LOGOS - Multilingual Translation Portal

21 - Equivalence - part four

«Y las narraciones que inventamos, de las que se apropiarán los otros, o hablarán de nuestra pasada existencia perdida y jamás conocida convirtiéndonos así en ficticios»¹».

"And the narratives we invent, which will be appropriated by others who, in speaking of our past existence, gone and never known, will render us fictitious"².

Now that we''ve gotten to this statistical measurement of the odds that a word be translated into another specific one, we realize that such a measurement is not very useful in practice because there is context, the fundamental element of translation that a computer is unable to consider.

But the equivalence-probabilities are, in fact, constantly affected by contextual and co-textual factors (by context we mean ''context of situation'', i.e. those elements of the extra-textual situation which are related to the text as being linguistically relevant: hence contextual. By co-text we mean items in the text which accompany the item under discussion: hence co-textual)³.

It remains therefore a statistical datum that, as such, can be useful only for knowing what happens in a given percentage of cases. On the basis of these statistics Catford proposes to create translation rules.

Provided the sample is big enough, translation-equivalence-probabilities may be generalized to form ''translation rules'' applicable to other texts, and perhaps to the ''language as a whole''-or, more strictly, to all texts within the same variety of the language (the same dialect, register, etc.)⁴.

Rules, extrapolated in this way, are the least useful generalization one can put forth concerning the translation process. They concern solely the odds that a single word, under given conditions, and in a given linguistic combination, is translated through use of a single other word.

A translation rule is thus an extrapolation of the probability values of textual translation equivalences ⁵.

I wouldn''t call it "translation rule", but, better, "probability of lexical output of a word in a given linguistic combination". It is something much more limited than a translation rule, and it is so limited because the undeclared aim is to command a computer to apply such a rule. One understands that when "human translators" are introduced:

For human translators the rules can make appeal to contextual meaning ⁶.

For non-human translators, i.e. computers, this is not possible, as we know. For a computer to be able to make out which context it must apply, it is necessary to input huge quantities of data to form an adequate database, and there must be careful routing as to the semantic area in which the work takes place. But Catford soon abandons the subject of odds and computers, for a theory that could be defined as "functional non-linguistic equivalence" ante litteram.

The SL and TL items rarely have ''the same meaning'' in the linguistic sense; but they can function in the same situation. In total translation, SL and TL texts or items are translation equivalents when they are interchangeable in a given situation..

It is no longer equivalence, nor linguistic correspondence: it is simple functional homology. There are no philological concerns for the prototext, pragmatically placed in the background, to the advantage of communicative functionality of the text, having - implicitly - a practical aim: being understood in a concrete situation, not expressing something extra-denotative in an artistic-literary situation.

The aim in total translation must therefore be to select TL equivalents not with ''the same meaning'' as the SL items, but with the greatest possible overlap of situational range ⁷.

Once granted the broadness of the situational range, the problem of equivalence is solved. Once more, someone tries to saddle a natural language with mathematical parameters:

translation equivalence occurs when an SL and a TL text or item are relatable to (at least some of) the same features of substance (the type of substance depends on the scope of the translation. For total translation it is situation-substance: For phonological translation it is phonic-substance: For graphological translation it is graphic-substance)⁸.

At last, we come to the key question of translatability and cultural specificity, of the implicit in culture:

In total translation, the question of ''sameness'' of situation-substance is a difficult one, and is linked to the question of the ''sameness'' or otherwise of the cultures (in the widest sense) to which SL and TL belong ⁹.

There are no two identical cultures, i.e. the statement seems a little too theoretical and abstract in order to have any repercussions on the practical side. Anyway, we are told nothing more regarding the translation of a text as translation of one culture into another culture.
Another subject that is touched upon is intersemiotic translation. Jakobson''s essay On linguistic aspects of translation into which this category of translation falls in its first, 1959 version, was published six years before Catford''s.

Translation between media is impossible (i.e. one cannot ''translate'' from the spoken to the written form of a text or vice-versa)¹⁰.

Since it seems impossible for Catford not to have read Jakobson, this statement has the flavor of polemics.

Conversion from spoken to written medium, or vice-versa, is a universal practice among literates; but it is not translation, since it is not replacement by items which are equivalent because of relationship to the same substance ¹¹.

Based on Catford''s complex - but maybe not as productive from the research point of view - definition, intersemiotic translation is not "translation".
In another chapter of the essay, Catford tackles another concrete problem going against any attempt to fall back on equivalence: the idiolect, expressive mode that is typical of an individual. His proposal is to find the ''equivalent idiolect''.

In such a case the translator may provide the same character in his translation with an ''equivalent'' idiolectal feature ¹².

How simplified Catford''s view of the extra lingual world is! Languages differ a little from one another, but fortunately there are ''same'' cultures within which individuals are all strange in the same way, for which reason the equivalent idiolects exist. The same goes for dialects:

Texts in the unmarked dialect of the SL can usually be translated in an equivalent unmarked TL dialect. When the TL has no equivalent unmarked dialect the translator may have to select one particular TL dialect ¹³.

Having also brilliantly overcome also the problem of dialect, for which reason if a character in the original speaks with a typical South Carolina drawl in the French translation, owing to geographic homology, he will have the typical cadence of the Lyon area, Catford passes to the delicate question of the historical connotation in a text.

Here, as in the case of geographical dialect, equivalence of absolute location in time is normally neither possible nor desirable ¹⁴.

It is no good for a text to have historical connotations (for certain terms) and, if it does, it is not ''desirable'' for the translation to remind us of them. But if two languages have a different history,

Here, if the TL has no equivalent register, untranslatability may result ¹⁵.

Untranslatability is the notion introduced whenever all other resources are exhausted. Research (as a whole) does not advance one iota, but the author''s image is enhanced.
Even for style there can be problems whenever a style does not already have an ''equivalent'' in the receiving culture. In this case,

[...] cultural factors may dictate the use of a non-corresponding style as translation equivalent ¹⁶.

Catford must have a closed text in mind (even if not specified), like a railway timetable or an instruction manual for the mounting of a tent. Because the problems that he poses always revolve around practical understandability, never center around expressiveness. Even when he starts speaking about the name of cultural words (realia), in this case of the sauna, he prefers a homological solution: abandoning any reference to the source culture and automatic insertion of an institution that, from the text''s functional standpoint, has the same aim. In this case: to wash up.

equivalence of material aspects of the institution are less important than equivalence in its major personal or social function ¹⁷.

The sauna can, therefore, be directly transformed into a bathroom.

Bibliographical references

CATFORD J. C. A Linguistic Theory of Translation, Oxford, Oxford University Press, 1965. ISBN 0-19-437018-6

JAKOBSON R. On linguistic aspects of translation, in On translation, edited by Reuben A. Brower, Cambridge (Massachusetts), Harvard University Press, 1959, p. 232-239.

MARÍAS J. Negra espalda del tiempo, Punto de lectura, 2000 (original edition 1998), ISBN 84-663-0007-7.

MARÍAS J. Dark Back of Time, New York, New Directions, 2001 (translated by Esther Allen), ISBN 0-8112-1466-4.

¹ Marías 2000, p. 14.
² Marías 2001, p. 11.
³ Catford 1965, p. 30-31.
⁴ Catford 1965, p. 31.
⁵ Catford 1965, p. 31.
⁶ Catford 1965, p. 31.
⁷ Catford 1965, p. 49.
⁸ Catford 1965, p. 50.
⁹ Catford 1965, p. 52.
¹⁰ Catford 1965, p. 53.
¹¹ Catford 1965, p. 53.
¹² Catford 1965, p. 86.
¹³ Catford 1965, p. 87.
¹⁴ Catford 1965, p. 89.
¹⁵ Catford 1965, p. 90.
¹⁶ Catford 1965, p. 91.
¹⁷ Catford 1965, p. 99.