The Translatability of Metaphor in LSP : Application of a Decision-Making Model

The pragmatic approach to translation implies the consideration of translation as a useful test case for understanding the role of language in social life. Under this view this article analyses the decision-making stage translators go through in the course of formulating a TT. Henee this article contributes both to enhance the status of translation theory and to explain some of the decisions taken by the Spanish translators of three English Manuals of Economics. In short, we have argued that the use of a 'maximax' strategy for translating English metaphors as Spanish similarity-creating metaphors can be attributed to subjective factors, especially to the translators' cognitive system, their knowledge bases, the task specification, and the text type specific problem space. As a result, we have also claimed that proposals for translating microtextual problems —for example, metaphors —can benefit from the study of the above-mentioned subjective factors since they allovv or inhibit the translators' choices in the decision-making stage of the translation process.


Introduction
Although the practice of translation is "as oíd as the tower of Babel" (Gentzler, 1993: 1), translation theory seems to have lacked full academic status, perhaps because it tended to show terrninological confusión and to rest mostly on untested assumptions.Holmes et al. (197 8:68) summarise this approach by clairning that since the discipline was born there has been a lack of "consensus regarding the types of models to be tested, the kinds of methods to be applied, the varieties of terminology to be used.More than that, there is not even likemindedness about the contours of the field, the problem set, the discipline as such (...)"• The oíd debate concerning the translator's priority for the source language (SL) or for the target language (TL) illustrates the confusión that has permeated the whole discipline, usually manifesting in terms of dichotomies, such as 'right ' versus 'wrong', 'form' versus 'content', 'art' versus 'science', 'theory' versus 'practice', 'translation' versus 'interpretation', etc. Table 1 offers a representative sample of some of these dichotomies and the scholar(s) introducing them.
Table 1-dichotomies concerning the translator's priorities for the SLr TL.SL on the translation process and will illustrate this claim by providing an integrated approach to a microtextual translation problem, namely, the translation of metaphors in LSP.
Our analysis distinguishes between the text, as the embodiment of meaning, and the translator, as the human mind who makes that embodiment possible.This distinction is mirrored in the dichotomy between the concept of 'equivalence' -understood here as "a matter of entities between texts" (Hatim and Masón, 1990: 35)-and the role of the translator in a process of cross-cultural commimication.To carry out this analysis, different assumptions have been taken and used as theoretical foundations of our research: 1. metaphor is a translation problem and in translating metaphor there is " little point in seeking to match target language words with those in the ST in isolation from a consideration of the of the writer's world view" (Hatim and Masón, 1990: 4); 2. metaphor is a cognitive tool (Lakoff and Johnson, 1980), which plays an important role in term formation and concept analysis (Thagard, 1992;Meyer et al., 1997; Fuertes-Olivera and Pizarro-Sánchez, forthcoming) 3. translating is an act of communication which takes place within a social framework.The translator as communicator seeks to maintain coherence by striking the appropriate balance between what is effective and what is efficient in a particular environment, for a particular purpose and for particular receivers (Hatim and Masón, 1997).4. the translator plays a central role in the translation process, especially under the tenetsof decisión theory (Levy, 1967(Levy, ,1988;;Jumpeltz, 1988;Wilss, 1981Wilss, ,1994)).Since 'decisión theory' plays a pivotal role in this article, we devote the next section to it.

Decisión theory: Translation as "decision-making"
Decisión theory deals with "decision-making", defined as" (...) part of the process which the translator goes through in the course of formulating a TT" (Shuttleworth & Cowie, 1997: 37).Although it was first introduced in the translation literature in the 1960s and 1970s, "the decisión making aspect of translation has been relegated to the fringe of research, and rarely has decisión theory been called upon to support translation theorists and translation practitioners with findings useful for systematically improving translator performance" (Wilss, 1994: 140).
The earliest decision-making research concentrated on the development of formal decision-making models which proved to be of little use in their application to the translator's performance.Those models were documented in operation research and aimed at "discovering the optimal strategy of anundertaking by gradually filtering outalternative, less efficient strategies (...)" (Wilss, 1994:137).One of the most successful of these models was developed by Levy 1 , a Czech scholar interested in applying a model called "The Game Theory" to Uterary texts.He drew a parallelism between the translation process and the actívity of game-playing originally suggestedby Wittgenstein(1953:1:23), and later refínedby Luce and Raiffa( 1957).Inhis view, translation "is a DECISION-PROCESS: a series of a certain number of consecutive situations -moves, as in a game -situations imposing on the translator the necessity of choosing among a certain (and very often exactly defínable number of) alternatives (...)" (Levy 1988: 38).In other words, a decisión means a choice to be taken between a number of possible solutions to a given problem encountered while translating a text.That choice wül influence subsequent choices by opening up or closing off other options dependent on the first selection made.Those choices, whether necessary or unnecessary, motivated or unmotivated are hierarchical rather than merely sequential.These decision-making choices seemed to be influenced by two types of factors: (i) objecüve ones, which are of a linguistic nature and (ii) subjective ones, which are of an extra-linguistic nature.Furthermore, Levy also argued that translators evalúate their work to accommodate the target audience's preferences and expectations, basically resorting to the so-called "minimax strategy" according to which "the translator resolves for that one of the possible solutions which promise a máximum of effect with a minimum of effort" (Levy 1988: 48).This strategy will, in turn, influence decisions taken during the translation process, so that "linguistic elements perceived as being non-native to the TL system may be avoided" (Shuttleworth andCowie, 1997: 106).
Apart from the major contribution of Levy, approaches such as 'the early translation studies' and the 'polysystem theory' have offered small contribution to "decision-theory".Only Popovic (1970) and Holmes (1988) seemed to have been interested on Levy's work, since their concepts of shifts and mapping are cióse to Levy's philosophy.Popovic (1970: 79) defined shifts as "ail that appears where it might have been expected" and stated that shifts represent "the relationsbip between the wording of the original work and that of the translation" (1970: 85).He acknowledged the existence of constraints, both linguistically and culturally.For example, the existence of norms, which will influence the decisions made during the translation process.The precise nature and distribution of these norms has been further investigated by Toury (1980;1995) and Leuven-Zwart (1989, 1991).Toury, for example, has distinguished two varieties of shifts: the obligatory (e.g.linguistically motivated) and the non-obligatory (e.g.motivated by literary or cultural considerations).Similarly, Holmes (1988) proposed the concept of mapping to account for the fact that not one, but two maps are used during the translation process; the first of these reflects the features which the translator abstracts from ST, while the second is created on the basis of the choices (linguistic, stylistic, rhythmic, and so forth) which he or she makes from the available options and reflects the shape that TT will ultimately take.Furthermore, the second map is influenced by a bierarchy of correspondences which inevitably emerge as certain choices are ruled out simply because other, more important ones have already been made.
Decision-theory is also the basis of some of the different approaches developed in the 1980s and 1990s in the field of Translation Studies under the ñame of "Think-Aloud Protocols" (TAPs).They tended to investígate the cognitive processes and psychological aspects of the act of translating.Translators were observed while they were at work, their decisions were recorded or videotaped and their TTs were scrutinised for insights which might reveal what went on inside the "black box" of the translator's mind (cf.Krings 1986aKrings , 1986bKrings , 1987;;Hónig 1990;Lorscher 1991Lorscher , 1992;;Fraser 1996, etc.).Although this approach has benefited from the recent interest on cognitivism, it has deíivered very little, and has been subject to severe criticism on different grounds: (1) it is a tentative model, useM only for forming rather than testing hypotheses (cf.Lorscher 1991:75); (2) it offers subjects' verbalisations which are incomplete or produces commentaries on processes which are to a largeextentunconscious (Krings 1987:163); (3)itconfusesthespokenandwritten modes of translation, each of which may entail different thought processes (Toury 1995: 237); and (4) it does not address the question of whether the very act of thinking aloud influences what goes on in the translator's head (Lorscher 1991: 71).
Finally, "decision-theory" is being used by proponents of the "corpus-based approach" to translation in their quest for studying translation and translating by means of identifying the distinctive features of the language of translation (cf.Laviosa 1998, Baker 1998, etc.).Munday (1998), for example, uses a variety of tools currently used in corpus linguistics in the analysis of shifts undertaken by the EngUsh translator of García Márquez 's Diecisiete ingleses envenenados.
Our article follows this line of research by analysing the decision-making stage the translator goes through in the course of formulating a TT.Our data were taken from different manuals of Economics, where different English metaphors were rendered novel metaphors into Spanish.Following Levy's formal generative model of translation we will concéntrate on the translational decisions made by analysing the subjective factors conditioning the translator' s decision-making strategies, and Wilss' pscycholinguistic factors (the translator's cognitive system, his or her knowledge bases, his or her task specification and problems specific to the particular text type).There are grounds to suspect that some or all of these subjective factors determine concrete decision-making strategies in the translation of metaphors in LSP.And that those strategies are unexpected and responsible for subsequent unnecessary decisions taken in the translation process: the resort to literal translations producing similarity-creating metaphors instead of existing conventional ones.

Data and methodology
In a previous article, Fuertes-Olivera and Pizarro-Sánchez (forthcoming) identified five different metaphorical scenarios typically used to conceptualíse the idea of 'infiation': ROBBERY, and INFLATION is A DISEASE.They also found that the Spanish translators of Economics manuals did not always follow the 'minimax strategy' proposed by Levy in their translations of some of the linguistic metaphors associated to the above scenarios.Instead of adopting necessary and motivated decisions, they found out that these translators have resorted to a kind of 'maximax strategy' (máximum effort for máximum effect), namely, they resorted to unnecessary' similarity-creatiag metaphors' (Indurkhya, 1992).This arricie follows this line of research by conceatrating on explaining some of the translation strategies taken by the Spanish translators of the three English Manuals of Economics analysed by Fuertes-Olivera and Pizarro-Sánchez: Samuelson and Nordhaus 's Economics (12* ed., 1985); Blanchard's Macroeconomics (1997); Mankiw's Principies of Economics (1998).
In other words, we propose to offer a model for translating metaphors from the point of view of translation as a process, since we have detected that most of the models already proposed ignore the subjective factors which determine the translator's decision-making process.In our view, the role of the translator in the translation process should be highlighted as a motivation factor introducing novel and innovative metaphorical expressions, thus creating the similarity between the target and the source domains.
Fuertes-Olivera and Pizarro-Sánchez classified their corpus into two sets: the English text and the Spanish translation.Then, they used the Wordsmith Tools to do a concordance of 'inflation', obtaining 814 and 740 occurrences respectively.Next, they aligned the texts and their translations and presented them to three English native speakers to whom it was explained what each metaphorical scenario indicated and who were asked if they considered the metaphorical expressions given to be linguistic manifestation of the metaphorical scenario.If the answer was positive, then they were asked to rate them according to their degree of metaphoric contení.In so doing, they followed Indurkhya (1992), who referred to it as the metaphoric-content continuum, and identified its two ends as novel-metaphorical and conventional respectively.Their next step was to present the Spanish translations of the metaphorical expressions previously identified by the English informants to three native Spanish speakers.To avoid possible interference, the three Spanish informants are professors of Economics with a poor knowledge of English and the English informants are professors of English with a poor knowledge of Economics.The Spanish informants had to decide if the metaphorical scenarios were found in the paragraphs given and were asked if the expressions in bold were metaphorical expressions corresponding to the metaphorical scenarios previously explained.When the answer was positive they were asked to rate these expressions as 'novel' or as 'conventional' in line with Indurkhya' s metaphoric content.The Spanish informants found examples [1], [2] and [3] below, previously identified as novel metaphorical expressions by the English informants, to nave been translated as novel metaphorical Spanish expressions creating a similarity: <! -Ll, S 70-> ALightingRod?Having reviewed the impacts of inflation, itmay now be clear that modérate mflations impose but modest costs.The inability of analysts to find major costs has led some to think that the aversión to inflation is a social phenomenon.Fearof inflation mav be a lightning rod that attracts popular and política! he awav from < --12, S 70-> ¿Un pararrayos?Una vez examinados los efectos de la inflación, tal vez sea ya evidente que las inflaciones modereadas solo imponen modesto costos.La incapacidad de los analistas para Encontrar grandes costos ha llevado a algunos a pensar que la aversión a la inflación es un fenómeno social.El temor a la inflación podría ser un pararrayos que desviara la ira popular y política de cuestiones más profundas como los conflictos de clase o las discrepancias sobre la distribución justa de la renta; y, quizá, a la larga, sea mejor insultar al IPC que al jefe o vecino.

Results and discussion
Our analysis has followed a linear methodology.First, we applied the four psycholonguistic factors-Le., the translator's cognitive factors, their knowledge bases (linguistic, referential, sociocultural and situational), the task specification, the problems specific to the particular text type -to the translation we are commenting.Second, we applied Levy' s model to the actual decision-making choices of these translations.Third, we explained the decisions taken in terms of these subjective factors.Finally, we concentrated on the importance of this pilot study for understanding the translatability of metaphor.

Analysis of subjective factors affecting the decision-making process SHE TMNSMKWS OK»aSIVl StfSlfM
One of the translator, Manuel Tonaría, is an Economies Professor in a Spanish University.On the other hand, Esther Rabasco, another translator, is a professional translator.Botli constitute a very proféssional and active team, who has translated most of the Manuals of Economies used by Spanish students.Boíh have a high scientific knowledge of 'infiation'.Both are aware of the authors' cognitive knowledge: Samuelson and Nordhaus think that infiation is caused by excess demand in the economy ('demand-pull infiation'), whereas Blanchard and Mankiw claim that infiation is the result of excessive money supply ('monetarism') together with two side factors: excess demand as Keynessians claimed and higher costs.______

2, THE TXMmJO®m& 83Í0WM&GE M$m
1. LÍNGUISTIC KNOWLEDGE BASIS Poor knowledge of the English language.Some of the renderings are simple literal translations.

REFERENTIAL KNOWLEDGE BASIS
High declarative knowledge of inflación as we have seen in the analysis of the translators' cognitive system.High cognitive knowledge of the issue..3-SQCIOCULTURALKNOWLEDGE BASIS High knowledge of the differences between the Spanish conceptualisation of the economic background and the English one.Ability to introduce new conceptualisations of technical terms by means of metaphorisation in the target language.
High degree of predictability of the adequacy and functioning of those new terms in the target language.
Good appreciation of sociocultural differences.2.4.SITUATIONAL KNOWLEDGE BASIS -High knowledge of the spatio-temporal setting of the ST.

TASK SPÉCIFÍCATIÓN
Not many hints at task specification.We will assume an easy instruction which can be expressed as: In this model all decisions are represented in a binary form, although the range of theoretical possibilities can be "1-n members".In our application of the model we will use a representation of 1-3 members : Novel metaphorical English expression -conventíonal metaphor, similarity-based metaphor or similarity-creating metaphor.
The situation ("an abstractíon of reality, which, in a formalised theory, would be expressed by means of a model" -Levy 1988: 38)) is "the search for an equivalent to a novel metaphorical English expression".As to Instruction I ("defíning the class of possible alternatives" -Levy 1988: 39) we have "it is necessary to fmd a Spanish metaphorical expressions for the English novel one".The paradigm ("the class of possible solutions" -Levy 1988: 39) we have used is made of three members, as above mentioned.
Instruction II ("directing the CHOICE among the alternatives"-Levy 1988: 39) is derived from the four psycholinguistic factors that determine the translation process.Although Levy' s model accounts for both definitional instructions (they give forms to the paradigm) and selective instructions (they govern the translator' s choice), we have concentrated on the four psycholinguistic factors above mentioned, in line with Wilss who prefers not to analyse definitional instructions which are objective determinants of a linguistic nature (aesthetic rules, syntactic forms, etc).The three alternatives are not equivalent.Henee, the choice is not random but context-bound, as we will show.Here this choice is limited to three possibilities.
Instruction III is determined by each of the subparts of the 4 subjective factors analysed in paradigm II.Our paradigm ni will show the different possibilities offered by the translator according to those subparts.
Once our translators have decided in fevour of one of the alternatives, they have selected their own choices in a number of consequent moves: grammatical forms (technical decisión), the interpretation of metaphorical scenarios (cognition), the formation and/or standardisation of new terms by means of a metaphorisation process (terminology).In other words, they have created the context for a certain number of subsequent decisions.It can, therefore, be said that the process of translation has the form of a game with complete information in which every succeeding move is influenced by the knowledge of their previous decisions and by the situation which resulted from them 2 .
The outeome of two different 'games' (e.g. the decisions resulting from the three alternative interpretations of the novel English metaphorical expressions) are three different translation variants, the distance among them may be measured by the number of differing decisions incorporated in the text, and what it seems more knportant to us, by the specific determinants that prompted those decisions.Let us turn now to the analysis of the three examples above mentioned (see pages 73-75):

Subjective factors influencing thedecision-making process
Our analysis shows that the translator's cognitive system influences the decision-making of the three examples considered here, as the perceptual processings in text comprehension of the two Spanish translators are clearly visible in the three cases.Thus, the Spanish informants used by Fuertes-Olivera and Pizarro-Sánchez judged adequate the new metaphorical scenarios introduced by the Spanish translators of the novel English metaphors analysedhere: INFLATIONIS ASTORM, INFLATIONISAWAR, and INFLATION ISA RACE.What they seem to imply is that the Spanish translators nave in mind the cognitive systems of the English authors.In addition, these results highlight the essential significance of the translators' experience in the field of translation (cf.Wallsten, 1980: xiii).In view of this, and as Wilss states, the studies on the translation situation "may make it difficult for the translators to avoid arbitrary (subjective) decision-making and to develop a feeling for methodological considerations as a prerequisite for objective (or near objective) decisionmaking" (Wilss 1994: 139).
As to the translators' knowledge bases we can suppose that their poor linguistic knowledge seems to be responsible for the three choices made.Their referential, sociocultural and situational knowledge bases account for the choices made.This can be explained in view of the fact that the conceptual mapping of the second and third examples are closer between the Spanish and English culture.The translators' high cognitive knowledge of inflation has allowed them to introduce new conceptual mappings in the target language.Their world knowledge has allowed them to introduce similarity-creating metaphors in the target language.
The task specification, especially the translation instruction and the client needs are responsible for the three choices made.The translators nave adopted their decision-making strategies to satisfy the translational needs and wants.As Wilss claims "Apart from standardly operative one-to-one correspondences, a translator has to decide which coping strategy to adopt in order to satisfy translational needs and wants.It is, therefore, important for any translator to learn the degree to which a specific decision-making move will lead to a desirable or undesirable outcome (...)" (1994: 142).In this sense, we can affirm that the use of a strategy, like the maximax one, is clearly didactic-oriented.
The text type specific problem space has preved to be an essentially relevant factor of the translators' choices.One of the particular problems of translating scientific and technical translation is the translation of technical terminology and the need to respect both the referential function of language and the conventions of technical language (cf.Jumpeltz 1988;Mateo, 1993;Arribas Baño et al. 2000).This is crucial to understand the ultímate outcome of term standardisation and term formation in LSP.The translators, on the other hand, seem well aware of the characteristics of economics academic textbooks: high frequency of terminology, médium degree of difficulty and high frequency of paralinguistic communication features in the text (cf.Arntz, 1996).We can conclude that this subjective factor is of crucial transcendence in the translators' decision-making process and ultímate outcome of that process, in our case "term standardisation" and "term formation".Furthermore, we have observed that there are both "cióse problem spaces" which "(...) do not compel the translator to unravel their meaning by activating hermeneutic (divinatory) capabilities, but require referential knowledge in a particular scientific and technological domain" (Wilss, 1994: 136) and "open problem spaces" which are typically found in literature.The latter type of text type problem spaces relies on the translator's intuition.In our case we have found that the three cases analysed introduced simüarity-creating metaphors in the target language instead of similarity-based metaphors which could have been rendered literally and consequently, easüy understood by the readers.This intuitive motivation behind the translators' actual choice can justify the use of a "maximax strategy" ("máximum effort": the use of a similarity-creating metaphor for "máximum effect" didactic aim).
In conclusión, we can claim that subjective factors are of immense relevance in the renderings of English novel metaphorical expressions in Spanish-translated economics manuals.The translators' cognitive system, their knowledge bases, the task specification and the text type specific problem spaces are of equal importance in accounting for the decision-making process involved in this concrete translation situation.The translators' knowledge of the nature of the text type is responsible for a further outcome (or move as in a game) of the translation: term standardisation and text formation.The text type space problems, especially the use of "open space problems" of intuitive nature can account for the use of the " maximax strategy", as illustrated in this article.
A figuralist or fiínctionalist view that offers a heuristic account of the figure by claiming its cognitive function (Snell-Homby 1988;van Besien & Pelsmacker 1988;Pisarska 1989;Rabadán Álvarez 1991;Kurtz 1995).This last view has been essential to understand the role of metaphors in specialised fields.
In addition, a new approach sponsored by Meyer et al. (1997) and by Fuertes-Olivera and Pizarro-Sánchez (forthcoming) seems to promote a hybrid approach by defending both the aesthetic and the cognitive role of metaphor.This mixed approach is analysing metaphor according to three main translation difficulties: cultural specificity, structural constraints and its cognitive role.To thera we would like to add a new one: the translator's role in the decision-making process, especially when dealing with LSP texts in which the referential function must condition most of the decisions taken.This would imply that the models for translating metaphors typically commented (cf.Newmark 1980Newmark /1988;;van den Broeck 1981;Dobrzyñska 1995), seem to be incomplete, since they rely heavily on conditional probabilities ("if we nave X we will transíate Y").With LSP, more factors should be considered.Samaniego Fernández (2000), for example, proposes a model that accounts for cultural references, semantic associations, communicative purpose, functional relevance, linguistic constraints, degree of informativity, interpretation, register, texttypes, metaphor typology and contextual constraints.In other words, her model provides a good study of a large number of objective factors intervening in translation as a communicative event.Her descriptive approach provides an excellent account of translation as product that can be complemented with the analysis of the four psycholinguistic factors that determine the decision-making process we propose here.
Our analysis can, therefore, complement objective proposals for translating metaphor.Similarly, we can use the same paradigm to solve other microcontextual problems which also seem to be rooted in subjective factors: i.e. semantic vagueness, semantic ambiguity, syntactic complexity, rheme/theme distribution, word-plays, ironic text elements, opaque formulations, collocations, morphological idiosyncrasies or innovations, lexical gaps, etc.The translator's role in the decision-making process, then, needs to be studied properly to see the underlying motivation behind decision-making strategies, no matter if these strategies seemto be unconscious (Montero-Martínez etal. 2001).

Conclusions
In Une with Hatim and Mason's suggestion that both the ST and the TT are evidence records of a communicative transaction, this article has examined the decision-making stage the translator goes through in the course of formulating a TT.Henee, our article contributes to emphasise the important role of translators, especially when dealing with microtextual problems such as the translatability of metaphor.To enhance the status of translation theory, both objective and subjective factors must be considered because objective factors cannot totally explain certain translation strategies, since translators also commit themselves to making irrational decisions.For example, the Spanish translators of English Economics manuals analysed in this article have produced novel Spanish metaphors instead of conventional ones.We have argued that this outcome can be attributed to some subjective factors, especially to the translators' cognitive system, the translators' knowledge bases, the task specification, and the text type specific problem space.In addition, we have also maintained that there seems to be a direct link between certain text type problem spaces ("open" ones) and concrete motivated and unnecessary translation strategies, like the "maximax" one.Finally, we have argued that actual proposals for translating microcontextual problems, i.e. metaphors, idioms, collocations, etc. can benefit from the study of the subjective factors that allow or inhibit the translator's choices in the decisionmaking stage of the translation process.Studies of translation as a process, then, can complement studies of translation as a product, thus offeringr an integrated view of translation as both process and product.

Notes
deeper issues such as class conflicto or disagreements about the just distribution of income.And perhaps, in the end, it is better to rail at the CPI than at vour boss or neighbour Example [2] < !-Ll, S 64-> (...) But over time, the quantity of money in the economy starts growing faster and faster.At about the same time, infiation also takes off (...) Example[3] < i-Ll.S 250-> Stop-go dríving of the British and other economies during the 1960s and 1970s proved ineffective in controlling infiation and exacted a cruel price in terms of productivity and living standards.