7.2.4 Discussion of Toury's work

Toury's methodology for DTS seems to be an important step towards setting firm foundations for future descriptive work. Gentzler lists four aspects of Toury's theory that have had an important impact on translation studies:

1. the abandonment of one-to-one notions of correspondence as well as the possibility of literary/linguistic equivalence (unless by accident);

2. the involvement of literary tendencies within the target cultural system in the production of any translated text;

3. the destabilization of the notion of an original message with a fixed identity;

4. the integration of both the original text and the translated text in the semiotic web of intersecting cultural systems.

(Gentzler 1993: 133-4)

Nevertheless, Toury's TT-oriented position is questioned by Hermans (1995: 218) in a review of Toury's earlier (1980) book. Certainly, Toury's stance risks overlooking, for example, ideological and political factors such as the status of the ST in its own culture, the source culture's possible promotion of translation of its own literature and the effect that translation might exert back on the system of the source culture. These are areas which will benefit from employing concepts from reception theory, notably con­sideration of the way in which a new literary work influences its audience (see chapter 9).

Furthermore, criticisms which Genztler makes of the earlier polysystem work (see section 7.1) can also be levelled at Toury: there is still a wish to generalize (or even overgeneralize) from case studies, since the 'laws' Toury tentatively proposes are in some ways simply reformulations of generally-held (though not necessarily proven) beliefs about translation. It is also debatable to what extent a semi-scientific norm/law approach can be applied to a marginal area such as translation, since the norms described are, after all, abstract and only traceable in Toury's method by examining the results of the often subconscious behaviour that is supposedly governed by them.

One might also question whether the translator's decision-making really is sufficiently patterned as to be universalized. Hermans (1999: 92), for example, asks how it is possible to know all the variables relevant to transla­tion and to find laws relevant to all translation. Toury's two laws themselves are also to some extent contradictory, or at least pull in different directions: the law of growing standardization depicts TL-oriented norms, while the law

118 Systems theories

of interference is ST-oriented. Findings from my own descriptive studies (Munday 1997) suggest that the law of interference needs to be modified, or even a new law proposed, that of reduced control over linguistic realisation in translation. This would bring together some of the varied factors which affect the translation process and make the concept of norms and laws in transla­tion more complex than is suggested by some of Toury's studies. These factors include the effect of ST patterning, the preference for clarity and avoidance of ambiguity in TTs and real-life considerations for the translator, such as the need to maximize the efficiency of thought processes and the importance of decision-making under time pressure (compare Levy's minimax strategy, discussed in chapter 4). When taking real-life consider­ations into account, it is worth noting that systems theorists in general have restricted their work to literary translation. However, Toury's inclu­sion of sociocultural factors in and around the translation process might well lend itself to the examination of the translation of non-fiction or

technical texts.

Toury's ambivalence towards the notion of equivalence has also been dis­cussed by Hermans (1999: 97), who furthermore highlights (p. 77) the confu­sion inherent in Toury's proposed terms 'adequate' and 'acceptable' because of their evaluative connotations in other contexts. Hermans prefers 'TT-oriented' and 'ST-oriented'. Finally, the ad-hoc nature of the ST-TT map­ping inevitably means that Toury's model is not fully objective or replicable. The alternative is Holmes's (1988a: 80) suggestion of an extensive 'repertory of features' approach, even though this is, as we have seen in chapter 4, potentially 'arduous and tedious'.