Translation and Its Problems: an Inter-professional Perspective and Lessons for Accounting

concepts are common in accounting because accounting language is a highly specialised ‘dialect’ or variety of a ‘natural’ language (such as English). Such special varieties, characterised mainly, but not exclusively, by specialised terminology, occur where communication serves specific purposes. In the linguistics literature such varieties are referred to (inter alia) as special ‘registers’ (Salmond, 1998), ‘languages 2 Expressions lacking a precise definition, such as “probably”, “likely”, “remote”. 3 This assumption contrasts with the universalist “cloak theories”, which assume that any thought can be expressed in language and translated (Bruner, Goodnow and Austin, 1956). We believe, however, that this is a view often held by monolinguists who are unaware of difficulties of translation. APIRA 2010: The power of words and the power to translate Page 7 of 45 05/07/2010 7 for specific purposes’ (e.g. Engberg, 2006), or ‘jargon’ (Crystal, 2003; Allan, 2006; see also Evans, 2009). Such registers can facilitate communication among members of a group, such as a profession, by allowing economical, efficient and precise expression (Crystal, 2003; Allan, 2006) and capturing distinctions that ‘ordinary’ (everyday) language does not make (Allan, 2006). According to Mills (1989, p. 21): “As with other professional fields of knowledge, accounting in both its theory and practice is, and has been throughout most of its recorded history, peculiarly dependent on a specialized vocabulary or terminology, both to transact its business its business as expeditiously as possible and to differentiate it from other disciplines”. The translation of such specialized terminology can be particularly problematic. This is illustrated by Alexander and Nobes (2007, p. 186) with a simple example comparing the overlapping, but not equivalent meanings of the English terms depreciation, amortization and impairment with the German Abschreibung (any writedown, incl. depreciation and impairment) and the French amortissement (depreciation, amortization) and dépréciation (one-off write-down, e.g. impairment). This shows effectively that what may not be an obstacle to understanding in ‘everyday’ communication could lead to significant misrepresentations and misunderstandings in a highly specialized context, such as accounting. Such a lack of exact equivalence is also explored, in the context of the Swedish implementation of IFRS, by Dahlgren and Nilsson (2009), who find, for example, that Swedish accounting does not have “a concept corresponding to ‘income’ (encompassing both ‘revenue’ and ‘gains’)” (Dahlgren and Nilsson, 2009, p. 16). With reference to this example, they suggest (ibid, p. 21): APIRA 2010: The power of words and the power to translate Page 8 of 45 05/07/2010 8 “... one can easily see that the translation does not adhere to the principle of conceptual identity. The reason behind this is of course that the structure of Swedish accounting differs from the structure of UK or US accounting. Two professional contexts clash and the translation collapses. Since Swedish accounting has its roots in Germany, German accounting is faced with identical problems ...” As an additional problem, some (e.g. minority) languages, may suffer from a lack of standardized, and therefore inconsistent use of accounting terminology (see e.g. Fuller-Love, 1998). Prior literature in accounting that deals with the translation of specific accounting terminology also includes inter alia Rutherford (1983), Parker (1989), Zeff (1990), Nobes (1993), Alexander (1993), Evans and Nobes (1996), Aisbitt and Nobes (2001), but much of this is limited to one or few concepts (such as that of the true and fair view (TFV)) and/or is descriptive or conceptual in nature. The findings suggest for example that translations of TFV into the official language versions of the EU member states are not, as a rule, literal translations of the English original (Rutherford, 1983; Nobes, 1993; Alexander, 1993; Aisbitt and Nobes, 2001) nor applied equivalently (e.g. Nobes, 1993, Aisbitt and Nobes, 2001). Kosmala MacLullich (2003) finds that a variety of translations appear in Polish, which reflect a lack of consensus on the concept’s role. She argues that this is due to the fact that the Polish translations are incompatible with the meanings associated with TFV in English. Albu et al. (2009) explore perceptions of TFV in a transitional economy with a Roman law system (Romania), and find that this differs between auditors on the one, and preparers and regulators on the other hand. Many of the languages in the Eastern European transitional economies also do not have definite and indefinite articles. Alver et al. (1997, p. 44) suggest that: APIRA 2010: The power of words and the power to translate Page 9 of 45 05/07/2010 9 “For languages that do not provide for both the definite article and the indefinite article in their grammatical structures the introduction of the precept of ‘a true and fair view’ may well present formidable difficulties.” Meaningful literal translations of TFV are not possible (Van Hulle, 1993), and the translation of TFV as well as its national implementations demonstrate that countries managed to impose their own culture on an alien concept (Nobes, 1993; see also Parker, 1989). Archer and McLeay (1991) investigate whether accounting concepts are similar enough to permit equivalent translation but conclude that, at least for audit reporting, that is not the case, not only because of a lack of semantic equivalence but also because of pragmatic idiosyncrasies in language use. In other words, there is no translinguistic register of accounting (ibid.; see also Evans, 2004). Evans (2004) links problems of translation to the Sapir-Whorf hypothesis (cf. Sapir, 1929) which, in its weak form (‘linguistic relativity’) suggests that people who speak different languages perceive and think about the world differently. This is supported also by Monti-Belkaoui and Belkaoui (1983) who found that professional concepts are interpreted differently by unilingual speakers of different languages as well as by bilingual speakers. This would suggest that translation between one language and another is at the very least problematic and sometimes impossible. Literature exploring the interpretation of uncertainty/probability expressions has also been extended to cover translation. Davidson and Christman’s (1993, 1994) findings suggest for example that such terms in English permit more precise interpretation than their French translations. Doupnik and Richter (2003) find inter alia that the translation of such expressions from English to German “results in significant 4 The strong form, which suggests that thinking is determined by language, is not now generally accepted by linguists. APIRA 2010: The power of words and the power to translate Page 10 of 45 05/07/2010 10 differences in interpretation”, which they suggest may be due to poor translation or a lack of an available equivalent term in German. Doupnik and Richter (2004) extended this work by looking at interpretations of uncertainty expressions within extracts of International Accounting Standards, i.e. in context. They related this work to research on accounting and culture (Gray, 1988) and conclude that cultural differences may lead to a lack of equivalent interpretation of uncertainty expressions and thereby to differences in the application of International Accounting Standards. A similar study was conducted by Doupnik and Riccio (2006), with subjects from Brazil and the US. They also find some (albeit limited) support for the assumption that culture affects the interpretation of probability expressions. In summary, the key issues arising from prior accounting literature on translation are as follows: (i) Accounting language represents special registers of natural languages. For example, meanings and usages in everyday English differ from the meanings and usages of accounting discourse (e.g. ‘material’, ‘reserve’, ‘conservatism’). The same applies of course to other languages. (ii) There is no transnational register of accounting – i.e. accounting concepts differ internationally. (iii) It is not possible to achieve exact equivalence in translation. Translation is especially problematic when it concerns particularly culture-dependent concepts, such as ‘true and fair view’. (iv) Translation is also particularly problematic where cultures, and accounting subcultures, differ considerably. (v) Translation of ‘alien’ concepts often leads to a shift in meaning. (vi) Even near-equivalent translations (such as those of probability expressions) are often interpreted differently. 5 However, Chand and White (2006) find no significant differences in the interpretation of IAS/IFRS (i.e. not specifically probability expressions) between different ethnic/cultural groups in Fiji and conclude that different cultural values may be overridden by professional influence. APIRA 2010: The power of words and the power to translate Page 11 of 45 05/07/2010 11 These issues are important in the context of the internationalisation of accounting, through the work of the IASB and IFAC, and the increasing international adoption of their standards. To achieve equivalent implementations, high quality translations of the standards and guidelines are required into the languages of the respective countries. An additional complexity is created by the fact that the standards have legal standing in some jurisdictions. This is the case, for example, with International Financial Reporting Standards, which are incorporated into EU law and binding on listed companies for the preparation of their group accounts since 2005. However, European-wide application of IAS/IFRS may not be sufficient to ensure equivalent quality of financial reporting, because local traditions and cultures, including legal and political systems, financial markets, corporate governance arrangements, auditing and enforcement of regulation are likely to continue to exert an influence (see e.g. Ball, 2006; Nobes, 2006, Zeff, 2007). An additional problem is the translation of IFRS from the original English into other languages (Wong, 2004; Nobes, 2006; Zeff, 2007). The IASCF’s translation process is undoubtedly rigorous, but given the findings identified by prior research, not all problems have been eliminated, and it is likely that some never will. Having offered a brief review of some accounting perspectives, albeit with a scoped or limited focus, which have examined translation issues in accounting and how translation is undertaken, this research project follows by offering the following exploration of the literature outside of accounting in order to examine how other professions and disciplines position their philosophy or approach to translation; and in some cases how these diverse professions overcome, or at least address, translation problems where a need of equivalent interpretation is essential. It is offered in the 6 See http: //www.iasb.org/Use+around+the+world/IFRS+translations/IFRS+translations.htm APIRA 2010: The power of words and the power to translate Page 12 of 45 05/07/2010 12 hope that an understanding of this literature will be valuable also for the discipline of accounting.