《Processing figurative language in a multilingual task.doc》由会员分享，可在线阅读，更多相关《Processing figurative language in a multilingual task.doc（21页珍藏版）》请在三一办公上搜索。
1、Processing figurative language in a multi-lingual task: Translation, transfer and metaphorAbstract This paper reports on a study of metaphor comprehension and production in a multi-lingual setting using a translation task. Subjects were asked to translate sentences to and from their native language
2、(Turkish) and their second language (English). The proportion of metaphors transferred between the languages during translation was examined. The results do not support forward transfer between L1 and L2, but rather indicate a translation advantage from L2 to L1. An activation-based framework is pro
3、posed to explain the experiment results and in general, metaphor processing in multi-language environments. The study indicates that figurative language processing by bilinguals or in multilingual tasks could be an interesting area of research. 1. Introduction One of the active areas in modern cogni
4、tive science and linguistics is the study of metaphor. Recent research on metaphor has emphasized the psychological aspects of metaphor as well as social, cultural and linguistic aspects (e.g., Lakoff & Johnson 1980, Murphy 1997, Verbrugge 1984). Although there is certainly room for further explorat
5、ion, cross-cultural and cross-linguistic studies of metaphor and other figurative aspects of language have also received some attention (Buchowski 1996, Souto Silva 2000, Van Brabant 1986). On the other hand, abundant research has been carried out on second language learning, processing, and product
6、ion. Within this literature, a subfield that has received much attention is that of interference between the languages a person speaks. Interference often takes the form of forward transfer, in which strategies from the native language (L1) are used to learn and process the second language (L2). Bac
7、kward transfer is where the reverse pattern occurs. Forward transfer has been demonstrated several times, mainly within tasks and domains that are syntactic and phonetic; with semantic priming the results have been less consistent (see Odlin 1989, Gass & Selinker 1994 for good overviews; see also Jo
8、nes & Tetroe 1987, Ritchie & Bhatia 1996). Translation has also been studied in bilingualism and second language research (de Groot 1992, de Groot et.al. 1994, Gekoski 1968, Kroll & Stewart 1993, Sholl et.al. 1995). Early studies have reported a translation advantage (usually measured temporally and
9、 thereby using short pieces of text as stimuli) from L2 to L1. This is argued to be due to translation being automatic from L2 to L1 (e.g., Gekoski 1968) or translation being conceptually mediated from L1 to L2 but lexically mediated from L2 to L1 (Sholl & Stewart 1994, Sholl et.al. 1995). In this p
10、aper, the term transfer is used in a broad sense that encompasses both forward transfer from L1 to L2 and the effects of translation advantage (or, backward transfer) from L2 to L1. The term also refers to any and all aspects of language, not limited to phonology, syntax and semantics, but also incl
11、uding pragmatic and sociological aspects. I believe using the term to refer to such an array of phenomena is still appropriate, even though part of the literature uses the term for a concept narrower in scope, because as noted by Lado: individuals tend to transfer the forms and meanings, and the dis
12、tribution of forms and meanings in their native language and culture to the foreign language and culture - both productively when attempting to speak the language and to act in the culture, and receptively when attempting to grasp and understand the language and the culture as practiced by natives (
13、Lado 1957, p.2; also cited in Gass & Selinker,1994, p.53). There are few studies focusing on semantic, pragmatic and figurative aspects of second language comprehension and production (Francis 1996, Kirstein 1972, Opoku 1987, Souto Silva 2000). Translation of metaphors across languages has been stud
14、ied within the context of poetry, literature and religion, but a processing account has not been provided. Linguistic transfer of figurative aspects of language is clearly an area open for further exploration. This paper studies metaphor comprehension and production in a multi-lingual setting using
15、a translation task. The goal is to examine whether, under translation, metaphors exhibit forward transfer, or whether they show backward transfer (i.e., whether there is a translation advantage from L2 to L1). 2. Methods Subjects: Ten native speakers of Turkish, aged 21-51, participated in this stud
16、y. These subjects were chosen to represent native speakers of Turkish (henceforth L1) who are competent enough in English (their second language, henceforth L2) for the purposes of the translation task. All used Turkish as their dominant language, and none had professional translation experience or
17、training. Individuals with translation experience were excluded from the study to increase the chance of translated sentences giving us an idea about metaphor processing, and not about the execution of learned skills. All subjects had completed all or part of their higher education in English, and i
18、ndicated that they read books and magazines, and watching movies in this language. Stimuli: The sentences for translation were taken from a text on .Istanbul.s Coffee Houses. originally written in Turkish and translated professionally into English (Sungur 1999a, 1999b). These texts were taken from T
19、urkish Airlines. Skylife magazine. After a detailed analysis of the texts, 10 pairs of sentences were chosen. Two stimuli sets were composed by dividing up the pairs of sentences so that each set contained 5 Turkish sentences and 5 English sentences. Each subject was randomly assigned to one of the
20、stimulus sets. Therefore, each subject translated 5 sentences from English to Turkish and 5 from Turkish to English. All pairs but one contained metaphors in both languages. Below is a list of the 10 sentence pairs, with the metaphorical expresions they contain identified. 1. Turkish: .Gnesli gnleri
21、n devri bitip, yagmurlu havalar saltanatini ilan edince, Istanbulun insanlari da kendilerine siginacak yer aramaya baslar. Metaphor: DURATION IS REIGN CHANGE IS SUCCESSION Literal Gloss: .When the reign of the sunny days is over and rainy weather declares sovereignty, the people of Istanbul start lo
22、oking for shelter. English: .When the sunny days of summer pass and the weather turns cold and rainy, the inhabitants of Istanbul look for somewhere to take refuge. Metaphor: TIME IS SOMETHING THAT MOVES CHANGE OF STATE IS CHANGE OF DIRECTION 2. Turkish: .Masmavi g.kyznn kani kaynatan .zgrlk duygusu
23、, yerini gri havalarin melankolik ruh haline birakmistir. Metaphor: INTENSE EMOTIONS ARE HEAT CHANGE IS REPLACEMENT Literal Gloss: .The freedom feeling of the blue sky that boils one.s blood has left its place to the melancholy mood of gray weather. English: .The sense of freedom which lent vitality
24、 under blue skies is replaced by gloomy lethargy under lowering grey clouds. Metaphor: PROPERTIES ARE POSSESSIONS (which can thus be lent) CHANGE IS REPLACEMENT 3. Turkish: .Sadece soguktan ka.mak i.in degil, vakit ge.irmek, dinlenmek, kimi zaman da bir iki dostun muhabbetine kulak vermek zere mekan
25、 degistirmek gerekir. Metaphor: TIME IS SOMETHING THAT MOVES ACTION (listen/hear) IS THE FACILITATOR/ENABLER (ear) OF ACTION Literal Gloss: .It is necessary to go different places not only to escape the cold, but also to pass the time, relax, or to give ear to the conversations of a few friends. Eng
26、lish: .Now it is time to leave the open air, not just to escape the cold, but to pass the time, relax, and enjoy a spot of conversation. Metaphor: TIME IS SOMETHING THAT MOVES A SMALL AMOUNT OF CONVERSATION IS A SMALL THING IN SPACE (a spot) 4. Turkish: .Kahveler yazin da hizmet verirler, ancak psik
27、olojik ihtiya.lara ayak uydurmak icin, dumani tten demli .aylar ve fokurdayan nargilelerle kislik ortamlarini hazirlarlar mdavimleri i.in. Metaphor: TO ADAPT IS TO MATCH WALKING PACE Literal Gloss: .Coffehouses are open in summer too; but to match feet with the psychological needs, they prepare wint
28、er environments for their customers, with steaming well-brewed tea and humming water pipes. English: .Although open in summer too, it is in winter that these really respond to psychological needs, with steaming glasses of hot strong tea, and the sound of bubbling nargiles. Metaphor: NONE 5. Turkish:
29、 .Kahveler, Istanbulun karmasik dokusunu yansitan aynalar gibidir. lkenin siyasi durumuna iliskin degerlendirmeler, futbol karsilasmalari i.in atesli kritikler, okuldaki, isteki durumlara iliskin son dedikodular kahvelerde konusulur. Metaphor: TO EXEMPLIFY IS TO REFLECT LIKE A MIRROR INTENSE CONVERS
30、ATIONS ARE HEAT Literal Gloss: .Coffee houses are mirrors of Istanbul.s complex texture. It is here that people assess the political situation, carry out heated critiques of soccer games and exchange latest gossip from school or work. English: .Coffee houses are mirrors of Istanbuls complex social t
31、exture, the place for heated discussions of politics, the latest football results, and the latest gossip from school or work. Metaphor: TO EXEMPLIFY IS TO REFLECT LIKE A MIRROR INTENSE CONVERSATIONS ARE HEAT 6. Turkish: .Hikayenin sonrasi ise, giderek n kazanip zirveye oturan bir hanendenin .yksne b
32、enzer. Metaphor: WORKING FOR SUCCESS IS STEADY CLIMBING LONG-TERM PURPOSEFUL ACTIVITY IS A JOURNEY CONTROL IS UP ACTION IS SELF-PROPELLED MOTION CHANGE IS MOTION Literal Gloss: .The rest of the story resembles that of a ruler who incrementally gains fame and finally sits at the summit. English: .The
33、 rest of the story resembles that of a dynasty which steadily climbs the hill of power and fame until it settles on the summit to rule unchallenged. Metaphor: WORKING FOR SUCCESS IS STEADY CLIMBING LONG-TERM PURPOSEFUL ACTIVITY IS A JOURNEY CONTROL IS UP ACTION IS SELF PROPELLED MOTION CHANGE IS MOT
34、ION 7. Turkish: .Kahvehane .esitleri, zamanlarinin sosyo-kltrel yapisina uygun bir yelpaze olusturmustur. Metaphor: GRADED DIVERSITY IS A FAN Literal Gloss: .The types of coffee houses have constituted a socio-cultural fan for their times. English: .The varieties of coffee houses have reflected the
35、socio-cultural spectrum of their times. Metaphor: GRADED DIVERSITY IS A SPECTRUM 8. Turkish: .Istanbulun kahvehanelerinin once renkleri, sonra hayalleri, en son da kokulari karisip gitmistir hayatin akisina. Metaphor: WEAKENING THINGS BLEND AWAY LIFE IS A FLOWING BODY OF WATER Literal Gloss: .First
36、the colors, then the dreams, and finally the smells of Istanbul.s coffee houses have blended away and gone into the flow of life. English: .The coffee houses of Istanbul today are but a faint shadow of their counterparts in earlier centuries. Metaphor: WEAKENING THINGS ARE SHADOWS 9. Turkish: .Diger
37、leri ise msterilerinin tercihlerine g.re kabuk degistirmis simdinin kahveleridir. Metaphor: CHANGING APPEARANCE IS CHANGING SKIN Literal Gloss: .The remaining are modern coffee houses that have changed skin to adapt their customers. preferences. English: .The remainder have changed in step with the
38、times to suit a new type of customer. Metaphor: TO ADAPT IS TO MATCH WALKING PACE 10. Turkish: .Sonradan ithal Fransiz caflerinin egemenligine inat, Istanbul onlari eski bir dostun hatirasi gibi .zenle saklar. Metaphor: WIDESPREAD EXISTENCE IS SOVEREIGNTY OLD BUT LOVED THINGS ARE OLD FRIENDS Literal
39、 Gloss: .Despite the sovereignty of newly imported French cafs, Istanbul protects them like memories of an old friend. English: .Despite the newfangled cafs which have mushroomed in fashionable districts, Istanbul obstinately protects its old friends. Metaphor: APPEARING SUDDENLY IS SURFACING LIKE M
40、USHROOMS OLD BUT LOVED THINGS ARE OLD FRIENDS Procedure: Subjects were provided with the original sentences and asked to translate them into English if the sentence was in Turkish, or into Turkish if the sentence was in English. They were allowed to use as much time as they needed. However they were
41、 encouraged not to try to make their translations perfect, and to rely on their instincts when in doubt. 3. Results Analysis was carried out to measure whether there was a difference in the proportion of metaphors that were carried over into the target language during translation. In doing so, the a
42、vailability of the metaphor in the target language was taken into consideration. The number of metaphors translated by the subjects into Turkish and into English were counted, as well as the metaphors that did not make it into the translated sentences and new metaphors which emerged in the translati
43、ons. Taking into consideration whether a translation of each metaphor into the target language is at all available, chi-square tests were run on the distributions of translated and non-translated metaphors in the two target languages. The results indicate that a significantly greater proportion of m
44、etaphors were translated literally when subjects translated from their L2 to their L1, namely from English to Turkish, as shown in table 1 and figure 1. Overall, when the metaphorical usage was possible in the target language, subjects translated 49% of the metaphors from Turkish to English and 78%
45、of those from English to Turkish. The chi-square test of independence was significant at p 0.001. Table 1. Translation of metaphors in each direction Metaphors translated Translatable metaphors Percentage Turkish to English (L1 to L2) 41 83 49% English to Turkish (L2 to L1) 56 72 78% Figure 1. Trans
46、lation of metaphors in each direction Metaphors translated 0% 20% 40% 60% 80% 100% L1 to L2 L2 to L1 Further analysis was carried out to demonstrate that the result generalizes across subjects. The proportion of metaphors translated over the total number of those that can be translated into the targ
47、et language was calculated separately for each of the ten subjects and languages. The paired t-test for the means was significant at p 0.01. We will briefly outline some of the qualitative results: For sentence pair 1, subjects tended to carry over one of the RULING OF SEASONS metaphors to English. But the reverse was also true; about half of them carried over the MOTION OF TIME metaphor into Turkish. In the translated languages both metaphors are available but infrequent. In sentence pair 2, all subjects used the expression .left its place to. in their Turkish tr
