Translation and Semantics in Theory and Practice

Translation and Semantics in Theory and Practice

Robert de Beaugrande

Universidade Federal de Paraíba

1. Symmetry and asymmetry

My title might appear to express some comfortable symmetries, but, on closer examination, important asymmetries emerge. For most of its very long history, translation has been regarded as an eminently practical concern of its practitioners, whereas, until the 1960s, ‘translation theory’ remained an obscure and barely explored domain or discipline positioned uncomfortably in academic institutions somewhere between literary studies and linguistics. In fact, in both literary studies and linguistics, some leading standpoints claimed or at least implied that translation is in theory impossible, however much it may be practised. Notable instances include Benedetto Croce’s (1902) aesthetics suggesting to Edward Sapir (1921: 222) that we might imagine ‘a work of literary art can never be translated’ insofar as the ‘effects’ due to ‘the formal “genius”’ of a ‘language’ or to ‘the colour and texture of its matrix’ ‘cannot be carried over without loss or modification’; and Noam Chomsky’s (1965: 201) avowal that linguistics cannot be expected to specify any ‘reasonable procedure for translating between languages’ because ‘an encyclopedia’ of ‘extralinguistic information’ would be demanded.

On the side of semantics, the situation is just the reverse. With a few exceptions (e.g. the work of J.R. Firth and M.A.K. Halliday) semantics has been regarded as supremely theoretical issue to be explored at some level of abstraction well removed from the practical concerns of meaningful activities in ordinary communication. Many primary notions, such as ‘denotation’, ‘reference’ and ‘referent’, ‘intension’ and ‘extension’, and ‘semantic features’ or ‘sememes’, indicate a preference instead for categorising meanings in terms of deterministic or ‘logical’ relations between words or expressions and real or possible states of the world.

The asymmetry between translation and semantics might accordingly be viewed as one of the many consequences of a massive asymmetry between theory and practice (Beaugrande 1997b, 1997c). The otherwise surprisingly low regard for the work and achievements of practising translators can be readily understood as one reflex of the very general low regard for practitioners in contrast to theoreticians in a ‘modernised’ and ‘specialised’ society. Translating is thus considered by uninformed outsiders to be a fairly mechanical and tedious activity of substituting word after word inside phrases or sentences; problems and obstacles are rarely significant and can be overcome by mere ingenuity or minor ‘sacrifices of fidelity’ to the ‘literal meaning’, though the concept of ‘fidelity’ urgently needs to be reassessed (Hurtado 1990). This simplistic image is exploited by institutions like publishing houses to justify stingy remuneration for the work of translators, and an ominously self-serving tolerance for hasty or careless results.

In sharp contrast, the production and elaboration of ‘theories’ has typically been entrusted to prestigious experts who are insulated from the day-to-day exigencies of practice. In at least some disciplines, among them ‘theoretical linguistics’ widespread in the U.S. and its intellectual colonies, the degree of abstraction from practice actually appears to constitute a criterion for ‘scientific authority’.

The widespread disconnection between practice and theory is therefore a predictable consequence of sharp disparities in the social status and working conditions of practitioners versus theoreticians. In social and institutional terms, the cost of this disconnection has been staggering: blocked communication and consultation, massively inefficient implementation, mistrust and alienation, and much more. But the disconnection does meet a major goal of the elites within officially ‘democratic’ societies to evade the responsibility of putting inclusive theories like ‘universal public schooling’ or ‘equal opportunity’ into inclusive practices (Beaugrande 1997a). Instead, we find a symptomatic disconnection between inclusive theories and exclusive practices. Meanwhile, public discourse of government administration, mass media ,and so on work tirelessly to assert and defend the equality and fairness of the social and economic system and to assure the victims of inequality and unfairness that they alone and individually are to blame and deserve just what they got (Afheldt 1994).

The growing tensions and inequalities in recent years are virtually all variations of a single complex theme: who shall control the parallel evolution of social and economic theories versus practices within a global society professing its fervent allegiance to ‘democracy’, ‘free enterprise’, ‘free market’, ‘human rights’, and so forth? The endemic failure to put these inclusive theories into inclusive practices is kept out of public awareness by the incessant messages that ‘progress’, ‘economic growth’, and ‘technology’ offer incontestable evidence that the society as a whole is achieving a ‘better life for all’, whereas in reality the gains are concentrated among an elite whose size is shrinking in proportion to the whole population (Martin and Schumann 1996). In practice, the mantras of the ‘free market’ and ‘free enterprise’ now designate the ‘freedom’ of dominant economic forces to maximise their profits and flexibility without responsibilities toward ordinary individual citizens, whose ‘freedom’ now consists of being ‘free’ to compete, under the conditions dictated by the corporations, for a shrinking supply of precarious down-skilled jobs that may soon be sacrificed to ‘rationalisation’.

This dangerous situation cannot be significantly changed until the citizens of a ‘modern society’ become genuinely aware of the constant need to choose between inclusion and exclusion. Those whose profession and expertise are expressly centred upon language and discourse in turn face the vital challenge of utilising all available means, especially through public discourse and education, to cultivate a widespread sensitivity toward the strategies of discourse itself for including rather than excluding (Beaugrande 1997a, 1997b).

2. Language and discourse in theory and practice

To take up the challenge, a productive outlook might be to define ‘language’ itself as a theory of human knowledge and experience, and to define discourse as the set of practices for working out the theory (cf. Halliday 1994). But we should immediately qualify these definitions by stipulating that (a) a language is the richest and most comprehensive theory in the whole history of ideas; (b) unlike a scientific theory, the theory within a language allows for a huge set of competing alternatives without specifying which ones might (let alone must) be true or false; (c) a language as theory cannot be verified or refuted by controlled experiments, but only examined regarding the recommendations and proclivities it entails for expressing things in some ways rather than others; (d) the theory can always be modified by the practices of discourse without being at all discredited or refuted. In turn, discourse practices are not neatly circumscribed by the language as theory. Instead, they partially confirm it and partly move forward its evolution to accommodate a rich variety of contexts and circumstances.

Yet discourse also constitutes a ‘theory’ about the objects and events constituting situations in the world, including the situation wherein the discourse itself occurs; a neat correspondence between what the discourse says and how the objects and events may be occurring ‘in reality’ cannot be taken for granted. On the contrary, the several participants may each be seeking to negotiate a version of ‘reality’ which holds adaptive value for their own goals, e.g., by giving the impression that those goals are natural, sensible, necessary, and so on. Still, testing the theory of the discourse against the practices with objects and events in a real situation is far more feasible and commonplace than testing the theory of the language against the practices of discourse on any large scale.

This layering of relations among theory and practice, as suggested in the graphic shown here,

suggests that all our conceptions of language and of all its factors or components should be grasped in terms of a dialectical interaction between theory and practice, and not as pertaining only to theory, as in formalist views of linguistics, or else only to practice, as in simplistic views of translation. Within a true dialectic, the theory should account for practices, whereas the practices should tune the constraints upon the construction and evolution of theories.

With a view to a new ‘semantics’, we should explore the dual status of meaning as theory and practice. The meaning of a given word or expression needs to be grasped as one element situated within the total language that constitutes a theory of knowledge and experience, rather than just as one free-standing label or pointer for one equally free-standing object or event in the world. We can then explore how these meanings mutually constrain each other on the plane of the discourse and not just how any one meaning might be defined in isolation in terms of ‘reference’, ‘denotation’, and the other familiar terms of conventional semantics. Again, we can assume that the several participants in a discourse seeking to negotiate a version of ‘meanings’ which hold adaptive value for their own goals, e.g., by making them seem obvious, transparent, commonsensical, and so forth. Much of the negotiating is done without deliberate awareness to accommodate the remarkable speed and efficiency of discourse processing, and relies on the self-organising convergence of multiple constraints (see Beaugrande 1997a for details).

These important principles point to a rather unfamiliar conclusion: to an impressive degree, a language is always in the process of being created and negotiated. So the correspondence between theory and practice is continually undergoing evolution whenever people participate in discourse, but modern linguistics has often conveyed a rather different impression by equating language (‘langue’, ‘system’, ‘competence’ etc.) with order, whilst equating discourse (‘parole’, ‘speech’, ‘performance’ etc.) with disorder. We thus find Saussure (1966 [orig. 1916]: 14, 9, 11, my italics) announcing that ‘language is a well-defined object in the heterogeneous mass of speech facts’, whereas ‘speech cannot be studied’, nor indeed can it be ‘put in any category of human facts, for we cannot discover its unity’. The tenor was the same when Chomsky (1965: 3f, 201, my italics) defined ‘language’ in respect to ‘an ideal speaker-hearer in a completely homogeneous speech-community, who knows its language perfectly’, and asserted that ‘much of the actual speech observed consists of fragments and deviant expressions of a variety of sorts’ as a reason why the ‘observed use of language’ ‘surely cannot constitute the subject-matter of linguistics, if this is to be a serious discipline’.

The italicised items project a dualism whereby an orderly, unified system continually produces disorderly and disunified manifestations when speakers use the language; presumably, hearers perform the reverse transformation. Yet already as a mere thought-experiment, such a mode of operation seems both frankly implausible and utterly counter-productive. A far more plausible assumption would be that the order of discourse is simply not of the same mode as the order these linguists have attributed to ‘language’. In effect, they have viewed ‘language’ as a theory permeated by a purely static and ideal order abstracted away from the apparent diversity and dynamic evolution whereby language is put into practice during discourse.

2.1 Implications for ‘translation’

For the purposes of the present discussion, two major implications emerge, one about ‘translation’ and one about ‘semantics’. The first implication is that translation is not at all a merely practical activity of substituting words but rather a vastly complex process of co-ordinating two dialectics between theory and practice. Skilled translators continually assess the current stretch of discourses in the source language and the target language in respect to the theory and practice of the relevant discourse communities. The process is necessarily far beyond the reach of any conventional linguistic theory that theorises about ‘language’ disconnected from discursive practices, especially when the theorising is devoted to constructing a highly idealised mode of order existing nowhere but within those same theories.

A corollary of this first implication would be that translating is a process of creating dynamic connections, many of them novel, between dual modes of order; but these are precisely modes of order which have not been adequately accounted for by the theories of language either in linguistics or in literary studies, just because the order is continually being created on line. This factor might explain why a number of early studies of ‘translation theory’ were not very informative for the practical concerns of professional translators (cf. Beaugrande 1978). On the one side, the linguistic studies tended toward premature idealisation, abstraction, and formalisation, especially when pursuing the prospects for developing mechanised algorithms. Little attention was devoted to the necessity for real translators to generate novel modes of order, or at least to generate familiar modes of order in novel degrees of delicacy and detail.

On the other side, the literary studies tended, on the contrary, to present unsystematic or anecdotal lists of special cases and colourful problems together with ostensibly ingenious solutions. Here, novelty was highlighted; but, in keeping with the ‘vitalist’ tradition of attributing creativity to some unfathomable or even sacrosanct ‘inspiration’, literary translators were inclined to portray themselves in an intensely personal and esoteric dialogue with the literary author, and to marginalise the experience of the target language-reader (Beaugrande 1978).

Within the approach advocated here, our leading agenda should be to establish by careful bottom-up studies of skilled translating the nature of the strategies that are both systematic and innovative. This agenda would demand that we greatly intensify our efforts to derive the theory out of extensive practices, whilst testing our theories for their empirical applicability to practices. In recent years, this agenda has begin to take shape in earnest (e.g. Lederer 1983; House and Blum-Kulka (eds.) 1986; Krings 1986; Hatim and Mason 1990; Lörscher 1991; Strolz 1992), following upon a series of works which programmatically highlighted ‘theory and practice’ (e.g. van Hoof 1962; Holmes (ed.) 1970; Grähs, Korlen, and Malmberg, (eds.) 1978; Kelly 1979; Königs 1979; Garcia Yebra 1982; Delisle 1984; Dedecius 1986; Snell-Hornby (ed.). 1986;. Garcia, P. 1990) or which departed from conventional linguistics by foregrounding the ‘text’ as the most valid unit of translation (e.g. Beaugrande 1978; Neubert 1985; Nord 1988).

However, what still needs to be worked out would be a comprehensive science of text and discourse where translation can be assigned a strategic position: ‘interlingual discourse’ as a linguistic, cognitive and social activity (Beaugrande 1997a). Virtually all of the theoretical or practical issues and concerns that confront us in translation also confront us in other modes of discourse and deserve to be approached in correspondingly general terms:

(a) What operations are required for data to be ‘translated’ from any modality into another?

(b) What standards or measures can determine whether and how well those operations have succeeded?

(d) How far are those modalities compatible with language?

(e) How far does this compatibility vary from one language to another?

(f) What operations are required for this compatibility to be maintained when translating a discourse from one language into another?

(g) Again, what standards or measures can determine whether and how well those operations have succeeded?

Viewed in such general terms, questions about ‘translation’ open out into very large questions of how data can be represented, accessed, and transferred in or among all types of modalities. We would be highly unrealistic to demand that the discipline of ‘translation studies’ or ‘translation science’ (‘Übersetzungswissenschaft’) answer such questions on its own when they have so seldom been raised in other disciplines with a view to genuinely reconciling theory with practice. And we would be highly unfair to charge the discipline with failing to achieve the status of a true ‘science’ when the ‘scientific status of its close neighbours in linguistics and literary studies remains profoundly uncertain (cf. Beaugrande 1988a, 1991a, 1997a).

The rational alternative would be to co-operate in a large-scale inquiry about what sort of a ‘science’ could represent the evolutionary dialectics between theory and practice within human interaction and discourse in general and within translation in particular. Only then could we determine how far ‘translation’ in the ordinary sense is similar to or different from other operations of representing, accessing, and transferring data. We might also be able to describe what sort of implicit theories get applied during the practices of skilled translators to favour successful operations; and what sort of implicit theories get projected onto the real-world context of situation by the discourse itself in either the source language or the target language.

If, as I have suggested above, a language is always in the process of being created and negotiated, we may anticipate a brighter picture than has typically emerged from ‘purely linguistic’ studies of translation, which naturally tend to emphasise the formal differences among languages, which in turn create a misleading impression that the languages are incompatible. In return, the factors which favour compatibility are largely functional and therefore much harder to capture precisely with ‘purely linguistic’ methods. This dilemma applies especially when ‘theories’ are expounded calling for ‘equivalence of meaning’ when our conventional theoretical concepts of ‘meaning’ are themselves far too static and deterministic.

2.2 Implications for ‘semantics’

We thus arrive at our second major implication, the one about ‘semantics’. Just as the practical aspect of translation have been unduly stressed, so too have the theoretical aspects of semantics. The most intractable problems within semantics have probably arisen from preconceived and inappropriate notions of how ‘science’ should proceed, especially as interpreted by the ‘science’ of ‘modern linguistics’ (cf. Beaugrande 1984, 1988b, 1991b). Semantics was led to theorise at high degrees of idealisation and abstraction, and yet to postulate that, in its ultimate essence, meaning is fully determinate and stable, without seeing the deep contradiction implied. Semantics has typically followed modern linguistics in tending to equate language (‘langue’, ‘system’, ‘competence’ etc.) with order, and discourse (‘parole’, ‘speech’, ‘performance’ etc.) with disorder, and has therefore readily assumed that the essence of meaning could not be derived from realistic observations of meaningful practices in authentic discourse. Instead, meaning was to accounted for by means of ambitious schemes of composition and classification, e.g., to label and enumerate all the basic elements of meaning, called ‘semes’ (‘sememes‘, semantemes’ etc.) or ‘semantic features’ in hopeful analogy to phonology and phonetics with their ‘phonemes’ and ‘phonetic features’. Yet the analogy was badly misconceived; the sounds of any given language constitute a system that is far more determinate and stable, as well as reassuringly compact and realistic; the system comprises a modest number of phonemes, and each one can be consensually defined in terms of its articulation, e.g., as the ‘voiced dental stop’ /d/. So phonology and phonetics succeeded in reconciling theory and practice, and could reasonably claim that their own ‘theory’ of the sounds of the language corresponded reliably with the implicit ‘theory’ that guided the articulatory and auditory practices of the community of speakers.

None of these reassuring conditions holds in semantics. No one has remotely succeeded in defining the size and status of the system of elements, let alone enumerating them all. Nor do we find any consensus about how the elements are to be labelled: if we use language, then we are dealing with the meanings of our language-labels, and not with meaning itself; if we invent another mode of representation, then we risk merely producing a ‘non-natural’ language of labels whose content is at least tacitly relayed through a ‘natural language’ like English, much like the operations in traditional cryptography like Morse code.

In consequence, semantics has been permeated by the aspiration of define and describe ‘meaning’ in reference to some formal or logical system on a deep and abstract plane thoroughly removed and purified of the ‘heterogeneous mass of speech facts’ that has intimidated linguists since Saussure. Their conception of ‘language’ was increasingly situated on a similar plane, e.g. in terms of a ‘generative grammar’.

And yet some influential linguists conjectured that set apart from language, meaning would be hopelessly indeterminate. For Saussure 1966 [1916]: 111f), ‘language takes shape between two shapeless masses’, namely between ‘the indefinite plane of jumbled ideas and the equally vague plane of sounds’; ‘without the help of signs we would be unable to make a clear-cut consistent distinction between two ideas’. Yet how language does this was thoroughly obscured when Saussure insisted that ‘the choice of a given slice of sound to name a given idea’ is ‘completely arbitrary’, for there are no necessary relations between sound and meaning’ (1966: 189, 113). This move didn’t really satisfy him, as we can see from his gloomy conjecture that ‘the mechanism of language is but a partial correction of a system that is by nature chaotic’ (1966: 133).

The same reasoning can be traced in the discourse of Louis Hjelmslev, who resolutely sought to build upon Saussure. His conception of the Danish term ‘mening’, translated into English as ‘purport’, was said to ‘have no possible existence except through being substance’ for a ‘form’: ‘the content-form’ ‘is independent of, and stands in arbitrary relation to, the purport’ (1969 [orig. 1943]: 54, 52). ‘Linguistic form’ ‘lays arbitrary boundaries on a purport-continuum’ that ‘depends exclusively on this structure’; otherwise, the ‘purport’ ‘exists provisionally as’ ‘an unanalysed entity’, and ‘subjected to many different analyses’, it ‘would appear as so many different objects’ (1969: 74, 50f). So Hjelmslev metaphorically portrayed ‘purport’ as an ‘amorphous “thought-mass”‘ ‘formed in quite different patterns’, like a ‘handful of sand’ or a ‘cloud in the heavens’ (1969: 52, 57).

What these linguists and their many followers failed to see is that language does not immediately and definitively impose ‘clear-cut consistent distinctions’ upon ‘meaning’ (or ‘purport’), but rather is always in the process of constructing meanings during discourse. By excluding discourse (‘parole’, ‘speech’, ‘performance’ etc.) or even equating it with disorder, linguistics and semantics stopped short of the most perspicuous phase in the evolution of meaning and cut themselves off from the most productive source of determinacy and stability: it is precisely in discourse where people reach the greatest certainty and agreement about what is meant, and not at all in the hypothetical ‘decomposition’ of word-meanings into ‘semantic features’.

3. A lesson for us?

The lesson for us today might be that if we want to discover and describe meaning, then we must go seek it where it actually subsists and persists: when participants use discourse both to theorise about their own world and to control or manage their practices in that world. We could acknowledge from the start that, since meanings are always being constructed and negotiated, the same must hold when meanings are investigated and described. We must be suitably sensitive to the types of interventions entailed in our own modes of inquiry. This factor can never be fully neutralised, but we might bring it under control if we adhere to some rational methodological principles. First, we should work chiefly with authentic discourse data and thus act in our capacity as members of a real language community who knows the language fluently but never ‘perfectly’. Second, we should expand and back up our own ‘competence’ in that capacity by working with very large corpora, which represent the discursive practices of a large population, though we can never cover all of those practices nor all of the language community. Third, our corpora should distinguish among significant language varieties, e.g. English in England versus English in Botswana. Fourth, our corpora should allow us to distinguish among relevant discourse domains insofar as these help determine the construction of meanings, which is doubtless far more commonplace than conventional semantics would like to admit.

If we adhere to these methodological principles, we can confidently expect to find that many of the central problems of semantics, such as the potential ‘indeterminacy of meaning’ caused by the presumed ‘heterogeneity of speech’, simply do not obtrude in authentic discourse data. To indicate this effect, I shall cite some data from the South African Corpus of English, recently established at the University of Port Elizabeth under the supervision of Chris Jeffries and Linda Pearce Williams and now containing about 2 million words of authentic contemporary discourse from a variety of sources (cf. Beaugrande and Pearce Williams, in press). There, we find the term ‘translation’ itself used to mean not just a relay among languages [1], as most people would predict on the basis of their unaided intuition, but also a range of practical operational, causal, or rhetorical relations between ‘policy formulations’ and ‘workable systems’ [2], between the ‘absence of land reform’ and ‘social instability [3], or between the ‘diversity’ of ‘democracy’ and the ‘diversity’ of mass electronic communications [4] (my italics).

[1] the questionnaires were translated into all three of the languages and sent the appropriate one to the appropriate linguistic group

[2] And you can see on that 148 as well as externally, including Government and private sector and other development agencies. An ability to translate complex policy formulations into workable systems and to oversee their implementation is vital.

[3] Business must accept that the absence of effective land reform will in itself translate into long term social instability.

[4] The demand for diversity of voices in a democracy needs to be translated into a diversity of radio and TV stations

In contrast, ‘semantics’ occurs much less often and usually to mean devious practices of using language to ‘confuse’ people rather than to define precisely what you mean, e.g., when your ‘stated intention’ gets translated into ‘half-hearted’ implications [5]:

[5] confused by the government’s stance, which has mostly been marked by semantics. Its stated intention to rationalise state assets, for instance, implies privatisation, but of a half-hearted sort.

Such data suggests a commonplace belief that the participants in ordinary discourse are normally clear about what is meant, and that ‘semantics’ can be deployed for nit-picking in order to mean something other than what you seem to be saying, or to twist about the meanings of other participants to your own advantage. In his memoirs, Jimmy Carter (1982) recalls Menachem Begin’s obsession with ‘semantics’ during the formulation of the Camp David treaties as a tactic for whittling away at the real commitments of the state of Israel.

‘Semantics’ in the sense of a serious ‘study’ or a ‘science of meaning’, as defined in contemporary linguistics, hardly ever appears in the large-corpus data I have examined, presumably because the general public can sustain folk-notions of ‘grammar’ as ‘correct usage’ and of ‘phonetics’ as ‘correct pronunciation’ but cannot sustain a folk-notion of what such a ‘study’ or ‘science of meaning’ might be concerned with or why anyone would need it when your common sense should suffice to tell you what words mean. This factor seems to have promoted pejorative uses of ‘semantics’ in data like [5] to designate a domain where pedantic or devious speakers try to doctor up the meanings of words in terms of take advantage of ordinary speakers.

Paradoxically, the determinacy of meaning for discourse data persists even though much of the data are innovative or even unique in fine detail. I see little doubt that the several meanings of ‘translate’ can be distinguished with reliably delicacy in their respective contexts back in [3-5], even though each data sample was novel at least to me and may be unique even in a much larger corpus than the one I queried.

Similar benefits might be expected for theory and practice of translation. To translate these corpus data into German, I would use the term ‘übersetzen’, which is given as the equivalent of ‘translate’ in all bilingual dictionaries, only for [1a] but not for [2a-4a]:

[1a] die Fragebögen wurden in alle drei Sprachen übersetzt und jeweils an die entsprechende linguistische Gruppe versandt.

[2a] Die Fähigkeit, komplexe Formulierungen von Richtlinien in praktikable Systeme umzusetzen und ihre Ausarbeitung zu überwachen, ist entscheidend.

[3a] Der kommerzielle Sektor muß zur Kenntnis nehmen, daß ein Ausbleiben von wirkungsvoller Agrarreform zu langfristiger sozialer Instabilität führen wird.

[4a] Der Ruf nach einer Vielfalt der Sprechergruppen soll durch eine Vielfalt der Radio- und Fernsehsender verwirklicht werden.

Rendering ‘semantics’ in [5] obliges the translator to interpret from the context, just because the ordinary use of the term is so fuzzy and mutable, e.g. [5a]. Certainly, German ‘Semantik’ would hardly work.

[5a] verwirrt durch die Haltung der Regierung, die zumeist durch eine umschweifende/nichtsagende Redeweise gekennzeichnet worden ist. Die erklärte Absicht etwa, den Staatsbesitz zu rationalisieren, deutet auf Privatisierung hin, jedoch von einer halbherzigen Sorte

The next step for translation theory and practice is clearly to make available parallel bi- or multilingual corpora, such as the English-French-Spanish corpora developed at the University of Lancaster. And as soon as possible after, these should be specified for relevant discourse domains.

4. Conclusion

I have tried to suggest why the relation between ‘semantics’ and ‘translation’ has been somewhat asymmetrical, due to their respective outlooks upon theory and practice. I have also suggested that the ratio between theory and practice urgently needs to achieve a new balance in both disciplines by adopting more realistic and dynamic view of language and discourse being worked out as theories and practices in their own right. The most promising source for a fresh initiative of this kind surely consists in detailed examination of the regularities in data assembled in very large corpora; and recent trends are pointing in just this direction.

Note

* This paper was composed in September 1997 at the Gaborone Sun Hotel, where the author was quartered pending his accommodation in university housing and where he did not have access to his reference libraries and data banks except for what was stored in his laptop.

References

Afheldt, H. 1994. Wohlstand für niemand? Die Markwirtschaft entläßt ihre Kinder. München: Kunstmann Verlag.

Beaugrande, R. de. 1978. Factors in a Theory of Poetic Translating. Assen: Van Gorcum, and Amsterdam: Rodopi.

___. 1984. Linguistics as discourse: A case study from semantics. WORD 35, 15-57.

___. 1988a. Critical Discourse: A Survey of Contemporary Literary Theorists. Norwood, N.J.: Ablex

___. 1988b. Semantics and text meaning: Retrospects and prospects. Journal of Semantics 5, 89-121.

___. 1991a. Linguistic theory: The discourse of fundamental works. London: Longman.

___. 1991b. An agenda for text semantics: Meaning and parametric adjustment. Folia Linguistica 25/2, 1-35.

___. 1997a. New foundations for a science of text and discourse. Greenwood, CT: Ablex.

___. 1997b. Society, education, linguistics, and language: Inclusion and exclusion in theory and practice Linguistics and Education.

___. 1997c. Theory and practice in applied linguistics: Conflicting, estranged, or cyclical? Applied Linguistics

___, and L. Pearce Williams. In press. Discourse and ‘democracy’: Some signals from the South African Corpus of English.

Carter, J. 1982. Keeping Faith. New York: Bantam.

Chomsky, N. 1965. Aspects of the Theory of Syntax. Cambridge: MIT Press.

Croce, B. 1902. Estetica come scienza dell'espressione e linguistica generale. Milan: R. Sandron.

Dedecius, K. 1986. Vom Übersetzen: Theorie und Praxis. Frankfurt: Suhrkamp.

Delisle, J. 1984. L‘analyse du discours comme méthode de traduction: initiation, théorie et pratique. Ottawa: Editions de l'Université.

Frawley, W. (ed.) 1984. Translation: Literary, Linguistic, And Philosophical Perspectives. Newark: Univ. of Delaware Press.

Garcia Yebra, V. 1982. Teoría y práctica de la traducción. Madrid: Gredos.

Garcia, P. 1990. Aspectos teóricos y prácticos de la traducción: Alemán-español. Salamanca: Universidad de Salamanca.

Grähs, Lillebil; Korlen, Gustav; Malmberg, B. (eds.) 1978. Theory and Practice of Translation. Bern: Peter Lang.

Halliday, M.A.K. 1994. Language in a changing world. Sidney: Australian Association of Applied Linguistics.

Hatim, B. and I. Mason. 1990. Discourse and the Translator. London: Longman.

Holes, C. 1994. Designing English-Arabic dictionaries. In R. de Beaugrande, A. Shunnaq, and M.H. Heliel (eds). Translation, Education, and Culture: An Arabic Perspective. Amsterdam: Benjamins.

Holmes, J.S. (ed.) 1970. The Nature of Translation : Essays on the Theory and Practice of Literary Translation. The Hague: Mouton

Hoof, H. van 1962. Theorie et pratique de l'interprétation, avec application particulière á l'anglais et au français. Munich: Max Hüeber.

House, J., and S. Blum-Kulka (eds.) 1986. Interlingual and Intercultural Communication: Discourse And Cognition In Translation. Tübingen: Gunter Narr .

Hurtado Albir, A. 1990. La notion de fidélité en traduction. Paris: Didier Érudition.

Kelly, L. 1979. The True Interpreter: A History of Translation Theory and Practice in the West. Oxford: Basil Blackwell.

Königs, F. 1979. Übersetzung in Theorie und Praxis: Ansatzpunkte für die Konzeption einer Didaktik der Übersetzung. Heidelberg: Julius Gross.

Krings, Hans P. 1986. Was in den Köpfen von Übersetzern vorgeht. Tübingen: Gunter Narr.

Lederer, M. 1983. La traduction simultanée: Expérience et théorie. Paris: Minard.

Lörscher, Wolfgang. 1991. Translation Performance, Translation Process, and Translation Strategies: A Psycholinguistic Investigation. Tübingen: Gunter Narr.

Martin, H.P. and H. Schumann. 1996. Die Globaliserungsfalle: Der Angriff auf Demokratie und Wohlstand. Reinbeck: Rowohlt.

Neubert, A. 1985. Text und Translation. Leipzig: Verlag Enzyklopädie.

Nord, Christiane. 1988. Textanalyse und Übersetzen: Theoretische Grundlagen, Methode und didaktische Anwendung. Heidelberg: Julius Groos.

Sapir, E. 1921. Language. New York: Harcourt, Brace, & World, 1921.

Snell-Hornby, Mary (ed.). 1986. Übersetzungswissenschaft — Eine Neuorientierung: zur Integrierung von Theorie und Praxis. Bern: Francke.

Strolz, B. 1992. Theorie und Praxis des Simultandolmetschens: Argumente für einen kontextuellen Top-Down Ansatz der Verarbeitung und Produktion. Vienna: U of Vienna habilitation thesis.