"I managed to play with them while also dealing with the wallpaper people"¹.

In the previous unit, we have seen that often in translation choices imply chains of repercussions: every choice implies other choices – or impossibility to choose – and sometimes this can occur even without the translator’s full awareness. Such choices can be objective, i.e. determined by the kind of linguistic material used. But they can also be subjective, therefore depending on the translator’s memory, on her ideology and her aesthetic canon.

Every translatant, i.e. every word located in a translated text, can be used in the attempt to reconstruct its origin. It is a matter of backtracking along the decision process that led to the choice of the translatant starting from the prototext. Such process is called by Levý "generative model". However, it is very important to distinguish it from Chomsky’s generative view because, but for the name, it has nothing to do with that linguistic approach. The organization of Levýs reasoning refers, if anything, more to the semiotic world than to generative linguistics:

The decision processes in translation have the structure of a semiotic system, having its semantic aspect (i.e., a repertory of units defined through their relation to their denotata), and its syntax (i.e., rules for combining these units – whether by units we mean paradigms or instructions) (1966: 1179).

Levý depicts words according to a recognoscative model, which is used by the reader to trace back the possible decisions of the translator. Such a model provides for lexical units be organized hierarchically, from the most generic to the most specific one. Let us examine Levý’s example (1966: 1174):

to exist
/						\
to move						to rest
/				\		/	|	\
to move as a whole					to move in parts	to sit	to stand	to lie
/	/	|		\
to walk	to ride	to fly		to drive
					|	\
					to drive	to be driven

This is the scheme followed by the author for the drafting of the prototext, and this is also the scheme followed by the translator in order to get to the possible translatants. Such a specifying path can, however, especially in case of incomplete awareness by the translator, be brought to a higher or lower level as compared to the author, causing a specifying or generalizing translation, respectively.

In our concrete case, the example is the English sentence:

His Lordship jumps into a cab, and goes to the railroad.

The translation

Her Excellency jumps into a cab and has himself brought to the railroad.

is specifying because, on the basis of the scheme that we just showed, it has one more level of specification (the one distinguishing to go and to be transported) as compared to the original, that only stated "goes".

Levý calls this specifications "surplus decisions". Since, however, the discussion following Levý’s article has centered on the terminology comparing generalization and specification. Because it is very useful, I think it important for the readers of this course to know that such surplus decisions are, exactly, definable as "semantic specifications".

Specifications and, more generally, translation decisions, can be necessary or unnecessary, motivated and unmotivated. They are motivated if made on the basis of contextual needs (of both linguistic and cultural character). They are necessary if the metatext language does not have some (grammatical, or syntactical, or semantic or cultural) categories, present in the source language of the prototext.

The broader the semantic segmentation in the source language when compared to that of the target language, the greater the dispersion of translation variants becomes (1966: 1175).

And, I add, the greater the necessary rate of specification when translating. Let us suppose that, in the following diagram, in the left column the source language is depicted, while in the right column the receiving language is depicted:

source culture	receiving culture
make	produce
	manufacture
	constitute
	create
	form
	earn ecc.

In the source culture, only one word is present to denote the meaning covered, in the receiving culture, by a series of more specific words. Consequently, translation is necessarily specifying. By contrast,

 

the finer the lexical segmentation of the source language in comparison to that of the target language, the more limited is the dispersion of translation variants (1966: 1175).

I add in this case too that in this scenario the number of translation variants is lower because translation is generalizing, and one translatant matches the semantic field that in the source language was covered by more than one lexical units:

source culture	receiving culture
produce	make
manufacture
constitute
create
form
earn etc.

Another way Levý synthesizes the translation process in the form of chains of decisions is that according to which, being an interpretive and, at the same time, creative process, there are two pertinent kinds of choice:

1. a choice within the semantic field of the word (or the group of words) of the prototext in order to find the interpretation (meaning) that makes sense in the context and co-text and eliminate the meanings that should be narcotized in the context and co-text;
2. a choice, from a sampling of target language words, of the translatant that best satisfies the expression of the meaning located during the choice made at 1.

The last aspect of translation as a decision process Levý deals with is the pragmatic one, i.e. of the practice of translation. Levý demonstrates that he is not a follower of the trends of his time, in which the "translation theorists" tended not to consider practical aspects. Since translation activity is inserted in a social context, in which both the time used to complete a translation variant, and the work lucrativity variant (connected to the previous one) are very important, it does not make any sense thinking of translation activity as a job in which the problem can be solved ignoring the time required to solve it.

For this reason, Levý suggests to consider the strategy followed by most translators a minimax strategy: i.e. aiming at producing the most effect with the least effort. This means that the translator does not pursue the best, ideal solution; she instead is contented of

a form which, more or less, expresses all the necessary meanings and stylistic values, though it is probable that, after hours of experimenting and rewriting, a better solution might be found (Levý 1967: 1180).

In Levý’s opinion, the translator aims at the minimally admissible result according to her own linguistic and aesthetic standards. She therefore focuses on the reaction of her reading public (model reader), trying to figure out in what percentage it can be composed of philologists, language purists, distracted or superficial readers etc. The example made this time by Levý is the French translation of the English words

not a little embarrassed

The translators considers two possibilities:

1. pas peu embarrassé

2. très embarrassé

With the decision 1, the stylistic trait of understatement (litotes) is preserved, but the risk is run that most French-purism-sensitive readers will be annoyed by this feature, a supposed Anglicism.

With the decision 2, the stylistic trait of understatement (litotes) is not preserved, but the risk is not run that most French-purism-sensitive readers will be annoyed by this feature, a supposed Anglicism.

Therefore, in Levýs opinion, the translator each time mentally calibrates her model reader and ends up choosing the solution creating the least dissent among most readers. This attempt, even if not well developed, to apply semiotic and mathematical models to the translation process is, in my opinion, very interesting; it should deserve a deeper elaboration, leading maybe to the creation of a more articulated theory.

 

Bibliographical references

CANETTI ELIAS Die gerettete Zunge. - Die Fackel im Ohr. - Das Augenspiel, München, Carl Hanser Verlag, 1995, ISBN 3-446-18062-1.

CANETTI ELIAS The Tongue Set Free. Remembrance of a European Childhood, translated by Joachim Neugroschel, in The Memoirs of Elias Canetti, New York, Farrar, Straus and Giroux, 1999, ISBN 0-374-19950-7, p. 1-286.

LEVÝ JIRÍ, Translation as a decision process, in To Honor Roman Jakobson. Essays on the occasion of his seventieth birthday, Den Haag - Paris, Mouton, 1967, vol. 2, p. 1171-1182.

---

¹ Canetti 1980: 58