"But since Mother, while saying these words, gazed at the book in her hand,I grabbed the opportunity and said: ‘Yes, I would like to. I’ll need it at school in Vienna’"¹.

Having examined the problems regarding translation loss and its management, we now face one of the questions that traditionally concerns the production of a translation loss more directly: wordplay translation. The text that most exhaustively deals with this problem is undoubtedly the monumental work by Dirk Delabastita (1993).

As in the case of poetry, the first point to be made on wordplay concerns the taboo about its supposed untranslatability. Many authors have spoken of untranslatability, while translations with translated wordplay continued to be published, even if this self-evident contradiction made no bells ring for any of them.

Delabastita argues against that axiom on two points. The first concerns the exaggerated importance attributed to the anisomorphism of natural languages. The advocates of untranslatability hold that it is founded on the fact that a pun has a metatextual charge that is complementary to the textual one. Let us examine a concrete example (Delabastita 1993: 347):

You come most carefully upon your hour,

Where carefully has two meanings that are relevant in the context:
precisely, attentively
anxiously, full of care or anxiety

The very fact that the double entendre is suggested by the author (Shakespeare, in this case) has metalingual implications. Because the reader, who in the absence of rules can devote herself to the primary sense suggested by the words in the context, when she meets a word like the ‘carefully’ in this example she faces a junction from which, unlike from the usual situation, originate two hermeneutic paths that are equally relevant. Neither of the two mentioned meanings can be narcotized in favor of the other, because they both suggest very plausible interpretations.

The reader, therefore, is induced to ask questions about which interpretation to choose and, moreover, is forced to (metalinguistically) wonder about the ambivalent nature of the word in question. But the metalingual charge is partially transferable. It is not systematically repeatable (for example, it is not at all easy to find a word in a language other than English having the same semantic ambivalence), nonetheless it may be repeatable. Two natural languages being anisomorphic does not imply that they cannot have a common feature. They have some shared features, as it is proved, for example, by the fact that Whorf, wishing to study languages that were as little isomorphic as possible to American English, turned to the Hopi language, that has no connections to the Indo-European branch, granted that Indo-European languages have shared traits. Such similarities would have invalidated the results of his research.

The second point against the axiom of wordplay’s untranslatability axiom focuses on its textual function. In order to understand such an argument one should be familiar with two semiotic terms: synfunction and autofunction.

"Autofunction" is the function that a sign element has within itself, independent of the context. "Synfunction" is the systems meaning of a sign element, i.e. its contextual meaning.

Those who say that wordplay is untranslatable imply that wordplay is autofunctional, i.e. that it has a value in itself. But if you consider the synfunctional aspect of wordplay, which is often very important:

Set aside, therefore, the untranslatability prejudice, Delabastita inspects the different outcomes of the translated wordplay, and locates nine categories.

The first category is called PUN > PUN, i.e. it is the outcome that from a pun produces a pun. Of course, the debate on the different perception that a translated pun can have on the model reader of the metatext. Said difference in perception is very difficult to be evaluated. Here is an example of a pun translated with a very similar pun:

The naked truth of it is, I have no shirt.

Where «naked» has the double sense of «pure, sincere» and «not covered by clothes». The same double sense we find in the Italian and in the Netherlandish version (Delabastita 1993: 195):

La nuda verità è, non ho la camicia.
De naakte waarheid is, dat ik geen hemd heb.

This is a very fortunate case, because the adjective in the metatext has a polysemy that is very similar to the one of "naked".

The second category is PUN > NON-PUN. In this case, the prototext’s wordplay has a text fragment without any wordplay as an outcome. Let us see this example of Delabastita’s:

Peace!
Be to me and every man that dares not fight.

Where «peace» has the double sense of "silence" and "absence of fighting". In the Italian and the Netherlandish versions the two sense are rendered in a very different way, without any pun:

Calma!
Pace a me e a tutti gli uomini che non osano lottare.
Rust!
en vrede voor mij, en voor iedereen die niet durft te vechten.

In these cases, the translation loss is not so much in the semantic content, as in the missing conveyance of the linguistic attitude, of the absence of the spirit of double entendre.

The third category outlined by Delabastita is PUN > PUNOID, i.e. the outcome of a pun translated with a pseudo-wordplay. Let us see an example with a French version:

I do adore thy sweet grace’s slipper.
Loves her by the foot.
He may not be the yard.

Where "yard" has a double meaning of "unit of measure" and "penis". The French version examined in Delabastita 208 ignores such a double meaning:

In these pseudo-wordplays, clearly (differently from the previous category) the translator recognized the wordplay and was inclined towards an explicitating rendering.

The following category is called PUN > ZERO. In the metatext, simply, there is no fragment that could be considered a translatant of the prototext wordplay. It is a case of omission.

Another category, named "direct copy", consists in transcribing the wordplay in the language in which it was expressed in the prototext, with no explanation, leaving the metatext reader to face the pun on her own.

There is still the possibility of transference, consisting in the covert use of a semantic of the prototext language even in translation. Such a category is usually called "semantic calque", and consists in coining (or simply using) a word in the receiving language that semantically is composed through the same procedure as the word to be translated.

There is the possibility of the NON-PUN > PUN solution, i.e. adding a pun not present in the original, usually as a compensative device (see the unit 10 in this part of the course). Delabastita distinguishes this category from the ZERO > PUN category, in which the wordplay is created from nowhere, through the addition of verbal material that in the original does not exist. In the NON-PUN > PUN category, however, the prototext material is present, but does not have a play on words as an outcome.

The last category is the one where in the place of the pun there is a sort of metatextual apparatus, a footnote for example. But we will deal with that in the next unit.

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.

DELABASTITA D. There’s a Double Tongue. An investigation into the translation of Shakespeare’s wordplay with special reference to Hamlet, Amsterdam, Rodopi, 1993, ISBN 90-5183-495-0.

¹ Canetti 1999: 76.