The semantic model

As it has been shown, the transformational model views the process of translation as a series of transformations through which the translator reaches the core/kernel structure, which is common or very close to each other in the various languages, and then through another series of transformations he/she arrives at the target language surface.

It has not been explicitly stated, but it has been assumed all along that meaning (“Bedeutung”, “znachenie”, “signification”) or - as scholars who consider meaning language-specific refer to it - sense (“Sinn”, “smisl”, “sans”) remains unchanged in the meantime.

5.6.1. The antecedents of the semantic model

The followers of the semantic model, treated as a branch of the transformational model, approach translation from the point of view of meaning or sense. Research on semantic theory at the end of the 1960s exerted significant influence on the linguistic ideas related to translation. The ability of the speaker to decide about a sentence whether it has one or several meanings, or to decide about two sentences whether they have common meanings or not, that is whether they are synonymous plays a significant role in translation as well.

Translation scholars were also inspired by the theory of semantics developed by Katz and Fodor (1964), based on Chomsky’s theory of generative grammar. In Nida’s 1964 book the classic diagram of the meaning of bachelor is reproduced, with the semantic markers +human, +animal, 4-male and distinguishers [one who has never married], [young knight serving under the standard of another], [one who has the first or lowest academic degree], [young fur seal when without a mate during breeding time] (Nida 1964: 39).

It became more and more fashionable to assume that the translator might be breaking down the words of the source language sentences into such semantic constituents. In other words, in source language analysis the original text is not traced back to basic lexical units and grammatical structures but to basic semantic constituents, to basic meanings. These basic meanings form a particular system irrespective of the situation and the context: some of them will be relevant (cen­

tral) in the given context and others will be secondary (peripheral).

5.6.2. The process of translation according to the semantic model

The transfer from one language to another, according to this theory, does not hap­

pen through kernel structures consisting of basic lexical units and grammatical structures but through a semantic deep structure consisting of some system of basic meanings.

The literature on translation theory contains abundant examples for semantic analyses presumably conducted by translators (first of all in Nida 1964, Nida and Taber 1969), but unfortunately these examples always illustrate the semantic analy­

sis of words or word classes (cf. Komissarov’s English-Russian comparison of the semantic structure of verbs of spatial position: 1973: 51-9), and only rarely discuss how translators break a complete sentence down into basic meanings (Catford

1965: 39).

They also fail to explain the organisation and hierarchy of basic meanings, the nature of this assumed semantic deep structure, and the way these basic meanings are organised within this semantic deep structure.

5.6.3. Melchuk and Zholkovskiy’s semantic model

Melchuk and Zholkovskiy’s “smisl^>tekst^>smisl” (sense -> text -» sense) model (1965) reveals more about this assumed semantic deep structure. They start out from the idea that during the process of translation the translator first understands the text to be translated and then he/she expresses what he/she has understood in the given language, i.e. he/she expresses the sense of the text. But can sense be defined? In their view, it is in fact the possession of the sense that has to be defined which implies that the speaker can express the same idea in different ways and the listener can perceive the identity between the senses despite the formally different statements.

Thus, according to the authors, the sense of the text refers to the common ground that can be found in all texts intuitively regarded identical with the given text. They intend to describe this common content-related invariant with the help of a special semantic language, the so called basic language.

Accordingly the process of translation (e.g., from English into Russian) is as follows: in the first phase the translator switches from idiomatic English to the English basic language (independent sense analysis), in the second phase he/she switches from the English basic language to the Russian basic language (this is what can actually be considered translation), and in the third phase from the Rus­

sian basic language he/she switches onto the idiomatic Russian language (inde­

pendent sense synthesis).

What does the basic language contain into which the translator paraphrases surface sentences of the source language? The lexis of the basic language is a com­

posite of three types of elements: (1) predicates, (2) nominals, (3) adjuncts. The lexical elements of the basic language can be subdivided into constant elements and changing elements, that is, into lexical functions, which are the lexical corre­

lates of the given word.

5. Linguistic models of the process of translation

Lexical correlates (abbreviated in Latin) are the following:

a) substituting correlates, e.g.,

synonyms: storm -» Syn -» tempest antonyms: good -> Anti -» bad b) connecting correlates, e.g.,

rain -» Incep -» begin

mourning -> Magn -» deep

sleep -» Fin -» wake up

engine -* Fin -» shut off

5.6.4. Lexical functions

Lexical functions make the paraphrasing of the source language possible, i.e. allow for the explication, expression, and clarification of the semantic relations below the surface of the sentence. Lexical functions contribute to both the deconstruc­

tion of the source language meaning to its elements and also to the reconstruction of the target language meaning.

Every language has to be capable of expressing that something starts (Incep), finishes (Fin), operates (Oper), causes something (Cans), liquidates/eliminates something (Liqu), goes wrong (Degrad), etc., and all of these relations are expressed differently in the different languages.

Magn (rain)

English: heavy rain

Russian: silnij dozhd’ (‘strong’) Magn (losses)

English: heavy losses

Russian: tyazholie poteri (‘weighty’) Magn (prison terms)

English: heavy prison terms

Russian: dlitel’nie sroki zaklucheniya (‘long-term’) Oper (talk)

English: give a talk

Russian: chitat’ doklad (‘to read’) Oper (look)

English: give a look

Russian: brosit’ vzglyad (‘to throw’)

If, for instance, one has to translate to any language expressions such as: meat becomes stinkybread becomes dryy wine turns acidy apple becomes rotteny milk turns soury engine

fails, clock is out of order, fish taints quickly etc. then the translator does not need to think about how to translate the verbs to become stinky, to become dry, to turn acid, to turn sour, to fail, to taint etc., but how the given foreign language expresses the Degrad function besides meat, bread, wine, milk, engine, clock, and fish.

The finite number of universal lexical functions (cca 60 according to Melchuk 1997) allows for the systematisation of basic meanings and provides a basis for the deconstruction and reconstruction of meanings in any language. Besides lexical functions, Melchuk and Zholkovskiy’s paraphrasing system of rules also contains syntactic rules, some kind of filters, which can filter out the inappropriate solu­

tions, on the way towards the target language surface.

Of course, it is not only Melchuk and Zholkovskiy who approach the process of translation from the point of view of meaning or sense, but it is these authors who have achieved the most in the formalisation of sense relations and the precise modelling of the sense -►text -►sense process.

5.6.5. The advantages and disadvantages of the semantic model

The semantic model reflects many aspects of the process of translation correctly.

In the process of translating the translator does in fact go back to sense relations rather than to basic lexical units and grammatical structures. It is also self-evident that equivalences of basic lexical units and grammatical structures - be it intralin­

gual or interlingual equivalence - can only be determined on the basis of identity of sense. But the question may be raised whether it is possible or necessary to per­

fectly formalise sense relations or not. Is linguistics able to do so, and does the theory of translation need this in order to model the process of translation?

The semantic model, even in its present form, can explain several translational operations. Just to take one example, why is it that the English verb make enters the translator’s head “at one side” and the Russian verb navodif leaves his/her head “at the other side”, despite the fact that these verbs never appear together in the dictionary. This may be explained as follows: both are connected to the nouns order/poryadok by the semantic function Caus.

However in cases where the source language form does not play a role in find­

ing the target language form, where the target language equivalent has to be found on the basis of the situation or the communicative goal, the semantic model could not provide satisfactory explanations even if it were worked out in more detail.