3.1 Contrastive Phonology

At the level of phonology, contrastive analysis can help to specify differences between the sound systems of the languages in question. To isolate the segmental units, various systems of phonetic/phonological description and classification compete, depending on the slate of the art of linguistic scholarship for each particular language pair. The transcription symbols devised by the fPA are employed in many (but by no means all) monolingual and interlingual dictionaries. Thus, both Collins dictionaries adopt [rj] as in ['kr>ntru:st] and [э] as in [kon'trast] to distinguish the

stressed and unstressed variants of English o\ German and French share with English

the pronunciation [o] but not ho]. However, both phonemes are transcribed by symbols

59

whose significance may not be obvious to the dictionary user .<,..> Recent contrastive 'suprasegmentaf or phonotactic analysis has produced interesting new-evidence (Ternes 1976, 4 explains the apparent 'singing'quality of Welsh speakers of English and Swiss speakers of Gentian in terms of inversely correlated stress and intonation patterns), but most interlingual dictionaries are still notoriously deficient in reflecting the prosodic facts of connected speech.

Much less is known about the pragmatics of sound systems, e.g. how such features as strong emphasis, ridicule, emotion and attitude are conveyed in different languages.

Progress in contrastive linguistics may lead to new advances in bilingual lexicography here, perhaps even to the development of new genres of interlingual dictionaries of pronunciation.

3.2. Contrastive Graphology

Some languages contrast sharply in their use of (especially non-alphabetic) writing systems.<...> An example of how this affects the lack-of-Ш between sound and spelling may be gleaned from the specimen entries: in word-initial position we note (disregarding loan orthographies) that French uses the letter с to represent the sound [k] as in contraste. German uses к as in Kontrast, while English uses either с as in contrast or к as in kindred. Dictionary users may or may not be familiar with such conventions. Whether special bilingual forms of the spelling dictionary are likely to become available, to reflect the paradigmatic, syntagmatic and pragmatic facts of written language, cannot be predicted here.

3.3. Contrastive Lexicology

At the level of lexicology, anisomorphic relationships arc even more awkward. 'Meaning discrimination⁵ is the lexicographical technique which is supposed to clarify the various multiple relationships between words (lexemes)

60

and their meanings (senses). Jn the monolingual dictionary, semantic similarity is shown by paraphrase or definition or the use of synonyms, in the bilingual dictionary equivalence is achieved by metaphrase or translation or the use of glosses. For the contrastive analysis of these complex lexico-semantic relations, a 'componential approach' is often suggested as a solution: just as phonological units may be broken down into distinct phonetic features, lexemes are held to be capable of analysis into senses as 'complexes of semantic features' (James 1980, 91). Thus a contrastive analysis of the different senses of a word like contrast ('comparison', 'opposition', 'distinction', etc.) would not only be regarded as revealing their equivalent(s) in other languages (French contrast?, opposition, etc. and German Gcgenuberstellung, (iegensafz, Unferschied, etc.), but also seen as improving the arrangement of the information within the dictionary entry itself However, in view of the known indeterminacy of interlingual interference in the language acquisition progress (cf. Nemser/Vincenz 1972) and the apparent inability of existing dictionaries to capture the complexities of 'directionality' and 'multivergence', such a straightforwardly direct application of contrastive linguistics to bilingual lexicography will be very difficult to execute. Could it explain, e.g., the fact that contrast(e) in English (and French) has a greater semantic range than German Kontrast and therefore requires more glosses and a larger entry in the English-German dictionary? It is possible that further advances in contrastive lexical semantics, semotactics and pragma-lexicology may bring limited benefits to bilingual lexicography. Suggestions have been put forward, e.g., for a 'contrastive dictionary of synonyms' (Snell-Hornby 1984, 278) or for better codifications of 'false friends' and technical terminology.

3.4. Contrastive Grammar

The contrastive analysis of the word, its formation from smaller building

blocks (morphology) and its functioning in sentences (syntax) is fairly well developed

for many pairs of languages, using a variety of theoretical models. Thus, word classes

61 are marked by the labels n and v, as are subclasses of verbs (note that English contrast, but not all its equivalents, can function either as w or vt). Systemic differences in degrees of collocability are sometimes indicated by means of examples, e.g. in contrast with to par opposition d_y par contraste avec im Gegensatz in Kontrast zu.

But many problems remain unsolved; does a phrase like and now, by way of contrast have literal or idiomatic equivalents, and could they be found by the means of an as yet inadequately constituted contrastive pragma-grammar? Some specialised dictionary genres are open to interlingual adaptation in this area, e.g. the dictionary of 'constructions' and 'idioms'.

3.5. Contrastive Textology

If the objective of bilingual lexicography is to find lexical units of the target language which, when inserted into the context, produce a smooth translation (Zgusta 1984, 147), we need to know how discourse is structured into the text and how this works in different languages. There have been several attempts, since Vinay/Darbelnet 1958, to develop a contrastive linguistics at the level of the text, but none have resulted in models that can be easily applied to the design of the interlingual dictionary. The problem seems to lie, once more, in our inability so far to model the bilingual text competence of the translator-lexicographer. Can a systematic analysis of 'parallel texts' (Hartmann 1980, 37) help us explain the various communicative strategies that are at work when functional equivalence is produced⁰ How does such a model incorporate semantic, syntactic and pragmatic dimensions? At its most advanced and ambitious, contrastive discourse analysis also needs to address itself to problems of culture-specificity, which might in turn lead to a new generation of dictionaries of cross-cultural information.