped open against the pillow.
Harry moved the tip of his eagle-feather quill down the page, frowning as he
looked for something that would help him write his essay, “Witch Burning in the
Fourteenth Century Was Completely Pointless— discuss.”
The quill paused at the top of a likely-looking paragraph. Harry pushed his round
glasses up the bridge of his nose, moved his flashlight closer to the book, and read:
Non-magic people (more commonly known as Muggles) were particularly afraid of
magic in medieval times, but not very good at recognizing it. On the rare occasion
that they did catch a real witch or wizard, burning had no effect whatsoever. The
witch or wizard would perform a basic Flame- Freezing Charm and then pretend to
shriek with pain while enjoying a gentle, tickling sensation. Indeed, Wendelin the
Weird enjoyed being burned so much that she allowed herself to be caught no less
than forty seven times in various disguises.
Harry put his quill between his teeth and reached underneath his pillow for his
ink bottle and a roll of parchment. Slowly and very carefully he unscrewed the ink
bottle, dipped his quill into it, and began to write, pausing every now and then to
listen, because if any of the Dursleys heard the scratching of his quill on their
way to the bathroom, he’d probably find himself locked in the cupboard under the
stairs for the rest of the summer.

capítulo 5 • 113
In Portuguese:

O Correio-Coruja
Harry Potter era um menino bastante fora do comum em muitas coisas. Para co-
meçar, ele detestava as férias de verão mais do que qualquer outra época do ano.
Depois, ele realmente queria fazer seus deveres de casa, mas era obrigado a fazê
-los escondido, na calada da noite. E, além de tudo, também era bruxo.
Era quase meia-noite e Harry estava deitado de bruços na cama, as cobertas puxa-
das por cima da cabeça como uma barraca, uma lanterna em uma das mãos e um
grande livro encadernado em couro (História da Magia de Batilda Bagshot), aberto
e apoiado no travesseiro. Harry correu a ponta da caneta de pena de águia pela
página, franzindo a testa, à procura de alguma coisa que o ajudasse a escrever
sua redação, “A queima de bruxas no século XIV foi totalmente despropositada —
discuta”. A caneta pousou no alto de um parágrafo que pareceu a Harry promissor.
Ele empurrou os óculos redondos para a ponta do nariz, aproximou a lanterna do
livro e leu:
Os que não são bruxos (mais comumente conhecidos pelo nome de Trouxas)
tinham muito medo da magia na época Medieval, mas não tinham muita capaci-
dade para reconhecê-la. Nas raras ocasiões em que apanhavam um bruxo ou uma
bruxa de verdade, a sentença de queimá-los na fogueira não produzia o menor
efeito. O bruxo, ou bruxa, executava um Feitiço para Congelar Chamas e depois
fingia gritar de dor, enquanto sentia uma cocegazinha suave e prazerosa. De fato,
Wendelin a Esquisita gostava tanto de ser queimada na fogueira que se deixou
apanhar nada menos que quarenta e sete vezes, sob vários disfarces.
Harry prendeu a caneta entre os dentes e passou a mão embaixo do travesseiro
à procura do tinteiro e de um rolo de pergaminho. Devagar e com muito cuidado,
retirou a tampa do tinteiro, molhou a pena e começou a escrever, parando de vez
em quando para escutar, porque se algum dos Dursley, a caminho do banheiro, ou-
visse sua pena arranhando o pergaminho, ele provavelmente ia acabar trancafiado
no armário embaixo da escada pelo resto do verão.

114 • capítulo 5
We can see in this small fragment of the book that there are some adaptations
(highlighted) which are essential due to the fact that the languages present
different structures, especially when we take reference into consideration. In
English, pronouns make a better job than in Portuguese. Because of this, we can
have further reference, but in Portuguese, the description must be narrower.

One more part:

Harry, who happened to be in the room at the time, froze as he heard Ron’s
voice answer.
Ron was yelling so loudly that Uncle Vernon jumped and held the receiver a foot
away from his ear, staring at it with an expression of mingled fury and alarm.
“WHO IS THIS?” he roared in the direction of the mouthpiece. “WHO ARE YOU?”
“RON — WEASLEY!” Ron bellowed back, as though he and Uncle Vernon were
speaking from opposite ends of a football field.
Uncle Vernon’s small eyes swiveled around to Harry, who was rooted to the spot.
“THERE IS NO HARRY POTTER HERE!” he roared, now holding the receiver
at arm’s length, as though frightened it might explode. “I DON’T KNOW WHAT
And he threw the receiver back onto the telephone as if dropping a poisonous
spider. The fight that had followed had been one of the worst ever.
Uncle Vernon had roared, spraying Harry with spit.

capítulo 5 • 115
In Portuguese:

Harry que, por acaso, se achava na sala àquela hora, gelou ao ouvir a voz do
amigo responder.
Rony gritou com tanta força que tio Válter deu um salto e afastou o fone a mais de um
palmo da orelha com uma expressão em que se misturavam a fúria e o susto.
— QUEM É QUE ESTÁ FALANDO? — berrou ele em direção ao bocal. — QUEM
— RONY... WEASLEY! — berrou Rony em resposta, como se ele e tio Válter
estivessem falando de extremidades opostas de um campo de futebol.
Os olhinhos de tio Válter se viraram para Harry, que estava pregado no chão.
— NÃO TEM NENHUM HARRY POTTER AQUI! — vociferou ele, agora segurando
o fone com o braço esticado, como se receasse que o aparelho pudesse explodir. —
E atirou o fone no gancho como se estivesse se livrando de uma aranha venenosa. A
briga que se seguiu foi uma das piores da vida de Harry.
GENTE COMO VOCÊ! — berrara tio Válter, salpicando Harry de cuspe.

In this part, we can observe some cultural aspects related to translation

(highlighted) and how much a translator needs to know about both languages/
cultures in order to produce a successful translation.

Now, some examples from “Harry Potter and the Order of the Phoenix”

116 • capítulo 5
The hottest day of the summer so far was drawing to a close and a drowsy silence
lay over the large, square houses of Privet Drive. Cars that were usually gleaming
stood dusty in their drives and lawns that were once emerald green lay parched and
yellowing; the use of hosepipes had been banned due to drought. Deprived of their
usual car-washing and lawn-mowing pursuits, the inhabitants of Privet Drive had
retreated into the shade of their cool houses, windows thrown wide in the hope of
tempting in a nonexistent breeze. The only person left outdoors was a teenage boy
who was lying flat on his back in a flower bed outside number four.
He was a skinny, black-haired, bespectacled boy who had the pinched, slightly
unhealthy look of someone who has grown a lot in a short space of time. His jeans
were torn and dirty, his T-shirt baggy and faded, and the soles of his trainers were
peeling away from the uppers. Harry Potter’s appearance did not endear him to
the neighbors, who were the sort of people who thought scruffiness ought to be
punishable by law, but as he had hidden himself behind a large hydrangea bush this
evening he was quite invisible to passersby. In fact, the only way he would be spotted
was if his Uncle Vernon or Aunt Petunia stuck their heads out of the living room
window and looked straight down into the flower bed below.
On the whole, Harry thought he was to be congratulated on his idea of hiding here.
He was not, perhaps, very comfortable lying on the hot, hard earth, but on the other
hand, nobody was glaring at him, grinding their teeth so loudly that he could not hear
the news, or shooting nasty questions at him, as had happened every time he had tried
sitting down in the living room and watching television with his aunt and uncle.
Almost as though this thought had fluttered through the open window, Vernon
Dursley, Harry’s uncle, suddenly spoke. “Glad to see the boy’s stopped trying to butt in.
Where is he anyway?”
“I don’t know,” said Aunt Petunia unconcernedly. “Not in the house.”
Uncle Vernon grunted.
“Watching the news . . .” he said scathingly. “I’d like to know what he’s really up to. As
if a normal boy cares what’s on the news — Dudley hasn’t got a clue what’s going on,
doubt he knows who the Prime Minister is! Anyway, it’s not as if there’d be anything
about his lot on our news —”
“Vernon, shh!” said Aunt Petunia. “The window’s open!”
“Oh — yes — sorry, dear . . .”
The Dursleys fell silent.

capítulo 5 • 117
In Portuguese:

O dia de verão mais quente do ano estava chegando ao fim e um silêncio modorrento
pairava sobre os casarões quadrados da rua dos Alfeneiros. Os carros, em geral reluzentes,
estavam empoeirados nas entradas das garagens, e os gramados, que tinham sido verde-
esmeralda, estavam ressequidos e amarelos – porque o uso de mangueiras fora proibido
durante a estiagem. Privados das atividades de lavar carros e cortar gramados, os habitantes
da rua dos Alfeneiros haviam se recolhido à sombra de suas casas frescas, as janelas
escancaradas na esperança de atrair uma brisa inexistente. A única pessoa do lado de fora
era um adolescente deitado de costas em um canteiro de flores à frente do número quatro.
Era um garoto magricela, de cabelos pretos, a aparência macilenta e meio doentia de
alguém que cresceu muito em pouco tempo. Suas jeans estavam rotas e sujas, a camiseta
larga e desbotada, e as solas dos tênis se soltavam da parte de cima. A aparência de Harry
Potter não o recomendava aos vizinhos, que eram do tipo que achava que devia haver uma
punição legal para sujeira e desleixo, mas como ele se escondera atrás de uma repolhuda
hortênsia, esta noite ele estava invisível aos que passavam. De fato, a única maneira de
localizá-lo era se o tio Válter ou a tia Petúnia metessem a cabeça pela janela da sala de
estar e olhassem diretamente para o canteiro embaixo.
No todo, Harry achava que devia receber parabéns pela ideia de se esconder ali. Não
estava, talvez, muito confortável, deitado na terra quente e dura, mas, por outro lado,
ninguém estava olhando para ele, rangendo os dentes tão alto que o impedia de ouvir o
noticiário, nem disparando perguntas incômodas, como acontecera todas as vezes em que
tentou se sentar na sala de estar para ver televisão com os tios.
Quase como se tais pensamentos tivessem entrado pela janela aberta, repentinamente
Válter Dursley, o tio de Harry, falou:
– Fico contente de ver que o garoto parou de se meter aqui. Por falar nisso, onde será que
ele anda?
– Não sei – respondeu tia Petúnia, desinteressada. – Aqui em casa não está.
O tio grunhiu.
– Assistir ao noticiário... – comentou com severidade. – Gostaria de saber o que é que ele
está realmente aprontando. Como se um garoto normal se interessasse por noticiário; Duda
não tem a mínima ideia do que está acontecendo; duvido que saiba quem é o primeiro-
ministro! Em todo o caso, não há nada sobre gente da laia dele no nosso noticiário...
– Válter, psiu! – alertou tia Petúnia. – A janela está aberta!
– Ah... é... desculpe, querida.
Os Dursley se calaram.

118 • capítulo 5
In this example, the only adaptation we found is that in the original book,
we don’t have “street” written; on the whole, the translation is very similar to
the original.
One more example:

What?” said Harry blankly.

“He left!” said Mrs. Figg, wringing her hands. “Left to see someone about a batch of
cauldrons that fell off the back of a broom! I told him I’d flay him alive if he went, and
now look! Dementors! It’s just lucky I put Mr. Tibbies on the case! But we haven’t got
time to stand around! Hurry, now, we’ve got to get you back! Oh, the trouble this is
going to cause! I will kill him!”
“But —”
The revelation that his batty old cat-obsessed neighbor knew what dementors were
was almost as big a shock to Harry as meeting two of them down the alleyway. “You’re
— you’re a witch?”
“I’m a Squib, as Mundungus knows full well, so how on earth was I supposed to help
you fight off dementors? He left you completely without cover when I warned him —”
“This bloke Mundungus has been following me? Hang on — it was him! He
Disapparated from the front of my house!”
“Yes, yes, yes, but luckily I’d stationed Mr. Tibbies under a car just in case, and Mr.
Tibbies came and warned me, but by the time I got to your house you’d gone — and
now — oh, what’s Dumbledore going to say? You!” she shrieked at Dudley, still supine
on the alley floor.
“Get your fat bottom off the ground, quick!”
“You know Dumbledore?” said Harry, staring at her.
“Of course I know Dumbledore, who doesn’t know Dumbledore?
But come on — I’ll be no help if they come back, I’ve never so much as Transfigured
a teabag —” She stooped down, seized one of Dudley’s massive arms in her wizened
hands, and tugged.
“Get up, you useless lump, get up!”
But Dudley either could not or would not move. He was still on the ground, trembling
and ashen-faced, his mouth shut very tight. “I’ll do it.” Harry took hold of Dudley’s arm
and heaved: With an enormous effort he managed to hoist Dudley to his feet. Dudley
seemed to be on the point of fainting: His small eyes were rolling in their sockets and
sweat was beading his face; the moment Harry let go of him he swayed dangerously.
“Hurry up!” said Mrs. Figg hysterically.

capítulo 5 • 119
In Portuguese:

– Quê? – exclamou Harry sem entender.

– Ele saiu – respondeu a Sra. Figg, torcendo as mãos. – Saiu para ver alguém a propósito
de uma remessa de caldeirões que caiu da garupa de uma vassoura! Eu disse que o
esfolaria vivo se ele fosse, e agora veja o que aconteceu! Dementadores! Foi uma sorte eu
ter posto o Sr. Tibbles no caso! Mas não temos tempo para ficar parados! Corra agora, você
tem de voltar para casa! Ah, a confusão que isso vai provocar! Eu vou matar aquele homem!
A revelação de que sua vizinha gagá com mania de gatos sabia o que eram dementadores
foi quase um choque tão grande para Harry quanto encontrar dois deles na travessa.
– A senhora é bruxa?
– Não consegui ser, e Mundungo sabe muito bem disso, então como é que eu ia poder
ajudar a espantar os dementadores? Ele deixou você completamente descoberto e eu
o avisei...
– Esse tal Mundungo andou me seguindo? Espere aí... foi ele! Desaparatou na frente
da minha casa!
– Isso mesmo, mas por sorte eu tinha mandado o Sr. Tibbles ficar debaixo de um carro,
só por precaução, e ele veio me avisar, mas quando cheguei você já tinha saído de
casa... e agora... ah, que é que o Dumbledore vai dizer? Você! – gritou ela para Duda,
ainda inerte no chão da travessa. – Levanta essa bunda gorda do chão, anda logo!
– A senhora conhece Dumbledore? – perguntou Harry olhando fixamente para
a velhota.
– Claro que conheço Dumbledore, quem não conhece Dumbledore? Mas vamos logo.
Não vou poder ajudar você se eles voltarem. Eu nunca consegui transfigurar nem um
saquinho de chá.
Ela se abaixou, agarrou o braço maciço de Duda com as mãos enrugadas e puxou.
– Levanta, seu monte de carne inútil, levanta!
Mas Duda ou não podia ou não queria se mexer. Continuou no chão, trêmulo, de cara
pálida, a boca hermeticamente fechada.
– Eu faço isso. – Harry agarrou Duda pelo braço e puxou. Com enorme esforço,
conseguiu pô-lo de pé. Duda parecia prestes a desmaiar. Seus olhos miúdos giravam
nas órbitas e o suor gotejava em seu rosto; no momento em que Harry o largou, ele
balançou precariamente.
– Anda depressa! – disse a Sra. Figg, nervosa.

120 • capítulo 5
In this part, we can see some omissions in the Portuguese translation,
probably to keep the dialogue flowing fast, as it is in English.

5.1  Examples of translations from Portuguese to English

Conto “A benfazeja” de Primeiras estórias.

Tradutora: Barbara Shelby
The intention to bring Guimarães Rosa is to show how a translator dealt
with one of the most difficult authors of the Portuguese language. It is possible
to observe the difficulty to create the same meaning created in a genius form
by him.

Vive-se perto demais, num lugarejo, às sombras frouxas, a gente se afaz ao devagar
das pessoas. A gente não revê os que não valem a pena. Acham ainda que não valia a
pena? Se, pois, se. No que nem pensaram; e não se indagou, a muita coisa. Para quê?”
(ROSA, 1968, p.125)

“People live so close together in a little place like this, like impotent shadows, that they
take other people’s habits for granted. And when someone doesn’t seem interesting
enough to bother about, they stop noticing him at all. But do you really believe that
she wasn’t worth troubling about? If you do – well, it’s because you’ve never really
thought about her at all, or wondered much about her, either. After all, why should you?”
(SHELBY, 1968, p. 57)

“Vocês todos nunca suspeitaram que ela pudesse arcar-se no mais fechado extremo,
nos domínios do demasiado?” (p. 125)

“None of you ever suspected, did you, that she might have taken on a burden too heavy
for anyone to bear?”(p. 57)

capítulo 5 • 121
“Como? O amor é a vaga, indecisa palavra.” (p. 127)

“If you ask how that was possible, I will simply answer that ‘love’ is the vaguest, most
indefinite of words.” (p. 61)

Some parts of “The Zahir”, by Paulo Coelho:

Ela, Esther, correspondente de guerra recém-chegada do Iraque porque a invasão do

país deve acontecer a qualquer momento, 30 anos, casada, sem filhos. Ele, um homem
não identificado, aproximadamente 25 anos, moreno, traços mongóis. Os dois foram
vistos pela última vez em um café na rua Faubourg Saint-Honoré.
A polícia foi informada de que já haviam se encontrado antes, embora ninguém
soubesse quantas vezes: Esther sempre comentara que o homem – cuja identidade
ocultava sob o nome de Mikhail – era alguém muito importante, embora jamais tenha
explicado se era importante para sua carreira de jornalista ou para ela, como mulher.
A polícia iniciou um inquérito formal. Foram aventadas as possibilidades de sequestro,
chantagem, sequestro seguido de morte – o que não seria absolutamente de se
estranhar, já que seu trabalho a obrigava a estar frequentemente em contato com
pessoas ligadas a células terroristas, em busca de informação. Descobriram que
sua conta bancária indicava saques regulares nas semanas anteriores ao seu
desaparecimento: os investigadores consideraram que isso poderia estar ligado a
pagamento de informação. Não havia levado nenhuma roupa, mas, curiosamente, seu
passaporte não foi encontrado.
Ele, um desconhecido, muito jovem, sem nenhum registro na polícia, sem nenhuma
pista que permitisse sua identificação.
Ela, Esther, dois prêmios internacionais de jornalismo, 30 anos, casada. Minha mulher.
Sou colocado imediatamente sob suspeita e detido – já que me recusava a dizer meu
paradeiro no dia do seu desaparecimento. Mas o carcereiro acaba de abrir a porta e
dizer que sou um homem livre.

122 • capítulo 5
Por que sou um homem livre? Porque hoje em dia todos sabem tudo de todo mundo,
basta desejar a informação e ela está ali: onde o cartão de crédito foi usado, quais
os lugares que frequentamos, com quem dormimos. No meu caso, foi mais fácil: uma
mulher, também jornalista, amiga de minha mulher, mas divorciada – e, portanto, sem
problemas em dizer que estava dormindo comigo –, se ofereceu para testemunhar a
meu favor ao saber que eu tinha sido preso. Deu provas concretas de que eu estava
com ela no dia e na noite do desaparecimento de Esther.
Vou conversar com o inspetor-chefe, que devolve minhas coisas, pede desculpas,
afirma que minha rápida detenção foi feita com base na lei, e que não poderei acusar
ou processar o Estado. Explico que não tenho a menor intenção de fazer isso, sei que
qualquer pessoa está sempre sob suspeita e sendo vigiada 24 horas por dia, mesmo
que não tenha cometido nenhum crime.
– Você está livre – diz, repetindo as palavras do carcereiro.
Pergunto: não é possível que algo realmente tenha ocorrido com minha mulher? Ela já
havia comentado comigo que, por causa de sua enorme teia de contatos no submundo
do terrorismo, vez por outra sentia que seus passos estavam sendo acompanhados de

Now, in English:

Her name is Esther; she is a war correspondent who has just returned from Iraq
because of the imminent invasion of that country; she is thirty years old, married,
without children. He is an unidentified male, between twenty-three and twenty-five
years old, with dark, Mongolian features. The two were last seen in a café on the Rue
du Faubourg St-Honoré.
The police were told that they had met before, although no one knew how often:
Esther had always said that the man—who concealed his true identity behind the
name Mikhail—was someone very important, although she had never explained
whether he was important for her career as a journalist or for her as a woman.

capítulo 5 • 123
The police began a formal investigation. Various theories were put forward—kidnapping,
blackmail, a kidnapping that had ended in murder—none of which were beyond the
bounds of possibility given that, in her search for information, her work brought her into
frequent contact with people who had links with terrorist cells. They discovered that, in
the weeks prior to her disappearance, regular sums of money had been withdrawn from
her bank account: those in charge of the investigation felt that these could have been
payments made for information. She had taken no change of clothes with her, but, oddly
enough, her passport was nowhere to be found.
He is a stranger, very young, with no police record, with no clue as to his identity.
She is Esther, thirty years old, the winner of two international prizes for journalism, and
married. My wife.
I immediately come under suspicion and am detained because I refuse to say where I
was on the day she disappeared. However, a prison officer has just opened the door of
my cell, saying that I’m a free man.
And why am I a free man? Because nowadays, everyone knows everything about
everyone; you just have to ask and the information is there: where you’ve used your
credit card, where you spend your time, whom you’ve slept with. In my case, it was even
easier: a woman, another journalist, a friend of my wife, and divorced—which is why she
doesn’t mind revealing that she slept with me—came forward as a witness in my favor
when she heard that I had been detained. She provided concrete proof that I was with
her on the day and the night of Esther’s disappearance.
I talk to the chief inspector, who returns my belongings and offers his apologies, adding
that my rapid detention was entirely within the law, and that I have no grounds on which
to accuse or sue the state. I say that I haven’t the slightest intention of doing either of
those things, that I am perfectly aware that we are all under constant suspicion and
under twenty-four-hour surveillance, even when we have committed no crime.
“You’re free to go,” he says, echoing the words of the prison officer.
I ask: Isn’t it possible that something really has happened to my wife? She had said
to me once that—understandably given her vast network of contacts in the terrorist
underworld—she occasionally got the feeling she was being followed.

It is clear that when the text is simpler, the translation seems to be easier
and much closer to the original.

124 • capítulo 5
“Eleven minutes”

Era uma vez uma prostituta chamada Maria. Um momento. “Era uma vez” é a melhor
maneira de começar uma história para crianças, enquanto “prostituta” é assunto para
adultos. Como posso escrever um livro com esta aparente contradição inicial? Mas,
enfim, como a cada instante de nossa vida temos um pé no conto de fadas e outro
no abismo, vamos manter este início: Era uma vez uma prostituta chamada Maria.
Como todas as prostitutas, tinha nascido virgem e inocente, e durante a adolescência
sonhara em encontrar o homem de sua vida (rico, bonito, inteligente), casar (vestida
de noiva), ter dois filhos (que seriam famosos quando crescessem) e viver em uma
linda casa (com vista para o mar). Seu pai trabalhava como vendedor ambulante, sua
mãe era costureira, sua cidade no interior do Brasil tinha apenas um cinema, uma
boate e uma agência bancária. Por isso Maria não deixava de esperar o dia em que
seu príncipe encantado chegaria sem aviso, arrebataria seu coração e partiria com
ela para conquistar o mundo. Enquanto seu herói não aparecia, só lhe restava sonhar.
Apaixonou-se pela primeira vez aos onze anos, quando ia a pé de casa até a escola
primária local. No primeiro dia de aula, descobriu que não estava sozinha em seu trajeto:
junto com ela caminhava um garoto que vivia na vizinhança e frequentava aulas no
mesmo horário. Os dois nunca trocaram uma só palavra, mas Maria começou a notar
que a parte do dia que mais lhe agradava eram aqueles momentos na estrada cheia de
poeira, sede, cansaço, o sol a pino, o menino andando rápido, enquanto ela se exauria
no esforço para acompanhar-lhe os passos. A cena se repetira por vários meses. Maria,
que detestava estudar e não tinha outra distração na vida exceto a televisão, começou
a torcer para que o dia passasse rápido, aguardando com ansiedade cada ida à escola
e, ao contrário de algumas meninas de sua idade, achando aborrecidíssimos os fins de
semana. Como as horas demoram muito mais a passar para uma criança do que para
um adulto, ela sofria muito, achava os dias longos demais porque lhe davam apenas
dez minutos com o amor de sua vida e milhares de horas para ficar pensando nele,
imaginando como seria bom se pudessem conversar. Então aconteceu. Certa manhã,
o garoto veio até ela, pedindo um lápis emprestado. Maria não respondeu, fez um ar de
irritação por aquela abordagem inesperada e apressou o passo. Tinha ficado petrificada
ao vê-lo caminhar em sua direção; tinha pavor de que soubesse quanto o amava, quanto
esperava por ele, como sonhava em pegar sua mão, passar diante do portão da escola
e seguir até o fim da estrada, onde – diziam – se encontravam uma grande cidade,
personagens de novela, artistas, carros, cinemas e um sem-fim de coisas boas.

capítulo 5 • 125
In English

Once upon a time, there was a prostitute called Maria. Wait a minute. ‘Once upon a time’
is how all the best children’s stories begin and ‘prostitute’ is a word for adults. How can I
start a book with this apparent contradiction? But since, at every moment of our lives, we
all have one foot in a fairy tale and the other in the abyss, let’s keep that beginning. Once
upon a time, there was a prostitute called Maria. Like all prostitutes, she was born both
innocent and a virgin, and, as an adolescent, she dreamed of meeting the man of her life
(rich, handsome, intelligent), of getting married (in a wedding dress), having two children
(who would grow up to be famous) and living in a lovely house (with a sea view). Her father
was a travelling salesman, her mother a seamstress, and her hometown, in the interior of
Brazil, had only one cinema, one nightclub and one bank, which was why Maria was always
hoping that one day, without warning, her Prince Charming would arrive, sweep her off her
feet and take her away with him so that they could conquer the world together. While she
was waiting for her Prince Charming to appear, all she could do was dream. She fell in love
for the first time when she was eleven, en route from her house to school. On the first day
of term, she discovered that she was not alone on her way to school: making the same
journey was a boy who lived in her neighbourhood and who shared the same timetable.
They never exchanged a single word, but gradually Maria became aware that, for her,
the best part of the day were those moments spent going to school: moments of dust,
thirst and weariness, with the sun beating down, the boy walking fast, and with her trying
her hardest to keep up. This scene was repeated month after month; Maria, who hated
studying and whose only other distraction in life was television, began to wish that the days
would pass quickly; she waited eagerly for each journey to school and, unlike other girls her
age, she found the weekends deadly dull. Given that the hours pass more slowly for a child
than for an adult, she suffered greatly and found the days far too long simply because they
allowed her only ten minutes to be with the love of her life and thousands of hours to spend
thinking about him, imagining how good it would be if they could talk. Then it happened.
One morning, on the way to school, the boy came up to her and asked if he could borrow a
pencil. Maria didn’t reply; in fact, she seemed rather irritated by this unexpected approach
and even quickened her step. She had felt petrified when she saw him coming towards her,
terrified that he might realise how much she loved him, how eagerly she had waited for him,
how she had dreamed of taking his hand, of walking straight past the school gates with
him and continuing along the road to the end, where - people said there was a big city, film
stars and television stars, cars, lots of cinemas, and an endless number of fun things to do.

126 • capítulo 5
01. Now, it is your turn:
a) Durante o resto do dia não conseguiu concentrar-se na aula, sofrendo com seu
comportamento absurdo, mas ao mesmo tempo sentindo-se aliviada, porque sabia que
o menino também a havia notado e o lápis não passara de um pretexto para iniciar uma
conversa, pois quando ele se aproximara ela percebera uma caneta em seu bolso. Ficou
aguardando a próxima vez, e durante aquela noite – e as noites que se seguiram – ela
passou a imaginar as possíveis respostas que lhe daria, até encontrar a maneira certa de
começar uma história que não terminasse jamais. (Paulo Coelho)
b) “No need to tell us he’s no good,” snorted Uncle Vernon, staring over the top of his
newspaper at the prisoner. “Look at the state of him, the fi lthy layabout! Look at his hair!”
He shot a nasty look sideways at Harry, whose untidy hair had always been a source
of great annoyance to Uncle Vernon. Compared to the man on the television, however,
whose gaunt face was surrounded by a matted, elbow-length tangle, Harry felt very well
groomed indeed. The reporter had reappeared. “The Ministry of Agriculture and Fisheries
will announce today —”
“Hang on!” barked Uncle Vernon, staring furiously at the reporter.
“You didn’t tell us where that maniac’s escaped from! What use is that? Lunatic could be
coming up the street right now!”
Aunt Petunia, who was bony and horse-faced, whipped around and peered intently out
of the kitchen window. Harry knew Aunt Petunia would simply love to be the one to call the
hot line number.
She was the nosiest woman in the world and spent most of her life spying on the boring,
law-abiding neighbors. (J.K. Rowling)

COELHO, Paulo. Onze Minutos. http://www.martinsfontespaulista.com.br/anexos/produtos/
_____________. The Zahir. http://www.kkoworld.com/kitablar/Paulo_Koelyo_Zair_eng.pdf
____________. Eleven Minutes. https://raisuman123.files.wordpress.com/2013/05/eleven_minutes.

capítulo 5 • 127
Rowling, J. K. Harry Potter e o Prisioneiro de Azkaban. Tradução: Lia WYLER
___________. Harry Potter e a Ordem de Fênix. Tradução: Lia WYLER.

Capítulo 1

a) Every organization, as soon as it gets to any size (perhaps 1,000 people), begins to feel a
need to systematize its management of human assets. Perhaps the pay scales have gotten
way out of line, with apparently similar-level jobs paying very different amounts; perhaps
there is a feeling that there are a lot of neglected skills in the organization that other
departments could utilize if they were aware that they existed. Perhaps individuals have
complained that they don’t know where they stand or what theirfuture is; perhaps the
unions have requested standardized benefits and procedures. Whatever the historical
origins, some kind of central organization, normally named a personnel department, is
formed to put some system into the haphazardry. The systems that they adopt are often
modelled on the world of production, because that is the world with the best potential
for order and system.
b) We all tend to complain about our memories. Despite the elegance of the human memory
system, it is not infallible, and we have to learn to live with its fallibility. It seems to be socially
much more acceptable to complain of a poor memory, and it is somehow much more
acceptable to blame a social lapse on ‘a terrible memory’, than to attribute it to stupidity
or insensitivity. But how much do we know about our own memories? Obviously we need
to remember our  memory lapses in order to know just how bad our memories are.
Indeed one of the most amnesic patients I have ever tested was a lady suffering from
Korsakoff’s syndrome, memory loss following chronic alcoholism. The test involved
presenting  her  with lists of words; after each list shewould comment with surprise
on her inability to recall the words, saying: ‘I pride myself on my memory!’ She appeared
to have forgotten just how bad her memory was’.

Substitution and ellipsis

Identify examples of substitution and ellipsis in this text:
c) The human memory system is remarkably efficient, but it is of course extremely fallible.
That being so, it makes sense to take full advantage of memory aids to minimize the

128 • capítulo 5
disruption caused by such lapses. If external aids are used, it is sensible to use them
consistently and systematically - always put appointments in your diary, always add
wanted items to a shopping list, and so on. If you use internal aids such as mnemonics,
you must be prepared to invest a reasonable amount of time in mastering and practicing
them. Mnemonics are like tools and cannot be used until forged. Overall, however, as
William James pointed out (the italics are mine): ‘Of two men with the same outward
experiences and the same amount of mere native tenacity, the one who thinks over his
experiences most and weaves them into systematic relations with each other will be the
one with the best memory.’
d) This conflict between tariff reformers and free traders was to lead to the “agreement
to differ” convention in January 1932, and the resignation of the Liberals from the
government in September 1932; but, until they resigned, the National Government
was a genuine coalition in the sense in which that term is used on the continent:
a  government  comprising independent yet conflicting elements allied together,
a government within which party conflict was not superseded but rather contained - in
short, a power-sharing government, albeit a seriously unbalanced one.
e) The number of different words relating to ‘camel’ is said to be about six thousand. There
are terms to refer to riding camels, milk camels and slaughter camels; other terms to
indicate the pedigree and geographical origin of the camel; and still others to differentiate
camels in different stages of pregnancy and to specify in-numerable other characteristics
important to a people so dependent upon camels in their daily life (Thomas, 1937)
“others” = “other terms”
f) There were, broadly, two interrelated reasons for this, the first relating to Britain’s economic
and imperial difficulties, the second to the internal dissension in all three parties.
“the first” = “the first reason” and “the second” = “the second reason”
g) These two forms of dissent coalesced in the demand for a stronger approach to the
Tory nostrum of tariff reform. In addition, trouble threatened from the mercurial figure
of Winston Churchill, who had resigned from the Shadow Cabinet in January 1931 in
protest at Baldwin’s acceptance of eventual self-government for India.
h) These two sets of rules, though distinct, must not be looked upon as two co-ordinate and
independent systems. On the contrary, the rules of Equity are only a sort of supplement or
appendix to the Common Law; they assume its existence but they add something further.
i) The clamour of complaint  about teaching in higher education and, more especially,
about teaching methods in universities and technical colleges, serves to direct attention
away from the important reorientation which has recently begun. The complaints, of
course, are not unjustified. In dealing piece-meal with problems arising from rapidly

capítulo 5 • 129
developing subject matter, many teachers have allowed courses to become over-
crowded, or too specialized, or they have presented students with a number of apparently
unrelated courses failing to stress common principles. Many, again, have not developed
new  teaching methods to deal adequately with larger numbers of students, and the
new audio-visual techniques tend to remain in the province of relatively few enthusiasts
despite their great potential for class and individual teaching.
j) When we look closely at a human face we are aware of many expressive details - the
lines of the forehead, the wideness of the eyes, the curve of the lips, the jut of the chin.
These elements combine to present us with a total facial expression which we use to
interpret the mood of our companion. But we all know that people can ‘put on a happy
face’ or deliberately adopt a sad face without feeling either happy or sad. Faces can lie,
and sometimes can lie so well that it becomes hard to read the true emotions of their
owners. But there is at least one facial signal that cannot easily be ‘put on’. It is a small
signal, and rather a subtle one, but because it tells the truth it is of special interest. It
comes from the pupils and has to do with their size in relation to the amount of light that
is falling upon them.

Capítulo 2

a) Well, in a classroom basically there aren’t many things. Normally there’s a black
board, where teachers write and teach their classes, an eraser, chalk or markers,
because some boards are white and require markers. There are the student’s desks,
too. Because I studied in a military school, there were flags and other things related
to militarism in my classroom, but generally these things don’t exist in typical schools.
In college, there are just these things: desks; some colleges sometimes have books
and a place to put them, sometimes a TV, a VCR to play movies, a projector, but
aside from these, there aren’t many things in a classroom.
b) Well, people describe this color in different ways: pumpkin, orange, I call it orange.
Here is black, which is a basic color. This one is rose or pink. Actually, people call it
pink because it’s a bright rose. This is white – a color that everybody wears... Blue,
can you see it? The blue. Here there’s a light rose, and I think that basically there is
still the red that you can see. I don’t know if you can see the red. It’s so small, but also
an important color. Here, the color of the hair – brunette – almost black or brown,
and here brown which, is also a color.
c) Well, in Portuguese we call the four basic math operations addition, subtraction,
multiplication and division. The verb for addition would be to add, “I can add things”

130 • capítulo 5
or “to add things”, and this is an universal concept that I believe in Portuguese and
in Brazil is more or less the same as in other places, in other countries of the world,
at least in western civilization. Regarding subtraction, we have the verbs to subtract,
to diminish or to take away, which all mean the same thing. For division, we have
the verb to divide. Maybe other verbs with the same meaning might be to split or...
Well, that would be the basic idea, to divide; the basic verb would be to divide. For
multiplication, we have the verb to multiply... I also think this verb would mean the
same thing as the verb for multiplication in Portuguese. For example, in addition,
we use the “plus” sign for mathematics operations, so we can say “three plus three
is equal to six”. Relating to subtraction, we use the “minus” sign, so we say “five
minus two is equal to three.” These are things that are common in daily speech in
Brazil. For multiplication, the sign is “x”, and we orally say “times”, so “two times two
is equal to four.” And for division we have “divided by”, that is to say “nine divided by
three is equal to three” or “nine over three is equal to three.” These would be some
basic examples.
d) My father was an accountant and he worked in accounting for a supermarket for
many years. Now that he is getting older, he started making jewelry. He decided to
make trinkets, and it’s really funny because he is the macho type, do you understand?
So, he decided to make trinkets and he is doing them really well. And my mom was a
Portuguese teacher. Then she took another course and began to work as a student
counselor. She really enjoyed working with education, and then... Now she is retired.
She wrote a book about education and how to... It is to help students interact with
their teachers and their parents.
e) My mother is pretty tall, she has curly dark brown hair with a tint of red, she is fairly
thin, and she is medium height. My younger sister is also very tall, taller than me. She
is a little bit chubby, she has blonde hair, freckles, and very light skin. My older sister
is a little bit shorter, dark-skinned, has a round face, but a very beautiful smile. She
is a little bit chubby. I’m tall, my hair is naturally curly, very curly, although today it’s
straight. I have brown eyes because nobody in my family has light eyes, blue eyes,
for example, and I’m dark-skinned. My father is also very tall, he has short curly dark
hair, very curly hair, brown eyes and is dark-skinned. My grandparents are a little
bit different. My grandmother is paler, she looks European. She has blonde hair –
actually, light brown hair – and very light skin.

capítulo 5 • 131
Capítulo 3

01. A dynamic equivalence, as defined by Nida, is to reproduce “in the receptor language the
closest natural equivalence of the source-language message…”(Nida and Taber, 1969: 12).
The key words are “closest”, “natural” and “equivalence”. By “closest”, he indicates that owing
to the impossibility of absolute equivalence, the “closest” equivalence is the most ideal one.
Either the meaning or form should not sound “foreign”. The essence of dynamic equivalence
is the receptor’s response, in Nida’s own terms, “the degree to which the receptors of the
message in the receptor language respond to it in substantially the same manner as the
receptors in the source language” (Nida and Taber, 1969: 68).
02. Formal correspondence is a term used in contrastive analysis, while translation
equivalence belongs to the metalanguage of translation. In principle, perhaps, the two terms
could be discussed separately in their two disciplines, and it is indeed possible to imagine a
theory of translation which would operate with the concept of equivalence defined without
reference to formal correspondence, just as it is possible to imagine contrastive analysis which
would rely on the concept of correspondence established without the use of translation. In
practice, however, both terms have been found necessary by students of translation and by
contrastive analysis.
03. The difference between language-based and text-based (or system-based and
equivalence-based) in formal correspondence is seen in the fact that while the former type of
correspondents stand in a one-to-one relationship, the relationship in the latter type is one-to-
many. Typically, a given formal element of the source language, when used in different texts
produced in different communicative situations, will have several target-language formal
elements which will correspond to it in translated target texts. But one should be aware
that, precisely for that reason, the formal elements which are correspondent in translationally
equivalent texts are never matched in totality, as they would be if parts of the systems of the
two languages were contrasted. Rather, they are matched in those of their meanings with
which they participate in the particular source and target texts.
04. Semiotics (the science investigating the general properties of sign systems) distinguishes
the following types of relations: semantic (sign to object), syntactic (sign to sign) and
pragmatic (sign to man). One of the most essential requirements imposed on translation is
that the two texts (the original and its translation) should be semantically equivalent. The goal
of translation is to produce a text, bearing the same relation to the extra-linguistic situation
as the original.
05. While cohesion concerns the surface relations that organize and create a text, coherence
is the network of conceptual relations which underlie the surface text as perceived by the
language users. The mere presence of cohesive markers, such as linkers or lexical chains, is
not sufficient to create a coherent text.

132 • capítulo 5
Capítulo 4

01. Adaptation: Albir describes adaptation as a “technique whereby one cultural element is
replaced by another which is typical of the receiving culture. This technique is very useful when
translating advertisements, slogans, etc., which employ a number of different linguistic processes.
In these cases, the most important thing is the actual meaning of the message rather than the
words making it up.” Linguistic Amplification: According to Albir, “this translation technique
adds new linguistic elements in the target text. It is the opposite of the linguistic compression
technique.” This is usually about using a paraphrase to explain a word that has no equivalent
in the target language. Compensation: Compensation, on the other hand, is a “translation
technique whereby a piece of information or stylistic device is moved to another location in the text, 
because it does not have the same effect if maintained in the same place as in the original text”.
This process is intended to compensate for the losses that a text suffers when it is translated.
The technique is especially useful when it comes to wordplay: If the translator cannot directly
adapt a pun, for instance, which tends to happen quite often, then they will try to create another
play on words in another part of the text. Elision: The fourth technique of literary translation
described by Albir is elision. Elision is a process that “involves removing items of information
in the original language text so that they do not appear in the target text. As with the linguistic
compression technique, elision is the opposite of the amplification process.” It is frequently
the case that the literary translator is obliged to condense the information contained in certain
passages being translated. To do this, some items which are not considered essential must
be removed as their elision will improve the stylistic quality of the translated work. Borrowing:
Borrowing is a technique frequently used in literary translation, but which can also be applied
in medical and business translations, for instance. For Albir, this translation technique involves
“using a word or an expression in the original text and placing it as it is, with no modification,
in the target text.” This can be an expression taken from a third language (e.g., Latin), or an
expression which is familiar to speakers of the target language, or even an untranslatable
expression which is not worth explaining.
02. From this perspective, saying that the translator has authorship is also saying that all
authors work translationally. And if that means that translators, like all authors, transform
texts, bring newness into the world, have complex productive cognition processes churning
within them as they work, and are all different, then I have no qualms about the proposition at
all: Translators are indeed subjective in their minds and creative in their writing, as any piece
of empirical research should be able to show. “Authorship”, however, can be understood in
several senses. It concerns not just creativity or individuality, but also ethical responsibility, a
point that has been overlooked by many of the literary ideologies.
03. Personal Answer

capítulo 5 • 133
Capítulo 5

a) For the rest of the day, she couldn’t concentrate on her lessons, tormented by her
own absurd behaviour, but, at the same time, relieved, because she knew that the
boy had noticed her too, and that the pencil had just been an excuse to start a
conversation, because when he came over to her, she had noticed that he already
had a pen in his pocket. She waited for the next time, and during that night - and the
nights that followed - she went over and over what she would say to him, until she
found the right way to begin a story that would never end.
b) — Nem precisa dizer quem ele é — riu-se tio Válter, espiando o prisioneiro por cima
do jornal. — Olhem só o estado dele, a imundice do desleixado! Olhem o cabelo dele!
E lançou um olhar de esguelha, maldoso, para Harry, cujos cabelos despenteados
sempre tinham sido uma fonte de grande aborrecimento para o tio. Comparado ao homem
da televisão, porém, cujo rosto ossudo era emoldurado por um emaranhado que lhe chegava
aos cotovelos, Harry se sentiu, na verdade, muito bem penteado.
O repórter reaparecera. “O Ministério da Agricultura e da Pesca irá anunciar hoje...” —
Espere aí! — berrou tio Válter, olhando furioso para o repórter, — Você não disse de onde
esse maníaco fugiu! De que adiantou o alerta? O louco pode estar passando na minha rua
neste exato momento!
Tia Petúnia, que era ossuda e tinha cara de cavalo, virou-se depressa e espiou com
atenção pela janela da cozinha. Harry sabia que a tia simplesmente adoraria poder ligar para
o telefone do plantão de emergência. Era a mulher mais bisbilhoteira do mundo e passava a
maior parte da vida espionando os vizinhos sem graça, que nunca faziam nada errado.

134 • capítulo 5

capítulo 5 • 135

136 • capítulo 5