Trans Internet-Zeitschrift für Kulturwissenschaften 13. Nr. Mai 2002

Translation as Process Record

The scientific paper as a temporary verbal link

Stuart Sillars (Bergen/Norway)


Fundamental to a traditional concept of translation is the idea of equivalence - the idea that there are a series of concepts, entities or world views that lie beneath language that can be moved across from one system of linguistic markers into another. Recently, however, the concept has widened. The practice of translation has been extended to include 'newly texted'(1) materials such as advertising and promotional copy produced by translators with considerable deviation from the original words or underlying styles and assumptions because of differences between the original and the target culture. Thus the idea of translation has shifted ground, and both the practical role of the translator and the theoretical basis on which her or his work may or may not rest have undergone similar changes. At the same time, ideas that question the referentiality of language, especially the deconstructionist notion of language as self-contained and self-corroding, have destabilised the idea of language in literary, aesthetic and philosophical frames.

In the light of these shifts it may be valuable to consider a model of translation which touches on all the elements central to the above discussion but presents them in a different relation and with a different emphasis: the discourse of the international scientific paper. Most particularly this model is suggestive about a network of relationships: those between language and the material world; between language and the idea of process; and between literary and non-literary language. These come together to offer a significant insight into post-structuralist ideas of the innate inadequacy of language.

For many decades, the established language of the scientific paper has been English - perhaps a mid-Atlantic or North American English, or an English made very specific in its dialect, but a language still recognisable in its vocabulary and syntax to native speakers of a variety of englishes throughout the world. Perhaps this is in response to the insistent demands exerted in the paper's production. Frequently - if not invariably - the product of collaboration between academics of various levels of experience, perhaps working in different countries or even continents, the work undergoes a process of negotiated statement during its writing which may well include aspects of the new, broader transcultural definition of the concept of translation. Presented to a journal for peer review, it will probably undergo further revision as part of the process of evaluation. When published, it may well be read by a small, intimately related scientific community which nonetheless represents considerable linguistic diversity. Yet much may hang on its clarity in presenting its material: the comparatively recent controversy surrounding the claimed discovery of cold fusion by Fleishman and Pons - something significantly revealed outside the usual framework of the reviewed journal article - laid bare the consequences, in terms of the attainment of major grant awards and research facilities, to say nothing of international prestige for the host institutions, that may rest on the first presentation of findings. The paper, then, is eminently practical in its consequences; and, since a large proportion of scientific papers are concerned with inscribing the design and results of practical experiments, be they with old-fashioned test-tube work, contemporary dry laboratories, linear accelerators or computer simulations, there may be little doubt that the focus and concern of the papers is eminently practical too.

The particular kind of scientific paper with which I am here concerned addresses process - the process of demonstrating experimentally the truth of a particular proposition about the material world. The purpose of publishing is to establish this truth in a very specific way, by enabling the replicability of the process. This has not inconsiderable repercussions on the concept of language and translation that it may present: and this is the key difference in the scientific paper's identity. For all these reasons of production, revision and readership, the scientific paper of this sort is not written in one language, or several in the case of international collaborations, subsequently to be translated into English; instead, it moves straight from the laboratory or screen into a specialised language of its own. Or rather, two languages. One is the genuinely international discourse of mathematical formulæ which, though it relies on the Greek and Arabic alphabet for its signs, has a syntagmatic structure that, whether or not it may have certain parallels with linguistic structures in various tongues, is without question untranslatable because it is universally accepted within the scientific community: there is, for instance, no other way of saying H + O = H2O that parallels its concise precision and uses a alternative set of signs that is universally current. Thus the procedure already has a language of process, the only equivalent of which may be found in the signs used in symbolic logic. The question that underlies this concept, of whether or not the formula or equation can exist beneath or independent of the symbolic language that inscribes it, is as unanswerable as it is irrelevant in this context: what is central is the proposition that the language of the formula is universal because it transcends barriers presented by verbal structures.

The second language is provided in the verbal armature that holds the formulæ together - to explain the next physical stage of a process, limitations of circumstance, reasons for a particular choice or direction, or to suggest likely causes. The armature is formed in the second of the scientific paper's two modes: international scientific English. It makes direct assertions - 'x is a stable compound'; it makes extensive use of the passive voice, so that personal responsibility is displaced by a universal order of things; and often its common syntax extends beyond shared phrases to larger organising structures. Of these the most common is that known by its initials as IMRAD - Introduction, Methods, Results and Discussion.

I would suggest that, to all practical ends, the language system that constitutes this larger armature operates in a manner identical to that of the mathematical formula. Just as the language of scientific symbols and syntagma is universal, so is that of scientific discourse - the language that explains, justifies, orders and links. Consequently, the process of writing and reading the scientific paper exists in an intellectual and linguistic space quite different from those presented in the traditional patterns of translation. In these, there is an equation that has been summarised in this way:

Between the resultant text in L2 (the target-language text) and the source text in L1 (the source-language text) there exists a relationship, which can be designated as a translational, or equivalence relation.(2)

With the scientific paper, a quite different model obtains. Instead of being conceived in one language and then moved across - translated - into another, the concept has its origin in the special dialect or language adopted by the scientific community, international scientific english. Or rather, it has its origins in the concept or process. What has happened is that one key phase of the translation process has been erased, and we are left with what we might call a primal act of translation or equivalence between process and language.

This drastically reduced model of translation in production has its mirror in the process of reading or reception. Once again, this is concerned not with moving the linguistic codes across to the reader's mother tongue but with assimilating the data - from both mathematical formulæ and linguistic codes - directly into the reader's consciousness at the level of material actuality and process. This has a very significant consequence: since the aim of this kind of publication is to establish the replicability of the experiment, the actual nature of both language and formulæ is rapidly made irrelevant, lost in the process itself when it is duplicated by the reader, either conceptually or physically. As a result, the immediacy of process in the material changes inscribed in the paper is mirrored in the immediacy of the process by which it is stated and read. This has a further consequence on the conceptual status of the paper: its stress on materiality and process paradoxically reveals the temporary nature of its statement as one stage in the reconceptualisation or replication of the process, and its ultimate absorption into a larger cycle of knowledge or research. This gives us the most intriguing paradox of all: language that seems to be most intimately concerned with the actual, tangible world is given a degree of impermanence that mirrors, in a different dimension, the slipperiness of the relation between signifier and signified that so much preoccupies the postmodern consciousness. This ganglion of ideas and relationships can be developed in ways that illuminate the historical growth of language in practice, the concept of language in theory, and the process that the scientific paper represents. All these directions have much to suggest about the ways in which we classify language.

Whatever philosophers, literary theorists or translation theorists think about the relation of language to the material world - if, indeed, we believe in the existence of such an entity, rather than conceiving language as determining rather than recording whatever materiality may be out there - it is nonetheless the business of many scientists to uncover and elucidate the workings of this material world. As a result, the terminology of the preceding paragraphs is not only valid but essential. Further, the existence of such a pattern allows us to explore the relation of idea, material and language that some theories of linguistics and translation seek to model. In this, the work of a particular group of writers and thinkers becomes important: the group set up in the 1650s and subsequently known as the Royal Society, whose own experimental deliberations were recorded in the writings of Thomas Sprat. The work for which he is best known, The History of the Royal Society, appeared in 1667. This is often praised as a model of clarity in narrative, but it is not Sprat's work that is my main concern. His purpose is to chronicle, to place experimental philosophy within the context of a developing narrative line of the kind more recently rejected along with all other grands récits. In so doing his writing, leaving aside its precision of style and the object of its focus, is little different from that of Defoe: it orders, selects and presents along a temporal datum and is concerned with precision to the degree of offering elegance. In a very different way, another text is striking: John Wilkins' Essay towards a Real Character and a Philosophical Language.(3) Produced at the order of the Royal Society in 1688, this establishes an alternative set of relationships between language, the material world and our idea of sequentiality.

The introductory part of Wilkins' Essay sets out to present a schematic view of language as an inevitable part of the order of the material world, in a remarkable act of pre-Linnaean categorisation. In this it should not be confused with other attempts to address the materiality of meaning. The 6-volume Words and Phrases Legally Defined,(4) for instance, is clearly different, since it is an anthology of meanings based upon legal precedent - 'a work of judicial interpretations' (I, vii), as it proclaims itself, in terms delightfully ambiguous for the linguist. Here, words derive meaning through usage and consent: in Wilkins' text, they stand revealed with their proper peculiarity in a scheme of things as inevitable as the orders of species and genera. It should not surprise us that Wilkins makes use of these very terms when defining the nature of language and more specifically when advancing his core thesis - the 'real character' of the title.(5) This is a system of symbols which allow any language to be read from them.(6) If this sounds absurd, then let us remind ourselves that this is precisely what happens with mathematical symbols: the cross with radiused, enclosed extremities is read as åtte, huit, acht, otto, eight and so on. What is of particular interest in Wilkins' writing is that it draws together key elements of the scientific paper and recent ideas of how languages function: both Chomsky's notion of the deep structure of languages and the scientific paper's use of the passive, and both the scientific community's use of the symbol and the idea of reading the same symbol in many languages. All these ideas return to what we might call a creationist view of language, in which quite literally (in at least two senses), in the beginning was the word: the concept of an inevitable, pre-Babel relation between what we have come to regard, with poststructuralist mistrust, as Les Mots et Les Choses. (7)

In asserting the relation between language and materiality, the new system proposed by Wilkins stresses immediacy and transparency at a direct structural level, in a sense offering us a model for scientific communication. But it does so at a price: those elements of intrinsic pattern and internal relation that literary communities have traditionally valued as writerly style are effectively ignored. In so doing, it locates itself on one side of a debate that in different ways underlies the production of language in most periods but which is most particularly active from the end of the sixteenth to the middle of the seventeenth centuries. We can hear the debate in action by listening to the structures of two texts which exemplify respectively what we might term a consciously literary mode of discourse and one concerned far more with direct exposition. What is intriguing here is the nature of the difference in comparison with the dates of composition: that which we would expect to follow a model of scientific order and precision, coming as it does from the period of the English scientific revolution of the mid-sixteenth century and dealing with the creation and its consequences is the literary model; that which comes from a far earlier period, the great flowering of literary language in the 1590s, and which discusses a similar problem of cosmic order, is closer to what we might term a scientific pattern of statement. The first is the opening of Milton's Paradise Lost (published 1667 but written earlier); the second, the opening of Richard Hooker's Lawes of Ecclesiastical Polity (published 1594).

Milton's structure is complex. It depends on a Latinate principle in which a final main verb governs a subject often presented several lines away, the intervening spaces filled with a range of clauses which, rather than being clearly subordinate, all relate to elements essential to the relation between subject and verb. Punctuation is so complex and idiosyncratic as to have been the subject of at least one book, (8) many articles and countless footnotes, the use of the colon being far stronger than we would accept today and, often because of a use of conjunctions at the start of a sentence, the full stop or period much weaker. The opening period of Book I, for example has its first main verb at the beginning of line 6, a colon at line 10 providing a pivot between two large blocks equating to sentences, and the first period at line 16. This is, however, immediately negated by the use of 'And' to start the next sentence, making clear that it is in effect not a discrete unit. It is only at the end of line 26 that we have the ending of what is in all practical dimensions the first sentence.

Statistics of this sort are less important than the reasons behind them. Conceptually, the clauses move around each other in a celestial dance of mutual clarification that mirrors the complexity of the theme and inscribes the dignity of the process it captures: the process of creation, which is of course both that of the World and that of the Word, the Logos and the Poem. In a sense it might appear that there is a parallel with the scientific paper here in the shared stress on procedure; but there are two key differences. First, the mutual modification of the clauses ensures that the conceptual relations are not so much sequential as simultaneous, so that the process is held in abeyance within their circling. The second difference is one of function: whatever other reasons for its writing have been advanced, we may I think agree that Paradise Lost was not written with the view of enabling the replication of the process it describes.

Contrast that with Hooker's Laws of Ecclesiastical Polity. The sentences are in many cases just as long as those in Milton's epic, but their construction is wholly different. Instead of circling around each other to achieve a simultaneity of dependence, their clauses are presented in a sequence governed not by time but by logic, and signalled by the use of appropriate conjunctions and relative pronouns to lead us through the relationships, with verbs invariably presented next to the nouns they govern. The form is a linguistic reflection of the elaborate hierarchical structure that it is the work's purpose to unfold and which, elsewhere in the book, is revealed in diagrammatic form. In this model, the parallel with the scientific paper exists through the creation of a sequence of clearly identified concepts which are presented with the inevitability of symbols naming material elements: it is a Boethian inscription of symbolic logic. And reading it as though it were a scientific paper offers us some insight into how it must have been written - and perhaps read by its original readers: as a statement of truths as undeniable as those of the material world investigated by later generations of experimental philosophers, a neo-Platonic reality, full of grace and truth, being progressively unveiled, rather than a concept, figured forth by the inventive faculties, being proposed in a stative, not a dynamic manner.

Reading these two texts in the light of scientific papers and their particular complex of translation, symbol and form is instructive: it reminds us of how ideas are structured in two worlds - not those of science and art as so often, after Snow,(9) suggested, but those of different orders of belief, idea and language. Instead, the two may coexist within the larger idea of discourse, most particularly that of fiction. A certain kind of fiction - most often, though not always or necessarily, popular fiction - is read as process. The language is frequently described as transparent, but a more complete description of the process involved would be that, instead of being seen through as this implies, it is rather discarded by the reader as soon as possible to allow involvement with plot and event; and the process of its organisation, the polish of the telling, if you like, is a process metaphoric of that which it narrates, the whole relationship forming a remarkably similar model to that of the scientific paper. The other kind - the Miltonic model, perhaps - offers a wholly different relation between language and idea, one that is far more interpenetrative and, in a sense, fluid and open, since concepts and words revolve around each other within its complex forms.

This notion of process is, I think, the most significant insight that is available from the convention of the scientific paper. In his study of the assimilation of change in the scientific community, T. S. Kuhn argues that scientific change rests on the notion of the 'paradigm shift'(10) . What is really important, however, in such changes - what makes them happen, rather than what makes us recognise them when they are there - is the very processes that are revealed in the scientific paper. These embrace both the way that the paper is written and read and the procedures that it conveys, the one a metaphor of the other.

In the process - quite literally so - the scientific paper reveals a model of language that is deeply paradoxical. As the medium in which the process - experiment or whatever else - is conveyed, it is fundamental, and has a kind of truth that most linguists and theorists would today reject as unutterable and unattainable. But because its whole function is to disintegrate into the replication of the process it presents, it is temporary and in a key sense irrelevant. Writer in Tokyo uses temporary English to produce it: reader in Berlin uses temporary English to read it. The real equivalence is not that between L1 and L2, in Werner Koller's terms, but that between two enactments - the second either in the reader's laboratory or the reader's mind - of the process that it temporarily and awkwardly encrypts. Add this to the earlier conclusion - A + B = C - and we realise that even at its apparently most referential, language is impermanent and an approximation. It is pertinent to remark here that one of the landmark texts of literary theory, Jacques Lacan's lecture 'The agency of the letter in the unconscious or reason since Freud',(11) makes use of equations to explain the workings of metaphoric an metonymic structures. And this leads us once again to an awareness that the language of critical analysis of language and literature and that of the scientific paper may not be the polar extremes that we intuitively assume them to be. In this context, the stress in language on the processes it encodes reveals its own inadequacy, and offers us the final paradox about the nature of expression in scientific writing: the less elegant, the less satisfying the language is, the more immanent is the process, and therefore the more satisfactory the paper. It is a case not so much of More is Lessing but of Better is Worsing. The rawness of the value of language is again revealed, and the Saussurean model of slippage is given another application, here in splendidly paradoxical terms: it is the very closeness of the language to its referents in which its inadequacy and its transience stands revealed as its major significance.

© Stuart Sillars (Bergen/Norway)

TRANSINST       table of contents: No.13


(1) Werner Koller, 'The Concept of Equivalence and the Object of Translation Studies'. Target: International Journal of Translation Studies, 7:2 (1995) 191-222. Quotation from p. 196.

(2) Ibid.

(3) Reproduced in facsimile, Menston: Scolar Press, 1968.

(4) Second edition, ed John B. Saunders. London: Butterworths, 1969.

(5) The terms appear in a large chart which organises language in 'Genus's [sic], Differences, Species', p 433.

(6) The system makes use of geometrical symbols which, because they bear no resemblance to the phonetic structure of words, may be applied to several languages once the kind and order of the word is established. Wilkins uses the system in action to present the Lord's Prayer and the Creed (421-34).

(7) Michel Foucault. Les Mots et Les Choses. Trans.: The Order of Things: An Archaeology of the Human Sciences New York: Vintage, 1994. It is interesting that the translated title not only rejects the complex relationship of the French original but imposes a specific chronological relationship and discipline on the subtitle.

(8) Treip, M. A. Milton's Punctuation. London: Methuen, 1970.

(9) The Two Cultures and the Scientific Revolution. Rede Lecture, 1959. Cambridge: Cambridge University Press, 1959.

(10) The Structure of Scientific Revolutions. Chicago: University of Chicago Press, 1962. The term is discussed and defined on pp 84-5.

(11) L'instance de la lettre dans l'inconscient ou la raison depuis Freud. Sorbonne, Paris, 9 May 1957, translated by Alan Sheridan in Jacques Lacan: Écrits. London and New York: Norton, 1977, 146-178.

For quotation purposes:
Stuart Sillars: Translation as a Process Record. In: TRANS. Internet-Zeitschrift für Kulturwissenschaften. No. 13/2002.

TRANS     Webmeister: Peter R. Horn     last change: 01.05.2002     INST