a discussion of the discrepancy between theorists and practitioners within the field of translation, and a tentative solution to the problem

Transcription

1 HANS-JØRGEN BIRKMOSE THE YAWNING GAP a discussion of the discrepancy between theorists and practitioners within the field of translation, and a tentative solution to the problem Aarhus School of Business 2006

2 Hans-Jørgen Birkmose (258486) Supervisor: Annemarie Backmann 180,448 characters (82,0 standard pages) Summer 2006

3 The Yawning Gap What can we reason, but from what we know? Alexander Pope

4 Contents 1. Introduction The Theorist/Practitionist Discrepancy The Proposal Methodology Methodological Considerations Theoretical Trends of Translation Introduction The Source Text Approach Jakobson and Nida Recent Initiatives Two Practitionist Views The Target Text Approach Functionalist Translation Descriptive Translation Studies The Translator Oriented Approach Resistance Lawrence Venuti Feminist Translation Post-Colonial Translation More Translator Oriented Approaches Hermeneutics Translation and Relevance Theory Summation Translating Edgar Allan Poe Introduction A History of the Translation Process The Pre-Translation Period The First Round of Translations The Second Round of Translations The Third Round of Translations 50

5 3.3 The Examples Linguistic Problems Cultural Problems Summation Discussion and Conclusion The Story So Far The Translator s Reality and the Theoretical Trends The Translator s Reality and the Theoretical Trends Resumé in Danish/Resume på dansk 102 References 104

6 1 1. Introduction 1.1 The Theorist/Practitionist Discrepancy During the last 40 years or so translation theory has come a long way towards establishing a framework that may facilitate and secure sustained high quality performance in all aspects of translation, from training programmes to professional practice. However, translation theory has yet to reach its ultimate audience, namely the practitioners of the field who seem to carry on business oblivious of or indifferent to the progress theorists make. Moreover, the practitioners can do so in good conscience since they show consistently good results all things considered and disregarding the cases of hilarious blunders as the mere coffee break conversation pieces they are. This discrepancy is well known and commonly recognized, but one has to look long and hard to find anyone suggesting a way to overcome it. To expect it to go away completely is unrealistic (and, who knows, such an outcome might even, if achieved, entail the loss of a vital dynamic principle). It is unrealistic because at the end of the day the realities of practitioners and theorists are incompatible: theorists strive to find and express lasting facts in abstract, formulaic terms while practitioners are engaged in a short-term, highly concrete event, ie. rendering a text in one language in another language. Still, this fact/event duality is present I dare say across the humanities, and regular outbursts of raillery aside theorists and practitioners usually in sober moments recognize each other as quite indispensable. One would imagine therefore that more work was put into effecting a similarly fruitful relationship within the field of translation. In Chesterman and Wagner (2002) two people, one from each camp, try to negotiate a reconciliation, and headway is made in so far as both combatants (and the reader) come away somewhat the wiser, but

7 2 apart from this instance more often than not the distance between practitioners and theorists is regretted or shrugged off as a simple fact of life. Defeatism being a rare trait among both translation practitioners and theorists I have come to wonder why it is so, and immodest as it may seem I intend with this text to present my reflections on the issue and propose a tentative solution to the problem. Some time ago I translated Edgar Allan Poe s short stories into Danish knowing next to nothing about translation, and when it became known I was faced with eager theorist enquiries about my skopos, strategy and so on. I met these questions in what I believe to be true practitionist fashion, that is to say with a great deal of suspicion. There may even have been both apprehension and disdain in my response; after all, I had done without their help, and were they now to challenge me, probe into my private affairs and (God forbid!) pass judgment on me? It did not help matters that even when I knew them to be sympathetic and honestly interested I could not answer their questions. At my most accommodating I mumbled something about a drawn-out trial-and-error search for the right words, but requests for more technical elaborations were met I fear with a blank stare. They then concluded that I was a natural, used my gut instinct or whatever, and I resented that as much as being asked in the first place because it gave (me) the impression that they saw me as working in some happy-golucky manner without any real clue as to what I was doing. That was of course not the case. It is evidently possible also in translation to make quality decisions consistently without an underlying, utterable rationale. The sheer volume of texts translated in this way and ranging from flawless to brilliant speaks for itself and cannot be argued away, ignored or dismissed as pure luck or extraordinary talent. Nevertheless, the frustrating feeling of being treated like a dilettante triggered questions in my head that induced me to look more closely at my own work process and to study translation theory systematically in order to know mine enemy. I am happy to say that I found most of my thoughts and experiences mirrored or described in the literature; that

8 3 certainly whetted my appetite, but the sense that as a practitioner I could easily do without all the theoretical back and forth never left me, a fact that induced me to put my thoughts to paper in Birkmose (2003). In the end it led me here. Having then at the time of writing both a theoretical and a practitionist outlook has made it a point of first importance for me to resolve the conflict between the two. Great motivation though they may be, personal issues remain questionable elements in theoretical contexts, but as suggested initially I soon discovered that the problem had a global aspect and was universal rather than personal. Otherwise we might well have ended here. As it was I was permitted to pursue the matter on grounds of common interest, and on the following pages I will outline my proposal for a solution to the problem and the method by which I expect to test the proposal against the knowledge generated by writers on translation over the last many years, granting I hope equal attention to practitionist pragmatism and theoretical detachment and, in the end, as little as possible to the personal preferences that set it all in motion. 1.2 The Proposal The title of this thesis is a quote from two separate sources, and it was chosen because the phrase was employed by two people seemingly as far apart as can be when working within the same field. Theorist Gideon Toury (1995: 2) uses the phrase in an attack on the supposedly more prescriptivist branch of translation theory, and in Chesterman and Wagner (2002: 1) practitioner Emma Wagner uses it to lament an in her eyes inordinately high level of theorist abstraction. Toury as we shall see later advocates descriptivism whereas Wagner (ibid: 4) bluntly asks the theorists to get off the high horse and provide some concrete advice and guidelines, even doctrines. Yet when one tries to isolate the points of view of practitioners and theorists in order to define the problem area(s) it soon becomes clear that there are conflicting opinions on either side. It has already been hinted above that

9 4 theorists are methodologically torn between prescriptivism and descriptivism, but it will also become apparent in the course of this text that there are practitioners who contrary to Wagner s appeal find theorist activity of the more prescriptive variety stifling. In fact Wagner herself in the long run modifies her point and states (ibid: 67, my emphasis) that anything practical, concrete and non-coercive would be welcome. Thus there is a certain incongruity between the requirements some theorists deem imperative for a good translation and the unquestionable quality of translations nonetheless produced in negligence or ignorance of these same requirements. At the same time the reverse conflict can be observed: some practitioners express a desire for more immediately applicable answers from theorist circles who on their side refuse to be drawn into discussions that are in their view fundamentally about right and wrong, thus inherently moralistic and, it follows, theoretically invalidating. When it is the case that various fractions of the practitionist community demand more or less prescriptive theory respectively, and when it is the case too that various fractions of the theorist community do in fact supply more or less prescriptive theory respectively it seems clear to me that the reigning confusion (or its physical representations, the evidence of a theorist/practitionist discrepancy) owes to a logical flaw at the outset, at the very foundation of the craft. The flaw cannot exist but at the foundation; were it an error at any other point its consequences would have been local, confined to a specific area rather than having the global ramifications it factually does. I see theories ultimately as complexes of facts where, ideally, one layer of facts after the other can be peeled off until the basic layer of the theory in question is discovered and laid bare. Once at that point the process can be inverted and layers of facts can be placed one on top of the other until, ideally, the complex of facts is restored.

10 5 Similarly I see any practical event as a complex of activities where, ideally, it is possible to retrace the main steps of the process and follow them in either direction. As an event translation may therefore also be likened with a series of layers that can be reduced or reconstructed at leisure. Given these many parallels it seems reasonable to suppose that if theory and practice are ever to see eye to eye they must have at least one layer of perfect correspondence. There has to be a spot of common ground acceptable to both. The occurrence of higher level differences is of far less (global) significance since inevitably both theory and practice will diversify increasingly the further away from basis they get, making it comparatively easier to isolate and analyze any problems in as it were their local environment. All theoretical trends agree that a translation is a target text derived by a translator from a source text. Ample evidence will later on show that the importance, role and status of each of the three constituents is a matter of dispute; in fact the trends are split three ways on the question, each taking one of the three elements as their basic layer, but that has no bearing on the case at this juncture. We need to go to extremes to find writers on translation who do not see translation as a source text translator-target text relation. What about the practitioner s reality? I agree that translation is a source text translator-target text relation, but when I ask myself to describe my process that is not the first thing on my mind. I write. I produce text. I translate, true, but surely in any logical sequence generics come before specifics or in other words writing (text production) is the superordinate, translating the subordinate. Put plainly, in the process I am primarily concerned about the result of my efforts, about the quality, contents and progress of my own work; the independence of the text produced may be arguable (the question will be taken up later), but sitting at the computer, in the at-screen situation, my principal understanding is that I am producing a text in its own right, one that never was before and for

11 6 which I alone am responsible. I say therefore that the translator s reality is the writing process. The text will show that this is no out-of-the-way point of view. The time has come then to formulate my proposal: I put it that the practitioners and the theoretical trends within the field of translation differ fundamentally in their perception of translation because all theoretical trends work and study on the basis of the source text translator target text relation whereas practitioners perceive translation as the production of text, peeling off thereby an extra layer; put differently practitioners reduce the matter beyond the point the theoretical trends have defined as ne plus ultra. Hence the divide. With this text I will seek to render probable that this divide can be overcome by taking the translator s reality as the starting point. I will supplant no theoretical knowledge in the process; my aim is to supplement it, and I shall proceed at all times with a view to ensure both that the validity of current and established theoretical knowledge remains unchallenged by this concept and that the practitioners experiences are duly respected and considered. The process by which I hope to achieve this will be outlined in 1.3, but there are a number of points I need to make clear first in support of my core concept. As I see it all writing originates in one of three scenarios: Mode 1: The text producer writes on the basis of an idea of his or her own (a novel, a newspaper article, a thesis ) Mode 2: The text producer writes on the basis of a second party s more or less well-defined guidelines (ghost writing, speech writing, manuals ) Mode 3: The text producer writes on the basis of an entire text, typically converting it from or into a foreign language (translation) I define writing as the text producer s release of an interpretation (be it of physical, intellectual or emotional origin) of a subject matter that the text producer for whatever reason needs to concretize in words.

12 7 The definition contains four elements, the text producer, the subject matter, the interpretation and the release, all of which are integral parts of the writing process. A few words on each of these are necessary. I regard the text producer as a subject, that is as constituted by body, character, feelings, mind, thoughts, emotional responses and so on. This definition is intended to encompass every conceivable personal circumstance, and in all its lack of discipline it will suffice for the argument here. Since for natural reasons the text producer can never completely shed nor surpass what constitutes him or her as a subject all writing is conditioned by the text producer s subjectivity. By subjectivity I mean each subject s unique set of traits. Thus the word subjectivity (or its derivatives) cannot be understood in the sense of personal bias; bias is a potential trait of the text producer s subjectivity and may as such occur textually as instrument or error, but that is a different matter. I see the subject matter as composed of two elements: a logical universe and the models of reality we apply to it. The logical universe comprises the subject matter s properties as they are regardless of the way we happen to perceive them: the sun is hot and immobile whether we see it as a deity or a scientifically describable celestial body, and the earth retains the same set of features whether we see it as flat, round or ellipsoid. I am aware that also concepts like warmth and immobility derive from a human propensity to measure everything by our own scrawny yardstick, but rather than let myself be dragged into quicksand discussions about life, death and the meaning of it all I will lowering the level of abstraction just a trifle - say that this propensity is precisely why our models of reality must be considered part of a text s subject matter. A model of reality usually comes with an expected discourse that will direct the text producer towards some text forms sooner than others: insisting that the

13 8 earth is ellipsoid rather than golf ball spherical virtually demands a marked precision also in the rest of the discourse. There being usually more than one model of reality available the form of the text is equally negotiable. In the nature of things the logical universe, once known, is less flexible and only allows the text producer to arrange its properties as parts of the text s contents. The text producer s decisions about the text s form and contents being based on these two elements the subject matter must, if only as an idea, be known before actual text production can be said to take place: the text producer is always writing about something, and that preposition is at the same time an indication that once the subject matter is identified, no matter its nature, the text producer begins to interpret it, that is to turn it into text. The interpretation is the text. I call the text so to stress that all texts are interpretations, but being no particular friend of abstract terminology I shall as far as possible prefer the word text in that sense from now on. That, conversely, texts are interpretations is evident in every aspect of them: the choice of words, the order of priorities as well as the organization of the material and the various constituents are all marked by the text producer s subjective intervention. The subject contingency of any text makes it possible to describe the level of subjective intervention in a text as subjective (conceived in a personal vein or manner) or objective (containing less or no expressly personal traits). This is common procedure in genre analysis. There is of course any number of shades between the two points, but their common root, the text producing subject, is evident in the fact that at the farther ends they tend to approach one another. I shall illustrate this graphically later, cf. Fig 1 (p. 12), but to give an example: a work of art is the definitive subjective expression, but it is a trivial fact that for a piece of literature to function as such the author must refine his or her subjectivity to a point approaching objectivity, and it is therefore not wholly unreasonable to talk about literature as objectivized subjectivity. If anyone wants an example of subjectivized objectivity they may look to Jacques Derrida. Still, this is

14 9 argumentative and better left for later; I shall instead turn to the release, to the at-screen process itself, where the textual refinement of the text producer s subjectivity takes place. The release is the axis of rotation in my definition of writing. The word means the final letting-go of the text (which must be the objective of any determined writing process), but it also signifies the work in progress towards that objective, a double meaning that mirrors that of interpretation which likewise denotes both a process and its end. Since these concepts are two sides of the same coin in this context the terms are aptly chosen. The criterion for the successful execution of any text is that it is released from the text producer s subjectivity (his or her private sphere, if you will) and in its final form selfexplanatory, autonomous. Otherwise the text producer will not be able to understand his or her own text after a while without relying on memory or resorting to guesswork. To bring about the release the text producer creates from the beginning a distance between his or her private sphere and the emerging text. I trust this applies generally for I notice that text producers always distinguish between my text and me, preferring the former and doing their utmost to remove or at least touch up the latter. The conflict is resolved only in the end, and I may be anticipating events here, but this could be the first of the many dichotomies with which we are so richly blessed within translation theory. Be that as it may. In spite of all, text producers do not have free rein. Text production is governed by two absolute rules: any text must have form, and any text must have contents. Text producers occasionally foster ideas that defy belief; that is their prerogative, but even the most extravagant approaches succumb to the two absolutes. Form and contents both depending on the subject matter it follows that a text producer does not enjoy perfect freedom; it would be more correct to say that s/he has a certain latitude as regards the subject matter s models of reality and a certain latitude as regards its logical universe.

15 10 It will be a key point of this thesis to define these latitudes. I shall describe the latitude as regards the models of reality in terms of the text producer s level of subjective intervention, and I shall describe the latitude as regards the subject matter in terms of the text producer s possible treatment of its logical universe. By defining the latitudes of both Mode 1 and 3 writers I hope to show that translation is text production and that its conditions are such that independent studies of translation are not only justified but indeed necessary. I will finish my remarks on text production in general by conceding that granting the word subjective two different functions may be terminologically unfortunate. Preserving the terms subjective and objective to distinguish between markedly personal and impersonal texts is nevertheless of some importance to me, and adding the word textual to them when ambiguity lurks in the shadows should prevent any confusion. The listing of the writing modes above is accidental, not hierarchical. The three text production modes have a number of properties in common: the planning stages and research phases have many identical traits, they all share the atscreen process, they overlap with regard to textual strategies and so on. It is also characteristic of all of them that neither necessitates much previous grounding. None of them requires specialist competence, only an ordinary understanding of the most basic mechanisms, and given a minimum of school training it is perfectly possible to produce text in any of the three modes spontaneously. Each mode has unique attributes as well. The difference between Modes 1 and is that Modes 2 and 3 as distinct from Mode 1 are based on objects in the outside world (eg. guidelines or the source text) whereas Mode 1 is bade on an idea, and the difference between Modes 2 and 3 is that in Mode 3 the object can only be a source text whereas in Mode 2 it can take a multitude of forms. I will not discuss Mode 2 in this text other than to illustrate the connection between the three modes. This starkly schematic outline will enjoy the benefits of elaboration later, but the differences between them should be clear enough

16 11 even now to justify the study of them also as separate entities. This is where translation theory comes into play as an independent field of study. It is customary even among scholars to make a distinction between literary and databased translation (legal texts and the like) and treat the two as if they were worlds apart. This problem is however not one of translation, and so I disregard it; having tried both literary and databased translation I find the two to be of one piece, and unless concrete examples call for it I shall make no differentiation between them forthwith: what goes for one goes as a rule for all. I make this assertion in no uncertain terms and consequently owe an explanation. I will present it while discussing one of the most hotly debated features of translation: the translator s level of visibility. Mode 3 is defined by the presence of a source text. Of the four elements that constitute writing (p. 7) the source text can only be the subject matter. So, provided my analysis of the writing modes etc. is correct the source text is the translator s subject matter and the translation a released interpretation thereof. In Mode 3 writing the level of textual subjectivity is given in advance not only by the genre but most likely also by attitudes and ideational values formulated already in the source text. With much determined in this regard the Mode 3 writer s subjectivity is expressed primarily in his or her level of textual visibility. In Fig 1 below I try to visualize this idea and demonstrate that the translator s conventionally acceptable level of textual visibility is largely identical with typically accepted levels of textual subjectivity in Mode 1 writing:

17 12 Literature Personal letters, diaries Sales promotion material Subjectivity Visibility Objectivity Invisibility Philosophy Legal texts Manuals Fig 1 The genres added to the circle are meant strictly for illustration of course and make no claim to being an exhaustive list, but they do suggest I believe a) that seen in this light literary and databased translation are graded notions of the same concept and b) that with the level of textual visibility the Mode 3 writer chooses a level of textual subjectivity the same way a Mode 1 writer does. The introduction of the translator s level of textual visibility as an expression of the translator s subjectivity establishes the translator as the writer of a text in its own right. As a result a Mode 3 text is as unique as Mode 1 and Mode 2 texts, and so the general description of text production made earlier eg. the absolute rules (p. 9) - applies to translation also, allowing the translator as much (and as little) latitude as regards the subject matter as other text producers. As much and as little latitude however is not necessarily the same latitude, but I will as promised (p. 10) return to this in due course. What happens in the translator s or any other text producer s black box during the writing process is anybody s guess. To my knowledge no one has as yet fathomed the wonder of it. Text production can be described through its origins and outcome, but the intricacies of the intervening period of creation have on occasion driven the most matter-of-fact of men to suggest divine inspiration for want of more pedestrian explanations. I have coined the phrase informed intuition to if not answer then cover this question. I find the idea of the writing process as an encounter between the (in Mode 3 cases) translator and the translation very appealing, and to me informed intuition illustrates well enough the unverbalized and brittle yet inexplicably certain knowledge one gains of a

18 13 second party over time, an experience that as I will show resembles to a T my relationship with a translation. So, if only for the sake of argument I will adopt this point of view, stressing thereby the unconditional centrality of the translator s writing process in the translation chain of events. Undoubtedly all these issues so scetchily presented raise many questions. I shall elaborate on them later in the text and see if I can bring what is at the moment no more than a practitioner s private musings into alignment with prevailing theoretical trends. 1.3 Methodology The objective is to introduce the translator s reality as the first layer of translation theory. As should be clear by now my suggestions are strictly speculative and offered as a proposed way out of an apparent deadlock. As a result the validation of my proposal rests mainly on its capacity for adaptation: can it with all its qualifications co-exist with the theoretical trends without overthrowing or harming accumulated theoretical knowledge? My proposal and its qualifications must in other words be consistent with all the various trends, ie. there can be no contradictions between my proposal and the trends. To that end I will in Part 2 map the theoretical area of translation. As far as I know chartings of this kind have been few 1 so I will do it myself. I will show that the theoretical area is composed of three pillars, in that all the trends take their point of departure in either the source text, the translator or the target text, and I will introduce the different perceptions of translation against which my proposal and its qualifications may be measured. This is not the place for a critical assessment of the trends, and none will be made; they are to stand intact at the end of this text so I shall take them one and all at face value. 1 Mary Snell-Hornby s integrated approach springs to mind, but it is in my view weakened by her design to bring the various trends together, an implicitly moral aim. To all appearances it has never met with much praise anyway. Nord (1997) pays it little or no attention, Toury s (1995) comments are dismissive, and out of three brilliant sensors of current trends, Baker (1998), Bassnett (1980/2002) and Munday (2001), only Munday makes any reference to it

19 14 First I will discuss the source text oriented approach to translation. Besides a summary of the beginnings of modern translation theory represented by Roman Jakobson and Eugene Nida this section will contain an account of more recent developments of the approach and introduce Umberto Eco and Giovanni Pontiero as practitioners who both promote close source text renderings. Next I will present target oriented translation theory. Reiß and Vermeer s Skopos theory will be considered. Christiane Nord s further work on the concept will also be discussed, and in this section of the text I include Gideon Toury s ideas as well along with a word or two on James S. Holmes. Thirdly I will give examples of translators that see translation as a vehicle for expressing or promoting their own attitudes and ideologies. I will concentrate on Lawrence Venuti, on post-colonial and on feminist translators. They all share a strong belief that translators have an activistic role to play in influencing sociocultural values. I will also give an account of George Steiner s theory-sceptical stance which will be followed by a look at Radegundis Stolze s more traditionally elaborated hermeneutical approach. Lastly I shall touch upon Ernst-August Gutt and his efforts to see translation in a relevance theoretical light. The translator oriented trends are divided into two sections for purely practical reasons. Owing to the range of trends included time forces me to make certain restrictions both concerning the trends and the test of my proposal; I will focus on the basic ideas of the various trends and elaborate on them only to make sure that the rationale behind them is clear; furthermore I will make a cross-section of the trends and see if I can discern a few generally recurring issues that may help structure the discussion later. Finally, to ensure that details and main points are treated each according to their significance the ensuing test will be focused on the translator, that being the constituent of translation closest to the writing

20 15 process. These restrictions will not be permitted to efface the distinctive traits of the various trends, nor do I intend to neglect any possible contradiction. In Part 3 I will describe my work on the translation of Edgar Allan Poe s short stories. Clearly it will not be an exhaustive account of all aspects of the process, but I will outline the stages of it and give examples of the practical problems. The ambition is to give as diverse an impression of the work as possible, from the tedious and trivial to the intricate and exploratory, and the material will serve to elaborate on my proposal in general and the concept of the translator s reality in particular. The very personal tone of Part 3 is by no means accidental; it may be seen as representing the practitioner s voice as much as Part 2 observes theoretical detachment. Parts 1 and 4 then are supposed to strike the balance. Part 4 will contain and conclude on a discussion of all the aspects presented in Parts 1, 2 and Methodological Considerations Translators have for centuries handed down their experiences and insights for the benefit of later generations, typically in essayistic form and rarely, if ever, claiming scientific status. All the same these texts remain favoured and oftenquoted references likely to be used for tipping the proverbial scale on any question related to the area. Many are the times when eg. Cicero, St. Jerome or Dryden has been sent into the field to carry home a point. The essayistic form is thus a traditional and accepted way of sharing knowledge within this field. On top of that, it is not uncommon to label such writings theories. Under this heading Schulte and Biguenet (1992) collect a number of contributions by writers discoursing about the nature of translation. Venuti (2000) is another example. I mean no criticism of the editors by making this point, but while the observations and arguments of these texts are invariably interesting and illuminating they hardly qualify as theories unless of course we accept a very broad definition of the term. Chesterman (1997: 117) argues that a translation in itself is a theory and seen from a particular angle it may well be so, but in what fol-

21 16 lows I shall use the word in the narrower meaning of a reasonably comprehensive, coherent and cohesive set of preferably supported, but certainly falsifiable assumptions. As some of the approaches in Part 2 do not qualify as theories by that definition I shall make a distinction between theorists and theoretical writers, and I shall when I refer to the four perspectives under one heading prefer the term theoretical trends. I ask of no one that they call my remarks here a theory. Quite the contrary, with any luck they will add up to a hypothesis. Nonetheless, personal accounts are still in use and regarded as a valid approach. One variant of the method, introspective assessment, is employed in various guises under controlled conditions (cf. Chesterman 1997: 136), and another can be found in Venuti (1995) and Eco (2001 and 2003) that all three are based partly on personal experience. What distinguishes the latter method from the classic, deeply personal method is that Eco and Venuti put their findings into a global perspective and balance their own observations against the general body of insights existing in the world. It is along this line I intend to proceed; admittedly, the method is not entirely above criticism. The risk is not lost on me that I may more or less inadvertently create an environment that suits my purposes, but the levels of manipulation and personal bias should be checked by my expressed desire to weave the proposal into the general pattern, and in keeping with my stated objective that will needless to say be a primary concern in my attempt to bridge the yawning gap. A final word. I used part of the material in Part 3 also in Birkmose (2003). Comparing that paper with this text will reveal factual discrepancies, and I beg that in such cases this text be considered the authoritative version.

22 17 2. Theoretical Trends of Translation 2.1 Introduction In this part of the text I will introduce the major theoretical trends of the last three or four decades. The objective is to introduce their basic perceptions of translation and thus lay the foundations for the discussion later of my proposal and its qualifications. Note that, unless expressly stated, the definitions stipulated in Part 1 do not apply in Part The Source Text Approach Source text oriented theorists give the source text first priority; they determine its form and contents chiefly through analyses of words, grammar, syntax and text structure and try subsequently to match it as closely as possible in the target language. As it is generally agreed that the very nature of language prohibits any second rendition of a text from being more than an approximation of the original, equivalence is a key concern Jakobson and Nida Jakobson (1959) discusses (in Venuti 2000) the possibility of absolute equivalence between any two linguistic signs. Based on Peircean semiotics, according to which the meaning of a sign is always translated into other signs in an act of interpretation, Jakobson makes (ibid: 114) a tripartition of translation as intralingual (an interpretation in the same language), interlingual (an interpretation in another language) and intersemiotic (an interpretation in another sign mode), concluding (loc. cit.) that neither type can produce perfect equivalents. His answer to this dilemma is what he (loc.cit.) calls equivalence in difference, and he sees (ibid: 115) the practical and theoretical resolution of the problem in the elaboration of differential bilingual dictionaries and grammars that compare and define the differencies and similarities of the two languages in question.

23 18 Nida (1964) also treats the issue of equivalence in terms of difference. Differences between source and target texts are to him inevitable because the translator frequently has to meet several, perhaps contradictory, requirements at once. The differences he says (ibid: 156) can be explained in three ways: (1) the nature of the message, (2) the purpose or purposes of the author and, by proxy, of the translator, and (3) the type of audience Explicable as they are, in Nida s view the differences remain evils and are permissible only to restore equivalence so in an attempt to define the borders of acceptable behaviour in this respect he formulates (ibid: 159) the dichotomy of formal and dynamic equivalence, with formal equivalence in its purest form being gloss translation (with maximum focus on the message) and dynamic equivalence being an attempt to establish (loc.cit.) complete naturalness of style, meaning that target audience reception is given priority. The dichotomy is not a choice of either-or. Nida speaks (ibid: 160) of grades between the two covering the spectrum of various acceptable standards of literary translating. One cannot overestimate the importance of Nida s ideas. Not everyone appreciates the school of theory he has come to personify, but by accenting target culture reception and the role of the translator he has contributed to a more nuanced understanding of the translation complex Recent initiatives Baker (1992) appreciates that translations can be made without formal training but makes a case (ibid: 2) for the acquisition of theoretical knowledge: (a) ( ) it prepares the student for dealing with the unpredictable, (b) it gives ( ) a certain degree of confidence which comes from knowing that his/her decisions are calculated on the basis of concrete knowledge rather than hunches or intuition, and (c) provides the basis on which further developments in the field may be achieved The words might have been written by Nida and indicate that the attitude towards the translator s role is unchanged, but together with Hatim and Mason (1990 and 1997) Baker (1992) also illustrates the direction linguistically based translation theory has taken since Nida.

24 19 Baker (1992: 4-5) concentrates on modern linguistics which no longer restricts itself to the study of language per se but embraces such sub-disciplines as textlinguistics ( ) and pragmatics, but she recognizes the importance of the sociocultural trends in current translation theory, cf. 2.3, and acknowledges (ibid: 4) that translation theory needs to draw on the findings and theories of other related disciplines in order to develop and formalize its own methods. With their emphasis on the translator as a mediator between cultures Hatim and Mason (1990: 13) highlight some of the issues Baker avoids for the sake of delimitation: the status of the source text as a social product, its intended readership, the socio-economic circumstances of its production, translation and reception by TL readers are all relevant factors in the study of the translation process The authors seek to encompass all aspects of the translator s task, and they absorb a multitude of translational trends in their world view. In Hatim and Mason (1997) they discuss and agree with the attitudes of Lawrence Venuti, cf , adopt the concept of Skopos, cf , and employ Toury s theoretically founded notion of the translation as a text in its own right, cf Even so, the point of departure has not changed: the cornerstone of the book is a model for text analysis, the terminology and conceptual tools are linguistic and the ideal translation continues to be what Nida (1964: 166) defined as the closest natural equivalent to the source-language message. As mentioned, not everyone adheres to this objective. Some even find it damaging, but before introducing these other perspectives it may be of use in the long term to discuss the views of two prominent practitioners, Umberto Eco and Giovanni Pontiero, who both in each their distinctive way lend a voice to the source text approach and at the same time touch upon the practitioner s view on translation and translation theory.

25 Two Practionist Views Eco and Pontiero Pontiero (1991) opens a lecture on literary translation apologetically (in Orero and Sager 1997: 17) by confessing his comparative unfamiliarity with theory, adding that he is often puzzled by slick, sophisticated theorising and clever jargon. His conclusion (ibid: 32) is no less straightforward: there is no need for a single detailed theory of translation, as there is no single theory of writing. Major distinctions can be made between literary, scripture, legal, etc. translation, but even in the field of literary translation, there is scope for as many theories of translation as there are objectives of writing Instead, in his estimation (ibid: 21) the translator needs a good knowledge of the source language and an excellent knowledge of the target language. Given a talent for writing and a firm grasp of two languages anyone can (ibid: 26) in principle translate. Other than that, the translator (loc.cit.) must be a reader; the ideal reader who reads closely, is vigilant and perceptive. As regards preparations for a concrete translation task Pontieri s (1991) observations (in Orero and Sager 1997: 22) are quite plain. The groundwork is done by careful and repeated reading of the source text which will provide information about plot, themes, style, lexicon, syntax etc. The aids (loc. cit.) used by the translator are all manner of dictionaries and linguistic reference works, not only bilingual works, but dictionaries of proverbs, idioms, usage in general of both languages Pontieri becomes evasive when next he has to describe the writing phase itself. The decision making process (loc.cit.) is described as an elusive process because it is unconscious and instinctive rather that [sic] conscious or logical. ( ) I make myself consider every jot and tittle of the original, and thereafter I find my perception and instinct to be of paramount importance Being a semiotician Eco is far more willing to go out on a limb in this respect. Preoccupied with word (sign) meaning and interpretation as he is Eco (2001) soon rejects, in keeping with general opinion, the idea of equivalent meanings;

26 21 he uses the standard objection that such an event would require a tertium comparationis which is not at our disposal. This leads him to a discussion of interpretation. Eco construes (ibid: 68) Jakobson s contention, cf , that all translations are interpretations as if Jakobson meant to say (loc.cit., author s emphasis) that translation is a species of the genus interpretation. He does so in the form of a rhetorical question and readily admits that others (he is referring to the hermeneutics, cf ) view the matter differently. His reading is important because he uses it (ibid: 73) to keep translation in check, so to speak: the universe of interpretations is vaster than that of translation proper. While he grants the translator room to move, cf. below, he is openly worried about the dangers of freedom out of bounds. So much so that he stresses (2003: 125) the difference between interpretation and translation yet again, and this time he names the adversary: Gadamer states that every translator is an interpreter, and I agree, but this does not mean that every interpreter is a translator In Eco s view the translator (ibid: 73) is entitled to change the literal meaning of a text at a) sentence level in order to preserve the sense and at b) text level, provided no key elements are altered in the process. Finally he concedes (ibid: 82) that the translator must take into account rules that are not strictly linguistic but, broadly speaking, cultural. Equipped with these fairly elastic guidelines translators are sent on their way. As far as the finer points of translation theory are concerned Eco (ibid: 182) shows complete faith in the practitioner s sound judgment: Semiotics, philosophy and cultural anthropology can discuss such discrepancies for years but a translator has continually to face them, here and now and every day. In doing so, translators avoid ontological problems ( ): they simply compare languages and negotiate solutions that do not offend common sense ( )

27 22 Refreshingly down to earth as this may sound its sheer simplicity will strike many as theoretically untenable, and even Eco would probably concur that it is. 2.3 The Target Text Approach Target text oriented theorists view the target text as an original product largely independent of the source text (Reiß and Vermeer et al., cf ) and as a study object in its own right (Toury, cf ). A crucial aspect is the perception of translation as a sociocultural more than a linguistic activity Functionalist Translation Theory In the 1970s and 1980s a number of German theorists (Nord 1997: 1-9), Katharina Reiß, Hans-J. Vermeer, Justa Holz-Mänttäri, Christiane Nord and others, looked for ways to establish a general theory of translation that dealt with all and not just the linguistic aspects of the process. They regarded the parameter of equivalence pivotal to source text oriented theory as arbitrary and felt that it chained the translator to the source text and obscured all other problems. In a radical move they reduced all matters related with the source text to factors of potential importance and proposed instead to make the function of the translation the defining question. Central in this development is Reiß and Vermeer s (1984/1991) notion of Skopos, of the purpose of the target text as the hub of all translation. With this Reiß and Vermeer believe to have found a concept sufficiently versatile to cover most instances of translation. Their reasoning is simple: human activities can be generally described as actions. All actions are carried out with a purpose, and all constituents of an

28 23 action are influenced by their given sociocultural embedding as well as the individual histories of producer and recipient, or, in short, by culture. Language is (ibid: 18) an element of culture (in a translational context, that is; language in itself is considered a different matter and of no consequence to this theory). The purpose of text production (ibid: 18) is communication: [man will] in Interaktion treten. A text is therefore seen as an offer of information ( ein Informationsangebot ) targeted at a recipient, and the purpose is achieved when the recipient perceives the offer of information as intended. It follows that for the communication to succeed the text producer must take the recipient s expectations and conditions into consideration. Translation is viewed (ibid: 95) as action of the second order because it is based on a primary action. For that reason translation is defined (ibid: 67) as an offer of information in the target culture about an offer of information in the source culture. The target text (for short) is the translator s interpretation of the source text; it is one of an infinity of possible interpretations and will be governed by the defined purpose of the translation. In speaking (ibid: 46) of interpretation Reiß and Vermeer refer to Jakobson, cf , and as they reject Gadamer (loc.cit.) their reading seems to concur with Eco s, cf The inherent uniqueness of any interpretation means that deviations from the source text are seen (ibid: 62) not in an evaluative light but as marks of a dynamic process in which a new (different) text has been written: Jede Rezeption realisiert nur Teile aller möglichen Verstehens- und Interpretationsweisen und neutralisiert und konnotiert jeweils andere Merkmale. Es handelt sich dabei nicht grundsätzlich um ein mehr o- der weniger (...), sondern um ein jeweils Anderes Reiß and Vermeer exclude homophonic translation from the Skopos theory. To them (ibid: 81) it is not information about information, and as a result it is beyond their definition of translation. This limitation makes it clear that when

29 24 they provide the translator with extensive authority they do so to assert the independence of the target text, not the translator. That distinction can be difficult to uphold, but as is also evident from the above quote the emphasis on target text functionality does not cut the ties to the source text. I shall return to this question later. A kindred spirit, Justa Holz-Mänttäri (1986), advocates too (in Snell-Hornby 1994: 348) the need for a more comprehensive theory: Ohne Theorie ist gar keine Aussage zu Translationsfragen möglich. Damit ist implizit festgestellt, daß die Reichweite der Theorie auch der Reichweite der methodologischen Aussagen über Translation festlegt Of further interest here is her study of the social structures of the translation process which Nord (1997: 13) singles out as her most important contribution to functionalist translation theory. With the target text producer (the translator) at the centre the act of translation is conditioned by a number of players: the initiator, the commissioner, the source text producer (perhaps), the target text user and the target text receiver. Holz-Mänttäri (1986: 354) stresses though that while the translator s interaction with these players influences his or her strategies the translator acts as an eigenständig und eigenverantwortlich handelnder Experte in einem Gefüge über-, neben- und untergeordneter Handlungen. While her charting of these activity structures was my principal reason for including her here a word or two on the source text are in order. Source text awe is considered a relic from the past, but almost in the same breath Holz-Mänttäri (1986: 355) explains that unconditional target text prevalence means weder Nichtachtung des Ausgangstexts noch Überbordwerfen anderer Ansätze. It means a new weighting of the various elements of the activity complex translation is. The objective is always the function of the target text, and rejecting (ibid: 362) all the text material of the source text is an option: