1.1 – Linguistic competence

3Within theoretical linguistics, the notion of competence is the notion of a native speaker’s knowledge of their language, which is sharply distinguished from their performance, “the actual use of language in concrete situations” (Chomsky 1965, 4). Speakers’ abilities to do things with words in actuality depend heavily on aspects of context (see Austin 1962; Hymes 1971, 1972a, 1972b) which lie way beyond the area that can be theorized from the Chomskyan perspective on language as primarily part of the natural world, rather than the social world, and of competence as an unconscious mental state reached at the end of a process of biological growth.

4This process of biological growth consists in interaction between (i) linguistic input data received by the language-acquiring individual and (ii) the default or initial state of the language faculty, which contains innate knowledge called Universal Grammar (UG) (Wearing 2006, 341). UG includes (i) “a set of innately endowed grammatical principles which determine how grammatical operations apply in natural language grammars”; and (ii) “a set of grammatical parameters which impose severe restrictions on the range of grammatical variation permitted in natural languages (perhaps limiting variation to binary choices)” (Radford 2004, 17). In addition, there is “a set of semantic primitives, out of which specific word meanings are constructed” (Carr 2006, 333).

5At the end of this process of interaction, we reach the steady state that is competence, which is characteristic of every normal adult native speaker of a given language. Variations in performance are exactly that: performance features, influenced by aspects of the social world, including other individuals, education, geographical location, social position, and so on. Competence is internal to each single individual; it is an I-language. It is not the shared, social, supra-individual entity that Saussure’s (1916) langue is often understood to be (Lyons 1996, 15), and which, for Chomsky, who of course does not deny its existence, is E-language. I-language is not primarily designed for use in communication, according to Chomsky (though of course he would not deny that it is used in communication), but for thinking (Carr 2006, 332).

6In linguistics, there is tacit agreement that this is what the term competence means within the discipline, whether or not a person agrees with the theory that the notion belongs within. And because everyone more or less agrees with the definition, people in linguistics mostly know what they are talking about when they use the term, and that is very useful. But it is not the way that the term competence is used in most other areas that use it as a technical term.

1.2 – Broader concepts of competence

7According to Wikipedia, in human resources parlance, competence is viewed as

8a standardized requirement for an individual to properly perform a specific job. It encompasses a combination of knowledge, skills and behavior utilized to improve performance. More generally, competence is the state or quality of being adequately or well qualified, having the ability to perform a specific role.

9According to the project “Promoting Social Competence” run by the University of Dundee in collaboration with the Scottish Executive, “Social Competence is possessing and using the ability to integrate thinking, feeling and behaviour to achieve social tasks and outcomes valued in the host context and culture.” [1]

10What distinguishes these definitions from the notion of Linguistic Competence can be depicted as in Table 1:

11In the following section, I will discuss some of the notions of translation competence that we find within translation studies, and attempt to determine whether they lean toward one, both, or neither of the two views of competence depicted in Table 1.

1.3 – Translation competence

12Even those discussions of translation competence in translation studies that strive towards the Chomskyan understanding tend to include aspects of, or be akin to, the Human resources and Social Competence definitions. For example, PACTE (2000, 100) claim to have borrowed the notion of translation competence “from the idea of linguistic competence”, but they define translation competence as including an array of knowledges, skills and abilities which vary between individuals and which would never find their way into the notion of linguistic competence.

13According to PACTE [2], there are six subcomponents of translation competence, which I list briefly here with just a smattering of what PACTE includes under each (PACTE, 2000, 101-102):

1. Communicative Competence in two languages, including linguistic, discourse and sociolinguistic competence.
2. Extra-Linguistic Competence composed of general world knowledge and specialist knowledge.
3. Instrumental-Professional Competence composed of knowledge and skills related to the tools of the trade and the profession.
4. Psycho-Physiological Competence, “defined as the ability to use all kinds of psychomotor, cognitive and attitudinal resources” including “psychomotor skills for reading and writing; cognitive skills (e.g. memory, attention span, creativity and logical reasoning); psychological attitudes (e.g. intellectual curiosity, perseverance, rigour, a critical spirit, and self-confidence)”.
5. Transfer Competence, which is “the ability to complete the transfer process from the ST (source text) to the TT (target text), i.e. to understand the ST and re-express it in the TL (target language), taking into account the translation’s function and the characteristics of the receptor”.
6. Strategic Competence, which includes “all the individual procedures, conscious and unconscious, verbal and non-verbal, used to solve the problems found during the translation process”.

14

Bell (1991) describes translator competence as a huge summation: target-language knowledge, text-type knowledge, source-language knowledge, subject area (‘real-world’) knowledge, contrastive knowledge, then decoding and encoding skills summarized as ‘communicative competence’ (covering grammar, sociolinguistics and discourse). Virtually everything that any kind of linguistics wanted to talk about was tossed into the soup. […] Hatim and Mason (1997, 204-206), working from Bachman (1990), present a traditional three-part competence inherited from linguistics (ST processing, transfer, TT processing) and then name a handful of skills for each of those heads. Hewson (1995) adds something called ‘cultural and professional elements’ (108), where the ‘professional’ part refers to ‘remuneration […] access to and use of proper dictionaries and data banks, access to equivalent material in the second language, practical knowledge of word-processors and peripherals, and so on’ (ibid.). […] Mayoral (2001, 109) insists on components including ‘common sense (above all), curiosity, ability to communicate, capacity for self-criticism, meticulousness, ability to synthesize, etc.’ Anything else? In Douglas Robinson’s Becoming a Translator (1997) we find serious attention to the real-world necessities of good typing speeds, Internet discussion groups, and working with a computer in a room at the right temperature [sic].

15Pym explains this proliferation of features in accounts of translation competence with reference to a perceived need to establish the discipline as something separate from linguistics and from language learning; another reason that people include so many features is probably that the task of the translator is indeed very complex. But are all of these desirable states and characteristics really parts of translation competence, or are some better placed in lists of more general competences, leaving translation competence to specify just those features that are peculiar to the translation task?

16De Groot (2000, 54) provides an interesting discussion of how to identify task specific components of complex tasks. Her discussion is framed in training-oriented terms, but I think it works equally well as a guide to concept inclusion in an account of task-specific skills:

17

Not all components that can be distinguished in a criterion task will need to be trained because a number of them may be mastered already at the onset of training. An example is visual word recognition, in translating written text. In fluent readers this process proceeds to a large extent automatically and effortlessly […] on the assumption that the typical trainee selected for participation in a translation training program will be a fluent reader, the inclusion of a visual-word-recognition component in such a program would thus be a waste of time and effort.

18It is my belief that all of the components listed by the scholars included in Pym’s entertaining article (2003), with the exception of the notion of transfer competence, may be prerequisites to translation, or desirable states which may enhance translation, which, however, do not make a translator. Let me now spend some time unpacking this notion of transfer competence.

1.4 – Transfer competence

19According to Toury (1984), transfer competence requires particular modes of socialisation (cf. Toury 1995, 246, 250). So however closely it may turn out to be possible to align the notion of transfer competence with a Chomsky-style account, it is unlikely that we shall achieve the Chomskyan elevation above the social. But if we can align other aspects of transfer competence to linguistic competence, the prospect of transfer competence as a category that encompasses aspects similar to those that characterise linguistic competence along with aspects similar to performance features presents itself invitingly. Perhaps the phenomenon of translation illustrates more clearly than any other the connections between competence and performance.

20Let us begin by imagining transfer competence as that which every adult translator has. Let us say that this competence has developed against the background of interaction between an initial state and relevant input.

21What, then, might the initial state be like? Presumably, it would need to include two or more languages in whatever measure, depending on when acquisition/learning of the various languages began. The input data would probably need to be translational: seeing translation, doing translation, and receiving feedback on translation, because there is plenty of evidence that just having two or more languages is not sufficient for someone to develop translation competence (see e.g. Toury 1984; Valdés 2003). So maybe we, researchers, could describe transfer competence as :

• the knowledge of the translational relationships between their languages that allows a translator to match languages appropriately when translating, as distinct from their ability to use their languages individually;
• an unconscious mental state reached through a process of cognitive development.

2.1 – The nature of the initial state: (developing) linguistic competence in two languages

22In the Chomskyan tradition I have modeled my theory of transfer competence on, relatively little has been said about bilingualism and bilingual acquisition; but there is a clear tendency not to distinguish these from monolingualism and monolingual acquisition in any fundamental way (Chomsky 2000, 169):

23

Even to speak of Peter as having the I-language L is a severe simplification; the state of any person’s faculty of language is some jumble of systems that is no more likely to yield theoretical understanding than most other complex phenomena of the natural world. Peter is said to be multilingual when the differences among his languages happen to interest us for one or another reason; from another point of view, everyone is multiply multilingual.

24This is not especially helpful from our point of view, because translational behaviour, properly speaking, can only proceed on the assumption that the bilingual is conscious of their languages as separate languages, however they happen to be organized in the mind. This may be something that translation studies can helpfully dwell on and feed into linguistics.

25The quotation from Chomsky strongly suggests that a major task for the developing translator is (i) to keep the languages apart in the first place to avoid malignant interference from one to the other and (ii) simultaneously to consider them together in order to be able (iii) to match a given stretch of one language with a contextually appropriate stretch of the other language. If this is right, then a more informative description of transfer competence than the one I have just provided might be this, where the notion of non-consciousness is beginning to disappear:

• a translator’s knowledge of their languages simultaneously as one system, and as at least separable, and as related (as distinct from their ability to use their languages individually);
• an “unconscious” mental state reached through a process of cognitive development.

26This view may seem to be in conflict with the Chomskyan position, but the two views can be reconciled with reference to insights from the neurolinguistics of bilingualism (Paradis 1985, 1987a, 2004; see also Ahlsén 2006 and Fabbro 1999).

2.2 – The neurolinguistics of bilingualism

27Like Chomsky, Paradis assumes no categorical difference between bilingualism and monolingualism. A bilingual has “two subsets of neural connections, one for each language, within the same cognitive system, namely, the language system” (Paradis 2004, 110), but (Paradis 2004, 112):

28

awareness of language membership is a product of metalinguistic knowledge. In online processing, language awareness is of the same nature and as unconscious as the process that allows a unilingual speaker to understand (or select) the appropriate word in a given context. The process of selecting a Russian word by a Latvian-Russian bilingual person is the same as the process that allows a unilingual Russian speaker to select among the indefinite, almost unlimited, possibilities for encoding a given message.

29Both processes involve relating the selected item to a single, language-independent conceptual component (Paradis 2004, 200):

30

the conceptual component of verbal communication is not language-specific and there is a single non-linguistic cognitive system, even though speakers group together conceptual features differently in accordance with the lexical semantic constraints of each language. The lexical items are part of the language system, but the concepts are not.

31According to Paradis’s Activation Threshold Hypothesis (1987b, 1993, 2004, 28-31), a linguistic item stored in a bilingual’s brain requires, at any one time, a certain amount of positive neural impulse to activate it, and this requirement is its activation threshold. Every time an item is activated, the threshold is lowered, but it rises again if the item is not activated for a while. Whenever one item is activated, all other possible alternative items are inhibited, that is, their activation thresholds are raised. For whole languages, the process is described as follows (Paradis 2004, 115):

32

When one language is selected for expression, the activation threshold of the other language is raised so as to avoid interference […]. However, it is not raised so high that it could not be activated by an incoming verbal stimulus that impinges on the auditory sensory system and sends impulses to the corresponding representation […] the unselected language is not totally inhibited. Its activation threshold is simply raised high enough to prevent self-activation, but not so high as to preclude comprehension.

33I think this accords well with the description of transfer competence I presented above.

34So how might transfer competence develop? And how unconscious is it?

2.3 – Natural translation

35The development of linguistic competence is commonly described in terms of a set of stages that the language acquiring infant and child goes through from the onset of the development in early childhood until its completion by very early adulthood. A similar account was given of the development of translation competence by Harris and Sherwood (1978) who explicitly model themselves on Chomsky (Harris & Sherwood 1978, 160) and who consider translation to be an innate skill, as their title suggests.

36Harris (1977, 96 [3]) suggests that the best way of investigating this skill is to study “the translation done by bilinguals in everyday circumstances and without special training for it”, in the same way that linguists at the time tried (and many still try) to lay bare UG by way of the study of naturally occurring spoken discourse, as opposed to written text. He says:

37

The scientific study of translation, translatology, is behind linguistics in the study of data. While linguistics has reached out to include all speech acts within its proper study, the data of translatology remain, almost exclusively, professionally authored texts. The proper study of translatology is all translation. Everyday speech should not merely be included; it should be given priority. Translatologists should first study natural translation, which may be defined as the translation done by bilinguals in everyday circumstances and without special training for it. Several studies have been done on bilingualism which have produced relevant data; more research must be done, however, with natural translation as the specific object of the study.

38There is an assumption here that speaking spontaneously relates to writing in the same way that interpreting relates to translation, and I am not sure that this is the case; I think the TT-ST relationship in each case makes quite a crucial difference; but let that lie for now. Harris also claims that all bilinguals are able to translate, a controversial claim indeed, though clearly its validity depends radically on what is meant by “translate”. No doubt all bilinguals can match up bits of their language, but research in Think Aloud Protocol Studies, and earlier, by Toury (1984, 89) suggests that the production of socially acceptable translations is learnt, not innate behaviour, as we can see from the fact that not all translators achieve it and also from the fact that what is socially acceptable translation changes over time. The behaviour is norm governed.

39There is also good evidence that even Natural Translation is not, actually, automatically coextensive with bilingualism. Valdés (2003) suggests that equally strong bilinguals (insofar as it is possible to measure this) interpret unequally well, and both teaching and practice of the activity can help them. As she puts it (2003, 173), “the unfolding of innate capacities is a function of bilingual speakers’ actual practice in translating or interpreting”.

40Valdés’ study (2003) includes an account of 25 children ranging in ages from 4 to 12 years who interpret for their families. The children’s abilities varied: adults would begin by selecting the oldest child in the family to interpret, because they would be taken more seriously, but (2003, 78)

41

Over the years, however, the families turn to other children in the family. This is especially the case when a younger child is identified as more able and more enthusiastic about interpreting […] In general, parents nominated children they considered to be good in both languages. Parents also selected the child who seemed to like translating.

42The purpose of the 1978 paper where Harris works with Sherwood is to test the postulate that all bilinguals can translate by showing that young bilingual children can do so because (1978, 155):

43

If all bilinguals, even nascent ones, can translate, and since the onset of bilingualism often occurs in infancy, we ought, so our postulate predicts, to find young children translating. The previous articles brought forward evidence that this is so: young children do translate. […] However […] one does not expect a natural skill to develop fully overnight. An attempt will be made to trace the stages that the young natural translator goes through.

44Harris & Sherwood identify the stages that the young translator goes through on the basis, largely, of their reading of the famous case studies of childhood bilingualism by Ronjat (1913) and Leopold (1949). The stages are:

45(i) Pretranslation observed by Ronjat in his French-German bilingual son, from the time the child was one year and two months old. Pretranslation involves automatic production of both realizations (one in each language) of one concept, as exemplified by the following examples from Ronjat (cit. Harris & Sherwood 1978, 166):

46(1) Father: Dis merci (French: “say thank you”)

47Child: Danke (German: “thank you”)

48(2) Mother: Was hat Papa im Mund (German: “what has father in his mouth?”)

49Child: Pfeife (German: “pipe”)

50(3) Father: Qu’est-ce que c’est que ça? (French: “what is it that it is?”)

51

Child: Pipe (French: “pipe”).

52According to Ronjat, in the first case, the child responds by reflex in German because he is most often asked to say “thank you” in that language. In the second case, the languages of inquiry simply trigger two linguistic realisations of the one concept. Harris & Sherwood (1978, 166) call this a “bilingual response”, and in neither case, according to Ronjat as well as to Harris & Sherwood, is translation involved. The child at this stage (co-extensive with the one-word stage of monolingual acquisition) can also be observed spontaneously rehearsing word pairs: œil/Auge (“eye”); Schiff/Bateau (“boat”), a playful, enjoyable use of the two languages which persists later (Harris & Sherwood 1978, 166).

53(ii) Interpersonal autotranslation means translation of one’s own utterance for the benefit of others (1978, 165). This is concurrent with the later stages of pretranslation, and was observed by Ronjat from when his son was aged one year and nine months. It is preceded by a stage during which the child autotranslates for itself; this stage can clearly be considered similar in nature and motivation to the playfulness with single words just mentioned.

54(iii) Transduction is defined as “communication in which the translator acts as an intermediary between two other people” (1978, 165). This was observed by Leopold in his daughter when she was three years and three months old. It is clearly what most people think of as translation.

55Harris & Sherwood (1978, 167) speculate about the relationship between these stages and monolingual language development and about which of the notions of innateness then prevalent in developmental psycholinguistics is relevant to their study: A weak sense of “a specialized predisposition in children to learn how to speak from the language they hear in their environment”; or the strong sense of “an inherited ‘theory’ of language (‘universal grammar’) which enables the child to speak sooner and more grammatically than can be accounted for by its chronology [i.e. age]” (1978, 168). In spite of the above mentioned parallel which Harris & Sherwood (1978, 160) draw between Natural Translation and “Chomsky’s sense of ‘competence’”, they now opt for the weak sense of innateness, because they believe that only this will allow them to speculate about what may be required for the special disposition to engage in Natural Translation over and above what is required for the acquisition of the languages as such. I think they are probably right to do so.

56They list the following as potential features (1978, 168-169):

1. the pleasure that young children derive from translating
2. a bilingual mental lexicon
3. a language-independent semantic store
4. an ability to conserve meaning across languages

57If the proposed stages were found to be supported, there would be certain advantages, especially with regard to the first stage: (i) a spontaneous pretranslation stage lets us off the difficult theoretical hook of accounting for first steps; if translation links are established between languages as they are acquired, we have a good basis for translation activity proper. And (ii) innately endowed pretranslation would provide a rationale for cognitively determined translation universals such as the under-representation in translations of features of the target language that are not shared by the source language (Tirkkonen-Condit 2004). The proposed explanation for this universal is that TL unique items (vis-à-vis SL [source language]) are under-represented in a translator’s mental lexicon while he or she is translating. Nothing in the source text is likely to trigger them. The idea of pretranslation strengthens this supposition, because it suggests that the lack of opportunity for linking to the other language during acquisition may result in a permanently weak activation of unique items, compared to terms that have counterparts in the other language.

58With regard to the four features that Harris & Sherwood deduce for the bilingual, translating child over and above its natural predisposition to acquire language, two (a bilingual mental lexicon and a language-independent semantic store) are accounted for within currently preferred theories of Linguistic Competence and the Neurolinguistics of Bilingualism. Recall the quotation from Paradis (2004, 112):

59

The process of selecting a Russian word by a Latvian-Russian bilingual person is the same as the process that allows a unilingual Russian speaker to select among the indefinite, almost unlimited, possibilities for encoding a given message.

60It is also compatible with Chomsky’s (2000, 169) understanding of everyone as multiply multilingual. The existence of a bilingual mental lexicon is not what distinguishes bilinguals from monolinguals.

61As for the language-independent semantic store, again, this is posited by Paradis (2004, 200):

62

the conceptual component of verbal communication is not language-specific and there is a single non-linguistic cognitive system, even though speakers group together conceptual features differently in accordance with the lexical semantic constraints of each language. The lexical items are part of the language system, but the concepts are not.

63I suppose that the ability to preserve meaning across languages is transfer competence, which we have repeatedly unpacked as the consciousness of the language systems as together, separate and related.

64The final thing we may need and want to pay attention to in addition in our account of the initial state that provides the bases for the development of transfer competence is the translation pleasure principle, if Harris & Sherwood are right. Perhaps we should describe the initial state like this, where the translation pleasure principle has been added to the description we set out with:

65The development of transfer competence takes place by means of interaction between:

1. translation related input data
2. an initial state consisting of (developing) linguistic competence in two or more languages + the translation pleasure principle

66Of 13 young interpreters Valdés interviewed, 2 spoke directly about their enjoyment in interpreting, and none “spoke of their experiences in a negative way” (ibid., 90).

67It is also the case that a small group of TAPs (Think-Aloud Protocols) researchers who have investigated translator attitudes and motivation suggest that translation quality is likely to be positively affected by a translator’s personal affinity and involvement with the translation task, while a disinterested translator is less likely to produce a high quality translation. Laukkanen (1996, 270) goes so far as to suggest that “negative attitudes may […] have an adverse effect on the quality of [a translator’s] work”, although this study is inconclusive about direct benefits of a positive attitude. For her part, Jääskeläinen (1996, 69) suggests that “affective factors, be they personal involvement, commitment, motivation, or attitude, play a significant role in translation as well as in other forms of human behaviour”, and Tirkkonen-Condit & Laukkanen (1996, 45) suggest that “the right affective frame of mind goes with creativity and success in translation”. Kussmaul & Tirkkonen-Condit (1995, 191) add that “Personal characteristics such as tolerance of ambiguity, flexibility, realism and intellectual curiosity seem to contribute to success of performance”.

68There is one problem left with the description of transfer competence, though, namely the use of the transfer metaphor. As I understand meaning, meaning is generated anew in each speech encounter so that there is nothing in fact to transfer. But this means that what I have so far called “transfer competence”, has to be called “translation competence”, because if meaning is ever new, it makes little sense to talk of transfer (cf. Reddy 1979).

69Translation competence, as it is now called, develops via interaction between

1. translation related input data
2. an initial state consisting of (developing) linguistic competence in two or more languages + the translation pleasure principle