Notes on Translation as Research

Structured as a companion piece to "Translation as Research: A Manifesto", this essay argues that there are compelling reasons for translation to be one of the ways in which academics, notably in Modern Languages, spend their "research time". It speculates on the nature of possible misgivings or hesitations on this point; these seem to include the fear that embracing translation may undermine the discipline's commitment to reading in the original, and apprehension at the difficulty of assessing translations as research, especially where aesthetic criteria are involved. Rather than being drawn into debates over the definition of "research", however, the essay argues, we should start by committing to the broad intellectual and cultural value of translations, and recognize that, in the UK context, the REF already provides a framework in which translation can be embraced as a research practice, as a form of scholarship and/or creative writing.

Against this backdrop, it is understandable if published translations tend to be stigmatized (even if there are arguments for studying parallel texts, and texts in translation). The issue is not only that translations may deprive students of a chance to improve their language skills, or may become a means of cheating in translation exercises. More importantly, those who read in translation undo the link between "language" and "content", whose inextricability is a matter of principle for Modern Languages.
All these points have possible ramifications for the status of translation in the realm of Modern Languages research, which is the central concern of this article. There is no doubt that in some cases a particular researcher is, through a combination of linguistic expertise and research specialism, particularly well placed to translate a particular text. Moreover, no one doubts that translations are intellectually and culturally valuable. All readers, including linguists, rely on published translations to engage with texts written in languages they cannot read, and translations allow important texts (and less important texts) to travel widely and across time. Clearly, a translation by a researcher in French Studies, say, is not usually intended primarily for students of French, but this need not be considered a problem: the translation may be of value to students and researchers in other academic areas, as well as in the wider world; and, within the home discipline, the act of translating feeds into teaching and research in other ways -not only, as I have already suggested, through the teaching of translation, but also because translating a text will alter and enhance the researcher-teacher's own understanding of it. Yet despite all that is attractive about translation as a practice, and about translations, many academics in Modern Languages hesitate to spend any "research time" working on a translation. I am not unusual in finding this regrettable, and other people -some of whom appear in my footnotes -have already worked to shift academic culture in this area. I began with generalizations about attitudes towards translation and towards Modern Languages because one impediment to any such shift, it seems to me, may be a perceived incompatibility between, on the one hand, the foundational commitment of the field of Modern Languages, in both research and teaching, to reading in the original, and on the other hand a commitment to translation as a form of "research". It is especially with regard to that perceived incompatibility that I will try to make a distinctive argument in this article. But I recognize that there are other, institutional impediments, which may be the first thing we need to shift. In the UK, some of these are associated with "research assessment" and/or the managerial culture around it.
With that context in mind, this article is designed as a companion piece to, and is organized around, a document entitled "Translation as Research: A Manifesto", which is being published simultaneously, with endorsements from a range of relevant bodies. 1 What follows will reproduce, in italics, the five numbered points that make up the manifesto, and will offer commentary after each point; and, insofar as the manifesto as such is the primary document, it probably makes sense to read that first. (It is two pages long and is available at www.modernlanguagesopen.org/index.php/mlo/article/ view/80.) Many of the points I will be making along the way are not original; indeed, for the practical purposes of the manifesto, it is important that some of them already enjoy wide support. But I believe it is a good moment to restate some of the arguments, especially in the UK, because of the rhythm of research assessment.
For some years now the Research Assessment Exercise (RAE)/Research Excellence Framework (REF) has played a very important role in UK universities in the allocation of resources, the formulation of policy, the production of symbolic capital and more fundamentally in influencing the way that UK academics spend their time; and it is likely to continue to do so for the next few years at least. One cycle has just ended, and the next, leading up to REF2020, is starting to take shape. It is for these reasons that the manifesto pays close attention to the REF, and this article, while trying to extend the discussion, follows its lead. I would not want this focus to be taken to imply general enthusiasm for "research assessment", but I must also make it clear that the manifesto and the present article do not attempt to attribute blame to the RAE/REF for whatever anxieties currently exist around the academic status of translation (even in the UK, let alone anywhere else). Indeed, it is important to the manifesto that -as I shall reiterate, and as "managers" need to know -translations are already acceptable in the REF. The focus on the REF is, then, purely strategic, and is based on the belief that in this 1 In the past, apprehensions -and some misperceptions -about the status of translations within the RAE/REF have inhibited some individuals from working on translations, or from submitting translations for assessment. (An example is given in the article, page 5.) The idea behind the manifesto is that translators and prospective translators in UK universities can take encouragement from the manifesto, and, if necessary, draw it to the attention of academic managers who need to know, as the essay indicates, that the REF provides a framework and a vocabulary through which translation is welcomed as a form of "research". The authority of the subject associations and other relevant bodies that have endorsed the manifesto is crucial, not least because they represent areas from which future REF assessors of translations will be drawn. The manifesto has a few brief endnotes; in the present article these are not reproduced as such, but are incorporated into the article itself or its own notes.
I should add that I hope this contribution will help strengthen the manifesto's case, but it also has an exploratory aspect, and some of the claims made here are more contentious than those made in the manifesto. No one and no organization endorsing the manifesto should be assumed to be endorsing everything I say in this article. particular area, the REF's power to shape academic culture in the UK can be put to good use. Evidently, the fundamental issues under discussion here go far beyond the REF, beyond the UK, and beyond the traditional European homelands of Modern Languages to areas including Classics, non-European languages, Theology, Philosophy, History, and Comparative Literature. All such disciplines, as I will emphasize in the final section, have a particular relationship to texts in their original languages, and so a particular relationship to translation. * * *

Translation should be treated as research by academics in and beyond Modern
Languages, and by those who facilitate, monitor and assess their research. In the US the MLA now advocates this view, and it matters for appointments, promotions and tenure. In the UK, "research assessment" is an additional factor. In practice this is an argument about how we spend our time (and are allowed and encouraged to spend our time) as academic researchers who work in and through languages other than English and on the cultures, societies and histories associated with them.
In 2011 the Modern Language Association of America (MLA) published "Evaluating Translations as Scholarship: Guidelines for Peer Review". 2 This followed a 2006 MLA report recommending that universities expand the range of activities valorized for the purposes of tenure and promotion. The 2011 guidelines assert: "Whether they ['faculty members who are scholars and practitioners of translation'] translate literary or scholarly works or other cultural documents, they are engaging in an exacting practice, at once critical and creative, that demands lexical precision; detailed knowledge of historical, political, social, and literary contexts; and a nuanced sense of style in both the source language and the target language. [...] Every translation is an interpretation; each one begins with a critical reading, then expands and ultimately embodies that reading." 2. In crucial respects, this is not an argument about the meaning of the word "research". In the MLA guidelines on "Evaluating Translations as Scholarship" the term "research" is never used, but there can be no doubt that the fundamental intellectual and professional issues are the same in the UK and the US. Nor Some people may need convincing that perceptions of or preconceptions about the RAE/REF have acted as a brake on translation. Susan Bassnett, an eminent figure in Translation Studies, served for ten years as pro-vice-chancellor of the University of Warwick. This doubly advantageous position did not allow her to submit translations to the 2008 RAE when she wanted to. At the end of 2007, when submissions had been finalized, she wrote: "I have been advised not to list the work of which I am most proud, in particular my poetry translations, and so I will be submitting some of my duller academic publications for the panel to scrutinise, well aware that they will be taken more seriously as 'research'. I do this in the full knowledge that what I am colluding with is both wrongheaded and foolish" (Bassnett,18; see also Higgins). As these remarks implied, although it has been possible to submit translations (and some academics have always done so), there has been a fairly widespread belief that translations will not be taken as seriously as other "outputs". Beliefs of that sort, and the academic/managerial culture of which they are a part, can be slow to change, in the UK as in the US, but the REF can serve as a catalyst (and by some accounts has already started to do so). 3 Some might feel that the idea of "research" does not naturally encompass the work of translation -or numerous other activities and "outputs" for which the label is now used. They could find support for their view, and a wider critique of research assessment, in an essay entitled "Against Prodspeak: 'Research' in the Humanities", written by Stefan Collini at an earlier moment in the development of research assessment in the UK, when the official language was less familiar, and had been less deeply internalized. 4 But for the immediate, pragmatic purposes of the manifesto, arguments over the meaning of "research" are a red herring. As noted in the manifesto, the intellectual and professional issues around translation are fundamentally the same in the US as in the UK, and the MLA guidelines addressing those issues happen not to use the word "research". So on the one hand, the word is not really necessary, conceptually; and on the other hand, as I have already emphasized, in the British academic context the REF already offers us a broad enough sense of "research" to encompass translation. I would add that it is impossible to believe that the REF's capacious conception of research has arisen from any truly settled or consensual definition of the term "research". Rather, the established approach in the UK implies some prior recognition that, as noted earlier, the RAE/REF has affected and will affect quite deeply how UK academics spend their "research" time, and also that universities may wish to support a diverse range of activities in disciplines such as English, Drama, Dance, Art, Design and Music, including musical composition, performance and creative writing. In other words, the broadness of definition is a means of avoiding a situation where the research assessment process makes it financially untenable for departments of music to support musical composition or for departments of English to support creative writing.
In this way, the REF's capacious definition of research also implies some recognition of the close links, in certain areas of academic activity, between research, creative practice and teaching. If translation were more eagerly embraced as research, it would, as suggested in the introduction to this article, help reinforce those links in Modern Languages, where expert knowledge of foreign languages is a fundamental tool of research as well as a central element of teaching; and it would make it possible for universities to offer more steadfast support to academics practising translation, just as they support musicians/academics practising composition, and just as they support creative writers.

Translation is intellectually and culturally valuable. In the words of the MLA guidelines, "Translation has been an indispensable component of intellectual exchange and development throughout recorded history [...;] the translation of a work of literature or scholarship -indeed, of any major cultural document -can have a significant impact on the intellectual community, while the absence of translations impedes the circulation of ideas". The far-reaching impact of translations is one of the reasons that translation is a good use of academic time and resources.
As I implied at the start of this article, the Modern Languages community, in the interests of promoting the study of languages, and the practice of reading in the original, has perhaps tended to downplay the intellectual and cultural significance of translations. In a survey of current attitudes presented in the MLA's "Report of the MLA Task Force on Evaluating Scholarship for Tenure and Promotion", 30 per cent of responding departments of all sorts (all in the US) believed that promotion and tenure committees regarded translations as "not important"; within that average were the responses from Modern Language departments in the sorts of universities UK academics would call "research-led", where the figure was 47 per cent. 5 This figure, it should be emphasized, concerns perceptions of perceptions; it does not necessarily offer a reliable sense of how translations are actually viewed by promotion and tenure committees, and certainly does not tell us how respondents thought translation should be viewed. But it is a worrying figure for advocates of translation, not least because in the US as in the UK, perceptions, and misperceptions, influence practice.
Again, the issues go beyond Modern Language disciplines. But those disciplines have the biggest stake in this debate, given their concentration of academics with combined linguistic and cultural/intellectual expertise. It seems likely that it is principally from those disciplines that academic translators will be drawn in the coming years, along with the REF panellists, tenure committee members and others who will assess their translations. 6 For anyone familiar with the REF, the MLA's use of the word "impact" in describing the importance of translations is striking. Research in Modern Languages does not always convert into "impact" in the restricted sense codified in the REF, and if translation can help the discipline here that has to be a good thing. Moreover, as the MLA's as it were spontaneous use of the term suggests, the notion of impact need not be understood only in the REF's particular sense. The MLA's emphasis on translations' contribution to intellectual exchange and the circulation of ideas puts translation close to the heart of the collective work of specialists in "foreign" languages, in their capacity as teachers, curators, mediators, interpreters and analysts of the cultures in which they specialize. None of us today assumes that an educated person should have read, or even pretend to have read, all the canonical works of European literatures in their original languages. None of us assumes that if you have familiarity with a handful of European languages, you have access to all that is important in the realm of literature. The list of languages one might learn and texts one might wish to read in the original and study is, for all practical purposes, infinite. In this context, the curatorial, mediating aspect of the work of linguists becomes more important, not 5 Apparently it is not unknown for US academics, especially in Modern Languages, to remove translations from their CV, fearing it will damage their chances of tenure. See "Why Translation?" by Catherine Porter, who was MLA president when she wrote that piece. 6 When it comes to any lingering anxieties around the place of translation in the REF, perhaps it is the attitudes of academic managers that matter most. But the main issue for managers is what is REFable, on which point they must be strongly influenced by what is taken by a subject community to be REFable; the intellectual basis of a REFable practice, or the definition of "research", is not their concern. less. A fundamental part of that work, for both teachers and researchers, consists in helping certain texts and ideas to circulate and recirculate, and helping them live a life in our culture. Translations will always be part of this process, and academics in Modern Languages have an important role to play, as researchers, teachers and prospective translators, in making sure that future translators are trained, that worthwhile texts are translated and that the translations are good translations. For all that, the argument for translation as research cannot and should not be made exclusively in terms of "impact", dissemination or teaching; it must also be an argument about translation as a scholarly and creative practice. Undoubtedly, not all translations are equally deserving of credit as scholarship (or, to pre-empt point 4ii, as creative writing). Without being entirely flippant one could say this is a good thing from the perspective of the REF or appointment committees, which need to create distinctions and to rank. The crucial point, in any case, is surely that translations may be of the highest intellectual value. In his essay "The Translation of Philosophy", Jonathan Rée writes: "In the English language in the twentieth century […] many of the finest translations are philosophical, and many of the greatest works of philosophy are translations" (Rée,231. See also Apter, 248-49). Rée's argument is not that many of the greatest works have been read in translation, but that certain translations are themselves great works of philosophy.

4ii. Translation is a form of creative writing. This view commands wide assent among creative writers and in academic circles, especially in the field of Translation Studies -itself readily embraced by the REF. Translations can be inventive, original works in their own right.
The idea that translation is or can be a form of creative writing finds ample support in Translation Studies. In the article "Translation and the Lipogram", for example, Kate Briggs discusses Georges Perec's novel La Disparition, composed entirely without the letter "e" and subsequently translated into numerous languages. As Briggs argues, there are affinities between translation and lipogrammatic writing: both are forms of "writing under constraint", as is writing a sonnet, and the constraint can be creatively productive. 7 It would be odd and inconsistent if (as some of Susan Bassnett's colleagues apparently feared) one had a good chance of a top ranking in the REF for an article arguing, or working on the assumption, that translations could have as much intellectual and academic value as critical articles, or as (other examples of) creative writing, but had a smaller chance of a top ranking if one followed through on that argument or assumption, and submitted a translation to the REF.
The practices of eminent creative writers also lend support to the idea that translations can be a form of creative writing. All poets have drawn on the work of other poets; many have written "versions" of existing poems, reimagining them and recasting them in their own language; and many (for instance, famously, Paul Celan) have created important original works by translating poems. The boundaries between these forms of poetic activity are blurred; and in terms of their levels of creativity or significance, it would be hard to generalize about the distinctions between them. Christopher Logue's War Music, a sequence of poems based on Homer's Iliad, is an acclaimed example of a "version". Logue could not read Greek, and drew his inspiration from existing translations -which does nothing to undermine the value of War Music as poetry. As such, and as a "version", it would have been readily admissible to the REF, if Logue had been employed by a university to teach creative writing, and it could have achieved a top grade. All this is reasonable. It would be perverse if an equally creative version or translation based on a good knowledge of Greek were less welcome as a prospective submission. There is no good reason why poems formed through translation (as distinct, though not very distinct, from "versions", and from other original works) should be disadvantaged in the REF, and no reason to assume they will be.

Translations are no harder to assess than various other forms of research, scholarship and creative work. Appointment committees, REF subpanels and so on already deal with a very wide variety of practices and "outputs"; and it is recognized that any difficulties they may pose to assessment do not constitute a good reason to inhibit the activities in question. The same unequivocal recognition needs to be extended to translation. In academic culture (before and far beyond REF 2020) translation should be treated as a fully legitimate form of research.
Some people may feel that it is hard to assess translation as research. It is clear in principle, however, that mere, or sheer, difficulty of assessment should not become a reason to discourage the activity. Comparisons with other practices that fall under the REF's definition of research again help make the point: in the company of musical compositions, say, translations surely do not appear uniquely difficult to judge, or to judge as research. More generally, one might add, radical disagreements over the originality, significance and rigour of works by eminent thinkers, over the value of different creative works, and over the very problems of evaluating them, are not only commonplace but are constitutive of large areas of academic work.
Even so, there are disagreements and disagreements. Some fall within the boundaries of a certain intellectual consensus, others do not. Evidently, the boundaries are blurred; when it comes to choosing REF submissions, many will suspect that the range of possible responses to an article about a controversial figure such as Derrida will be wider (and more dependent on the tastes and beliefs of the assessors) than in the case of an article about some other thinker. For such reasons, submitting an article on Derrida may feel relatively risky. All the same, if one does decide to submit the article, one simply submits it. In the case of a translation, however, one may be expected to add some sort of statement, as part of the scholarly apparatus, and/or in an independent document, making it clear how one's work involves -and counts as -research. At this point, one may be dealing with a deeper doubt about the very eligibility of translation as research or scholarship or creative writing, prior to or alongside any anxieties about the difficulties of assessing it.
In practical terms this sort of supporting statement can certainly be reassuring and helpful for people on both sides of the assessment process. Comparable documentation may be offered in other academic areas in the REF, such as Art and Design, and the same sort of practice is endorsed by the MLA's guidelines, which encourage academic translators to explain to their assessors the nature and value of the work that has gone into their translations. This may include an account of "any special challenges posed by the form, style, or content of the source-language text, along with examples and explanations of the solutions adopted in the translation process", and "the importance of the source text as a work of literature or scholarship or as a cultural document, and its potential impact".
These suggestions are sensible, but they leave room for doubt around certain issues of principle. The supporting documentation has a paradoxical quality, if it is an essential extra (what Derrida called a "supplement"), and the paradox raises questions about the status of translation as such. I am reminded of a passage in L'Ordre du discours: every discipline, Foucault wrote (35), recognizes true and false propositions, but beyond its boundaries, in territory metaphorically marked "here be monsters", lies "une tératologie du savoir". 8 When documentation is demanded of translations it is a bit like a passport for monsters: they may be allowed in, but without their paperwork they have no right of entry. And the experience of crossing the border can be intimidating.
The analogy is stretched, but the gesture of providing documentation is liable to be double-edged if the goal is intellectual parity for the work of translation alongside other forms of research. For various reasons I am not sure I would want to argue (or not yet) that such documentation should be abandoned, for the limited purposes of the REF, or in the situations envisaged by the MLA. But in principle, if one takes seriously the idea that translations may not only require but embody scholarship, and may be a form of creative writing, there is something wrong, at least in certain cases, with looking outside the translation itself for evidence of its scholarship, creativity, or "research content".
The issue is most clear-cut, I would suggest, at the "creative" end of translation practices, and is perhaps best approached by considering the sense of unease that may be felt in providing a "supporting statement" for any piece of 8 "A l'intérieur de ses limites, chaque discipline reconnaît des propositions vraies et fausses ; mais elle repousse, de l'autre côté de ses marges, toute une tératologie du savoir. L'extérieur d'une science est plus et moins peuplé qu'on ne croit : bien sûr, il y a l'expérience immédiate, les thèmes imaginaires qui portent et reconduisent sans cesse des croyances sans mémoire ; mais peut-être n'y a-t-il pas d'erreurs au sens strict, car l'erreur ne peut surgir et être décidée qu'à l'intérieur d'une pratique définie ; en revanche, des monstres rôdent dont la forme change avec l'histoire du savoir. Bref, une proposition doit remplir de complexes et lourdes exigences pour pouvoir appartenir à l'ensemble d'une discipline ; avant de pouvoir être dite vraie ou fausse, elle doit être, comme dirait M. Canguilhem, « dans le vrai »". In English, "The Discourse on Language" appears as an appendix to The Archaeology of Knowledge, and the passage (223-24) reads: "Within its own limits, every discipline recognises true and false propositions, but it repulses a whole teratology of learning. The exterior of a science is both more, and less, populated than one might think: certainly, there is immediate experience, imaginary themes bearing on and continually accompanying immemorial beliefs; but perhaps there are no errors in the strict sense of the term, for error can only emerge and be identified within a well-defined process; there are monsters on the prowl, however, whose forms alter with the history of knowledge. In short, a proposition must fulfil some onerous and complex conditions before it can be admitted within a discipline; before it can be pronounced true or false it must be, as Monsieur Canguilhem might say, 'within the true'." creative writing. 9 Poetry is a privileged example, because it is widely viewed (for good reasons, I believe) as unparaphrasable, and irreducible to propositional content; perhaps especially verse, whose prosodic qualities may carry semantic implications, but also create "musical" effects of other orders (as does music itself). This understanding of poetry, and for that matter of music, implies, among other things, that to any poet, musician or academic critic, any paraphrase or summary of the "content" of a piece of music or poetry, including its "research content", is liable to seem inadequate or misplaced in some fundamental way. 10 Nevertheless, poetry is "REFable" (as "creative writing"); and in that context, its "research content" has to be assessed. In this case too, for the purposes of assessment, the definitions on offer seem sufficiently capacious. Poetry may be considered to embody "a process of investigation leading to new insights, effectively shared", to quote again from the REF's definition of research. And it can be valorized for its originality (which may include formal originality), significance and "rigour"; all these terms can be used meaningfully in this context. But this example also makes it clear that the terms mean slightly different things in different areas and different cases; and in the case of poetry, these criteria will surely, inevitably, be aesthetic to a significant degree, because the fundamental value of the work itself is, to a significant extent, non-propositional and aesthetic.
Approaching things in this way implies that assessors are (and should be) free to give the best marks to what they judge to be the best poetry. Not everyone will be comfortable with this idea, or with the idea that this is already happening in the REF. But some such notion of value -however vexed -is always in play, albeit in complex and sometimes hidden ways, in academic work based around literature, film, music and so on, and it cannot be filtered out of judgements of creative work in the REF. Any value judgement of poetry that did not involve judgement of its aesthetic value would surely feel perverse, particularly if it ever meant in practice that more importance was placed on the supporting apparatus than on the thing it was meant to support.
It is for such reasons too -that is, on the basis of certain understandings of poetry, literature, creative writing and aesthetic value -that poetry 9 By "supporting statement" I mean here the sort of document that may be submitted to a formal assessment panel, in the REF or for tenure. The issues around translators' prefaces and afterwords are somewhat different, and have their own history; see Balmer. 10 Such an understanding of poetry also helps explain how it engages its readers, and can be endlessly reinterpreted, and so why it has been valued by teachers, among other readers. On this point, and the association of literature with invention and originality (as well as singularity and alterity) see Attridge. in particular is often described as "untranslatable". Used in that way, the expression does not mean it is impossible to create a good translation, but that important elements of the text (which may include its prosodic qualities) will always alter in translation, in ways traditionally described in terms of losses and gains. 11 Yet these obstacles to translation also serve as incitements to translation (as with the many translations of Perec's La Disparition); and it is especially in areas where these obstacles are formidable that translations deserve to be treated as creative works in their own right. In this respect, in the REF or anywhere else, such translations, like other creative works, need to be judged by aesthetic criteria, to a significant degree. Translations' similarity to, or continuity with, other creative works, notably from the point of view of the reader, does not alter the fact that in most cases it is surely desirable, and instructively so, for formal assessment of a translation to bring into play the translation's relation to the source text. This is not just a linguistic matter but also concerns the nature of the source and its originality / significance / rigour. The MLA guidelines accept this, stating, as we have already seen, that supporting documentation should address issues not only around the translation as such but also around "the importance of the source text as a work of literature or scholarship or as a cultural document, and its potential impact". Again, not everyone will like the sound of this (even if, as I have already suggested, UK academics have reasons to rejoice at the prospect of "impact"). In various academic fields, even within the humanities, the value of an analysis is now assumed, and often for impeccable reasons, to be wholly distinct from the value of the object analysed. We would not, however, expect a REF or tenure committee to give credit for the translation of an instruction booklet for a washing machine, say. Nor is the crucial difference between academic and non-academic material: the translation into English of, say, a recent French-language article reporting a laboratory experiment is unlikely to be ranked positively in the REF by either the Modern Languages panel or the relevant scientific panel, however momentous 11 In a previous article (see Harrison) I have criticized David Damrosch's efforts to define World Literature as that which "gains in translation", and what I see as his misrepresentation of the association of literature in general with the "untranslatable". Insofar as "World Literature" as a field of study has tended to valorize reading in translation, it risks blending smoothly with other factors that have driven down enrolments in Modern Languages in schools and universities, and increased students' reliance on translations.
The emphasis of the current article is different: accepting the need sometimes to read in translation, and the benefits of doing so, and accepting too the great intellectual and aesthetic value of certain translations, need not mean abandoning the commitment to reading in the original where possible. On untranslatability, see also Apter, Gaunt and Syrotinski; and, for some brief but thought-provoking remarks on the pressures around reading in the original and in translation, see Bush.
that article academically or scientifically, and however demanding for the translator its language and its points of reference. These counter-examples lend support both to the point made earlier, that translation cannot be justified as research solely in terms of "impact", and to the argument that in some forms of research/writing, "research" is not separable from writing itself. That argument applies, I would suggest, not only to various forms of creative writing but also to quite wide areas of "research" in the humanities -by contrast with various other areas, including the experimental sciences. In the case of that hypothetical science article, research can be separated from writing: if the original write-up is adequately translated, it will make no significant difference to the reader whether she or he reads (or hears) about the experiment and its results in French, English or any other language. The situation is very different with a poem, and with the translation of a poem -or with a poem that is a translation, or a retranslation, or something akin to a translation.
Although poetry may be a privileged example here, it is not exceptional: it sits at one end of a spectrum of writing practices where we have strong reasons to care about and dwell on the writer's particular uses of a particular language. Novels sit close by on that spectrum. Biographies, although in some respects akin to novels, are placed differently: biography may also be a form of creative writing, and aesthetic criteria may not be out of place; but as a genre it has a different order of obligation to historical facts, and so a different relationship to other, conventional forms of research. By the same token it also has a different relationship to its readers, who may be reading it primarily for factual information, and for separable -and falsifiable -"research content" in that sense. To the spectrum of different genres, in other words, corresponds a spectrum of responses. As one moves away from poetry on the spectrum, supporting documentation (in the case of writing submitted for formal assessment) and other forms of explanation or summary may look less out of place, and may more naturally form part of the published work itself -references to archives in the footnotes of a biography, for instance. By the same token, in areas and genres further from poetry, aesthetic considerations may have diminishing relevance, or be wholly inappropriate.
This brings us back to my starting point, back to the wider issues around teaching and research in languages (and not just Modern Languages), and to my final, broad claim. If the arguments I have just made are valid, then it appears that the commitment to reading in the original and the commitment to translation as an academic practice belong together. The areas where translation can and should be regarded as "research" are those in which reading in the original brings meaningful benefits, and is -or should ideally be -an integral part of research methodology.