Translators' English-Spanish metaphorical competence: impact on the target system

ELIA 3, 2002, pp. 203-218 TRANSLATORS’ ENGLISH-SPANISH METAPHORICAL COMPETENCE: IMPACT ON THE TARGET SYSTEM Eva Samaniego Fernández Universidad Nacional de Educación a Distancia (UNED) In spite of the large number of interlinguistic competences and skills generally mentioned in translation studies as being essential for translators, metaphorical competence does not figure as one of the relevant ones. And yet, it has a deep impact on the conceptual development of the target system, as we will try to demonstrate on the basis of a study and subsequent report carried out on a reduced corpus of target novel metaphors in newspaper texts published in Spanish and their corresponding source metaphors in English. Many of the translations of these metaphorical expressions are coherent with the principle of ‘transparency’; thus, their meanings in Spanish can be guessed at because they appeal to our recognition of underlying symbolism, even if the translations do not exist as common expressions in Spanish. Key words: novel metaphor, translation, metaphorical competence, cognitive linguistics, newspaper texts 1. Metaphor and translation 1.1. Metaphor Following the recent findings of cognitive linguistics, we define metaphor as an essential conceptual tool which consists in a structural mapping from a Estudios de lingüística inglesa aplicada 204 E. Samaniego Fernández ELIA 3, 2002, pp. 203-218 source conceptual domain on to a target conceptual domain. A conceptual metaphor consists of a target, a source, and a mapping (‘ground’) between them. The target conceptual domain (‘tenor’ in traditional approaches) is the domain to be understood metaphorically, whereas the source conceptual domain (‘vehicle’) is the domain in terms of which the target is to be understood metaphorically1 (Turner 1990: 465). Basic to this notion is the importance of the distinction between ‘metaphor’ and ‘linguistic metaphor’. Although this is now commonplace in linguistics, the term ‘metaphor’ is generally used as an equivalent of ‘conceptual metaphor’ (the major basis of our conceptual system), whereas the expressions ‘linguistic metaphor’ or ‘metaphorical expression’ mean the linguistic items chosen to realise a particular conceptual metaphor. Consequently, it is our conceptual system that is metaphorical in nature, not language per se, and language becomes one of the ways of expression of our cognitive models. Thus, we experience things in terms of metaphors and speak about things using metaphors because our concepts are metaphorically structured (Lakoff & Johnson 1980: 6; Hiraga 1994: 17). We believe that the approach adopted by cognitive linguistics and the proposals put forward by the semiotic theory of iconicity can be integrated in a single model (see Radwanska-Williams 1994: 23). In accordance with this integrative process, we follow Hiraga (1994), Radwanska-Williams (1994), Waugh (1994) and others, who treat metaphors as one subcategory of icons (the other two being images2 and diagrams3). These authors argue that linguistic motivation is iconic in nature; thus, just 1 For example, in the metaphor “his lips such rosebuds”, “lips” would be the tenor or target conceptual domain, whereas “rosebuds” would be the vehicle or source conceptual domain (in cognitive linguistics, the specific conceptual metaphor would be BEAUTY IS A FLOWER). The metaphor is from Kureishi (1990: 9). 2 “An image (…) replicates or reproduces a quality of the represented object” (Radwanska-Williams 1994: 28). 3 “A diagram represents relational qualities” (Radwanska-Williams 1994: 28). Translators’ English-Spanish … 205 ELIA 3, 2002, pp. 203-218 like an icon4 bears a structurally analogical relationship with its object, the cognitive structure of the source domain corresponds to that of the target domain in a metaphor. This means that there exists an inner relationship of similarity between the meanings of metaphors and the icons from which they derive insofar as the meanings of these expressions can be guessed at on the basis of that similarity. 1.2. Metaphor in translation studies Unfortunately, translation studies have inherited the terminological confusion prevalent in disciplines devoted to metaphor research. In general terms, “(...) most translation theorists evade the question of the definition of metaphor” (Pisarska 1989: 28). Some authors list the different forms of metaphor but do not define it (see Vázquez Ayora 1977; Snell-Hornby1988; Newmark 1988a), whereas others avoid the task by arguing that “to define metaphor does not belong to the proper task of translation theory” (Van den Broeck 1981: 74). A third group study the translation problems posed by metaphors without giving a hint of what they understand by metaphors (see Mason 1982; Van Besien & Pelsmaeckers 1988). Finally, some others get away with broad definitions, escaping commitment. There are exceptions, however; Rabadán Álvarez (1991), Menacere (1992) or Pisarska (1989) advance their own definitions. There are basically two major fields of metaphor research in translation studies: metaphor comprehension for non-native speakers and the nature of the interlinguistic inequivalence systematically posed by metaphors (and the solutions to it). It is noteworthy that, in spite of its relevance, the first aspect has not been studied in depth, ignoring the fact that second language speakers consistently find metaphors very difficult to interpret or outright opaque. This apparent disinterest in figurative language 4 “A sign which represents an object mainly by its similarity to that object” (Hiraga 1994: 6). 206 E. Samaniego Fernández ELIA 3, 2002, pp. 203-218 interpretation as applied to translation is probably a direct consequence of the traditional approach to metaphors, where they were seen as ornamental devices; seen in this light, metaphors would be a ‘nice addition’ to translators’ competences, but not an essential one. As for the second aspect, the nature of the interlinguistic inequivalence posed by metaphors, translation studies in general have gathered theories from linguistics or philosophy and applied them to translation in a very asystematic way; as a result, traditional (literalist) and cognitive (figuralist) approaches coexist at the moment. 1.3. The translation of metaphor Many efforts have been devoted in translation studies to the analysis of the variables that might have a bearing on the translation of metaphor. Among those most often quoted we find cultural references, communicative purpose, functional relevance, information burden, metaphor typology, cotext and context restrictions, degree of compatibility of the conceptual and formal structures of the two languages involved, prevalent synchronic norms, degree of lexicalization of the metaphor, translator’s competence, connotations, degree of anisomorphism between the source and target domains in both cultures, comprehensibility of the metaphor, cognitive role, etc. However, there are many other variables which certainly have a say in the translation process but whose nature is much more elusive, such as the reference material used by translators, the time pressure, the alterations introduced in post-translation revisions, the idiolect, the translator’s mood (Newmark 1993), client-imposed terms, etc. There are basically four positions in translation studies as regards metaphor translation: i) metaphors are untranslatable (see Nida 1964 or Dagut 1976 and 1987); ii) metaphors are fully translatable (Kloepfer 1981 and Mason 1982); iii) metaphors are translatable but pose a considerable degree of inequivalence (Van Den Broeck 1981, Rabadán Álvarez 1991, Toury 1985 and 1995, Newmark 1988a and 1988b) and iv) conciliatory Translators’ English-Spanish … 207 ELIA 3, 2002, pp. 203-218 approach, represented by Snell-Hornby (1988); for her, the range of renderings depends on the text type and ad hoc factors. There are a few researchers who have posited generalizations about what Dagut called ‘gradient of translatability’ (1987); most of these proposals use the degree of lexicalization of the metaphor as classifying feature. For Dagut (1976: 32 and 1987: 81-82), the translatability of any given source language metaphor depends on both the particular cultural experiences and the semantic associations exploited. Van Besien & Pelsmaeckers use the term ‘speculative’ to describe these broad principles of translatability. Snell-Hornby (1988: 59) inclines to the view that translation scholars should research on metaphor singularity, and additionally deal with the specific contextual features involved in the process of transfer. Newmark (1988a: 48-49) seems to be convinced that dead metaphors5 are the most ‘translatable’ ones, whereas stock6 and original7 metaphors would show a 5 Dead metaphors (also called ‘lexicalized metaphors’) are “those that are taken up and used by an ever-increasing number of other speakers, so that they gradually lose their uniqueness and peculiarity, becoming part of the established semantic stock of the language and being recorded as such in the dictionary” (Dagut 1976: 23). For Snell-Hornby (1988: 57), “the ‘dead’ metaphor is generally accepted as being one no longer recognizable as such”. According to Rabadán Álvarez (1991: 142), lexicalized metaphors are “aquellas que el hablante ha dejado de percibir como tales y que han pasado a formar parte del sistema lingüístico y cultural”. The following are examples of dead metaphors: “I was feeling at home “(Kureishi 1990: 146); “if they are a pain” (Kureishi 1990: 148); “the door to the future had opened” (Kureishi 1990: 15); “everyone was practically in heaven” (Kureishi 1990: 17). 6 A stock metaphor is “an established metaphor which in an informal context is an efficient and concise method of covering a physical and/or mental situation both referentially and pragmatically” (Newmark 1988b: 109). Rabadán Álvarez calls them “metáforas tradicionales” (1991); Newmark initially termed them “standard metaphors”, while Van Den Broeck (1981) calls them “conventional”. These are mainly metaphors coming from literature and assimilated through usage. Examples: “he is a good shepherd, who is rid now of the black sheep of his flock” (Swift 1992: 208 E. Samaniego Fernández ELIA 3, 2002, pp. 203-218 degree of translatability proportional to the proximity of the two polysystems involved (1988a: 109, 1988b: 49 and 106-113). For his part, Van Den Broeck (1981: 73) gainsays Dagut’s opinion: it is possible to make generalizations on metaphor translation, since otherwise the applicability of translation theory would be invalidated. For the author, “translatability keeps an inverse proportion with the quantity of information manifested by the metaphor and the degree to which this information is structured in a text” (1981: 84). Rabadán Álvarez (1991: 137) points out that novel metaphors are very difficult to translate, whereas stock metaphors are translatable if the systems involved are culturally close; lexicalized metaphors are the most translatable ones. There are a few exceptions: functionally relevant (foregrounded) metaphors, culturally-bound items not shared between both cultures, ‘burden’ of information supported by metaphors and translational norms. The more marked these features are, the more difficult it will be to render a metaphor (1991: 146). Eventually, we have the following principles of translatability: novel metaphors are not the most difficult to translate according to Dagut; the most translatable ones would be dead metaphors according to Newmark; Van Den Broeck and Rabadán Álvarez firmly believe that lexicalized metaphors are 53); “the night mist swirls round the battlements of Elsinore” (Swift 1992: 194); “forget your Klondikes” (Swift 1992: 129). 7 For MacCormack (1985: 136), ‘new’ metaphors (equivalent to what we have called ‘original’ metaphors) occur “(...) when an individual juxtaposes conceptual referents never before combined, producing both a semantic anomaly and a new conceptual insight”. Rabadán Álvarez defines ‘novel metaphors’ as “aquellas metáforas que presentan el grado máximo de violación de las reglas lingüísticas y literarias del polisistema sincrónico” (1991: 136). Examples: “he started to suffer the malnutrition of unalloyed seriousness” (Kureishi 1990: 208); “one of those oversized sardine cans (Lodge 1991: 5; it refers to a plane); “nature’s ceiling fan” (Lodge 1991: 93; it refers to the wind); “strapped into a pair of cramped dentist's chair” (Lodge 1991: 6; it r efers to the plane seats). Translators’ English-Spanish … 209 ELIA 3, 2002, pp. 203-218 the most translatable ones followed by novel and stock metaphors, whereas decorative metaphors would be more readily translatable. 1.4. Novel metaphor We have intentionally chosen the designation ‘novel metaphor’ (see footnote no. 7) as against other terminological alternatives due to its transparency. There are a number of options, however: ‘original metaphor’ (Newmark 1988b, Pisarska 1989, etc.), ‘similarity-creating metaphor’ (Indurkhya 1992), ‘poetic metaphor’ (Sweetser 1990), ‘private metaphor’ (Van Den Broeck 1981: 75), ‘creative metaphor’ (Sperber & Wilson 1986), ‘innovative metaphor’ (Pisarska 1989), ‘live metaphor’ (Larson 1989: 331-332) or, more simply, ‘new metaphor’ (MacCormack 1985: 136). There were several reasons for choosing novel metaphors in our study: first, while stock and lexicalized metaphors have been studied profusely, novel metaphors still lack an adequate cognitive and translational framework (Indurkhya 1992: 3). Second, one of our aims was to check whether novel metaphors were actually so frequent in newspaper texts as some authors believe them to be (see Newmark 1988b, Snell-Hornby 1988, Steen 1994: 73, etc.). Third, we were not interested in a study on frequency of occurrence, but rather in the kind of equivalence sought by translators. In our opinion, novel linguistic metaphors present the following features cumulatively: i) they integrate a figurative component; ii) they may manifest themselves in different linguistic forms and do not have length restrictions; iii) they are not ‘institutionalized’, that is, standardised, and have therefore not been incorporated into common usage, having no entries in dictionaries or reference works; iv) the receiver perceives them as metaphorical language and finds ‘novelty’ or ‘communicative surprise’ in them (they flout addressee expectations); v) they may be divided into three major groups according to the variable of ‘use’ (see Steen 1994 and Indurkhya 1992): ‘pure’ novel metaphors creating the similarity; novel metaphors based on a pre-existent similarity but showing a new mapping or being a new combination of cognitive domains and novel metaphors that are 210 E. Samaniego Fernández ELIA 3, 2002, pp. 203-218 a simultaneous exploitation of the literal and the metaphorical meanings (see Le Guern 1976: 123). 2. The sample 2.1. Compilation process and sample characteristics We have selected a total of 122 newspaper texts, which constitute the whole of texts originally published by The Guardian (U.K. edition) and subsequently translated and published in El Mundo from the 1st of January to the 31st of December 1995. This means that the sample includes only those texts whose original and translation were published within the selected period. These texts are part of a much more detailed study on the translation of metaphors developed in a doctoral thesis (Samaniego Fernández 2000). The total number of words in the binomials studied is well above 250,000 (more specifically 285,859). We are aware that we are not working with a large corpus (which is why we call it a ‘sample’ rather than a corpus), but following Pearson (1998), we believe that valuable results can also be obtained from reduced studies. The sample, then , includes 389 novel metaphors in Spanish, which correspond to 370 metaphors (whichever their nature, that is, not necessarily novel metaphors) in the source texts. The difference in number is due to the fact that 19 target text metaphors have as source text equivalent <Ø> and were consequently created in the TTs from ‘zero’ linguistic material in the STs. Two major criteria have been taken into account in the selection of the metaphors studied: use (non-recognition as known material by the speakers) and lexicography (non-appearance in dictionaries). It must be noted that the correspondences for these target text metaphors can have a varied nature in the source texts; we have come across four major occurrences: 1) the corresponding linguistic material in the source text is non-metaphorical; 2) the correspondence in the source text is a dead metaphor; 3) the linguistic material in the source text is a metaphor Translators’ English-Spanish … 211 ELIA 3, 2002, pp. 203-218 immersed in a process of lexicalization and 4) the linguistic material in the source text is zero (19 cases). In fact, only 364 Spanish metaphors out of the 389 correspond with metaphorical material in the source texts. 2.2. Sample study and major results obtained Interesting as a more detailed study would be, we have chosen one parameter of analysis only due to space restrictions: the use-related classification of the metaphors, which gives us insight into the cognitive and linguistic impact caused by translational procedures. As seen above (1.4.), we have classified the novel metaphors of the target texts into ‘pure’ novel metaphors, novel metaphors with an original phraseological occurrence, novel metaphors with an original meaning and novel metaphors with exploitation of the literal and the figurative. Notice that the possibilities are much wider in the source texts, where not all the metaphors are necessarily novel. The options in the STs, besides the four mentioned above, are: ∅ (absence of linguistic material), semi-novel metaphors (familiar but still surprising), dead metaphors and nonmetaphorical material. These are the results obtained: Figure 1: use-related metaphors (STs) 24% 2% 7% 27% 27% 7% 6% dead non-met exploit occurrence pure meaning semi-novel 218 E. Samaniego Fernández ELIA 3, 2002, pp. 203-218 Sperber, D. and Wilson, D. 1986. Relevance. Communication and Cognition. Oxford: Basil Blackwell. Steen, G. 1994. Understanding Metaphor in Literature. London: Longman. Sweetser, E.E. 1990. From Etymology to Pragmatics. Metaphorical and Cultural Aspects of Semantic Structure. Cambridge: Cambridge University Press. Cambridge Studies in Linguistics 54. Swift, G. 1992. Ever After. London: Picador. Toury, G. 1981. “Translated literature: system, norm, performance. Toward a tt-oriented approach to literary translation”. Poetics Today 2(4): 927. Toury, G. 1985. “A rationale for descriptive translation studies” in T. Hermans (ed.) The Manipulation of Literature. Sydney: Croom & Helm. 16-41. Toury, G. 1995. Descriptive Translation Studies and Beyond. Amsterdam/Philadelphia: John Benjamins. Turner, M. 1990. “Aspects of the invariance hypothesis”. Cognitive Linguistics 1(2): 247-255. Vázquez Ayora, G. 1977. Introducción a la traductología. Washington D.C.: Georgetown University Press. Waugh, L. 1994. “Degrees of iconicity in the lexicon”. Journal of Pragmatics 22: 55-70.