UDK 81'255.4:821.111.09-32 Joyce J. FROM DOUBLEVALENT TO MONOVALENT DISCOURSE: THE ROLE OF THE TRANSLATOR IN MEDIATING HETEROGLOSSIA AND HETEROPSIA IN A FICTIONAL NARRATIVE

Among the many unresolved issues in the field of translation studies is also the one pertaining to the question of who sees/speaks in the source and in the target text. To this effect, the paper explores, firstly, excerpts from James Joyce's Dubliners, in order to extrapolate the prevailing narrative strategies with respect to narrative perspective and their rendering in the Slovene translation. Secondly, the results obtained from-the analysTs of the selected segments 1r6m both the source and target texts-on the micro-structural level are compared with the effects that take place on the macro-structural level. The main objective, however, is to provide an adequate insight into those textual conditions that significantly govern the realisation of Bakhtin's notion of double-voiced discourse, particularly those which occur in a very limited stretch of language, sometimes even within a single phrase. The assumption is that such a narrative configuration is bound to present special difficulties for a translator of any text which includes a polyphony of voices and perspectives. Consequently, Bakhtin's postulation of heteroglossia is complemented with heteropsia, a medical term denoting unequal vision in the two eyes, which has been employed to signal a parallel perception of the object observed. Since these two rhetorical phenomena are most aptly realised by means of free indirect discourse, particular emphasis is placed on the variety of its function in a fictional narrative.


INTRODUCTION
Every translator of a fictional text, before embarking on the translation of their text, has to consider all the relevant discourse parameters, i.e. general linguistic and stylistic features as well as idiosyncratic peculiarities which make possible the realisation of the textual potentiality as to who sees and who speaks in the narrative.Our research, drawn on the theoretical and practical results of the contrastive analysis of the selected English prose texts and their corresponding Slovene translations, has revealed significant deviations especially on the axis narrator -narratee.This is largely due to the translators' inaccurate determination of the narrative mode(s) used, resulting in the displacement of the roles of the seer/speaker designated by the author of the original text.The greatest number of shifts in translation can be observed in those instances where the text either develops simultaneously on different narrative levels or where there is a comparatively weak signalling of shifting from one level to another, sometimes even within a single sentence or clause.Such narrative manipulation enables the author to introduce a variety of perspectives on the same issue and "juxtapose two sets of values, to imply a critique of the character's views without the direct judgement which an external perspective would produce" (Fowler 1989: 138).What ensues from the interplay of two or more different views might be called a kind of hybrid perspective, the realisation of which is left entirely to the reader.The case of bringing together the author's (objective) and the character's (subjective) perspective, which happens to be the most frequent situation produced by free indirect discourse (FID), gives rise to the emergence of the so-called double voice, within which one set of values, beliefs, etc. is involved in implicit dialogue with another (Fowler 1989: 140.)The concept of double voice seems to be a plausible suggestion as to who really speaks in FID, even though it significantly departs from the traditional notion, conceived already by Genette (1972), according to which the narrator is always the speaker, except in direct speech, where the speaking is performed by the characters.What Genette's theory fails to take into account is that, particularly in FID, the author attempts to imitate the speech of the character by using the kind of lexis, grammar, and other structural and stylistic peculiarities pertaining to the typical speech and emotive behaviour of that character, but presented in the auctorial past tense and third person singular (cf. Brinton 1980: 363).
The postulation of double-voiced discourse logically entails the existence of double:-viewed discourse (heteropsia).Consequently, two distinctive perspectives and their verbal manifestations can be emphasised to the point of presenting the reader with a set of values, beliefs, and world views, which are contrasting enough to motivate him/her to form an idiosyncratic opinion of the fictional world.Notwithstanding the seemingly even polarisation of the control of the speech/view activity between thenarrator and character in the case of FID, as suggested by Short and many other exponents (Leech and Short 1992: 318-351), there is reason to believe that the narrator, in spite of all, has a decisive advantage over the character in that s/he not only sees what the character sees, but s!he also sees the character himself.On this score, we tend to side with Leuven-Zwart's contention that "the narrator is always a focalizor, i.e. telling a story implies seeing the events, actions and characters which are its constituent parts ...Although it is not possible to tell a story without focalizing, it is possible to focalize without telling a story: a character may very well focalize without reporting what he sees" (Leuven-Zwart 1989: 176).Accordingly, the narrator's and the character's respective focalisation, when the latter's does not involve narrating, actually occur on separate levels, and should therefore be understood in hierarchical order.As this calls for a more differentiated and precise denomination of their functions, it seems appropriate to define focalisation as the process in which the point of view of the character is realised on the level of story.The term narrative perspective, however, ought to be reserved for that position on the level of discourse from which the narrator observes, comments on and qualifies the narrative.The main purpose of such delineation of perspective and focalisation is to provide some clarification, however arbitrary or even simplifying, of the perpetual issue concerning the use of FID as the most frequent means of implementing the double-voiced and double-viewed potential of literary expression.
The immense flexibility of FID, especially as far as voice and view are concerned, may indeed enable the author of the original text to freely manoeuvre in the space between the narrator and character.However, reasonable doubt may be raised as to whether any translator may find that a blessing in disguise.Rather and especially when it comes to a textual negotiation between so inherently different languages as English and Slovene, an occurrence of FID is bound to create an uproar in a target-language system.Fortunately, such helplessness will not happen inevitably and at all times, since many instances of FID are perfectly manageable in Slovene, which is probably true of any other language as well; the true concern aims at those precarious cases of the use of FID where, no matter how skilful or inventive the translator may be, the translated text will never hold water, owing to certain insurmountable systemic differences between the source and the target language.

GRAMMATICAL AND MODAL INDICATIONS OF FID
To begin with, excerpts from (1) to ( 4) have been borrowed from James Joyce's short story 'Eveline' from his collection of short stories Dubliners (D), to demonstrate some particularly hard-to-crack translation nuts (the emphases are mine, as in all the other excerpts): (1) She would not cry many tears at leaving the Stores.(D, 'Eveline',38) Slovene: Zato ker zapusca trgovino, ne bo prelila mnogo solz.(31) (2) People would treat her with respect then.(D, 'Eveline', 38) Slovene: Tedaj se bodo vedli z njo spostljivo.(31) (3) She would not be treated as her mother had been.All these illustrations demonstrate the use of would in the position of FID.The translation problem in these cases springs from the double function of would, that is from would as an auxiliary verb to form the future-in-the-past tense, and as a verb expressing conditional modality.The preservation of both functions in the target text is of vital importance here, since the suspense of the story actually draws on the main character's, Eveline's, dilemma whether or not she should elope with her boy-friend Frank to Buenos Aires.So, on the eve of her departure, "she weighs the pros and cons of this adventure as against the ties of her admittedly hard family life, looking after her violent father and two younger children after her mother died insane" (Fowler 1989: 139).For the reader of the original, the outcome of Eve line's decision is fairly uncertain and is resolved only at the very end of the story; the reader of the Slovene translation, however, in which all the would-structures have been transformed into the future tense, is put on the wrong track, as he is given clues throughout the story which clearly suggest that Eveline has more or less made up her mind -to take her chances.This vital discrepancy between the two texts must not be put down to the translator's incompetence or anything else, for that matter, since he has only played by the book of Slovene grammar.Admittedly, it would be possible to preserve the modality of the original also in translation, but not without the sacrifice of the grammatical future.Deliberating which of the two would-functions is more and which one less indispensable seems beside the point and already comes into the domain of literary hermeneutics.The only and sore point is that translation deficiencies are sometimes unavoidable, notwithstanding the congeniality of the translator.However, the gravity of the losses may vary, as the next case will prove.It is an excerpt from the close of the first chapter of Joyce's Portrait of the Artist as a Young Man (PAYM): (5) He was alone.He was happy and free: but he would not be anyway proud with Father Dolan.He would be very quiet and obedient: and he wished that he could do something kind for him to show him that he was not proud.(PAYM, 54) Slovene: Bil je sam.Srecen in svoboden.;vendar nikakor ni hotel kakor koli kazati ponosa pred patrom Dolanom.Hotel je biti zelo miren in poslusen: in zelel je, da bi mu mogel izkazati kako ljubeznivost in mu tako pokazati, da se ni prevzel.( 55) The expenditure of the future tense for the benefit of modality in translation (would not rendered iiifo Slovene as ald rioi want, and would be as wantea io be) is here not only a reasonable decision but also a justifiable one, because it is supported by the context.Having begun with the emphasis on the heteroglossia and heteropsia of FID, let us briefly consider what may be going on in this respect.In the case of 'Eveline', it appears that the main character's voice is recognisable throughout the story, but the indirect third-person report (as opposed to the direct report of stream of consciousness) prevents her point of view from becoming alienated from that of the author.Apparently, the two points of view must be running side by side, one (ofthe author) dominated by the would-structure denoting the future-in-the-past activity, the other (of the character) denoting, or rather, connoting modality of hesitation and indecision.
The other text, PAYM, effaces the auctorial control almost entirely by leaving the narrative stage to the character, Stephen Dedalus.An unreliable narrator, he speaks alone, by himself and for himself (after all, "he was alone"), therefore the translator's choice to adopt the character's modality of expression is absolutely warranted.In the end, we are left with one voice only, and that is Stephen's.
To these examples may be added one more.Again it has been borrowed from the Eveline story and it demonstrates how a grammatically marked FID situation in the original text may successfully be dealt with in translation even though the target-language system may lack equivalent grammatical apparatus.Eveline, the only provider in the family, has to hand out all her wages to her father and is then made to beg for money when she wants to go out: 200 (6) In the end he would give her the money and ask her had she any intention of buying Sunday's dinner.(D, 'Eveline', 39) Slovene: Na kraju ji je potem denar dal in jo vprasal, ali misli kaj kupiti za nedeljsko kosilo.(32) The underlined part of the sentence is evidently in disagreement with English grammar, however, it perfectly agrees with Gaelic, where "there is no difference in word order between a direct and an indirect question.Both are introduced by an interrogative particle showing that a question follows ... " (Hedberg 1981: 115).The anomalous wording of the interrogative clause has given way to the character's voice and thus signalled that FID is at work again.In order to recreate it in translation, the Slovene language would need to resort to its own idiosyncratic arsenal, in this particular case the implementation of the colloquial conjunction eel a instead of the standard ali (both denoting "if').The register would thus shift from formal to informal, implying that it is the character's voice we hear rather than the narrator's.Given the fact that the Slovene translator has failed to make good use of this possibility, whereby the FID has changed into a mere narrative report of speech or at best into indirect speech.And this, according to Short's dine of speech presentation, is almost exclusively in the narrator's control, meaning that instead of the character's intimate and subjective reflection we get objective information from the distanced narrator/author (Leech and Short 1992: 318-351).(This particular illustration may be of relevance to translators into the languages which resemble Slovene grammatical features.)

DOUBLE-VOICEDNIEWED ])ISCOURSE
The next field ofFID presentation is, translation-wise, a much less formidable task of rendering, albeit not completely devoid of complications.It concerns the Bakhtian pure 'double-voiced discourse' which serves two speakers at the same time and expresses simultaneously two different intentions: the direct intention of the character who is speaking, and the refracted intention of the author.In such discourse there are two voices, two meanings and two expressions.And all the while these two voices are dialogically interrelated, they-as it were-know about each other (just as two changes in a dialogue know of each other and are structures in this mutual knowledge of each other); it is as if they actually hold a conversation with each other (Bakhtin 1986: 324).This function ofFID may aptly be illustrated with the following excerpt from one of the Dublin stories, 'Clay'.Maria, a simple housemaid, looks at herself in the mirror; simultaneously, James Augustine Aloysius Joyce, born in Dublin in 1882 and educated at Jesuit schools and University College, Dublin, holds the mirror.This is what each of them sees, respectively and simultaneously: (7) She changed her blouse too and, as she stood before the mirror, she thought of how she used to dress on Sunday morning when she was a young girl; and she looked with quaint affection at the diminutive body which she had so often adorned.In spite of its years she found it a nice tidy little body.(D,'Clay',(112)(113) Slovene: Tudi bluzo je preoblekla in ko je tako obstala pred zrcalom, je pomislila, kako se je svoje dni oblacila ob nedeljskih jutrih za k masi, ko je bila se dekletce; in s cudno ljubecim custvom je pogledala drobceno tela.ki gaje bila tolikokrat lispala.Letom navkljub se ji je zazdelo.daje telesce kar cedno.( 90) Apparently, the narrator's manner of describing Maria' s body in the mirror differs immensely from the character's.Maria's idiolect is very specific: her word nice is the most frequently repeated adjective in 'Clay' -indeed it occurs twelve times here out of only twenty nine times in the whole collection of the Dublin stories.The juxtaposition of two disparate perceptions of the same object-the character's body, one given by the narrator and the other one by the character itself, serves as a basis for creating so-called "narrative irony", (Short 1991: 71-72) which in turn forms the basis for the characteristic modernist distance.From a translator's point of view, however, such an FID occurrence should not present too much difficulty, providing the target-language system has the necessary lexical tools in store.Slovene for one is definitely in command of such lexicon, even though the translator, admittedly, has not been completely successful here.The demarcation line between the narrator's presentation of perception and the character's represented perception is not as clear-cut in the translation as it is in the original text.However, as already pointed out, this sort of FID issue depends more on the translator's readiness to perceive an instance of FID and render it accordingly than on the capacity of the language of translation.

LEXICAL MARKERS OF FID
To round up, let me touch upon a very common and frequently used rhetorical device in FID: reiteration.As a rule, reiteration should be quite easily handled in translation, especially when it comes to repeating certain lexical items that constitute the ground of the character's idiolect.It is highly relevant for FID since it suggests "the unsophisticated nature of ordinary thought-processes such as would be reproduced in informal speech or writing" (Wales 1992: 41).Maria's fondness for the word "nice" has already been noted: the following items set forth all the twelve references to this adjective, merely for the sake of emphasis: 202 (8) The fire was nice and bright and on one of the side-tables were four very big barmbracks.(D, 'Clay', 110) Slovene: Ogenj je bil ziv in svetal in na eni pomoznih miz so lezale stiri struce.( 88 One does not need to be a speaker/reader of Slovene to observe that out of these twelve occurrences of the word "nice" as many as half of them have been replaced by their synonyms or substitutes.This is rather astonishing since the repetition is limited to such a short stretch of text and thus so obvious.But it has been noticed that quite a few translators (not only the Slovene ones) are curiously inclined towards 'polishing' the original whenever they feel an author has failed to comply with the principal rules of so-called 'elegant style'.Quite a different and pardonable matter occurs when there is a considerable textual lapse between the first and the second mention.Examples can be found in Ulysses, when Joyce first quotes verbatim the love that dare not speak its name (Joyce 1987: 41), a line from the poem '1\vo Loves' by Lord Alfred Douglas (Coote 1983: 262-264), Oscar Wilde's lover, and then repeats the quote on page 166.On this score, the Slovene translation offers two different versions of the same text simply because the translator assumingly lost track.The pragmatic consequences of effacing the figure of reiteration are aptly described in the words of Katie Wales: With repetition that occurs in close juxtaposition rather than over successive pages we are likely in Dubliners, as in Joyce's later prose, to be also in the presence of a marked subjectivity (of character), rather than an objectivity (of narrator).So the repetitions suggest then unsophisticated nature of ordinary thought-processes such as would be reproduced in informal speech or writing (Wales 1992: 41).
A translation which demonstrates a growing tendency towards the neutralisation of the informal or colloquial diction of the original, manifested in our case through the excessive use of reiteration, is bound to give rise to a greater objectivisation of narrative report and the shifting of perspective and focalisation away from the character towards the (omniscient) narrator.Unlike the reader of the original, who is inclined to assume a somewhat distant and sceptical position as regards the narrative information that s/he receives from the (unreliable) character, the reader of the Slovene text is more likely to trust the seemingly objective report of the author/narrator.In this respect, the former reader is confronted with a far less traditional text in that s/he can rely no longer on whatever information s/he gets from the narrator/author but has to instead realise the interpretive potential of the text entirely on his/her own.

BY WAY OF CONCLUSION
Translation problems of coming to terms with the phenomena of heteroglossia and heteropsia, as discussed in this paper, particularly in the light offree indirect discourse which appears to be their principal implementing tool, are but a drop in the ocean of their astonishing complexity.From my own experience with the English modernist texts in the Slovene translation, besides Joyce including also authors like Virginia Woolf, D. H. Lawrence and several others, the impression is that the translators have been by and large negligent of the specificities of the new style of writing and have sought instead to reproduce the original works in a more or less traditional fashion.To make the matter worse, most of these translations, questionable already when they came out, are nearly half a century old and absolutely nothing is being done towards remedy, even though the fresh blood of the incoming translators seems to be running along the right veins.For example, the very book titled Ljudje iz Dublina (Dubliners) was first published in 1955 and then reprinted in 1993 without any corrections because most of the publishing houses still prefer to cling to the old better-save-than-sorry policy.A sad case, indeed, but it is free direct mammon rather than free indirect discourse that makes the world go round.
She used to have such a bad opinion of Protestants but now she thought they were very nice people, a little quiet and serious, but still very nice people to live with.(D, 'Clay', 111) Slovene: Prej je imela tako slabo mnenje o protestantih, a zdaj je sodila, da so prav prijazni ljudje, nekoliko tihi in resnobni, pa vendar prijazni ljudje, s katerimi je kar moci ziveti.(89) 12) ... but the matron was such a nice person to deal with, so genteel.