The Role of Bilingualism in Translation Activity

Introduction

hroughout the world, translation is an everyday activity for many bilingual children. This fact inevitably raises a question in one's mind: Can every bilingual become a translator? In order to find an answer to this question, first the definitions of particular terms has to be made clear. The concept of bilingualism is a field of study in psychology and particularly in psycholinguistics, and different scholars attribute different roles to bilingualism in the general theory of translation.

We should clarify the terms through definitions in order to reduce the questions like: What is bilingualism? Who is a bilingual and who is a true bilingual? In his article "True Bilingualism and Second Language Learning" Christopher Thiery defines bilingualism and explains how it differs from true bilingualism (Thiery 1978).

"Bilingual: Having, speaking, spoken or written in, two languages" (Concise Oxford Dictionary).

He argued that the term 'perfectly bilingual' suggests two things: "the subject speaks both languages equally well; the subject has two mother languages." However, the first condition is useless in defining bilingualism because it is not possible to measure whether or not one can speak two languages equally well for lack of a criterion for comparison. When we consider the second condition, the way a language is acquired gains importance and at that point the definition of the term 'mother tongue' needs clarification. Thiery considers it as "the language (or languages) which the child has acquired by 'immersion,' i.e., by natural reaction to the sounds made by its environment in order to communicate with it" (1978: 146). Hence the important characteristic of the mother tongue is that it is not taught via another language; so no matter how well one speaks a language, if he has learned it by tuition, he cannot be considered a true bilingual.

As a result of these consideration, Thiery's article proposes the following definition for 'true bilingualism':

"True Bilingualism: a true bilingual is someone who is taken to be one of them by the members of two different linguistic communities, at roughly the same social and cultural level." According to this definition the subject is either rejected by his linguistic environment or accepted as a part of it. Here, the concept of cultural and social level emerges, which will be used in the following parts of the paper.

Paul A. Kolers (1973) discusses the relation between bilingualism and translation in his article "Translation and Bilingualism" through some real examples and case studies and mainly considering the lexical level of translation and the role of bilingualism. He uses bilingualism to examine some general questions in the use of language and to learn more about how the human mind handles different kinds of information. He emphasizes the human factor in translation by humans comparing it with machine translation. He points out that there is no satisfactory translation machine. The reason for this is the complex structure of languages and the fact that the meanings of words have more than one interpretation depending on the context in which they are used, and a machine cannot make this distinction.

Parallel to this argument, Catford (1965) argues that translation theory must be formed through a theory of meaning. His understanding of 'meaning' in a text can be analyzed at different levels and units; smaller units constitute the meanings of larger units. He assumes that "translation implies the substitution or replacement of textual material in one language by equivalent textual material in another language" (Catford 1965:20). The concept of equivalence is a problematic issue and, since it can be interpreted in different ways, we should clarify its usage here, in Catford's sentence. Stressing the importance of context, he argues that the equivalent meaning of a sentence can be entirely captured in a target language only at the sentence level.

One interesting point that he focuses on is the fact that the way bilingual people interpret words is different from the dictionary translations of the same words. Kolers explains this situation through a psychological approach: "Our words are commonly used in contexts, in situations that are defined both by their physical characteristics and by our habits, attitudes, dispositions, and intentions toward them. These cognitive and emotional conditions affect the way we interpret a word when we hear it or see it; they affect the meaning we give to the world" (Kolers 1973: 283).

In his article he points out that words denoting objects have rather similar meanings when translated. However, words that denote ethical or political ideas or emotions usually have different meanings in different languages. Although these types of words exist in all languages and cultures, the meanings people attribute to them differs from culture to culture. This explains the difficulty in translating culturally distant languages and why translations of bilinguals sometimes differ from dictionary translations. They adjust the meaning to the 'other' culture as they switch between languages. Even considering only the lexical level, we will see that culture plays a leading role in translation. The advantage of being a true bilingual as a translator becomes apparent here because being bilingual means being bicultural as well. That is to say, a 'true bilingual' is a member of two different linguistic communities, and that means s/he is also a member of two different cultures. In Kolers's article there is an example relevant to this topic. The word 'lamp' may have different interpretations according to the social and cultural context it is used. For a person living in an industrialized country, a lamp is a device attached to a wall by an electric wire and one turns it on and off by using a switch. On the other hand, someone living in the rural area of an underdeveloped part of the world would think of a device that can be carried and that can be lighted by applying a flame to a wick. From this example we can conclude that two words do not necessarily refer to the same thing even when they are given as dictionary equivalents of each other.

Kolers (1973) applied some tests to bilinguals in order to measure their brain function during the translation process. In one of the experiments, bilinguals were asked to recite the alphabet backwards. Half of the group tried it in English and the other half in their native language (French, German, Arabic or Korean). When they learned to do it in one language they tried to do the same thing in another language; however the results showed that learning to do one thing in one language does not necessarily mean that you will be able to also do it in the other language. French and German speakers were more successful in reciting the alphabet backwards in English than Arabic and Korean speakers because the names of letters of the first two languages are similar to English letters. This result shows us that, unlike music or natural sciences, languages have different characteristics that affect the translation activity and make it more difficult. If one were to do a word-for-word or phrase-for-phrase translation, that would yield the same result as a machine translation because there are no two languages that are so close that a word-for-word translation would create an equivalent product. Even closely related languages have some differences in idiomatic expressions or syntax (Malakoff and Hakuta 1991).

Think of a bilingual who has left his country and does not practice his native language anymore. After some time he will have difficulty in thinking in his own language and will find it easier to express his ideas in his second language. This has to do with getting used to one language, while the other language becomes "foreignized." The same explanation is valid in a situation where people who are fluent in language A but had their mathematical education in another language B prefer to calculate in language B despite being less fluent in that language.

Lambert considered translators as special people because they are more serious listeners when compared to other people. The reason for that is they must capture every detail of the speaker's message and convey it in another language without any loss. On the other hand, other people listen only enough to give an appropriate response. This difference as listeners makes translators special (Lambert 1978).

Lambert (1977 cited in Lambert 1978:132) relates this characteristic of translators to their bilinguality. "It may be that their bilinguality, a prerequisite for membership in the profession, has the effect of providing them with special forms of intelligence, sensitivity, and skills at finding out what is meant and what is implied."

Although translators are considered to be very careful listeners, some examples contradict this view. Some people can interpret while knitting or reading a book. So the question is how is the attention of translator divided? Maybe they are not such careful listeners. We know that interpreters use some phrases to fill in the gaps that occur during interpretation but how much filling can be used? Attempts have been made to answer such questions through a psychological approach to translation and its relation with bilingualism (Lambert 1978).

We have defined the terms 'bilingual' and 'true bilingual' above. Lambert uses different terms as he discusses the topic. He introduces the terms 'compound bilingual' and 'co-ordinate bilingual.' Actually 'compound bilingual' can be considered as another term for 'true bilingual' because it is defined as, "one who has learned his two languages simultaneously (e.g. from infancy on) and with interlocutors who used the two languages equally well, often, and interchangeably." In this article the term 'true bilingual' is used interchangeably with the term 'early bilingual.' On the other hand the word 'co-ordinate bilingual' is defined as, "one who had different acquisition settings for each language, different times of acquisition (the second language learned after infancy), and socio-cultural contexts (one language at home, the other outside the home)" (Lambert 1978:137,138). These definitions help to explain how bilingual language processing occurs and why differences in acquiring the language also affect the translator's manner of translation. A co-ordinate bilingual would have a different network of meanings of concepts, so his/her systems of languages are more independent from one another when compared to the relationship between the compound bilingual's language pairs. That characteristic of the co-ordinate bilingual results in more independent translation.

Another point that the type of bilinguality plays an important role in translating is that, as for compound bilinguals, if the meaning of a word is to be semantically shifted, this affects the meaning of any synonym for that word. That is to say, any kind of semantic change in a word affects its synonym. Co-ordinate bilinguals would not build such a link between a word and its synonym (Lambert 1978). Looking at this fact, we can conclude it is obvious that the brains of two different types of bilinguals work differently, and this can be explained by the difference between their language acquisition processes.

Lambert (1978) also used the term 'balanced bilingual'. This term evokes the idea of language dominance and equivalence, which is a delicate subject because measuring the language dominance of a bilingual is not an easy thing and the results are not considered meaningful by some scholars. Thiery (1978) mentioned this equivalence, using the term 'perfectly bilingual.' However, no one can speak a language perfectly, so in order to evaluate equivalence, one has to measure the dominance of one language over the other.

Lambert tested bilinguals for their language dominance. When applying the tests, it was assured that the frequency in the usage of the words used in the tests was similar in both languages. The things to be measured with these tests were the time taken to recognize words by the bilinguals in each language, their fluency in giving free associations of words in both of their languages, and the speed of reading words and translating them from one language to another. The results of the tests were interpreted according to bilinguals' time taken to recognize words. The ones who were balanced in their recognition of words were considered balanced in the other aspects as well (1978).

A critical view on the idea of measuring language dominance in bilinguals has been contributed by Thiery (1978). He argues that there is no way meaningfully to measure equivalence or dominance between the two languages of a bilingual. Considering the lexical level, if we are to measure how many words one knows in one language and how many in another, this would not lead us to a real result. What if one is interested in a particular field, and experienced it in one country but not the other? Thiery gives the example of sailing in England and not in France, then this person would know more words about sailing in English but that it does not mean that s/he speaks perfect English. Therefore, he finds direct comparison between two linguistic performances both meaningless and impossible.

Lambert (1978) also focuses on the time it takes for bilinguals to translate. He applies tests to 'balanced bilinguals' to measure the mental time it takes them to translate. (Of course the group of people he considers as balanced bilinguals is again decided by his method of measuring the equivalence of their performance in both languages.) In a thesis by Reynolds (1970 cited in Lambert 1978:135) active-affirmative (AA) sentences were given to subjects, who then practiced to transform each sentence to the passive (P), negative (N) and passive-negative (NP) form. The results showed that on the average translation took extra time but certain translations when coupled with certain transforms took a shorter time than the transforms without translations. Also some facts about particular languages were interesting. For example, French passive took a longer time to reconstruct in English because it is a more complex structure. When the subject was asked to repeat the original sentence, it took a longer time than his/her translating because the base sentence is not kept in the subject's memory in its original form. With the help of this experiment Lambert claims to have found an answer to the question of how bilinguals hold the original sentence in their temporary memory before the act of translation starts.

The innateness of translation as a natural skill has been studied by various scholars, who have come to different conclusions about this subject. One of the scholars who has taken up this topic is Brian Harris; in 1973 he claimed that translation is an innate skill that can be developed through guidance just like any other natural skill and supported this idea throughout various case studies which he applied to a group of young people who had not received any training in translation and who do 'natural translation' in Harris's terminology (Harris and Sherwood 1978). The term 'natural translation' is defined as: "The translating done in everyday circumstances by people who have had no special training for it" (155). Depending on this definition Harris and Sherwood stated that all children can translate in every culture and in every language starting from the moment they become bilingual. There are views that support Harris's ideas about natural translation; for example, Shanon (1887 cited in Malakoff and Hakuta: 144) argues that bilingual children interpret for adults in various situations in medical, legal, or administrative cases. Although the ability to translate is accepted as something natural, Harris points out the fact that no natural skill can develop overnight; this is why he focuses on the stages that the natural translator goes through. By claiming that "translating is coextensive with bilingualism" (155) he means that bilingualism and translation have a relation that is similar to the one between speaking a language and the ability to communicate. Toury (1995) claims although the 'predisposition' for translating is "coextensive with bilingualism," the unfolding of one's translation skill depends on 'interlingualism,' which is the ability of establishing a relation between the similarities and differences between languages. Another view supporting the existence of predisposition to translation comes from Wills (1982 cited in Toury 1995:245). He defines the term as "part and parcel of mankind's basic linguistic equipment."

In Harris and Sherwood's system, the stages that a natural translator goes through are the following:

Pretranslation: is translation using "mostly single words, which is hardly surprising since the monolingual child too may still be in the 'one-word sentence' stage at that age" (166). Pretranslation is before interpersonal auto-translation and transduction.

Autotranslation: when "a translator translates to others what he has said or written himself; and sometimes he translates to himself" (165). When the subject translates to himself, this is known as "intrapersonal translation" and when the subject's own words are translated to other people, this is known as "interpersonal translation."

Transduction: where "the translator acts as an intermediary between two other people" (165). Transduction can be grouped as intrafamily (within the family) or extrafamily (out of the family).

If we take a look at how Harris presents his evidence supporting the hypothesis of translation as an innate skill, we will see that he based his argument on one criterion, age. This is one of the aspects that Toury (1995) criticizes and points to the fact that the authors who focus on the succession of translation phases and factors other than age are not considered in the developmental period of Harris-Sherwood's model of 'natural translating.' Their understanding has been considered problematic because their approach to the age factor is biological, rather than linguistic. That means that, when measuring time, their starting point was birth instead of the start of bilingualism. That would not cause a big difference if the subject were brought up with two languages from the beginning; however, in the case of the studies Harris used, there is a gap between the biological and bilingual age of the subjects. What is more, the circumstances under which the translation occurs are given scarce attention. In addition to the circumstances, the subject's individual characteristics and behavior also affect translation, but these factors are also being ignored (Toury 1995).

Actually, the fact that Harris took 'age factor' as the central point in his argument limits its validity under different conditions and makes it a one-sided approach because he tries to link every result to the concept of age, whereas he could have considered other factors that normally affect translation whether or not it is 'natural' translation. It can be concluded from the case studies that as the age of the subjects increase, their translations become non-voluntary and socially functional. If we take one example from the case studies, one of the bilingual children whose languages are Italian and English, at the age of 8, helps the elder members of her family who spoke only Italian, and her translation is considered 'socially functional.' Being a bilingual and consequently a bicultural, occasionally she acts as a cultural inter-mediator, especially when the cultural differences create a barrier to communication. If she were younger, her translations would be considered more functionally redundant and spontaneous, which means she would translate even when nobody asked her to.

As we see, children do translate without receiving any special training in the field of translation.

It is obvious that not all natural translators translate in the same way, but what makes the difference then? Is it the personal characteristic of the translator or his/her knowledge of the two languages or other factors? Harris and Sherwood (1978) again relate this situation to the age factor and the increasing linguistic proficiency with increasing age. However, according to Toury (1995), the difference in different people's translations is due to the fact that the capacity to transfer ideas from one language to another, i.e., the 'interlingual capacity,' is different in every individual.

Harris and Sherwood (1978) emphasize that the word 'innate' has double meaning in psycholinguistics. In its 'weak' sense it means "a specialized predisposition in children to learn how to speak from the language they hear in their environment"; its 'strong' meaning is "an inherited 'theory' of language which enables the child to speak sooner and more grammatically than can be accounted for by its contact with the environment" (168).

Toury (1995) does not deny that there is a "predisposition" for translating, but he identifies translation with bilingualism--a sort of oversimplification. Some other factors such as personality, context, and environmental circumstances are necessary to apply the specialized predisposition to translation. At this point, we see the weakness of Harris and Sherwood's argument, which considers age the marking criterion when discussing translation ability. Toury points out that there is a gap between the innateness of translation and its emergence as an actual behavior and development as a skill in the hypothesis of Harris and Sherwood. We mentioned about the phases of natural translation (pretranslation, intra/interpersonal autotranslation and transduction). These phases show that the skill develops over time and the development of skill over time conflicts with the notion of natural translation because one can conclude that, with increasing practice, the act of translating loses its naturalness. Toury carries this conclusion to an extreme by saying that the term 'natural translation' may ultimately become a mere synonym for the term 'bilingual.'

At this point he introduces the term "native translator," which includes the importance of factors like social motivation for and social functions of translating.

Toury (1995) approaches the notion of translator generally from a socio-cultural perspective and argues that it is a sort of "socialization" that is experienced by beginning translators. It is inevitable to mention the existence of "environmental feedback" if we are talking about an interactional phenomenon such as socialization. Translation has been considered a way of communication by many theorists, including Toury. He defines translation as "a mode of communicative text production" (Toury 1995: 248). In this kind of socialization, feedback can come from the commissioner as well as the recipients or sometimes from the creator of the utterance. Toury suggests that, no matter from whom it comes, the feedback that the translator receives is "normative" because it concerns the way the translation is created. The norms in the feedback affect the relation between the source text and the produced translation because the feedback has an effect not only on the target culture and language, but also on the mode of new text production, i.e., the translation. The norms transferred by the feedback also determine whether the translation procedures are appropriate or not (Toury 1995).

Toury also mentions about the 'sanctions' that are involved in the notion of norm and the role of those sanctions on a translator's behavior. One receives 'negative sanctions' as a result of improper behavior and 'positive sanctions' as a reward for proper action (Toury 1995:249). He argues that at the beginning a novice receives feedback as a response to her/his final product. In this case the translator starts to respond to the feedback immediately, and after some time s/he creates her/his translation according to this feedback. Finally the translator starts to consider the 'potential responses'; at this level an internal control mechanism develops inside the translator.

We mentioned these negative-positive sanctions and how they affect the attitude of the translator, but it is impossible to think of one true way of doing a translation. That is why Toury says that there will always be more than one single option and there is no such thing as a 'universal criterion' of appropriateness and that criterion differs from one societal group to another, just like other matters like equivalence in translation and the concept of translation itself. During the socialization of bilinguals as translators, they adopt a certain strategy of translation which changes as the translator moves to a different subculture. This time the new culture has its own norms and sanctions. As soon as a norm is internalized by the translator, it can be applied to the production of spontaneous utterances even when the translator receives no sanctions. We know that Lambert (1978) mentions the fillings that interpreters use in order to fill the gaps that might occur while translating. Toury also states that the native translator develops strategies during their socialization period to cope with the possible problems which may occur in the process of translation. That ability of the translator helps him/her so much that it can even compensate for the unbalanced bilingualism.

Toury (1995) mentioned the terms 'generalization' and 'specialization' within the context of native translators. A process of generalization is said to realize as the translator internalizes the monitoring device as a strategy that consists of only one type of procedure used both for 'negative' and 'positive' sanctions. When another event which he refers to as 'specialization' occurs, the translator's translation competence and adaptability is reduced. On the other hand, when a novice is faced with different situations and different norms are applied to his/her translation, s/he translates in a more flexible manner and this means s/he gains adaptability (Toury 1995). As one can conclude from the explanation, specialization is the opposite of adaptability in terms of a translator's attitudes. It is also possible that the translator disregards or even changes the norms that are tried to be applied to his/her way of translating. Toury states the importance of one's social status in his/her attempt to change the norms in the culture s/he acts. This factor, however, is not enough to change the norms that have been adopted by that specific culture over the years. Toury suggests that one's professionalism is parallel to one's nativeness and "therefore lends it (nativeness) the power to change the very notion of translation for the society in question" (Toury 1995: 254).

He marks the difference between the notion of 'natural translation' and 'native translator':

Malakoff and Hakuta (1991) states that translation requires language manipulation at two levels. It must both convey the meaning of the source text and produce an appropriate target text. If one considers natural translation, the translator would have to go through four phases:

1. understanding the vocabulary in the original work,
2. understanding the message in the original work,
3. reformulation of the same message in a second language,
4. deciding on the adequacy of the produced text.

It is not only the meaning that the translator reformulates while translating, but also the correct sentence structure in the target language. From this particular information Malakoff and Hakuta conclude that translation is both a communicative and a metalinguistic skill. Its communicative part consists in the translator understanding the message that is aimed to be given in the source language and conveys it in the target language. While doing this, the translator considers the sentence structure and linguistic characteristics of the target language and this constitutes the metalinguistic part of translation skill. That is why translation proficiency requires both bilingual proficiency and metalinguistic knowledge.

In natural translation, which is translation by an untrained individual as Harris defines it, one cannot talk of the linguistic knowledge of the bilingual. This explains why bilingual children can convey the message of a source language text by translating it to the target language, although they are unable to use correct sentence structures and syntax. Malakoff and Hakuta (1991) suggest that translation strategy is a learned thing and, in order to make translation that conveys the meaning of a source text, one must use correct sentence structures in the target language. This view brings a critical approach to natural translation by suggesting that the outcome of a natural translation may not be grammatically correct and a natural translator lacks metalinguistic awareness.

The question whether or not any bilingual can translate obviously does not have one definite answer. The answer to this question depends on what is understood by the term 'translation.' We have seen that the translation is a subjective concept and as a result of the different approaches to this latter term it is possible to answer the question that was asked at the beginning of this paper in various ways.

It is a fact that bilingualism is a necessary condition for translation activity, but is it enough for translation proficiency? This question, discussed throughout this paper, leads us to other questions about the innateness of translation activity, the natural ability to translate and the place of training in this picture. We have seen that each scholar introduces his/her own terminology and that every term conveys a different meaning even though they sometimes seem to refer to the same idea, such as the terms 'natural translation' and 'native translator.' In the light of these different approaches I tried to take a closer look at the relation between bilingualism and translation and present the complex world of language from a cognitive perspective.

Catford, J. C. (1965). A linguistic theory of translation. Oxford: Oxford University Press

Harris, B. & Sherwood, B. (1978). Translating as an Innate Skill in Gerver, D. & Sinaiko, H. (Eds.), Language, Interpretation and Communication. New York: Plenum Press. 155-170.

Kolers, P.A. (1973). Translation and Bilingualism in George A.M. (Ed.), Communication, Language, And Meaning, Psychological Perspectives. New York: Basic Books, Inc. Publishers. 280-290.

Lambert,W.E. (1978). Psychological approaches to bilingualism, translation and interpretation in Gerver, D. & Sinaiko, H. (Eds.), Language, Interpretation and Communication. New York: Plenum Press. 131-143.

Malakoff, M. & Hakuta,K. (1991). Translation skill and metalinguistic awareness in bilinguals in E. Bialystok (Ed.). Language processing and language awareness by bilingual children. Oxford: Oxford University Press. 141-166.

Thiery, C. (1978). True bilingualism and second-language learning in Gerver, D. & Sinaiko,H. (Eds.), Language, Interpretation and Communication. New York: Plenum Press. 145-153.

Toury, G. (1995). Descriptive Studies and Beyond. Amsterdam: John Benjamins Publishing Company. 241-258.