9.  Chapter 9: Conclusions

The dissertation has attempted to combine practical criticism with translation theory so that there has been a two-directional deductive/inductive dynamic throughout the work. For this reason, it is helpful to divide the conclusion into four sections. 9.1 gives a quality assessment of the two translations of Tonio Kröger, Tristan and Der Tod in Venedig. 9.2 summarises the conclusions pertaining to translation criticism. 9.3 defines the strategic approach to literary translation and 9.4 describes the implications for the teaching of translation.

Both translators work within the narrow confines of what has been defined as the academic approach, in other words, the balancing act any teacher of translation goes through in order to produce a key to a set translation text in order to combine close fidelity to the SL text with a fluent TL text. At the level of mere information transfer, Luke essentially succeeds in this task. His versions of the three stories can be said to be competent, reliable and professional. Unlike Lowe-Porter, he rarely makes a lexical translation error or a grammatical mistake. In the appendices and in other quotations his translations are placed alongside Lowe-Porter’s for normative reference.

On the other hand, the Error Appendix has proven that Lowe-Porter’s translations fail even within the criteria of academic translation. The 187 errors in Appendix I (i.e. including those identified by Luke) consist of misreadings of German lexis and grammar at the surface level of meaning and, even worse, basic grammar mistakes in English grammar and usage. These are the kind of errors any teacher of translation is confronted with when teaching students with an inadequate knowledge of German and of the mother tongue. Indeed, some mistakes are even below normal student competence as has been seen in the Error Appendix with howlers such as “bath-hotel” for Badehotel (2. 5214) or in Buck’s examples such as “with big bones” for breitbeinig. The other chapters have shown that these errors are by no means harmless. Not only are poetic and stylistic effects lost in this version but there is a basic misreading and misrepresentation of Thomas Mann’s themes at the most elementary level.

The real scandal of the Lowe-Porter translations is not only the fact that translations of such poor quality have continued to be published uncorrected for over half a century (and will, no doubt, continue for some time to be the most widely read versions of Thomas Mann in the world) but also that Lowe-Porter still continues to have defenders. Opinions seem to be almost equally divided concerning the quality of her work. All translators make mistakes, but they are usually rare and relatively harmless as in the case of Luke’s versions, but to dismiss Lowe-Porter’s grossly inaccurate translations as “recastings” shows that common sense and basic linguistic competence are still criteria which cannot be ignored in the present debate on quality assessment.

The detailed analyses have shown that Thomas Mann’s prose has the same richness and density of poetry and that poetic, ironical and philosophical aspects are usually lost in the translations. Even though Luke’s translations are semantically reliable, they lack the poetry, humour and irony of the original and are often, in fact, dull. It has been argued that this failure to capture literary nuances is the inevitable result of academic translation. It is for this reason other strategies have been source-oriented suggested.

The starting part of the strategic approach is the realisation that for high literature and many other areas such as comedy or even marketing, the traditional academic approach fails because the semantic demands on the translator means that other aspects such as form, humour and wordplay are lost. It is at this point that a translation decision should be made concerning translating strategy. Present-day theory divides between source-oriented and target-language-oriented translation. The nomenclature varies from domesticating, communicative and Skopos-oriented to describe target-oriented texts to foreignising or semantic translations to describe source-oriented translations. These two strategies have been adopted with the suggested versions which by no means and by definition (i.e. in that two separate versions are offered) claim to be ideal translations of Thomas Mann. It is, however, claimed that it is better to produce either a fluent readable and enjoyable text in the target language or a very close text for the literary specialist rather than a compromise between these extremes which usually ends in dull versions following the academic approach.

The strategic approach has been developed to go further along these lines by suggesting there are many more possible strategies and each strategy is appropriate for differing types of text. The main departure from traditional ‘equivalence’ theories in translation is the redefinition of fidelity, which has always been assumed to mean semantic fidelity. Thus a faithful translation to any text normally refers to a semantically close translation. However, it has been shown that there are more important factors in certain types of translation. For example, for a translated comedy to be performed on the stage, it might well be more important for the translation to be amusing than to reflect every semantic item of the original with its equivalent. This has seen to be the case with Gotter’s highly successful translations for the eighteenth-century Gotha theatre. The techniques of compensation and ‘creative transposition’ are important in this area.

It has been argued that the proponents who claim that problem areas such as poetry, style, puns and dialect are ultimately untranslatable base their arguments on too narrow definitions of the process of translation. Indeed, they assume translation to be what has been defined as academic translation. The second-order semiotic approach of Levý has shown that Christian Morgenstern’s poetry can be successfully translated. This has also been implicitly the strategy of the various translators of Lewis Carroll’s Alice stories. The semiotic approach does not imply limitless, creative freedom. The analysis has shown that the translator should understand the semiotics of the original and then recreate a new text but along the same lines. Finally, as many semantic aspects as possible then need to be re-embedded in the text. Max Knight’s translations have been shown to vary considerably with regard to the success of their outcome.

Scientific equivalence-based theories of literary translation have been proven to be woefully inadequate. Not only has the formal refutation of Holmes’ use of mathematical models shown that equivalence theories fail even at the theoretical level but also the detailed analyses have revealed how rich in meaning and music great literary style can be and how far away we are from fully understanding these processes. The idea of encoding them in mathematical form is thus at the moment doomed to failure.

The strategic theory of translation derives some of its inspiration from Wittgenstein’s (1953) language-game theory. The translator should be playing the same language game as is played in the text. This has already been applied to poetry [page 197↓]and to comedy and also applies to philosophy where the main stress of fidelity is to the logical form of the argument. It has been seen that Thomas Mann has many philosophical passages embedded in dense literary prose and that the translation will come across with much greater clarity once the skeleton of the argument has been understood and displayed. The same language-game principles would apply to non-literary translation such as, for example, business letters, advertising and humorous speeches.

It has also been shown that great literary translation is possible. This area has been discussed within the parameters of Gentzler’s discussion of post-Derridean translation theory. It is at this level that the distinction between great poetry and great translation becomes blurred. It is no coincidence that the successful translators of literature have also been writers. The examples would seem to corroborate this view: Hölderlin’s translation of Sophocles, James Joyce’s translation of Finnegans Wake and Beckett in French translating Beckett into English. These examples alone refute the school that believes in the essential untranslatability of literature.

The area of dialect translation is another difficult area and relatively little has been written on this topic. Indeed, at the level of practice, most translators ignore dialectal features and most theoreticians claim that dialect is untranslatable. However, it has been shown that the strategic approach can be helpful even in this area. First, the translator needs to assess the extent of the dialectal features which may vary from light coloration to a new language. Secondly, the translator needs to find out the function of the particular dialect in the work which may be anything from sociolectal placement, regional coloration, exclusion, inclusion to humour and class dynamics or even any combination of both these and other functions. Thirdly, the encoding will depend on the type of translation which may range from total domestication (as has been seen to be the case with Gotter) to subtle metalanguage in that the translator explains the dialectal effects in an appropriate way in order not to disturb either the coherence or the tenor of the text. This is, however, an area where there is a great need for more research.

Many translation theorists are involved with the teaching of translation. The rejection of semantically bound equivalence-based definitions of translation for the translation of great literature does not imply that these approaches do not have their [page 198↓]uses. The exercise of translation both into and from the target language is, in my opinion, one of the most efficient ways of gaining a high level writing and reading competence in both the target language and the mother tongue. For first degree students, the academic approach is an excellent discipline particularly for regional studies-based texts, even though this strategy has been shown to be disastrously inadequate for the translator confronted with high literary texts.

The illusion for many is that if a linguist is highly competent in two languages he or she can translate anything in those languages. This is an illusion often held by literary publishers. The drastic effect of this mistake has been demonstrated by this analysis of the Lowe-Porter translations. The analysis is by no means intended as an attack on Lowe-Porter herself but on the whole publishing world and to a certain extent on certain academic and literary people who seem to be so blind with regard to the quality of literary translation.

As a corollary to the above, it can be seen that for the training of translators for an MA in translation studies, for example, other criteria than mere language competence would apply. (This is not to imply, of course, that everyone who studies literary translation will want to be a translator of some kind.) It has been shown that generally one can translate only as well as one can write so that the aspects of a literary translation course relevant for potential translators would not differ drastically from a creative writing course which has the aim of discovering the talents of its participants. One student may have a gift for translating plays and dialogues, another for humour and another may be a highly dextrous poet and so forth.

In conclusion, it is to be hoped that the Lowe-Porter debate will now be over and that a more creative definition of the translator’s role will have emerged as a result of this thesis. Literary translation is not the dull dictionary-bound activity suitable for pedants (even though the translator does often have to very precise), but is more akin to creative writing. Finally, it is to be hoped that the gap between creator and translator will be been at least partially narrowed as a result of this dissertation