is a specific methodological approach to data (Olohan 2004).
Toury (1995) makes a case for DTS as a valid scientific method within Holmes’s (2004/1988) pure TS, and further proposes the Theory of Norms for TS, explaining that translation ‘fulfil[s]; a function allotted by a community’, in a specific sociocultural context. As posited by other scholars, such as Venuti (1995: 306), translation is primarily perceived as a social, cultural activity – the target text is ‘the site where a different culture emerges, where a reader gets a glimpse of “a cultural other”’.
Borrowing from social sciences, DTS see translation as a social, norm-governed activity, closely linked with the translator’s behaviour. Hermans (1999: 80) specifies that the term norm does not only refer to regular patterns of behaviour but also to ‘the underlying mechanism which accounts for this regularity’. His criticism of Toury’s proposal is useful to highlight two aspects connected with DTS: (a) their social nature, and (b) their ability to make predictions about future translational behaviour:
The mechanism is a psychological and social entity. It mediates between the individual and the collective, between the individual’s intentions, choices and actions, and collectively held beliefs, values and preferences. Norms bear on the interaction between people, more especially on the degree of coordination required for the continued, more or less harmonious coexistence with others in a group […]. Norms contribute to the stability of interpersonal relations by reducing uncertainty. They make behaviour more predictable by generalizing from past experience and making projections concerning similar types of situation in the future. They have a socially regulatory function. (Ibid.)
The author further criticises Toury for not considering norms as templates, as had been done before in Literary Studies (§2.1). In Hermans’ view, norms are like (generic) marks and, if literary scholars had considered genre a template for writers, Hermans (1999) sees translational norms as catalogues for types of solutions to which professional translators may resort. In a similar vein, Martínez Sierra (2011) underlines that the existence of behavioural norms entails the possibility for different norms (translation options, solutions) to exist. In this respect, the choice ←17 | 18→authors make ‘is meaningful against the background of the choices they rejected’ (Olohan 2004: 146).
Norms effectively take over the focus of interest that in TS had previously been occupied by the notion of equivalence. As noted by Pym (1995: 160), equivalence ‘always implied the possibility of non-equivalence, of non-translation or a text that was in some way not fully translational’, an approach that diametrically contradicts the basis of DTS, in which the study of actual translations by definition assumes that the object of study is fully translational. Indeed, for Toury (1995: 61), ‘it is norms that determine the (type and extent of) equivalence manifested by actual translations’.
Toury (1995) suggests that the aim and scope of researchers in DTS is to ascertain rules, norms and tendencies, from more to less established patterns or regularities, in translational behaviour. As put by Laviosa (2002: 79):
This type of analysis is performed not to evaluate the quality of a given translation, but to understand the decision-making process underlying the product of translation and to infer from it the translational norms adopted by the translator.
Toury (1995) lists a typology of norms that govern all the phases of the translation process, from assignment to submission. The focus in these pages is on operational, textual-linguistic norms and the way in which they have been identified to date by means of corpus research. In the scholar’s words, ‘[t];extual-linguistic norms […] govern the selection of material to formulate the target text in, or replace the original textual and linguistic material with’ (ibid.: 59). The sources for the study of textual-linguistic norms may be textual, that is, the text under examination, or extratextual, that is, statements made by people connected with the text in hand. In the CoPP this could include interviews with screenwriters or TV producers, as well as with subtitlers.
Since the 1990s, researchers especially interested in the study of translations through textual sources have adopted CL as a methodology for TS. Baker (1996: 177) uses the certainty-marker will to guarantee that a ‘computerised corpus will reveal regularities’ in translation practice. The idea of regularity is emphasised in Olohan (2004: 16), who argues that corpus research is not merely concerned with ‘observable’ data, but rather with ‘what is regular, typical and frequent’.
←18 | 19→
Ultimately, the aim of translational corpus studies is to uncover ‘what is probable and typical in translation, and through this, [to interpret] what is unusual’ (ibid.). This approach has led to a substantive number of studies wanting to identify universals, tendencies in translation which could be observed for all language combinations and in all translations. However, the notion of universals has been challenged repeatedly in the literature (Toury 2004; Corpas-Pastor 2008). According to Chesterman (2004: 11):
What ultimately matters is perhaps not the universals, which we can never finally confirm anyway, but new knowledge of the patterns, and patterns of patterns, which helps us to make sense of what we are looking at.
Scholars seem to agree that the main advantage of Translation Studies undertaken with corpora is the possibility to gather more significant quantitative data (Martínez Sierra 2011; Díaz-Cintas 2008a). On the other hand, three shortcomings of corpus methodology are:
a) text alignment or KWIC concordances may hinder identification of suprasegmental phenomena (Malmkjær 1998; Laviosa 2002; Zabalbeascoa 2004);
b) it is difficult to draw the line between tendencies and norms, which depend on corpora representativeness (Tognini-Bonelli 2001; Corpas Pastor and Seghiri 2007, 2010, 2016; Martínez Sierra 2011);
c) for the specific case of AVT, ‘corpus investigations focusing exclusively on the verbal component are at risk of overlooking the importance of the other semiotic codes to the meaning making process in audiovisual products’ (Díaz-Cintas 2008a: 3).
By adopting a corpus-driven approach, the present study aims to uncover patterns of types of solutions, tendencies or norms that point at decision-making in the English–Spanish subtitling of a selection of
