Your Federal Quarterly Tax Payments are due April 15th Get Help Now >>

By the end of this long journey with Chomsky, Nida, Wilss and by 5x7V3m



        By the end of this long journey with Chomsky, Nida, Wilss and
     others… a number of conclusions concerning both Gentzler's claim and
     the utility of GTG to translation theory show up:

1-    When Gentzler put this claim, he meant Nida's 1964 and Wilss's 1982.
     To these two references, this study has added Nida and Taber's book as it
     is more or less an extension to 1964's, as far as GTG is concerned.
          The first disagreement one may make with Gentzler is his way of
     treating Nida and Wilss on equal terms.
          Nida did in fact say something that may or may not reflect his
     intentional or unintentional reliance on Chomsky. He spoke for example
     about kernels, transformations and deep structures. Not only Gentzler, but
     even the layman would suspect Chomsky's 1957 influence on his 1964's
     and GTG's general influence on 1969's.
          As for Wilss, he has almost said nothing that can be suspected in any
     sense. He keeps in fact on criticizing GTG as a model of little benefit to
     translation theory, and giving reasons for this uselessness.
         As was shown in Chapter Five, Wilss considers GTG as detrimental
     not to translation theory only but to linguistics as well. In spite of
     Chomsky's counterclaims, Wilss insists that the former's theory is too
     mechanistic to assume any degree of creativity, which renders it
     inapplicable to all fields of knowledge.
        Wilss, however, admits that the contemporary revival of interest in
     language universals has helped the 'science' of translation develop its
     methods and strategies. But interest in language universals, according to
     him, does not necessarily lead to interest in GTG: language universals are
     around three hundred years old whereas GTG is hardly fifty years.
     Moreover, GTG, says Wilss, is not as universal as it claims to be : it

restricts itself to studying individual languages. This interest did not lead
Wilss to attack Sapir-Whorf Hypothesis, as Chomsky did.
    In his marginal mention of Syntactic Structures, Wilss criticizes
those linguists who equated it "with a Copernican revolution in
    One of the reasons why Wilss distrusts GTG is the little support GTG
offered the 'science' of translation due to its lack of real explicitness.
    Another reason is that GTG premises are too formal to apply suitably
to the process of translation: GTG is "a strange cross between the
rationality of cognitive theory and speculative euphoria".
    A third reason is its mathematicness, which rendered it too rigid for
translation theory to benefit from.
  A fourth reason is that GTG restricts itself to competence, whereas
translation theory builds on performance.
    Lastly, GTG suffers from decontextualization, which deprives it from
the minimal applicability to translation methodology.
   As for his terminology, Wilss uses for example the term 'competence'
sometimes in the Chomskyan sense and sometimes in its dictionary sense,
i.e., ability: Chomskyan competence, he says, is too restrictive to fit
translational techniques.
   Besides, Wilss quite neutrally refers to Houston's replacement of
Chomsky's competence/ performance dichotomy by a tripartite one,
without bothering to show the confusion Houston's classification suffers
    It is hoped that this exposition has made it clear that Gentzler's claim
is NOT fair as far as Wilss is concerned in the least: he adopted a stiff
attitude toward Chomsky and his theory, he did not borrow any of his
techniques and finally he was not convinced with any tenet which
Chomsky proclaimed.

            What made Gentzler then put Wilss and Nida in the same dock?
     Why did he equate Wilss with Nida as far the application of GTG to
     translation theory is concerned?
2-    Most probably, the reason why Gentzler has fallen in the trap of
     equating Wilss with Nida that Nida and Wilss are perhaps the only
     translation scholars who talked about a 'science' of translation.
            In choosing his 1964 title, that is, Toward a Science of Translating,
     Nida might have been influenced in the wording by Richard's Toward a
     Theory of Translating in 1953.
            However, by 'science' , they meant two different things …..
            In 1964, Nida exerted great efforts in developing the process of
     translating into a 'science' by basing it on such fields as linguistics,
     psychology and anthropology and by directing it descriptively, i.e., be
     describing this process as explicitly and objectively as sciences normally
            "When we speak of the 'science of translating', we are of course
     concerned with the descriptive aspect; for just as linguistics may by
     classified as a descriptive science, so the transference of a message from
     one language to another is likewise a valid subject for scientific
     description" (Nida, 1964:3).
           As for Wilss, he uses the term 'science of translation' to refer to
     modern translation theory compared to the traditional one, which
     suffered, according to him, from the disease of prescriptiveness. The
     modern translation theory for him has become a science of translation as
     it delves into such basic areas as linguistic applicability, the degrees of
     translatability and the processes of decision-making in this activity. It is
     true, says Wilss, that this science is not as objective and abstract as
     mathematics, but it is more objective and enjoys more formalization than

   the traditional translation theory, in its attempts to balance itself between
   theory and practice.
       Wilss then distinguishes between specific and general sciences of
   translation, the general one being more theoretical than the former.
       Later on , however, Wilss favours the nowadays familiar term
   Translation Studies to his science of translation.

3- The third conclusion is perhaps more interesting than the previous ones:
   it concerns Nida's 'application' of GTG.
       Gentzler says that Nida read Chomsky's Syntactic Structures in a
   mimeograph form two years before its publication.
      Well, I think this not that important matter. What is more important is
   that Toward a Science of Translating appeared seven years after the
   actual publication of Chomsky's book. It is not plausible to assume that
   Nida did not read this book and Katz and Fodor's article thoroughly
   during these seven years.
       The question now is: did Nida plagiarize Chomsky as Gentzler
   alludes to, or did he establish his own model that happened to be similar
   to Chomsky's deep-structure / surface-structure one as he claims?
     First and foremost, Nida was concerned with Bible translation. This
   concern determined his translating techniques: he paid as much attention
   to meaning and meaning transfer as Chomsky paid to syntax.
      The way Nida analyzed meaning is completely different from
   Chomsky's: in 1957, Chomsky was not interested in the semantic
   distinctions which Nida made between such phrases as : my house, my
   frankness, my arrest,….. that Nida returned to their 'earlier' forms which
   he called 'kernels'.
      - I own a house                         - I was arrested
      - I was frank

     It is probable that Nida borrowed the term kernel from Chomsky, but
he viewed it in a completely different way: I was arrested is an earlier
form for Nida and a later one for Chomsky, since it is a passive
I own a house is by no means earlier, in the Chomskyan sense than my
house. As a matter of fact, Chomsky considers both this sentence and its
phrasal counterpart as kernels.
    Unlike Wilss, who stressed his disagreement with almost every claim
that GTG made, Nida did acknowledge from time to time the fact that a
generative grammar can be very useful to language description.
Chomsky's GTG is only one kind of generative grammar. It is generative
in the sense that it tries to be as explicit and as projective in its linguistic
exposition      as   possible.    Chomsky's      generative     grammar       is
transformational as it formulates transformational rules that convert
kernels into transforms according to Syntactic Structures and deep
structures into surface structures according to the Standard Theory on.
   A major split between Chomsky and Nida is what a kernel is. For
Chomsky it is the sentence that is arrived at via obligatory
transformational rules only. Nida , however, specifies it from a purely
semantic point of view: a sentence , or a phrase, whose meaning cannot
be derived from an earlier form. It is quite evident that Chomsky's
syntactic division of sentences into kernels and transforms is much more
definitive and easier to specify. Nida's classification, on the other hand, is
too subjective: there is no clear-cut justification why I own a house is a
kernel and my house is not.
  As we saw in Chapters Four and Five, transformation is unidirectional
for Chomsky and bidirectional for Nida, who built his sentence
classification on back-transformation that has no place in Chomsky's

  Similarly, Nida's lexical classification into object, event, abstract and
relation is specifically translation-oriented and does not cohere with
Chomsky's model in any way. The phrase: redemption of God, for
example, comes for Nida from its corresponding verbal construction: God
redeems. But what transformational rule converts that verbal kernel into
that nonkernel of-phrase? Nida does not specify it, probably because
there is no such a concretely specifiable rule.
   Building his transformational analysis on his lexical classification into
four functional or semantic units, Nida enters a vague impasse out of
which there is no way out: it is a real split between Chomsky and Nida.
   Personally, I trust Chomsky's model, since it enjoys clarity and
objectivity because it does not mix the lexical and the syntactic levels as
Nida does.
    However, what matters most for this study is that Nida's way of
building his transformational model on his own lexical classification is an
unrefutable evidence that Nida did not photocopy Chomsky's approach.
They have chosen two different aims: Chomsky's aim is syntactic
whereas Nida's is translation-oriented. So, their models are much more
divergent from each other than what those three scholars and perhaps
others have thought of.
   As a corollary , Nida's approach IS more authentic than what is said
about it. Whether this approach is practical or fruitful to the process of
translating or not is of secondary importance to the present study, which
opted to verify that claim.
   Anyhow, Nida's model is perhaps too limited in its scope of
application. It looks bright and inviting in its general exposition, but
vagueness envelop its three-stage translating process:
             analysis - transfer - restructuring

   One may ask how the decoded message is transferred for encoding.
Nida does not answer this question in 1964 nor in 1969: he does not
substantiate his model with rules that actualize this transfer.
   It is good that Nida and Taber tried hard to apply Katz and Fodor's
theory to translation. But they talked about determining the linguistic
meaning before dealing with the emotive one, without showing how to
split these two types of meaning, or how to rematch them in translating.
   It seems that the worst problem both Nida 1964 and Nida and Taber
1969 face is the confusion between lexical, syntactic and semantic levels.
   All in all , Nida did not plagiarize Chomsky: he tried both openly and
honestly to benefit from Chomsky's model in formulating his own
translation approach which diverges from Chomsky one quite sharply.
The accusation quoted in the abstrast of this study is neither just nor
soundly based.
4- If we are convinced that Nida and Wilss did not apply GTG to
translation theory as the three conclusions above say , shall we conclude
that GTG is principally inapplicable to this theory ?
     Well, I think it is POSSIBLE to locate certain areas of this grammar
that may enhance translation strategies.
   To end this study with what it started with, let us redefine translation
as the activity of transferring meaning from one language to another ….
Since meaning is the static and common element of this activity, GTG is
helpful so long as it deals with this 'commonness' of meaning , that is,
with what is common or universal among languages. A universal feature
does not necessarily benefit translation theory. The present conclusion
therefore will scan those universal features which the various versions of
GTG have developed, in an attempt to apply some of them to translation

a.    Since translation is concerned with the whole meaning of the sentence,
     not with its individual words, and since GTG takes the whole sentence as
     the starting point of its analysis, it becomes logical to assume that GTG is
     somehow more promising to translation than , say , structural grammar
     which starts its linguistic analysis with phonology.
b.    As the distinction which Chomsky established in 1957 between kernels
     and nonkernels proved groundless due to the evanescence of the
     distinction between obligatory and optional transformational rules in
     1965, one may rather resort to the deep structure as a common point of
     departure in translation. In a technique like Nida’s, one can return an ST
     sentence from its form, that is , from its surface structure to its deep
     structure , where meaning is somehow split from form. Since the deep
     structure assumes partial universality according to some, one can set out
     from it in formulating the TT sentence. But the concept of deep structure
     was demonstrably eliminated in 1993. So, we cannot reasonably rely on
     any GTG area after GTG dissociates itself from it.
c.    Instead, one can make use of the special attention both Katz and Fodor’s
     paper and Aspects gave to meaning. In translating a sentence that is
     lexically ambiguous, and in addition to the context of situation, which
     offers the main help, it is possible to break such a lexical item into its
     semantic markers (human , male, …):
         - They spoke about the bachelor
             To guarantee translation correctness, it is advisable to check
     whether these semantic markers go together with the grammatical
     markers (noun, verb, adjective, …) in both languages. The owe/madeen
     example in the previous chapter showed that they may not. Katz and
     Fodor’s assumption is that semantic markers are universal ones.
d.    Creativity is an essential feature both in translation theory and in GTG.
     In translation, it means the ability to modify the ST in order to produce

     the most successful TT. In GTG in general and in Aspects in particular, it
     means the ability to comprehend and produce novel sentences not
     encountered before.
              In spite of this difference in meaning, one can establish the
     principle that the linguistic creativity can lead to the translational one:
     once you understand any sentence that you did not come across before, so
     log as it complies with the rules of its grammar, you can certainly
     translate it correctly as far as comprehension is concerned.
e.      GTG pays more attention to categorial notions such as NP and V than
     to functional ones such as subject, direct object, … Aspects holds that the
     functional notions need not be stated since they are implicitly contained
     in the categorial description.
                    I think that this stand applies to English but not to Arabic,
     in which these functional classes determine the suffixes.
f.      The translating process can benefit from certain GTG innovations
     such as subcategorization features and selectional features or restrictions.
     A tranlator may pay more attention to the first type, because disobeying it
     resuts in a more serious deviation than disobeying the second one:
- Bill came the letter yesterday
- The wall smiled at last
g.      GTG classifies grammatical rules into univrsal and specific ones:
     every baby is equipped at birth with the universal rules that are the same
     for all languages.
           In a course for training translators, the course designer can take this
     distinction into consideration by focusing on the peculiarities of the
     foreign language the translator is trained on.
h.      Generative semantics with its classic example :
- John killed Sam
- NP-caues-NP-become-not-alive

     may be kept in the translator’s mind, just to secure correct rendering by
     analyzing certain words into their semantic primes which are supposed to
     be the same in all languages.
i.       Case Grammar by Fillmore and others can be of special relevance to
     translation theory as it assumes that case is a universal notion which
     should be observed in translation. What adds to the universality of case
     grammar is that case is not affected by the place of the NP in its sentence:
     the same case is attached to an NP whether it is at the beginning, in the
     middle or at the end of the sentence:
- John wrote a letter
- A letter was written by John.
            In translating these two sentences into Arabic or any other
     language, the case of John remains agentive in both the active and the
     passive sentences. Observing case therefore protects tranlation from
     going sideways.
j.       Chomsky’s trace theory is an area where English and Arabic do not
     behave similarly :
     “….when a transformation moves a phrase P from position X to position
     Y it leaves in position X a trace bound by P.” ( Chomsky, 1975: 95)
             This rule is supposed to be a universal one. However , a translator
     should realize that whereas the trace is implicit in English it is usually
     explicit in Arabic:
     [John is certain [John to win]]            John is certain t to win
                        ‫جون واثق انه سوف يفوز‬          )) ‫(جون واثق (جون سوف يفوز‬

     This syntactic difference between English and Arabic can cause some
     translation difficulties. In a sentence like:

      I have chosen here to write the Arabic sentences in this chapter in Arabic script
     because transcription may not be clear enough.

           -   Who did he say Mary kissed t?
 it is clear that the interrogative pronoun is the direct object of kissed. It
 left a trace in its original place at the end of the sentence. The sentence is
 clear and straightforward in its English form but not in its literal
 translation into Arabic:

                                      ّ ‫مار‬
                                   ‫من قال ي قبلت؟‬

       Grammatically correct as it is, the wording of this Arabic sentence
 tempts one to think erroneously that ‫ منن‬is the subject of ‫, قنال‬leaving the

 other verb without an explicit direct object.
            To do away with this ambiguity, one would rather be more
 liberal in translating it :

                                           ‫ّ مار‬
                               ‫من قبلت ي حسب قوله ؟‬

          Moreover, the unbound anaphora which trace theory treats may
 result in similar ambiguities in both English and Arabic :
     - Jill broke her hand

     - ‫جل كسرت يدها‬

          Both these sentences are ambiguous as Jill might have broken
 either her own hand or the hand of a certain female.
k.    GB is another universal theory which can benefit translation theory.
 The head-first / head-last parameter ought to be paid attention to when the
 SL and TL differ from this point of view. In translating from Arabic or
 English into Japanese , for example , one should notice that the English
 phrase 'to school' becomes 'school to' in that head-last language.
       As for C-command Principle, it is clearer in Arabic than it is in
 English, because of the type of inflection it results in. This implies that
 translation from English into Arabic demands more preparation than

translation from Arabic into English as far as this principle is concerned:
the first situation demands that a translator choose the correct case
ending. In English, a noun has of course the same ending for all cases.
      With respect to the three properties which         the binding theory
offers, English and Arabic are in full agreement, hence , there is no
translation difficulty in either direction:
      - Anaphors must be locally bound
      - Pronouns must be locally free of co-indexing antecedents
      - Referential expressions are free everywhere
        Concerning X-bar theory, it puts among other things the Specifier
of an NP as a counterpart of its N :

             Specifier                               N

              Mary's                             N         PP

                                                          to the problem

        In translating into Arabic, the elements included within the N do
not make a closed unit: the specifier intervenes within it:

                                          ‫ّ مار‬
                                 ‫- حل ي للمشكلة‬

        Concerning subjacency, English and Arabic sound similar:
movement in both languages cannot cross more than one bounding node
at a time , and as a result, there is no translation difficulty in either
language :
        - Which picture did Sam say that John would fix ?

                         ‫- اية صو ة قال سام سيثبتها حون ؟‬

     The important difference in the translation is that the relative
pronoun that disappears in the Arabic sentence because its antecedent
picture is indefinite.
     As for the Empty Category Principle, English and Arabic behave
similarly and translation therefore goes smoothly:
          - Frank wanted ................. to go

                                                    ‫ر ر‬
                     ‫- ا اد ف انك .................... ان يذهب‬

      The Theta Theory supplies the translator with better inight by
identifying the thematic role of every NP in the sentence. These roles are
universal by nature, which facilitates translating among all languages.
      However, sometimes it is not easy to identify these roles:
      - Thelma handed the manuscript to Louise.
      One in fact is not quite sure whether the NP Louise assumes a
beneficiary role or a goal one. Fortunately, this uncertainty does not
create a translation difficulty
      As for case theory, languages differ with respect to case realization:
in Arabic it is realized in nouns pronouns and adjectives; in English , it is
realized in pronouns only. This means that it is effective in translating
nouns from English into Arabic, not the other way round.
   However, translation from English into Arabic is easier if the subject
of its sentence is assigned by tense than if assigned by a preposition:
        - She may resign any moment
        - For her to resign is funny

     One of the important principles which Minimalism introduces is
Procrastinate. This principle shows that Arabic is similar to English but
not to French with respect to verb raising:
      -   John often kisses Mary

         ‫ّ مار‬
      - ‫جون غالبا ما يقبل ي‬

      - Jean embrasse souvent Marie
    As for the reconstruction process, a translator ought to notice that
Arabic and English behave similarly, as the lowering of a preposition of a
wh-phrase is optional in both languages:
      -   In which house does Emma live ?
      -   Which house does Emma live in ?
    Idiomatic ambiguity causes special translation problems. This is
Chomsky's example:
      -   John wondered which picture of himself Bill took
   A translator should realize that himself refers either to John or to Bill,
that take….picture is understood either literally or idiomatically and that
the idiomatic interpretation is possible only if the antecedent is Bill.
   Another area of Minimalism is word order. English is an SVO
language, Irish a VSO one. As for Arabic ,it allows both orders. It is
advisable then to translate into Arabic an English sentence as a nominal
one and an Irish sentence as a verbal one.
   Wh-in-situ is another feature which Minimalism surveys and which a
translator should notice before translating from one language into
another: the overt movement of the wh-phrase in simple questions is
required in English, barred in Japanese and allowed in Colloquial French.
Arabic happens to be similar to English, but translating from Arabic into
these three languages should take these peculiarities into consideration.

      The last point which Minimalism investigates in this study is c-
      -   ……..that John has bought the ball
      -   ……..das Hans den Ball gekauft hat
      One may notice that in German, inflection c-commands V although
it follows V and that V c-commands its direct object although it follows
that object. English has not got this complexity: inflection usually
precedes V and c-commands it and V precedes and c-commands its direct
object. Arabic is nearer to English, which makes translating from Arabic
easier into English than it is into German, as far as c-commanding is
                   x             x                x

      Finally, it is hoped that these modest remarks and suggestions
which this conclusion has contained do not clash with Chomsky's
warning that the existence of formal universals does not 'imply there must
be some reasonable procedure for translating between languages'.
      Most probably, Chomsky did not want to commit himself to any
practical implication of his linguistic findings. After all, Chomsky is not a
translation theorist. He is a linguist, the greatest linguist in the second half
of the twentieth century. Whether his linguistic school is applicable to
translation theory or not is beyond his concern, but certainly he would not
mind seeing others undertaking this task.


To top