Gardt subdivides his study (a 1987 Heidelberg doctoral dissertation) according to primary texts: there is a chapter on "Eveline", one on "Sirens", one on two poems from Pomes Penyeach, and one on Exiles. In each case he discusses at least two, but usually more, competing translations, covering all major and some minor German translators of Joyce in the process.
He starts off with a methodological chapter devoted to current trends in translation theory, adopting an approach which begins not with the translator translating but with the translator as reader. The translator-as-reader, just like any other reader, creatively acts on "impulses" (23) from the text, in the sense of Iser and Ingarden, to derive an overall understanding of it. Unlike any other reader he is reading and understanding it for the potential readers of the translation (31), which means he has to exercise special care. In today's version of the hermeneutic process we irrationally move "bottom up" to derive hypothetical "deep structures" and "top down" to test whether our assumptions check out on the surface. Once a consistent and cohesive interpretation (deep structure) has been arrived at, the translator's ultimate goal is to create an independent text which contains the same set of literary impulses as the original, allowing the target audience an equivalent reading experience.
In the case of "Eveline", Gardt mainly investigates strategies for rendering expressive sentence rhythm, diction, and leitmotifs. A [End of p. 160] few examples suffice to show that Goyert's (1928) translation is seriously defective, particularly in the areas of mind style and key words, whereas Zimmer's (1969) translation often displays great sensitivity to the text. Even at the very beginning, in the first sentence, Goyert's translation for "invade" ("herabsank") loses the crucial image; Zimmer's "eindrang" retains it (65). Later, for "she had begun to like him", Goyert translates "und dann hatte sie ihn auch bald geliebt", asserting exactly what the original does not imply. Goyert is also too formal in his presentation of the colloquial and idiomatic elements of Eveline's thoughts and blurs the distinction from genuine authorial passages (71). In fairness to Goyert, Gardt points out that Zimmer's translation profits from the increased temporal distance to the original and the body of secondary literature that has evolved.
The advantages of hindsight in translation are even more obvious in the case of Ulysses. In 1927, apart from a motley collection of basically unhelpful critical statements - Gilbert's book did not appear until 1930 - Goyert had almost nothing to go on except a four-day meeting with the author, whose command of German was questionable (94). His translation was granted the coveted authorisation, which subsequently served both as a selling point for the Rhein Verlag and as a cover for Goyert. But Joyce insisted on very extensive revisions for the 1930 edition, and Gardt quotes a spiteful Tucholsky as saying "Hier ist entweder ein Mord geschehen oder eine Leiche photographiert" (97). Forty years later, Wollschläger not only had access to the mass of published Joyce criticism but was able to shape his translation on the suggestions, support and advice of experts like Arno Schmidt, Klaus Reichert and Fritz Senn. He was also given four years to do it (98).
For the "Sirens" episode, Gardt contrasts Goyert's and Wollschläger's versions of musical analogies, expressive form, leitmotifs, neologisms, puns, and character diction. As expected, and more clearly than in "Eveline", Goyert is revealed as a reductionist who often defensively clings to the denotative content, loses most of the idiom and music of the original and produces a "normalized" text (127). Wollschläger, on the other hand, rises to the challenge and often creates exactly the stylistic effects that are at work in the original (according to our present understanding of it). There does come, of course, a point when translating Ulysses becomes a matter of ingenuity, sheer luck, and also utter hopelessness. When Bloom chuckles gleefully over his own "Pat is a waiter who waits while you wait", Goyert more or less gives up: "Pat ist Kellner, der bedient, während man wartet", whereas Wollschläger makes quite a creditable [End of p. 161] effort with "Pat ist ein Aufwärter, der aufwartet, während man abwartet" (136; but why not "... der aufwartet, während man drauf wartet"?). Neither translator can do anything with the "rows of cast steel" pun (117) though Wollschläger often tries to compensate for such unavoidable loss by providing a similar effect in the immediate context. Unfortunately, this practice may overstimulate the translator's creative energy and develop into a tendency to improve on the text. This is actually one of the few points of criticism that Gardt, after very careful deliberation, does level against Wollschläger (108, 158). (Would Wollschläger have gone as far as he did, one sometimes wonders, if Goyert had not stopped so short?)
On the poems "Alone" and "A Memory", both from Pomes Penyeach, Gardt takes into account no less than eight German translations - four each. His discussion covers a generous 54 pages, and since both poems are just a few lines long, he provides a substantial critical contribution on them - much more than for "Eveline" or "Sirens" where he works on excerpts and starts off from established or even hackneyed critical positions to which little additional insight accrues.
The difficulties in "Alone" are its associative imagery, its lyricisms and its complex mood. Despite this, all four translators do relatively well, even though Broch (1935) assimilates it to his own more expressionistic ways, Claes/Lohner (1957) tend to leave out the overtones of sex and guilt, Geilinger's (1967) version does not cohere well and Wollschläger (1981) perhaps slightly overaccentuates the erotic element.
"A Memory of the Players/in a Mirror at Midnight" (to give the full and semantically loaded title) is more interesting, both in its own right and because the translations diverge wildly. The problem can be narrowed down to an interesting ambiguity in the opening lines, which run:
For Exiles it is von Mettal (1919 - the first performance version) vs. Kremer (1956) vs. Reichert (1968). Gardt takes great pains to establish that Exiles exhibits the same thematic complexities we find in Joyce's other works, and bases his comparisons on an analysis of the handling of ideolects, dialogue technique, repetitions and word motifs. While the results (Reichert finishes first, von Mettal comes an honorable second, Kremer third) are again painstakingly worked out, one wonders if Gardt really required 69 pages to obtain them.
It is to Gardt's credit that he does not shy away from pronouncing judgment, positive or negative. He does it methodically, fairly and reliably. After all, there are good and bad translations and the target audience acts on recommendations or criticisms, not on analytical descriptions. Unfortunately, when Gardt praises a translation, he does not just say that it captures this and that impulse or effect of the original, he says this is so because the translator appears to have worked from a firm grasp or conception of the deep structure of the original. Similarly, if he criticises a translation, he does not simply say it is inaccurate or careless or superficial or without cohesion; from this he concludes that the translator appears not to have had a proper grasp of the text's deep structure. Intended as a reinforcement of and link-back to the theoretical basis, such phrases become devoid of meaning after a certain number of repetitions. Some quirks of the theoretical framework are too docilely accepted anyway; having to reconcile it with the findings of the comparative analyses leads Gardt to another questionable procedure which also stales with repetition. For each of the primary texts Gardt feels compelled to outline the scope of "possible" readings. This he does by pulling an uncalled-for and arbitrary number of half-hearted, more or less plausible hypothetical reader responses out of a hat. After that his analysis proceeds on the assumption that a translator should capture most (or is it all? some? the proper one?) of these responses. In view of this it is [End of p. 163] not a little surprising that his value judgments do come out convincingly and that the bulk of the investigation works as well as it does.
[August 22, 2000]