CC BY
38Journal of Siberian Federal University. Humanities & Social Sciences 2 (2014 7) 210-221
УДК 81'373.46
What Translators Do to Terminology: Prescriptions vs. Performance
Andrei V. Achkasov*
St. Petersburg State University 7/9 Universistetskaya nab., St. Petersburg, 199034, Russia
Received 02.11.2013, received in revised form 10.11.2013, accepted 18.12.2013
The article brings into focus variability of designation in bilingual terminology transfer. Onomasiological view of terminology equivalence presumes decontextualized coordination of concepts and linguistic labels and thus infers "one-to-one" inter-language terminology relation. This guiding principle is rarely applied in actual LSP translation process resulting in term forms variation. The gap between prescriptions and performance is traditionally accounted for as the arbitrary treatment of terminology by translators. Patterns of term variation in the reality of LSP translation depend on systemic, semantic and formal characteristics of terminology and, as a result, bear on terminology translation as a problematic concept.
Keywords: terminology, equivalence, LSP translation, terminology translation, variability of terms, transfer prescriptions, performance.
When I use a word it means just what I choose it to mean - neither more nor less.
Humpty Dumpty
Terminology affects translation. The issue has been more or less explicitly considered in research on both LSP translation and terminology work. One of the key points often brought to attention is that particular LSP texts cover segments of terminology systems, while specialized concepts, being by nature decontextualized entities, retain their systemic relations. The 'terminology stock' of a particular text builds up its 'information model' embedded in the terminology system of a related field, which is presented as a set of superordinate, subordinate and coordinate concepts. Thus bilingual or
multilingual terminology transfer should be based on prescriptive rules: the only proper way to 'translate terminology' is to 'identify equivalent concepts', i.e. to relate concepts as system-bound. Performative aspects of LSP translation contradict prescriptions, and, in turn, translators in many ways affect terminology. They often seem to be unaware and unconcerned about such rules and focus more on labels ('words') than concepts, use contextual and multiple designations or descriptive ways of concept representations within target texts, which confronts the nature on terminology. The latter issue has been the point
© Siberian Federal University. All rights reserved
* Corresponding author E-mail address: a_v_achkasov@mail.ru
of attention as well, predominantly as a matter-of-course though undesirable and non-systemic byproduct of terminology use.
Prescriptions and performance in LSP translation relate to the concepts of 'terminology translation' and 'translation equivalence' within the frameworks of translation theory and terminology studies. As a starting point I will only briefly refer to the key competing approaches.
Terminology vs. translation
The study of the problematic practice of 'terminology translation' traditionally operates under general semasiological assumptions, bearing on terminology as a subtype of lexical stock. The greater part of this research sets off with the highlights on performative aspects of translation in terms of inter-lingual and trans-systemic transformational 'manipulations' with terminology. The nomenclatures of such 'manipulations' vary due to the underlying theoretical assumptions concerning concept-designation dualism and particular LSPs. They include equivalent selection, concept identification, semantic and structural calquing, loan translation, decompositional and descriptive translation, borrowing, transliteration, quasi-defining etc. The other set of generalizations in such research regards terminology equivalence both in terms of the 'degrees of equivalence' and terminological equivalence as a subtype of a vaguer and more general concept of translation equivalence. Terminological equivalence is defined as semantic, notional, systemic, prototypical, conceptual, cognitive or decontextualized types of equivalence as opposed to functional, explanatory or contextual equivalence (Adamska-Salaciak, 2010).
Such conceptualization is the foremost and, presumably, the desirable outcome of semasiological insights into terminology translation, no matter that it is often carried
out as a case study of a particular LSP and is said to be equally concerned with the so called 'practical recommendations'. Theoretical output of this linguistic-centered research to a great degree overlaps with onomasiological view on terminology equivalence and at the same time overviews its practical applications that draw on terminology harmonization, multilingual information retrieval and automated LSP processing in general, including machine translation. The onomasiological approach draws on terminology equivalence in non-translation terms. In ISO 1087-1. 2000, equivalence is defined as "relation between designations in different languages representing the same concept" (ISO 1087-1. 2000: 9), lexical items or 'linguistic labels' being only one kind of designations. Thus, the concepts are treated as system-bound and language-independent, or, in other words, equivalent terms represent 'the same concepts', though have different 'linguistic labels'. Of course, concepts may overlap partially or have one-to-many relations, which again may be described in non-translation terms, such as, for example, 'inter-language synonymy' or 'inter-language quasi-synonymy, with a difference'. As follows, onomasiology is in fact little concerned with performative aspects of translation. To give it a translation perspective, we might say that this approach treats terminology as placeables, providing theoretical and operational basis for achieving pre-translation equivalence. Taking this as a theoretical assumption it would be proper to categorize transformational operations discussed in translation studies as either manipulations with linguistic labels or identifying degrees of inter-language synonymy. This allows to strip down the overall logic of conceptualizing about terminology translation by 'transformational operations'.
Related speculation on 'translation equivalence' has been intense and many, if not
all, reasonable interpretations have been given. In his tentative glossary A. Pym sums up the output of this speculation:
Equivalence: A widespread term for a relation that many believe in and no one can prove beyond the level of terminology. We should accept that equivalence has no ontological foundation, since translation problems allow for more than one viable solution. This means that, in the field of translation problems thus defined, equivalence is always "belief in the translation as equivalent of an ST". Recommendation: Always make it clear that equivalence means equivalence-belief, and indicate who is supposed to be holding that belief. (Pym, 2011: 81)
Following this definition, the vagueness of the concept seems not to be the case with 'terminology translation' or 'terminology equivalence', because terminology is (or is presumed to be) systemic and monosemic. Thus, we come up with a very plain concept terminology translation, which is, in fact, not translation by nature, but a process of substituting concept-coordinated linguistic labels. Translation and terminology, therefore, turn out to be contradictory concepts:
Terminology vs. translation: If a distinction must be made, let us propose the following: translation involves the obligation to select between more than one viable solution to a problem, whereas terminology seeks situations where there is only one viable solution. (Pym, 2011: 93)
LSP translators and translation-oriented terminologists, therefore, should be concerned with coordinating concept systems that exceed the terminology stock of particular texts and
represent decontextualized schemata. Prescriptive recommendations of this kind are typical:
In specialised-language translation, the translator has to structure terms of a given text by reference to a conceptual system. Therefore, it is highly important to identify the conceptual system a term is embedded in, independent from both the source language and the target language. (Edelmann, 2012: 2)
LSP translators are well aware that practical application of this guiding principle is hindered by inconsistences and current developments in terminology systems, trans-systemic differences (as in law or education terminology), lack of relevant sources, ambiguity of related concepts, time pressure etc. In fact, this principle is rarely, if at all, applied in actual translation. Nevertheless, some LSP researchers make even more uncompromising claims, that are not uncommon either:
There is one essential requirement for successfully transmitting a complex body of knowledge from one language community to another: There must be a target-language terminology that exactly mirrors the source-language terminology and that is precisely pegged to the source-language terminology so that it can be consistently applied by all translators and universally understood by all readers (Wiseman, 2002: 1).
This is fair. In theory. If terminology translation cannot be carried out as decontextualized coordination of concepts and linguistic labels, then the target text, again, in theory, would a priori lack essential informative qualities, no matter how functional it might be in other aspects. However, practice contradicts theory. Translators somehow seem to ignore or
override the fact that terminology essentially modifies the nature of translation, perform 'manipulations' with linguistic labels without systematic reference to concept systems, and produce bundles of target texts that are used in professional communication. Under such circumstances the achievability of terminological equivalence proper and the 'degree of termness' (Shelov, 1990) of items produced during translation is doubtful.
The LexALP case
In 1995 the Alpine Convention, an international treaty between the Alpine countries entered into force. The document "aimed at promoting sustainable development in the Alpine area and at protecting the interests of the people living within it" (Morandini, 2013: 6). The official languages of the Convention and the Implementation Protocols are French, Italian, German, and Slovene. The Protocols were drafted in one language, then translated into other official languages and submitted to the stakeholders for review and amendments (Chiocchetti, 2011: 553). In 2000, the four versions of the Convention and the Protocols were formally declared equivalent and "fully harmonised from the point of view of both language and style, without any modification of the content. <...> Nevertheless, it became clear during the following years that this formal process had left several gaps and inconsistencies between the four language versions" (Chiocchetti & Voltmer, 2008: 51). These gaps mainly related to various degrees of terminological equivalence and therefore in 2005-2008 the LexALP project ("Legal Language Harmonisation System for Environment and Spatial Planning within the Multilingual Alps") was carried out to harmonize the legal and scientific terminology of the Alpine Convention.
The harmonization work could not formally follow any prescribed order, such as that of ISO
860. The harmonization project started with analysis of the translations of all documents in four languages that had been formally declared equivalent. Therefore the linguistic component of harmonization had much more weight in this work than it is generally supposed to have. L. Voltmer accounts of the following illustrative case:
When extracting four corresponding strings from an AC text, one ends up for example with UVP-Recht (German), droit national applicable aux etudes d'impact (French), normativa sulla VIA (Italian), and pravni predpisi o PVO-ju (Slovene). The German string is a legal term with a precise meaning in law, but the expressions in the other languages are not. An explanation for this is that the Protocol on Transport was first written in German and then translated into the other languages. As a consequence, the German term would need a definition whereas the others would not. But then, can there be harmonized equivalents when there is only one term? In any case, for the benefit of translators, also a quartet of four corresponding phraseological units were harmonised in LexALP. (Voltmer, 2008: 81)
Further on, L. Voltmer makes a clear distinction between terminology as designation of concept with a precise meaning and 'phraseological units'that mean "anything but a legal concept" and "are used to indicate that a legal definition is not required" (Voltmer, 2008: 81). The question here is whether the German term, while being a term, should be harmonized as phraseological item or whether phraseology in other languages should be classified as terminology and supplied with definitions only because the draft document was written in German. Or, may be, the question is whether a concept designated by the German
term is essential to the concept system of the Alpine Convention. Anyway, in this situation comparing linguistic strings in different languages is a step that has to be taken at least at a pre-harmonization stage.
The participants of the harmonization process have made a few notable remarks concerning the opposite type of discontinuity in multilingual terminology. When analyzing texts in different languages, they registered differences in the meanings of similar concepts, designated by labels, that appear to be harmonized linguistically:
... terms may seem easy enough, but their legal definitions might diverge all the same. For example, the two Romance languages Italian and French seem to have a perfect correspondence for the terms protezione del clima and protection du climat (climate protection). Even in English the Latin etymology is obvious, and the meaning seems to be granted. Only when it comes to define the concept for the Italian legal system on the one hand and the French and Swiss legal systems on the other hand, then we find out about the profound difference. In Italy protezione means protection of an object from negative influence, for example protection of a child, a building or a computer. The protection of something as complex and constantly evolving as the climate is always labeled salvaguardia, a word with the same roots as safeguard. From this surprise at definition level we have to go back to the term level and couple the Italian salvaguardia del clima with the French protection du climat, even though the Italian protezione and the French sauvegarde do exist. Any uninformed user of the database would consider this, judging from the linguistic point of view and backed by any Italian-French dictionary as an error (Voltmer, 2008: 81).
A similar case is commented by E. Chiocchetti:
Some. terms are apparently easy to translate, yet at conceptual level there might be discrepancies... In such cases, it is mostly the meaning at national level that is applied to a term such as zona montana in Italian or Berggebiet in German (mountain area). In fact, from a purely linguistic point of view, the two terms are a perfect translation of each other. However, if the legal background is analysed, some important differences at conceptual level can be detected. In Italy mountain areas can be considered such starting from 600 m above sea level, in Austria and Germany only above 700 m. <...> The Protocol on Mountain Farming refers to zone montane in the Italian version and to Berggebiete in the German language version without defining the concept. Hence arises a problem of interpretation at national level (Chiocchetti, 2011: 537).
There is nothing unusual in such types of inter-language terminology relations and both cases may be accounted for as regular instances within ISO standards on terminology work. The first case appeals for harmonization at the level of designations (term level proper), the second case is a typical instance of 'minor differences' in concepts, that have to be reduced resulting in harmonized definitions. Both procedures may be agreed upon, performed and approved by the stakeholders only at the pre-translation phase. However, the case of the Alpine Convention is not typical: the text had been translated and formally declared equivalent before the harmonization work started. It could be objected, that harmonization basically starts only when there is a certain amount of translations in the related LSP and aims to normalize of professional communication in the specialized sphere. If so, 214 -
and following the procedures of harmonization, it would be relevant to build up the concept system of the Alpine Convention proper, and, as the first step, harmonize concepts, not designations. This could not be done within LexALP not only because it embraces fragments of various specialized spheres and is restricted to a very limited number of parallel texts, but also because the core of the 'specialized sphere' is multinational and relates to six national legal systems and four different languages. Some legal systems share the same language, while the others are multilingual. Thus, both inter-language harmonization and harmonization within one language was required at the concept level. If such a harmonization could be carried out, a completely new harmonized legal system would be built, just like international treaty terminology in general "might easily reach the level of secondary legislation" (Somssich R. Fazekas, 2012: 53) only partly overlapping with national legal systems.
Following this logic, no translation of the Alpine Convention that would be equivalent in a legal sense is possible. Indeed, going back to the example of 'phraseological equivalence', if the German legal term UVP-Recht is translated by 'the expressions in the other languages' that are not terms, then there is no point of translating the source term at all. It is pointless both ontologically and from a functional perspective of 'legal consequences', because the German text does imply a legal concept while the translations represent 'the expressions in the other languages' without any reference to the national legal systems. Such translation would make sense only if the 'the expressions in the other languages' referred to the German legal concept. Moreover, if documents are drafted, for example, in Italian or French, then there is a good possibility that they will operate with other national legal concepts that might have only 'phraseological equivalents' in other languages. Then 'the choice' of concepts
operating within the Convention will depend on the criteria of the 'original language priority'.
The power of linguistic labels
The situation as thus described implies that concepts should be defined prior to translation or even drafting. And yet the fact is there: not only were the Alpine Convention and the related Protocols translated before any systematic harmonization, but they were also formally declared equivalent. Translating without agreement on concepts and terminological equivalents is a daily bread of LSP translators in general. Knowledge in the related 'subject fields' is not of great help due to rapid changes in terminology and increasing interdisciplinarity of LSPs. The question that stems from this is what factors rule the 'terminology transfer' in translation, or how translators make their choices? The aspect of the quotations above that bears emphasizing and partly answers this question, is how the choice of terminological equivalents is discussed by the members of the harmonization project. L. Voltmer mentions that some 'terms may seem easy enough'. He describes the relation between the terms protezione del clima and protection du climat as 'a perfect correspondence' with a reference to their Latin etymology. E. Chiocchetti gives a similar remark mentioning that some 'terms are apparently easy to translate', and the terms zona montana and Berggebiet 'from a purely linguistic point of view <. .> are a perfect translation of each other'. Both translators and terminologists are, of course, well aware (or are supposed to be) that there nothing 'purely linguistic' about terminology equivalence, and that 'perfect correspondence' has nothing to do with Latin etymology. However, such analysis to a certain degree emulates the way translators come up with equivalent strings in translation. Thus, the 'power of linguistic labels' overrides rationale of relating concepts.
A few aspects of the process of selecting equivalent strings for source terms may be illustrated by the empirical data gathered at the SPbU Translation Laboratory. The results of the experiment are partly discussed in (Achkasov, 2010). The process of translation was logged by specialized software, including typing, deleting, using reference materials, such as electronic dictionaries, corpora and translation memory. In addition, screenshots were made on each mouse-click. Every step that translators made selecting equivalent terms (labels) was tracked and analyzed. The texts for translation on branding and terminology management include terms that have single and multiple equivalents available in open source glossaries. Equivalents for some terms, such as brand extension and brand stretching, may be identified only through concepts analysis as these concepts overlap and are often translated both as rastiazhenie brenda and rasshirenie brenda, though only one correlation is correct. The texts also include terms that have no conventional equivalents, such as terminology extraction, automatic term recognition, exact/fuzzy match, brand gap, brand ecosystem etc. The objective of the experiment was to measure time distribution in the process of deciding on terminological equivalents. The byproduct of this experiment is illustrative of the steps that translators undertake in identifying translation candidates.
Selection and in some cases coinage of linguistic labels are a series of operations registered by a keylogger. They predictably represent a 'label - to concept - to label' process. However, the process is often reduced to a plain 'label - to label' transfer and consists in consulting bilingual dictionaries and glossaries with subsequent verification of terms in a monolingual corpus. Translators generally preferred to use target terms that had similar formal structure or were characterized by semantic 'word-to-word
equivalence'. The same type of 'word-to-word' correspondence mattered when target terms were coined for unlabeled concepts, such as automatic term recognition or fuzzy match. The attempts to identify equivalent concepts for such terms in most cases were preceded by looking up general dictionaries for the term components such as recognition and fuzzy and subsequent experimenting with various word combinations. In the post-translation interview many translators could not explicitly explain their choice of both strings for unlabeled terms and of multiple equivalents presented in glossaries. A common explanation was that the term was verified statistically, i.e. had more hits in the target corpus. Concept analysis (searching for definitions and contexts) was registered when translators were dealing with designations that did not provide for intuitive 'on-the-spot understanding' of concepts (brand gap) or with overlapping concepts such as brand extension and brand stretching. But even in such cases the analysis was followed with looking up general dictionaries for single words equivalents (gap, extension, stretching). Thus, the choice of designations was driven to a significant degree by the 'linguistic labels logic'.
Diversity vs. uniformity
I must assume that in the discussion above I ignored an important issue, which I did intentionally. The point of the LexALP project was to build a new system of concepts. The concepts had to be internationally recognized, or, rather, the scope of application had to be both national and international. The new concept system was supposed to fit in, or, at least, not to clash with a number of national systems. On the contrary, the terminology in the experiment (the fields of branding and terminology management) does not need harmonization, partly because it is not national and partly because it has been and still is actively borrowed into Russian. So,
whatever Russian designations translators choose for new concepts, they, supposedly, will do. And yet, no matter how significant the differences in the transfer of legal terms in the LexALP project and terms in the technical domain of terminology management might be, in both cases translators seem to follow the 'linguistic labels logic'.
In fact, this is how new labels often appear bringing new concepts with them. Whenever new linguistic labels contradict patterns of term formation in a related LSP, it brings about dissatisfaction of the target LSP users and 'stylistic problems'. In such cases designations are subsequently modified and for some time many competing designations may coexist. It hardly causes any communicative problems albeit often provokes irritation of both LSP users and translators. The multiplicity of designation is as well a regular issue for criticism in research.
In the reality of LSPs the desirable relation 'one concept - one designation' is rarely the case. The diversity of designations is not solely the result of translation, it is an intrinsic feature of LSPs in general. A very illustrative example is provided by M. Rogers. The focus is on the instruction for a piece of medical equipment "in which a ST term - TT term relationship could reasonably be expected to have a textual equivalence probability of one <...> as a need for consistency in term selection and translation is implied by the purpose of the text" (Rogers, 2008: 104). A part of this equipment is a valve for which the original German text provides four designations: Ausatemsystem Schalldämpfer, Gerät, Schalldämpfer and Ausatemsystem. Similar sets of synonyms are found in French and English translations but the 'patterns of variation' are different in French (valve d'expiration de type silencieux, silencieux, dispositif, produit, valve d'expiration, vanne d'expiration, toutes les pièce) and English (the exhalation and muffling system, muffling system, exhalation system, device). Both
the number of occurrences of variants and the scope of their reference (different functions of the device) differ in the source and the target texts. (Rogers, 2008: 105).
However, uniformity of terminology is one of the prescriptive imperatives in LSP translation and translation-oriented terminology management: "Ensuring that translators and others keep an organized and easily retrievable record of term research and choices made ensures that the same term will be translated in the same manner whenever it appears in a similar context. Hence, terminology management promotes terminology consistency" (Gómez & Allard, 2012: 38). Surprisingly, in actual LSP translation the imperative to preserve uniformity of terminology has its flip side often bringing about linguistic, and hence, communicative problems. It is most visible when translators use terminology management systems with 'get term' function, i.e. automatic pasting of terms into the source segments as is shown in the example below (the source segment and the source segment with pasted Russian terms):
Cyclo-oxygenase-2 (COX-2) selective nonsteroidal anti-inflammatory drugs (NSAIDs) are as effective as acetaminophen and nonselective NSAIDs in treating of osteoarthritis, and are equally effective in reducing pain and inflammation and improving of joint function for patients with rheumatoid arthritis, when compared with nonselective NSAIDs.
циклооксигеназы-2 (COX-2) selective нестероидный противовоспалительный препарат (НПВП) are as effective as ацетаминофен and неселективный НПВП
in treating of остеоартрит, and are equally effective in reducing pain and воспаление and improving of joint function for patients with
ревматоидный артрит, when compared with неселективный НПВП
Getting target terms from a term base, no doubt, provides for terminology consistency and saves time. The task of a translator is to identify correct word forms and fill the gaps between terms. The problem is that translators are often inclined to keep terms as nominal collocations and preserve their positions in the sentence. As a result, the final version of a target sentence turns out to be a syntactically ambiguous utterance, though truly consistent in terminology. It is even more palpable when translation is done into a foreign language, which is a dominant practice in Russia. Inclination to keep terms as nominal groups as they are found in glossaries and term bases results in hardly readable linguistic chains such as ".. .project implementation in continuous and on-the-job training and retraining of highly qualified post-graduate specialists with independent research activities skills in different specialties and areas of training" etc. This is, of course in caricature, but the reality is not appreciable either.
In the 'ballpark' the variability of designation is admitted as a state-of-the-art though definitely undesirable reality: "In the initial stages of translation, variability of terms in the target language (in our case English) is an inevitable part of the decision-making process. Variability of terms is dangerous for the integrity of concepts." (Wiseman, 2002: 12).
Terminology translation as a problematic concept
The reason of term forms variation is not solely the arbitrary choice of translators but, as well, structural, pragmatic, functional, stylistic and other features of LSP. Thus, the 'linguistic labels logic' is not completely false, even in the perspective of terminology work. According to
prescriptive strategies for LSP translation and translation-oriented terminography, 'to translate a term' implies, first and foremost, identification of a related (identical) concept in the conceptual system of the target language. On the other hand, formation, standardization and harmonization of designations in general presume a certain degree of term motivation and LSP-dependent semantic and formal patterns. Translation-oriented terminography supplies prescriptive rules for the selection and formation of designations as well:
When specialized knowledge is transferred between language communities, the delimitation of concepts is not always identical in a given pair of languages. In comparative terminology, the process of term identification reveals any discrepancies, as proper designations may not be found in one of the languages. In such cases, the terminologist's role is to describe the gaps and propose designations to fill them. In order for the proposed term to be acceptable and valid, it must be based on sound knowledge of the target language's rules of lexical formation, must be harmoniously integrated into the existing set of terminology, and must be clearly presented as the terminologist's proposal. (Pavel, 2011: xviii)
If we take a closer look at similar prescriptive sets of rules that a translator should follow when 'translating terminology', they will be, again, concerned with designations, not concepts. One of such sets provided by V. Leichik and S. Shelov includes (1) identification of the equivalent term in the target language, (2) semantic convergence (creating a new target term by assigning a new meaning to a target word combination), (3) semantic and structural calquing, (4) loan translation, (5) using Greek and Latin components, (6) descriptive translation.
Only point (1) relates to concept analysis proper, while the other points imply manipulation with labels. These prescriptive rules are often unrealistic either.
Though prescriptive rules imposed by terminology work in many ways contradict performative aspects of LSP translation, it seems that an LSP translator should do some sort of terminology work, and to have competences of both a terminologist and a translator:
Terminography must not be confused with translation. Translators need specific terminology for specialized texts, but this does not imply that they themselves must develop the terminology, nor that they have to deal with all the terms in the special subject field in question. Working in terminology does not mean translating a term from one language into another based on supposedly equivalent designations, but gathering the designations that users of a language use to refer to a concept and ultimately, if necessary, proposing alternatives in those cases where speakers' designations are unsatisfactory. While translators are not terminographers, in daily practice the distinction between these two groups of professionals is often blurred. (Cabré, 1999: 115)
A translator is not commissioned to standardize or harmonize terminology, and is able only to manipulate linguistic labels, though it is not the same as 'to translate'. According to the quotation above, nobody seems to have the commission to 'translate terminology'. Translators must not develop terminology. Working in terminology does not mean translating a term from one language into another. Thus, the concept of 'terminology translation' turns out to be an aoristic and unstructured one, relating both to various manipulations with linguistic labels,
and 'identifying' target concepts. Producing 'equivalents', which are not equivalent in terminological sense, does not imply 'translating terminology'. Producing 'textual equivalents', on the other hand, appeals to functional and systemic equivalence through the complex task of 'producing texts' thus implying the potential non-achievability of "terminology translation" proper.
Conclusion
De rigueur scenarios for bilingual terminology transfer impose imperatives for coordination of concept systems and preserving uniformity of designations. Translation-oriented terminography and translator's practical guides address the criteria for the formation of target linguistic labels. Such criteria appeal to the general ideas of linguistic consistency and 'language rules', and, in turn and more or less explicitly, regard the systemic nature of terminology. Thus, onomasiological approach to bilingual terminology transfer implies the idea of pre-translation equivalence.
As has been shown, the reality of LSP translation contradicts theory. Regardless of the type of 'specialized sphere' and professional experience, translators often prefer 'label-to-label' terminology transfer, following LSP-dependent semantic and formal patterns of term formation. Such practice is not completely arbitrary, though it seems to disclaim essential principles of terminology work.
Thus the concept of 'terminology translation' is challenged as relating to the conflicting scenarios of manipulations with linguistic labels, and 'identifying' target concepts.
These issues have multiple competing and overlapping interpretations. My point was not to give another perspective of the concept of 'terminology translation', but to bring into focus the gap between performative aspects
of treating terminology and prescriptions on visible, it has not become an issue in translation how it should be done. Though this gap is very studies.
References
1. Achkasov A. (2011) Distribution of activities in the process of terminology search. [Raspredelenie tipov deistvii v protsesse terminologicheskogo poiska (na material log-dannykh)].
Universitetskoe perevodovedenie. Vip. 11. Materialy XI Mezhdunarodnoi konferentsii po perevodovediniiu "Fedorovslie chteniia", 20-23 oktiabria 2010. Proc. 11th Int. Conference "Fedorov Readings". Saint-Petersburg, 2010. PP. 48-66.
2. Adamska-Salaciak, A. (2010) Examining Equivalence. International Journal of Lexicography 23(4). PP. 387-409.
3. Cabré, T. (1999). Terminology: theory, methods and applications. Amsterdam & Philadelphia: John Benjamins Publishing Company. 247 p.
4. Chiocchetti, E. (2011) Harmonising legal terminology in four languages: the experience of the Alpine Convention. Proceedings of the XVI International Conference of Euskaltzaindia Pirinioetako hizkuntzak: lehena eta oraina (The Languages of the Pyrenees: Past and Present) 6-10 October 2008; Pamplona (E). PP. 533-546.
5. Chiocchetti, E. Voltmer L. (2008) LexALP - Harmonising Alpine Terminology. Chiocchetti, Elena / Voltmer, Leonhard (eds) Harmonizing Legal Terminology. Bolzano:EURAC. PP. 47-59.
6. Edelmann, G. (2012) Lexicography for specific purposes. Equivalence in bilingual and multilingual specialised dictionaries with reference to conceptual systems. WU Online Papers in International Business Communication, 2012/1. WU Vienna University of Economics and Business, Vienna. 15 p. Available at: http://epub.wu.ac.at/3508/ (accessed 1 October 2013).
7. Gómez, P., Allard, M. (2012) Managing Terminology for Translation Using Translation Environment Tools: Towards a Definition of Best Practices. University of Ottawa, Canada. [Theses]
8. Hammond, M. (1995) A New Wind of Quality from Europe: Implications of the Court Case Cited by Holz-Mânttàri for the U.S. Translation Industry. Translation and the Law Amsterdam/ Philadelphia, John Benjamins. PP. 233-245.
9. ISO 1087-1:2000. (2000) Terminology work - Vocabulary - Part 1: Theory and application.
10. Morandini, M. (2013) New perspectives of Alpine research. New perspectives of Alpine research. Dislivelli (Torino). PP. 6-12.
11. Pavel, S., Nolet, D. (2011) Handbook ofTerminology. Minister ofPublic Works and Government Services Canada. 154 p.
12. Pym, A. (2011) Translation research terms: a tentative glossary for moments of perplexity and disput. Translation research project 3. Intercultural Studies Group. Tarragona. PP. 75-110.
13. Rogers, M. (2008) Terminological equivalence: probability and consistency in technical translation. LSP Translation Scenarios. Selected Contributions to the EU Marie Curie Conference Vienna 2007. Mutra Journal 02, 2 . ATRC Group, Germany. PP. 101-108.
14. Shelov, S.D. (1990) Terms, Termability and Knowledge. Terminology and knowledge engineering. Vol. 1. PP. 108-111.
15. Somssich, R. Fazekas, F. (2012) Language-related aspects of international treaties concluded by the EU In: Studies on translation and multilingualism. Final reportfor the "Study on Language and Translation in International Law and EU law". European Commission. P. 35-64.
16. Voltmer, L. (2008) Experiences in Harmonising. Harmonizing Legal Terminology. Bolzano:EURAC. PP. 71-93.
17. Wiseman, N. (2002) Against Anti-terminology. Issued on the Chinese Herbal Academy Listserv. 12. Available at: http://www.paradigm-pubs.com/sites/www.paradigm-pubs.com/files/files/ Anti-Term.pdf (accessed 1 October 2013).
Как переводят терминологию: предписания и практика
А.В. Ачкасов
Санкт-Петербургский государственный университет Россия, 199034, Санкт-Петербург, Университетская набережная, 7/9
Предметом анализа статьи является вариативность способов терминологической номинации при переводе. Ономасиологический подход к терминологической эквивалентности предполагает необходимость координации понятий и лексических субстратов и, как следствие, однозначность межъязыковых терминологических соответствий. В практике перевода ЯСЦ этот принцип почти никогда не реализуется. Несоответствие практики перевода нормативным предписаниям традиционно рассматривается как результат некорректной и произвольной интерпретации терминологии переводчиками. В действительности вариативность терминологии в переводе ЯСЦ обусловлена системными, семантическими и формальными характеристиками терминов. Понятие «перевод терминологии» в таком контексте приобретает паллиативный характер.
Ключевые слова: терминология, эквивалентность, перевод ЯСЦ, перевод терминов, вариативность терминов, правила перевода, практика перевода.
