Article

"Pragmatic translation and literalism"

Peter Newmark
TTR: traduction, terminologie, rdaction, vol. 1, n 2, 1988, p. 133-145.

Pour citer cet article, utiliser l'information suivante :

URI: http://id.erudit.org/iderudit/037027ar
DOI: 10.7202/037027ar
Note : les rgles d'criture des rfrences bibliographiques peuvent varier selon les diffrents domaines du savoir.

Ce document est protg par la loi sur le droit d'auteur. L'utilisation des services d'rudit (y compris la reproduction) est assujettie sa politique
d'utilisation que vous pouvez consulter l'URI https://apropos.erudit.org/fr/usagers/politique-dutilisation/

rudit est un consortium interuniversitaire sans but lucratif compos de l'Universit de Montral, l'Universit Laval et l'Universit du Qubec
Montral. Il a pour mission la promotion et la valorisation de la recherche. rudit offre des services d'dition numrique de documents
scientifiques depuis 1998.
Pour communiquer avec les responsables d'rudit : [email protected]

Document tlcharg le 12 February 2016 12:42

Pragmatic translation and literalism

Peter Newmark

Defining pragmatism
I make the basic assumption that provided a source language text
contains no misstatements of fact, is competently written and has
to be fully translated rather than summarized or functionally reorientated, one's purpose in translating it is to be referentially and pragmatically accurate.
In this paper, I shall not deal with referential accuracy but concentrate instead on the pragmatic aspects of translation. As the term
pragmatic translation may be used in a variety of ways*, I should
like to state my own definition of pragmatic, which derives from
Charles Morris and ultimately from Charles S. Peirce.
I am using pragmatic as one of the two factors in translation:
pragmatic denotes the reader's or readership's reception of the
translation, as opposed to referential, which denotes the relationship
between the translation and the extra-linguistic reality it describes.
Aspects of the pragmatic factor
Characteristicsofthereadership.Thepragmaticfactorhastwoelements.
The first is relatively extra-contextual and relates to the reader's
characteristics, some of which (subject knowledge, linguistic level,
SL cultural familiarity) may be considered to be more relevant or
important than others (such as social class, age, sex and the time
elapsed since the writing of the SL text). Pragmatic translation is
largely tentative and presumptive, as opposed to referential translation
where the only assumption is that the readers are literate. In the
case of the illustrative text, only some of the SL or TL readers are
likely to be emotionally involved, those who feel strongly about some
leading statement, eg. how Europe reacts to Gorbachev. But in cases
where the whole readership is affected (pragmatic text types such
as publicity or public notices, for example), the translator must take
into account all aspects involving readership sensitivity in order
133

to stimulate the appropriate frame of mind in the reader. The success

of such a translation can only be assessed by the readership's subsequent
behavior.
Syntax, word order and stress. The second element to be considered
hinges on the language of the text itself and the readership's sensitivity
to it. Normally the syntax sets the tone of any pragmatic relationship:
action is pointed by verbs; description by nouns and adjectives or
adverbs of quality; dialogue by forms of address and tags; injunctions
by imperatives or rhetorical questions; urgency and speed by brief
sentences; leisureliness or meditation by long ones.
Stress, which is intimately connected both with syntax and word
order, is an essential element in pragmatic effect. Unusual word order
indicates stress, emphasis or liveliness. In the specimen text, for
example, the argumentative and strident tone is set by the negativepositive sequences: ... ... ni; non, celui qui tait, c'est... (11. 25). There is opening stress in the word order shifts in the c'est
sentences notably, but also in Non, celui qui tait... (1. 5); Finis les
heureux temps (1. 9); and Subtil, certes (1. 15), all admirably reproduced in No, the one who was... (1. 5); Gone are those balmy days
(U. 9-10); and Wily he may well be (1. 15). The sentences are bold
statements, self-contained but clearly interrelated. They are uncluttered
by clauses with the exception of die adjectival clause, the specially
privileged subordinate clause of French syntax, so often related to
the even more privileged emphatic c'est or ce sont. The two rhetorical
questions in paragraph 3 and the inversions help to force view
on the readership.
Note that I am here positing a universal word order. I assume:
(a)

that the natural sequence of a proposition is from given

(or old) to new information, from theme to rheme, which
is also the natural sequence of explanation or teaching; but
if rheme precedes theme, the degree of pragmatic expectation
imposed on the SL sentence should be reflected in the TL
equivalent;

(b)

that grammatically the natural word order of a sentence

is Subject-Verb-Complement, which follows the natural order
of thought;

(c)

that the main stress in this word order will be on the lexical
part or head-word of the complement, provided it contains
the new information or rheme (for example: in the sentence
No, the one who was on everybody's mind was Mikhail Gorbachev (U. 5-6), stress is on Gorbachev as it is in the French);

134

(d)

that every language has its own lexical, grammatical, wordordering resources for putting stress on a non-final group
of the sentence, which will normally entail a change or an
interruption in the natural word order;

(e)

that the reason why many languages (e.g. Latin or German)

do not follow the S-V-C natural word order is that its scribes,
(the clerks, the priests, the ideological hegemony) have imposed
an unnatural grammar on the illiterate masses.

Many of the above are intuitive statements, but I do not think there
is anything in the classical essays by Firbas (1979) and Greenberg
(1963) that contradicts them.
Given a natural sequence, we can state that in referential
translation the word order is normal; in pragmatic translation,
on the other hand, it is often upset by particular stresses.
Metaphor as a pragmatic factor
After the interplay of syntax, word order and stress, metaphor is
normally the most powerful pragmatic factor in translation. Metaphor
is language's main resource for conveying strong feeling. In many
European languages, sexual and scatological metaphors have replaced
religious ones to express the most vehement feelings. Original and
standard metaphors are strongest in taboo and colloquial language;
in scientific and descriptive language metaphor is equally important,
but it is likely to be spatial or temporal as an aid to reference.
Lastiy, metaphor is a means of conveying the world of the mind
and demonstrating the wealth of the life of the senses. The pragmatic
effect of metaphor in the most emotive types of texts (poetry, advertising, propaganda, metaphysical or religious writing) needs no illustration. And yet, in much translation, in the translation of poetry,
for example, metaphor is needlessly evaded or diluted. This goes
hand in hand with the fear of literalism.
The strongest metaphor in the specimen text, in my view, is:
enfoncer un coin (1. 12). To translate it using the verb 'to nick'
is rather neat but feeble, since a literal translation ('drive a wedge
into') would produce a perfect equivalence. Another strong metaphor
mettre au pied de ce mur-l (U. 39-40) is successfully and literally
rendered as push... up against this particular wall (1. 37). A third,
gommer ses... dclarations (1. 49-50), an eraser image that appears
with much use to be moving from 'delete' to 'moderate', is cleverly
given as back offfrom\l. 47). The more informal English journalese
lends itself to phrasal verbs, which are usually metaphorical. The
remaining metaphors are unremarkable, but incline to a more 'popular'
pragmatic effect. The English standard metaphors that are used are
135

warm, familiar, simple, homely, and on the whole more physical and
concrete than the French ones.
In this piece, syntax, word order and metaphor adequately take
care of pragmatic effect; here there is no need for SL and TL readerships to identify. But when dealing with persuasive and more universal
texts, the translator requires other pragmatic resources relating to
words as meaning and/or sound. On the one hand, unusual words,
unusual metaphors, unusual collocations, or neologisms may be used
to produce an arresting effect. On the other, alliteration, onomatopoeia,
assonance, rhyme, rhythm, and metre not only convey meaning but
may also appeal to the readership's senses.
The readership and the situation
Readership is like context: it can never be completely ignored, but
it is more important on some occasions than on others. If the readership
consists of one client, it is all-important, and you can normally elicit
his/her requirements in detail, establishing, for example, whether
technical or institutional terms have to be simplified. If your client
is a middle person, and you are translating for his/her customers
who are your readers, you should ask for all possible information
about the readership, but you may have to make certain assumptions
about their knowledge. At the other extreme, you may have a subtle
or dense expressive text, such as a poem where the putative reception
of the readership is irrelevant, and where you can only attempt to
assess your personal reception of the text. Between these extremes,
there are many variations and many compromises.
Literalism as a yardstick of translation
If one is discussing the full translation of a worthwhile text of some
importance, there can be no primary aim but accuracy, which may
itself be some kind of compromise between the referential (the content,
the matter) and the pragmatic factor (the style, the manner), and
if one looks for a yardstick, a general basis to judge a translation,
there is nothing concrete but literal translation. When you ask how
close, how faithful, how true a version is in relation to the original,
you can have nothing else in mind except the spirit of the original,
which is the reverse of concrete. Vinay and Darbelnet saw this thirty
years ago, when they wrote, at the conclusion of their great book:
On doit arriver ne s'carter de la littralit que pour satisfaire
aux exigences de la langue d'arrive... On ne doit pratiquer
la traduction oblique qu' bon escient, dans des limites nettement
dfinies. On doit rester littral tant qu'on ne fait pas violence
la langue d'arrive. On ne s'carte de la littralit que pour
136

des raisons de structure ou de mta-linguistique et on s'assure

alors que le sens est sauvegard. (Vinay & Darbelnet, 1958,
p. 268)
As I see it, the scientific principle in translation could not be
better expressed than it is here, though I think there is a corollary:
in an authoritative and/or a serious literary text, any violence done
to the norms of the source language has to be reflected in translation
by violence done to the norms of the target language.
Further, Vinay and Darbelnet rightly point out that whilst the
word is rarely the unit of accurate translation, it is wrong to suggest,
as did Gide, that one should translate sentences rather than words;
normally the sentence as a unit is much too wide. If that is so, it
is absurd to talk about translating texts rather than words.
If one accepts literalism in the above sense as the scientific principle
of a full translation, one has to make many qualifications, of which
the first is that translation is not a science but rather a craft and
an art, and is at the last revision, within certain narrow limits, often
just a question of taste. Secondly, the more concentrated, the more
authoritative, the more studied, the more important the language
of a text, the more meaningful become the individual words of the
text and pari passu their translation. In such a text, each word is
charged with meaning. The translator, like the writer beforehand,
is wrestling with words; words are so strong that the context falls
temporarily into the background. Words become like notes in chambermusic, expressively important vertically (paradigmatically) in their
associations and etymology, as well as horizontally (syntagmaticaUy).
In a run-of-the-mill text, on the other hand, the message is direct,
unadorned, but the language, like Benjamin's memorable cloak, is
looser, more otiose, more dispensable. One comes to a stage where,
as in the case of my deliberately chosen specimen text, one feels
one might as well translate literally but it is not a matter of importance, since one wants to be accurate, but a paraphrase would do
as well.
So, if one applies the literalism yardstick to the specimen text,
one notes how in most cases, not only the facts and the main syntax
but also the quality words that state the feeling of the text, have
been retained precisely to maintain the pragmatic quality: thus terms
such as pugnacious, particularly discreet, vulnerable, crucial, and
dangerous shine through from the original. While nicking Atlantic
solidarity (1. 13) is a neat but feeble version of the more literal
'driving a wedge into Atlantic solidarity', the translation of contradictoires (1. 45) as mutually exclusive (1. 43) is markedly an improvement
on the original. Here then, literalism is a form of control on the
style. The message is secure, accuracy in every verbal detail is

137

not important, but inaccuracy in detail is unnecessary rather than

elegant. The translator usually shows commendable restraint in resisting
the all-too-common temptation to convert straightforward literal
language into clichs, with the possible exception of the following:
hros
(L 1) translated as star, heureux (L 9) as balmy and accepter
lf argumentation (L 54-55) as take at his word.
Certainly there are informative texts where one is faithful not
out of loyalty to the author (his/her precise style is hardly worth
it), but simply out of loyalty to one's client and the readership (who
presumably want to know exactly what the author wrote).
Approaches to translating
Vinay and Darbelnet advocated direct translation where this could
be done without violence to the norms of the target language, and
only revert to indirect translation when necessary.
There are, in fact, two approaches to translation and to translating.
The first one may be called bottom to top. In practice, this is where
the translator starts translating immediately. It is a literalist approach,
where we plough ahead, usually sentence by sentence, stopping only
at places where literal translation won't do. My definition of a
translation problem is where literal translation fails and we have
to consider a number of choices, or procedures, guided by a reasoned
perception of a number of contextual factors, which can finally only
be confirmed by a reading of the whole text. More often than not,
the grammatical problems can be solved instinctively, or without
reflection, through readily available transpositions or shifts; lexical
problems (one word into one word won't go) require more deliberation
because of a larger number of choices; 'encyclopaedia' problems depend
on one's provisional assessment of the readership; problems of cohesion
derive from sentence and paragraph sequencing.
The second approach to translating and translation is top to bottom.
It is recommended by many a translation teacher, but practised by
few instinctive translators (except perhaps in cases where the text
is a technical maze). This approach consists in reading the text two
or three times before beginning to translate, and in assessing the
general before the particular problems of the text: first the topic,
Sien the language, readership, register, tone, down to a dwindling
number of other factors,figurativeor denotative.
Both approaches are valid. The top-to-bottom approach, based
as it is on the rather absurd premise of the text as the translation
unit, is at present overblown. The bottom-to-top approach, which
starts as literal translation and soon gives up, is more objective,
more specific, more scientific. It is the contrle of a translation,
138

the back translation test (BTT), which can easily be abused, which
can degenerate into translationese, but which is at present neglected
in the literature. But against the potential licence of pragmatics,
hermeneutics, the vouloir dire, the sub-text, the spirit, the gnie
de la langue, it is the only buffer, the only sense and common sense.
What I see as the most urgent objective in translation studies,
translation theory and, more pertinently, translation teaching is to
bring these two approaches and methods a little closer together.
The task is important, as it has a bearing not only on standards
of translation, but also on the way in which we understand the various
stages of translating and hence on translation teaching. I see this
as more profitable than objectively recording and scientifically
analysing what translators do (Harris, 1988).
Practical applications
The following steps may be helpful in at least bringing the two
approaches (and methods) closer together:
1)
2)
3)

4)
5)

relating them to text-types, readerships and other contextual

factors;
bearing in mind that stretches of text of whatever length
have meaning, and that the translator is concerned only
with meaning transfer;
in the critical third stage of revising, reading the translation
as an independent, autonomous, spontaneous text, as well
as reading it sentence by sentence, side by side, with the
original, and thereby not forgoing, as the interpreter has
to, the advantage of all the information that rereadings of
the original can continue to offer;
exploiting the contrasted insights of text linguistics as well
as of literal translation;
aiming for the closest possible translation, as Nida would
have it, which normally indicates some degree of intersection
between the two methods.

If a closer agreement can be reached on the approach, choices will

remain, but they will be a little narrower, and eventually translations
may become more accurate.
Conclusion
I have attempted to review some of the pragmatic factors in translation,
which are closely bound to the readership. These are factors that
influence the passage of the text into the target language literature
and contribute to its absorption into another convention and often
another tradition. This is not limited to the canonical or even alter139

native literature, but may extend to the media, computer language

and other forms of communication that are rapidly becoming internationalized. Thus excessive pragmatics tend to rob the target language
text of its translation^ character. Literal translation is one way
in which we may continue to preserve the genius or particular character
of the foreign language despite this process of assimilation.

University of Surrey
* The term was used in fact as the title of one of the sessions at
the 1988 CATS Conference in Windsor, at which this paper was
presented.

References
DELISLE, J. (1980). L'Analyse du discours comme mthode de traduction, Ottawa, University of Ottawa Press.
FIRBAS, J. (1979). A functional view of ordo naturalis. In Reno
Studies in English, vol. 13.
GRAMMONT, M. (1908). PetitTraitde versificationfranaise. Armand
Colin, Paris.
GREENBERG, J.H., Ed. (1963). Universals of Language. Cambridge,
Mass., MIT Press.
HARRIS, B. (1988). What I really meant by translatology. 7TR,
vol. 1, n0 2.
LEECH, G. (1981). Principles of Pragmatics. CUP.
VINAY, J.P. and J. Darbelnet (1958). Stylistique compare du franais
et de V anglais. Paris, Didier.
WILSS, W. (1982). The Science of Translation: Problems and Methods.
Tubingen, Gunter Narr Verlag.

140

L'OTAN face M. Gorbatchev

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42,
43

Le vrai hros du sommet de l'alliance atlantique, qui s'est

termin le jeudi 3 mars Bruxelles, n'a t ni Mme Thatcher,
pugnace comme son habitude, ni le prsident Reagan,
particulirement discret, ni le couple infernal MitterandChirac. Non, celui qui tait prsent, dans tous les esprits,
c 'est M. Gorbatchev.
En effet, le secrtaire gnral du Parti communiste
sovitique et sa nouvelle politique de dsarmement posent
problme l'alliance. Finis les heureux temps o il tait
ais d'anticiper les mouvements diplomatiques et stratgiques
de Moscou. L'adversaire est devenu infiniment plus subtil
dans l'art d'innover, de surprendre, d'enfoncer un coin
dans une solidarit atlantique toujours vulnrable aux gosmes nationaux.
Subtil, certes, mais M. Gorbatchev n'est-il pas sincre
lorsqu'il affirme vouloir soulager son pays d'une surcharge
militaire paralysante pour se consacrer la remise en ordre
d'une conomie chaotique? M. Mitterand a raison, lorsqu'on
lui pose la question, de refuser d'y rpondre. Qui peut
savoir ce qui est rellement dans la tte d'un homme qui
n'est, de toute faon, pas arriv au pouvoir par inadvertance?
Force est donc de le juger sur ses actes.
Deux tests seront cruciaux de ce point de vue pour
l'avenir des relations Est-Ouest. Le premier ne concerne
pas directement l'Occident, mais il devrait permettre de
se faire une ide de la volont de M. Gorbatchev de corriger
les erreurs du pass. C'est de l'Afghanistan qu'il s'agit,
dont l'vacuation par les troupes sovitiques ferait beaucoup
pour convaincre l'opinion mondiale que de nouvelles analyses
ont enfin cours Moscou.
La seconde matire de l'examen de passage auquel
il faut soumettre M. Gorbatchev, ce sont les ngociations
sur le dsarmement conventionnel. Ce sont les armes
classiques qui, en effet rendent la guerre possible, et il
serait dangereux d'encourager les rves de dnuclarisation
de l'Europe occidentale tant que ne sera pas efface la
supriorit du pacte de Varsovie dans le domaine de
l'armement conventionnel. C'est l'ide-force du sommet
de Bruxelles, et les Seize se doivent de mettre M. Gorbatchev
au pied de ce mur-l.
Ils n'y parviendront que s'ils russissent conjuguer
ouverture d'esprit et fermet. Les deux postures, illustres
ici par MM. Mitterand et Chirac, au niveau europen par

44
45
46
47
48
49
50
51
52
53
54
55

le chancelier Kohl et Mme Thatcher, ne sont pas forcment

contradictoires. Elles devraient mme tre complmentaires,
l'une ou l'autre prvalant en fonction du comportement
de Moscou.
Au-del d'arrire-penses lectorales, M. Mitterand
l'a compris qui a passablement gomm ses prcdentes
dclarations hostiles la modernisation des forces nuclaires
de TOTAN pour insister sur la ncessit de dfinir une
stratgie pour le dsarmement. M. Chirac aussi, qui affirme
aujourd'hui n'avoirplus aucune rserve l'gard des positions
du chef de l'Etat, qu'il souponnait d'accepter un peu vite
l'argumentation de M. Gorbatchev.
Le Monde, le 20 mars 1988

How to be open-minded and still firm

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28

The real star at the Atlantic Alliance summit which ended

on March 3 in Brussels was neither Margaret Thatcher,
who was as pugnacious as ever, nor Ronald Reagan, who
was particularly discreet, nor even that "infernal couple"
of Mitterand and Chirac. No, the one who was on
everybody's mind was Mikhail Gorbachev.
The Secretary-General of the Communist Party of
the Soviet Union and his new disarmament policy have
in fact presented the Alliance with a problem. Gone are
those balmy days when it was so easy to anticipate Moscow's
diplomatic and strategic moves. The adversary has become
infinitely more wily in devising new things, springing surprises
and nicking Atlantic solidarity which is still vulnerable
to considerations of selfish national interest.
Wily he may well bp, but isn't Gorbachev being sincere
when he says he wants to relieve his country of crippling
military over-expenditure and get down to sorting out
its economic mess? Franois Mitterand is right to refuse
to answer the question. Who can know what is really
going on in the mind o a man who, in any case, did not
gain power by inadvertence? We must then necessarily
judge him on his deeds.
Here, two tests will be crucial for East-West relations.
The first does not directly concern the West, but it should
permit us to get some idea of Gorbachev's determination
to correct the "mistakes of the past". This is Afghanistan,
where the withdrawal of Soviet troops will go a long way
towards convincing world opinion that new analyses are

29 at last being made in Moscow.

30
The second test to which Gorbachev will have to
31 be put is the egotiations on conventional disarmament.
32 These are conventional weapons which in fact make war
33 possible, and it would be dangerous to encourage dreams
34 of a denuclearised Europe so long as the Warsaw powers
35 maintain their superiority in conventional weapons. That
36 was the key idea at the Brussels summit and the 16 will
37 have to push Gorbachev up against this particular wall.
38
They will succeed here only if they successfully join
39 together in being open-minded and firm. The two attitudes,
40 illustrated in France by Mitterand and Chirac, and at the
41 European level by West German Chancellor Helmut Kohl
42 and British Prime Minister Margaret Thatcher, are not
43 necessarily mutually exclusive. They should even be
44 complementary, with one or the other prevailing, depending
45 on the line Moscow is taking.
46
Electoral considerations aside, Mitterand has realised
47 this and has somewhat backed off from his previous
48 declarations objecting to modernising NATO's nuclear forces.
49 He is now insisting on the need for laying down a "strategy
50 for disarmament". Jacques Chirac has also shifted his
51 position and is saying today that he ,has no reservations
52 about the attitude of Mitterand, whom he previously suspected
53 of being too ready to take Gorbachev at his word.
Guardian Weekly, March 1988