The Translator's Voice in Translated Narrative - Theo Hermans - Grifado

Download as pdf or txt
Download as pdf or txt
You are on page 1of 26

The Translator's Voice

in Translated Narrative
Theo Hermans
University College London

Abstract: When we read translated narrative, the original Narrator's voice is


not the only which comes to us. The Translator's discursive presence in the
translated text becomes discernible in certain cases, e.g. when the pragmatic
displacement resulting from translation requires paratextual intervention for
the benefit of the Implied Reader of the translated text; when self-reflexive
references to the medium of communication itself are involved; when 'contex-
tual overdetermination' leaves no other option. The ways in which the
Translator's discursive presence manifests itself are demonstrated on the basis
of different translations of the Dutch novel Max Haveiaar (1860).
Résumé: Les récits traduits ne reproduisent pas seulement la voix narrative de
l'original: la présence discursive du traducteur lui-même s'y manifeste de
plusieurs manières: lorsque le déplacement pragmatique occasionné par la
traduction requiert une intervention paratextuelle à l'intention du lecteur
implicite de la traduction; lorsque celle-ci doit rendre des références
autoréflexives au mode de communication; en cas de 'surdétermination ' par le
contexte. Des exemples sont pris dans différentes traductions du roman néer-
landais Max Haveiaar (1860).

1.

When Boris Yeltsin speaks through an Interpreter, do we really want to hear


the Interpreter's voice? We listen, surely, because we want to know what
Yeltsin has to say. To the extent that we are conscious of hearing the
Interpreter's voice, it is as no more than a minor distraction. We regard — or
better: we are prepared, we have been conditioned to regard — the Inter­
preter's voice as a carrier without a substance of its own, a virtually transpar­
ent vehicle. Anything that takes away from this transparency is unwelcome

Target 8:1 (1996), 23-48. DOI 10.1075/target.8.1.03her


ISSN 0924-1884 / E-ISSN 1569-9986 © John Benjamins Publishing Company
24 THEO HERMANS

'noise' in the information-theoretical sense of the term.


At the same time we know perfectly well that, unless we understand
Russian, the Interpreter's voice is all we have. It is all we have access to, all
we can make sense of on this side of the language barrier. Even when on radio
or television, for example, we can still hear Yeltsin's Russian in the back­
ground, we do not understand those words and so we shut them out, realizing
they are being broadcast alongside the interpretation only to authenticate the
Interpreter's re-enunciation. We are not unduly bothered about not having
direct access to Yeltsin's language because we know we can trust the
Interpreter's professionalism. We trust that the Interpreter's words are an
accurate and truthful copy or reproduction of Yeltsin's words. This trust,
underpinned as it is by a number of professional and institutional guarantees,
allows us to accept, or to project, the Interpreter's discourse as matching
Yeltsin's, as constituting the equivalent of it, as coinciding with it, as being, to
all intents and purposes, identical with it.
Of course, as soon as we stop to reflect on the various a-symmetrical
interlingual and intercultural processes taking place in the operation, we
realize we are entertaining an illusion. The translation never coincides with its
source, it is not identical or equivalent in any formal or straightforward sense,
and it remains to be seen how the notion of the one discourse 'matching' the
other is to be filled in. But the illusion is there, and necessarily there. It is part
and parcel of what we, in our culture, have come to understand by 'transla­
tion'. It fits, that is, our ideology of translation, which is backed up by a vast
amount of translation theory and history and by all manner of institutionalized
agreements governing relations between primary and secondary texts, up to
and including such things as intellectual ownership, copyright laws, autho­
rized translations, legally certified copies, professional codes of conduct, etc.
Translation, in this context, is delegated speech, and the delegate has no
executive powers.
The illusion, then, the necessary illusion, is one of transparency and
coincidence as exponents of equivalence. As the Interpreter's voice falls in
with, coincides with and in so doing — paradoxically — disappears behind
Yeltsin's voice, the physical experience of hearing two distinct voices speak­
ing more or less simultaneously is suppressed, or sublimated, and in practice
we consider the two voices to be wholly consonant. Even as we listen to the
one voice we are able to follow, we negate its presence because we recognize
its substantial and institutionally endorsed conformity to the primary enuncia-
THE TRANSLATOR'S VOICE IN TRANSLATED NARRATIVE 25

tion, which, we accept, has integrity, authority, and therefore primacy. We


conclude that 'Yeltsin has said so and so'. Because it has no substance of its
own, the re-enunciation in the other language assumes the quality of direct
quotation.1

2.

Although interpreting, as a rule, involves the simultaneous physical presence


of two individuals and two voices in the same space, there are good grounds
for arguing that very much the same factors, and the same illusion of transpar­
ency and coincidence, are at work in written translation, and in translated
fiction. Perhaps the illusion is even stronger here.
Consider for a moment a certified translation of a degree certificate. The
stamp issued by a third party declaring the translation to be a 'true copy' of
the original reminds the user that this is a translation: the stamp marks the
distance between original and translation (usually also identifying the sworn
translator by name), but at the same time as formalizing their differential
status it asserts that the copy is 'as good as' the original. Next, consider a
translation, say, of a safety regulation in a large international concern, which
first issues the new regulation in the head office and then has it translated into
a number of languages for use in its subsidiary plants. In each country this
version subsequently becomes one regulation among several, and the transla­
tors' interventions, including their proper names, are completely erased.
Whether the translators are the firm's own employees or the job was con­
tracted out to a translation agency, all concerned have every reason to take it
for granted that the translation constitutes a truthful copy of the original.
Translators, after all, like interpreters, speak in someone else's name and thus
they are expected to subscribe to what Brian Harris has called the 'true
interpreter' norm, or the 'honest spokesperson' norm, which "requires that
people who speak on behalf of others . . . re-express the original speakers'
ideas and the manner of expressing them as accurately as possible and without
significant omissions, and not mix them up with their own ideas and expres­
sions" (Harris 1990: 118). The translator is expected to observe total discre­
tion.
Although nowadays many works of translated fiction routinely carry the
translator's name and in this respect resemble the 'certified copy', in the way
26 THEO HERMANS

they are read they are mostly closer to the second case just mentioned. While
reading translated fiction, readers are normally meant to forget that what they
are reading is a translation. The translator withdraws wholly behind the
narrating voice. So whose voice comes to us when we read a translated novel?
Common usage is indicative. We tend to say that we are reading Dosto-
yevsky, for example, even when we are reading not Russian but English or
French or Spanish words. This blotting out, this 'erasure' of the translator's
intervention is paradoxical. In contrast to simultaneous or especially consecu­
tive interpreting, with two speakers sharing a given physical space, when we
read translated fiction we have only the translated text in front of us. The
primary voice, the authoritative originary voice, is in fact absent. And yet we
casually state it is the only one that presents itself to us. And in practice we do
so largely, perhaps wholly, on the strength of the hierarchy implied by the
order (and, more often than not, the size) of the names on the title page:

NOVEL by Writer X
Translated by Translator Y

The question arises: is the illusion of T am reading Dostoyevsky' all there is


to it? Does the translator, the manual labour done, disappear without textual
trace, speaking entirely 'under erasure'? Can translators usurp the original
voice and in the same move evacuate their own enunciatory space? Exactly
whose voice comes to us when we read translated discourse?
The question can be examined from a variety of angles, including an
ideological one. In what follows it is treated first and foremost as a
narratological issue, asking about the voice that produces the discourse that
we read, and therefore asking about the discursive centre from which the text
issues. Let us remind ourselves of the standard representation of narrative
communication:2
THE TRANSLATOR'S VOICE IN TRANSLATED NARRATIVE 27

This scheme represents the normal situation, without reference to translation.


What we read is a discourse produced by a Narrator. In translated fiction, who
exactly articulates the translated discourse? It is the same Narrator as in the
source text? Does this then mean that the translator, as Biographical Transla­
tor, is like the Biographical Author: another name on the title page, indicating
an entity located firmly outside the narrative discourse?
As it is, narratology does not distinguish between original and translated
fiction. The main narratological models currently available (Booth, Stanzel,
Genette, Rimmon-Kenan, Bal, Chatman, Prince) are designed to apply to
narrative texts in general, irrespective of whether they are original or trans­
lated. In what follows I will argue that these models overlook a presence in
the narrative text that cannot be fully suppressed.3 Translated narrative dis­
course, it will be claimed, always implies more than one voice in the text,
more than one discursive presence. It may be that in many narratives this
'other' voice never clearly manifests itself, but it should nevertheless be
postulated, on the strength of those cases where it is manifestly present and
discernible. And it is only, I submit, the ideology of translation, the illusion of
transparency and coincidence, the illusion of the one voice, that blinds us to
the presence of this other voice.

3.

The claim, then, is that translated narrative discourse always contains a


'second' voice, to which I will refer as the Translator's voice, as an index of
the Translator's discursive presence. The voice may be more or less overtly
present. It may remain entirely hidden behind that of the Narrator, rendering it
impossible to detect in the translated text. It is most directly and forcefully
present when it breaks through the surface of the text speaking for itself, in its
own name, for example in a paratextual Translator's Note employing an
autoreferential first person identifying the speaking subject.4 And then there
are shades and degrees in between.
As far as I can make out, the 'other' voice in translated narrative texts is
likely to manifest itself primarily in three kinds of cases. They all involve
what may be termed 'performative self-contradiction' in that the Translator's
self-denial runs into obvious, textually traceable contradictions. In each case
there is a certain pressure on the Translator to come out of the shadows and
28 THEO HERMANS

directly intervene in a text which the reader had been led to believe spoke
with only one voice. They are
(1) cases where the text's orientation towards an Implied Reader and
hence its ability to function as a medium of communication is
directly at issue;
(2) cases of self-reflexiveness and self-referentiality involving the
medium of communication itself;
(3) certain cases of what, for want of a better term, I will refer to as
'contextual overdetermination'.
In each case the degree of visibility of the Translator's presence depends on
the translation strategy that has been adopted, and on the consistency with
which it has been carried through. Since each of the three cases listed, and
certainly cases (1) and (2), involve a kind of communicative short-circuiting,
a fissure within the discourse which draws attention to the linguistic and
pragmatic dislocation that comes with translation, the resulting incongruity in
the translated text needs to be accounted for in one way or another. Some
translation strategies will effectively paper over the cracks and leave the
reader unaware of the other voice. My interest here is in those instances where
the translated text itself shows visible traces of a discursive presence other
than the ostensible Narrator. Before illustrating the case with reference to a
particular translation, a further word about the first two cases mentioned
above; the third will be elucidated later, with reference to a concrete example.
As for (1): Unlike interpreting, where speaker and interpreter more or
less simultaneously address a physically present if linguistically mixed audi­
ence, written translations normally address an audience which is not only
linguistically but also temporally and/or geographically removed from that
addressed by the source text. To the extent that a Reader is implied by and
implicated in the overall 'intent' (Chatman) and orchestration of narrative
fiction, translated narrative fiction addresses an Implied Reader different
from that of the source text, since the discourse operates in a new pragmatic
context. All texts are culturally embedded and require a frame of reference
which is shared between sender and receiver to be able to function as vehicles
for communication. The various forms of displacement that result from
translation (Folkart 1991: 347ff. speaks of a 'décalage traductionnel')
threaten this shared frame of reference. It is therefore not surprising to find
that it is precisely with respect to the cultural embedding of texts, e.g. in the
form of historical or topical references and allusions, that the Translator's
THE TRANSLATOR'S VOICE IN TRANSLATED NARRATIVE 29

Voice often directly and openly intrudes into the discourse to provide infor­
mation deemed necessary to safeguard adequate communication with the new
audience. As a rule, translations, and certainly modern translations of canoni­
cal literary fiction, stop short of reorienting the discourse so radically that the
orientation to the original Implied Reader disappears altogether. The trans­
lated text can therefore be said to address a dual audience, and thus to have a
'secondary' Implied Reader superimposed on the original one. This can lead
to hybrid situations in which the discourse offers manifestly redundant or
inadequate information, or appears attuned to one type of Reader here and
another there, showing the Translator's presence in and through the discor­
dances.
As for (2): 'self-reflexiveness' and 'self-referentiality' are used here as
rather broad terms covering various instances which have been discussed as
exemplifying untranslatability by, among others, Jacques Derrida. Obvious
cases are texts which affirm their being written in a particular language, or
which exploit the economy of their idiom through polysemy, wordplay and
similar devices. We are dealing, then, with instances in which language
collapses upon itself, as it were, or, as Derrida would have it, 're-marks' itself.
It is of course possible that the translated text solves the problems so dis­
creetly that no trace of a 'second voice' is left behind. But sometimes
translations run into contradictions and incongruities which challenge the
reader's willing suspension of disbelief; or the translated text may call on the
explicit intervention of a Translator's Voice through the use of brackets or of
notes, and they then remind the reader of this other presence continually
stalking a purportedly univocal discourse.
Examples of linguistic self-referentiality are not hard to find, and they
are not restricted to narrative texts. In translations of many of Derrida's essays
meta-linguistic notes or comments are added by translators in their attempts to
cope with the French puns. Nowhere is this more in evidence than in
'Survivre: Journal de bord', where the lower band of the page, the 'journal de
bord', openly challenges the translator to render the French puns ('pas de
méthode', 'point de méthode', the series 'écrit, récit, série', etc.); again and
again the English translation has recourse to square brackets in an effort to
keep up, overtly displaying the translated nature of the text by showing the
translator's hand (see e.g. Derrida 1991: 256ff.).
Derrida's discussion, elsewhere,5 of a rather different instance of lan­
guage affirming itself concerns the final chapter of Descartes' Discours de la
Méthode, where in the French original the author declares that his book is
30 THEO HERMANS

written in French and not in Latin. The Latin translation omits this embarrass­
ing sentence, to avoid the self-contradiction of a statement in Latin declaring
it is not in Latin but in French. Derrida regards this as a case of institutional
and statutory untranslatability, which is a perfectly valid observation. For the
reader of the Latin version, however, the omission is not detectable and
therefore does not reveal the translator's presence in the translated text as
such. In translations into languages other than Latin, where the sentence is
translated, the self-contradiction may be less glaring, but it is still obvious
enough. When the English version has: "And if I write in French . . . rather
than in Latin . . . it is because . . ." (Descartes 1968: 91), the anomaly of
reading an English text which declares, in English, that it is actually in French
creates a credibility gap which readers can overcome only by reminding
themselves that this is, of course, a translation. But in so doing those readers
also realize that the voice producing the statement cannot possibly belong to
Descartes, or to Descartes alone. There is, clearly, another voice at play,
duplicating and mimicking the first one, but with a timbre of its own. Derrida
himself exploits the paradox in his 1984 address "Ulysse gramophone",
which opens with the sentence: "Oui, oui, vous m'entendez bien, ce sont des
mots français", and leads to an inevitable if self-conscious self-contradiction
in the English translation: "Oui, oui, you are receiving me, these are French
words" (Derrida 1992: 256).

4.

In the following pages I will illustrate these points in more detail, with
reference to several translations of a single book, the Dutch novel Max
Havelaar by Multatuli, first published in 1860. With its complex narrative
structure Max Havelaar provides instances not only of cultural embedding
and linguistic self-referentiality, but also of 'contextual overdetermination',
all of them, to varying degrees, bringing the Translator's voice to the textual
surface. I will refer to two English translations, by W. Siebenhaar (1927) and
Roy Edwards (1967, re-issued as a Penguin Classic 1987); and to the French
and the Spanish translations by Mme Roland Garros (1968) and Francisco
Carrasquer (1975), respectively. There are other translations, including older
renderings into English and French as well as versions into German and other
languages, but the four considered here will do for the purposes of the present
THE TRANSLATOR'S VOICE IN TRANSLATED NARRATIVE 31

exposition. The Dutch text referred to is the standard scholarly 'historical-


critical' edition (1992), which is based on the novel's fifth and last edition
(1881) to appear during the author's lifetime.6
The Biographical Author of Max Havelaar is Eduard Douwes Dekker
(1820-1887), who wrote under the pseudonym Multatuli. The novel was
dedicated to Everdine Huberte van Wijnbergen.7 In its barest essence the
book tells the story of Max Havelaar, a Dutch civil servant in the colonial
administration in the Dutch East Indies in the 1850s. Witnessing the exploita­
tion of the local population by the native elite, he protests in vain to his
immediate superior. When he ignores the administrative hierarchy and brings
a charge against the local ruler he is relieved of his post and resigns in disgust.
The novel's unusual richness, however, springs at least in part from its
narrative structure and its use of different narrators. The opening chapters
employ the homodiegetic narrator Batavus Droogstoppel, an Amsterdam
coffee broker and the epitome of Dutch petit-bourgeois values and hypocrisy.
One day Droogstoppel runs into a former schoolmate, the destitute Sjaalman,
who leaves him a bundle of documents which Droogstoppel reckons he can
use for a book on the coffee trade. However, he delegates the writing to a
young German trainee in his firm, the romantic Stern who, assisted by
Droogstoppel's son Frits and probably helped by Sjaalman himself, writes an
altogether different book, which is being composed as we read. Stern thus
becomes the narrator of the dramatic story of Max Havelaar and his wife Tine
in the East Indies. As the novel continues, Droogstoppel occasionally breaks
into Stern's discourse to voice his disapproval of the way Stern is handling his
material. In the Havelaar episodes, meanwhile, the Stern-Narrator dishes up
other stories as well as documents purporting to authenticate Havelaar's
conflict with his superiors. The novel's concluding pages contain the main
surprise, as another first-person Narrator identifying himself as Multatuli
suddenly takes over from Stern, dismisses Droogstoppel, and addresses the
Dutch King with an impassioned plea to stop the oppression of the natives in
the East Indies.
The different narrative levels in Max Havelaar may be represented
schematically as follows:
32 THEO HERMANS

DROOGSTOPPEL (in Amsterdam)


= I-Narrator
'you' addressee = the Dutch reader
→Droogstoppel delegates narration to
STERN (in Amsterdam)
= I-Narrator2
'you' addressee = the Dutch reader
→ story of Max Havelaar, wife Tine etc.
(in Dutch East Indies)
Final pages: 'I, MULTATULF takes over as Narrator from Stern
and Droogstoppel
= I-Narrator3
'you' addressee = the Dutch King

As it happens, there is in Max Havelaar an instance similar to the Descartes


case mentioned above. In the Dutch text Max Havelaar quotes a poem in
French which he says he wrote himself and which is also reproduced in
French (MH: 127). As Havelaar is Dutch, the poem serves as a signal to the
reader, suggesting Havelaar's literary talent and linguistic abilities,
favourably contrasting him with the dreary Droogstoppel figure. The French
translation, unable to translate the French poem into itself, merely quotes it
(Garros: 186). But it adds a Translator's Note: "En français dans le texte
(N.d.T.)".8 The reason for rupturing the discourse and intervening by means
of a paratextual note is clear enough: the surplus value signalled in the
original by the use of a different language risks being lost on the French
reader. The note itself, however, with or without its curious ambiguity ("dans
le texte" is obviously redundant if it means the French text, so it must refer to
the Dutch original), reminds its readers that this is a translated text — and at
the same time alerts them to the presence of an altogether different voice,
breaking through the narrative voice ostensibly established by the discourse.
In contrast, a poetic dialogue in German also reported by the Narrator as
having been penned by Havelaar (MH: 157-160) is left, unannotated, in
German in the French translation (Garros: 225-228), revealing not a glimpse
of the Translator's presence — except in the dissonance between a Note
identifying the language of a French poem in a French text and remaining
silent on how to read a poem in a foreign language.
THE TRANSLATOR'S VOICE IN TRANSLATED NARRATIVE 33

5.

Before going on to consider examples of different ways in which, and degrees


to which, the Translator's presence makes itself felt in translations of Max
Havelaar, it may be useful to point out that they all concern instances where
the presence of an enunciating subject other than the Narrator becomes
discernible in the translated text itself. This means we are not looking at cases
where only the comparison with the source text will show the intervention of
a (biographical) translator, however revealing these manipulations, additions
or omissions may be from an ideological or other such point of view. In the
Havelaar narrative, for example, the Narrator at one point describes the layout
of Havelaar's house in the Indies. In the Dutch, and in the French translation,
this is done by means of numbers: imagine a rectangle divided into twenty-
one compartments, three across, seven down, etc. (MH: 136; Garros: 198). In
both the Spanish and the English versions diagrams are added as visual
illustrations, without paratextual comment (Carrasquer: 190; Edwards: 224).
Since the Spanish and English texts themselves provide no discernible trace
of a linguistic subject other than the Narrator as the source of the diagrams,
and consequently the intervention by (presumably) the biographical translator
can be detected only through a comparison with the source text, such in­
stances are not relevant to the argument being pursued here.
Hence, although the cases highlighted below mostly issue from source
text cruxes, they do not focus on the retrospective comparison between
translation and original. Rather, as different translations opt for different
solutions, some bring the Translator's discursive presence clearly into view
while others do not, or to a lesser extent. In each case the assumption is that,
given the dominant conception of transparent translation in modern fiction,
the reader's awareness of reading a translation lies dormant, leaving intact the
notion that (with the exception of embedded narrative and character speech)
there is only one Narrator speaking at any one time. As long as there are no
markers in the text suggesting another voice, all is well. The following
instances, however, concern places where another presence insinuates or
parachutes itself into the text, breaking the univocal frame and jolting the
reader into an awareness of the text's plurivocal nature.
34 THEO HERMANS

5.1

A couple of examples will suffice to illustrate the rather obvious case of


cultural references which, as a result of the displacement brought about by
translation, threaten to be left in a vacuum and prompt the Translator to
rupture the narrative frame by means of paratextual Notes.
In Chapter 9 of Max Havelaar, in a passage marking a sudden transition
from Stern as Narrator (of the Havelaar story) to Droogstoppel as Narrator (of
his own story), reference is made to Abraham Blankaart, a fictional character
from the Dutch novel Sara Burgerhart (1782) by Betje Wolff and Aagje
Deken. The name is first dropped in the Stern narrative, and taken up by
Droogstoppel. That its recognition requires a Dutch cultural frame of refer­
ence is evident from Droogstoppel's comment that his own son Frits has
obviously been lending young Stern a hand, for, as he says, "this Abraham
Blankaart is much too Dutch for a German" (MH. 95). Ironically, in a note
added by Multatuli himself to
the fourth edition (1875) and retained in subsequent editions (Multatuli 1992,
II: xxxvii), he wondered how many of his own generation still recognized the
literary allusion (MH 263), suggesting that even within a specific cultural
continuum shared frames of reference are time-bound phenomena.
Both English translations and the Spanish version at this point break into
the text to insert a Translator's Note explaining the reference (Siebenhaar:
122-123; Edwards: 134-135; Carrasquer: 155-156).9 The French translation
retains the name, without an explanatory note; a second occurrence of the
name, in the next paragraph, has been omitted (Garros: 142-143). This seems
consistent, and at first sight does not allow the reader a view of the
Translator's discursive presence. But before this passage other instances have
occurred where Translator's Notes, identified as such, explained linguistic or
cultural issues arising from the source text (Garros: 36, 38, 58, 60, 87,
108, . . .). It is precisely the inconsistency in the provision of paratextual
information designed to safeguard the shared frame of reference which
creates a disparity at the discursive level: the helpful voice which came to the
French reader's rescue on other occasions remains inexplicably silent here.
The deliberate silence of the voice, withholding information which is ex­
pected, signals the presence of a discursive subject different from either of the
two homodiegetic Narrators whose words we read. As we are reading a
translation, this 'differential voice' can only be the Translator's. 10
THE TRANSLATOR'S VOICE IN TRANSLATED NARRATIVE 35

A similar inference can be made, incidentally, with respect to the proper


names in several of the translations. Max Havelaar is set in a particular place
and time: the Netherlands and the Dutch East Indies in the 1850s. The proper
names of the Dutch characters are recognizably Dutch. Some, like Havelaar,
Bastiaans, Busselinck, Verbrugge, are proper names in the conventional
sense; others are more Dickensian, motivated names, holding out an invita­
tion to the reader to activate their latent semantic load. Roy Edwards' English
translation copies most of the names intact, including e.g. the suggestive
'Droogstoppel' (paratextually annotated as meaning 'Drystubble', 'Dryas­
dust', which is the way the name was rendered in the two previous English
translations). With a name like 'Slimering' (Dutch: 'Slymering'), luck is on
the Translator's side, as minimal adaptation preserves the satirical intent
without revealing the Translator's presence. But the pompous preacher
Wawelaar appears in the text as 'Parson Blatherer', creating a sudden incon­
gruity within the pattern of proper names which stretches the reader's willing­
ness to suspend disbelief: an obviously Dutch character, in a Dutch setting,
with so transparent and apposite an English name, yet without explanation?
Given the otherwise consistent discourse of the Narrators delivering the story,
the source of the disparity must surely be attributed to another discursive
presence.

5.2

The name of the character Sjaalman in Max Havelaar is a special case. This is
not a proper name at all, as Droogstoppel, the Narrator at this point, declines
to tell us the person's real name (which the reader will infer later). It is first
introduced as a descriptive term ("the man with the shawl, or scarf), then
contracted into 'de Sjaalman' ("the Shawl-man"), and Droogstoppel subse­
quently decides to continue to call the man 'Sjaalman' as if that were his
name. The first occurrence of the designation is accompanied by
Droogstoppel's metalinguistic comment to the effect that his son Frits prefers
the English word 'shawl' to Droogstoppel's good old Dutch 'sjaal'. It is the
first in a series of linguistic quibbles through which Droogstoppel character­
izes himself as a pedant:
In-plaats van een behoorlyken winterjas, hing hem een soort van sjaal over
den schouder — Frits zegt: 'shawl, maar dit doe ik niet — alsof hy zoo van
de reis kwam. (MH: 10)
36 THEO HERMANS

The translations render the passage as follows:


Instead of a suitable winter coat, he had a kind of shawl hanging over his
shoulder — Frits says 'Châle': he is learning French, but I keep to our good
old language — as if he had just come from a journey. (Siebenhaar: 12-13)
Instead of a decent winter coat he had a sort of scarf dangling over his
shoulder — we call it a sjaal in Dutch, so Frits has to call it a 'shawl', which
isn't even right, just to show off his English — as if he — the scarf-man, I
mean — was just back from being on the road. (Edwards: 27)
Aunque en lugar de llevar gabán, como habría sido lo propio, colgaba de sus
hombros una especie de bufanda or largo chal — Frits dice panuelo de
cuello, pero yo no paso por ahi — como si acabara de llegar de algún viaje.
(Carrasquer: 22)

Au lieu d'un pardessus convenable, une sorte de châle couvrait ses épaules,
comme s'il rentrait de voyage. (Garros: 36)
Neither Garros (on this page) nor Carrasquer upset the reading process, and
hence the fictional universe, as neither text leaves the language the reader is
engaged in. Garros omits Frits' use of a foreign term, and Carrasquer employs
an intralingual variant. As Sjaalman's descriptive designation turns into a
proper name, the Spanish version continues with 'Chalman'. At that stage the
French translation runs into a linguistic problem: since "homme au châle" will
not double up as a proper name in French, the transition from designation to
name cannot take place within the terms first set out for it. The problem is
solved by cutting into the text's linguistic homogeneity: the next occurrence
has: "Un jour, le Sjaalman que j'avais devant moi était avec nous" and the
novel term 'Sjaalman', more than halfway between designation {'le
Sjaalman') and proper name ('Sjaalman' capitalized), is elucidated in a
Translator's Note: "Littéralement: homme au châle (N.d.T.)" (Garros: 38).
But the linguistic transgression, the Note and the fact that the reference is to
the source language of a translation, break the narrative frame. The agent
responsible for this discursive act cannot be the Narrator.
Siebenhaar solves the 'sjaal/shawl' problem by substituting 'shawl/
châle', English/French for Dutch/English, culturally not dissimilar as a pair,
and so the reader's awareness that in fact Droogstoppel is Dutch and speaks
Dutch, remains largely dormant (not wholly, as the reader knows that
Droogstoppel's 'our good old languae' is Dutch, not English). Edwards' triad
'scarf/sjaal/shawl' does remind the reader of Droogstoppel's native language
— and more than that, since Dutch is not only the language of Droogstoppel
THE TRANSLATOR'S VOICE IN TRANSLATED NARRATIVE 37

as character and (homodiegetic) narrator, but also, inescapably, the source


language of the translation as a whole. The occurrence, in the translated text,
of a metalinguistic statement bearing coincidentally on both the fictional
world (Droogstoppel as a fictional character) and its framing context (a novel
translated from the Dutch) highlights a paradox. As readers, we accept the
convention, which also operates e.g. in historical fiction or in narratives set in
foreign lands, that we read words in one language while we know that the
fictional characters, to the extent that they resemble lifelike individuals
existing in identifiable locations, must have spoken a different language. In
this crux in the English Max Havelaar the metalinguistic reference draws
attention to that convention itself." This is no doubt partly because it involves
the source language of the whole work, as the Translator's paratextual Notes
remind us time and again. A further reason is that the 'sjaal/shawl' issue is
only one of a series of similar instances of linguistic point-scoring between
Droogstoppel and his son, some of which are presented within the terms of the
English language while others draw on Dutch. These disparities within the
discourse itself prevent the conventional suspension of disbelief and bring
into focus the linguistic as well as pragmatic displacement consequent upon
the act of translation.
The following passage provides additional illustration. This is Droog­
stoppel again, taking on Frits on a point of language:
. . . en dat ook de suikerraffinadeurs — Frits zegt: raffineurs, maar ik schryf
nadeurs. Dit doen de Rosemeyers ook, en die doen in suiker. Ik weet wel dat
men zegt: geraffineerde schelm, en niet: geraffinadeerde schelm . . . (MH:
29)
. . . and that also the sugar-raffinadeurs — Frits says refiners, but I write
raffinadeurs; this the Rosemeyers do also, and they are sugar people; I
know, one speaks of a refined scoundrel, and not a raffinadeur scoun­
drel . . . (Siebenhaar: 36)
. . . and that the sugar refiners, raffinadeurs as we say in Dutch — Frits says
raffineurs but I say raffinadeurs; the Rosemeyers do too, and they're in
sugar. I know one talks about a geraffineerd scoundrel, and not a geraffina-
deerd scoundrel . . . (Edwards: 54)
. . . y que también los refinadores (Frits dice 'refineros', pero yo hago como
los demás, incluidos los Rosemeyer, y eso que trabajan en el ramo del
azúcar; ya sé que se dice un carterista 'refino' y no refinado, . . .
(Carrasquer: 53)
. . . et que les raffineurs de sucre et les marchands d'indigo dussent y joindre
les leurs. (Garros: 61)
38 THEO HERMANS

The Spanish translation again solves the linguistic problem within its own
linguistic terms and the French version has suppressed it altogether; neither
allows a glimpse of the Translator's discursive presence. Both English ver­
sions go beyond their own language. The Siebenhaar translation has Droog-
stoppel using a non-existent English word (the OED does not list
'raffinadeur') coupled with Frits' 'refiners'. Apart from the psychological
inconsistency in Droogstoppel as a character/narrator who normally insists on
solid Dutch values and is here found using a term which the English reader
might conceivably (if erroneously) take to be French, there is little to suggest
another voice speaking alongside the Narrator's.
In the Edwards version Droogstoppel specifically refers to Dutch as his
own language, bringing in three terms: 'refiners', lraffinadeurs' and
'raffineurs'. Curiously, Droogstoppel's discourse here short-circuits itself, as
he writes, in English, 'sugar refiners', only to declare in the same breath, "but
I say raffinadeurs", a Dutch word. The issue of whether scoundrels are
'geraffineertf or 'geraffinadeerd' will not only be lost on the English reader
(both forms contain Dutch morphemes), but will draw further attention to
Dutch as the real language of Droogstoppel's fictional world — and as the
language of a book which, when the fictional narrative is shattered in the final
pages, closes with a direct address by a Dutchman (see below) to the Dutch
King. This pulls very closely together the different levels of language which
are involved: the language of the fictional characters and narrators, the
language of Multatuli's address to the King, and the source language of the
translation. What is striking, however, is not so much the paradoxical nature
of this situation — convention keeps that under control — but the linguistic
anomaly in Droogstoppel's own usage. That anomaly points to a different
discursive locus, not reducible to Droogstoppel's speech act.

5.3

Max Havelaar is a complex novel not only because of its different narrative
levels, but also because it plays, ingeniously and decisively, on the distinction
between fiction and non-fiction. A crucial aspect of this is the way in which
the narrative levels are linked. Some of these links are immediately obvious,
others are established piecemeal, through association, inference, and a num­
ber of subtle shifts and hints.11 In this way it gradually becomes clear to the
reader that Max Havelaar, who only appears as a character in Stern's narra-
THE TRANSLATOR'S VOICE IN TRANSLATED NARRATIVE 39

tive, is in fact the same person as the Sjaalman who later (but, as we read, in
the opening chapters) runs into Droogstoppel in Amsterdam and leaves him
the documents from which Stern constructs the Havelaar story. Stern also
copies the correspondence detailing Havelaar's conflict with his superiors.
However, when in the book's final pages Multatuli himself takes over, he
radically changes the perspective, emphasizing the factual truth underlying
the fictionalized Havelaar story (including the factual truth of the correspon­
dence) and declaring the novel form to have been a mere ploy, since the book
is really a tract written in self-justification. Together with a number of other
indications (detailed in Sötemann 1972: 67-69, 276-280) this leads to the
inescapable conclusion that Havelaar, Sjaalman and Multatuli are all the same
individual, thus linking the most deeply embedded narrative level, the
Havelaar story, with the framing discourse and the name on the cover and title
page.
In the narrative, Havelaar's wife is never called anything but Tine,
although she has a handkerchief with the letter E in the corner, her grandfa­
ther is a 'Baron van W.', and we learn her initials: E.H.V.W. In the
Droogstoppel narrative Sjaalman's wife is never mentioned by name, but
there are certain parallels with Tine. The book's dedication to 'E.H.v.W.'
(manuscript and first three editions) and, from the fourth edition onwards, to
'Everdine Huberte Baronnesse van Wynbergen' (MH: 1), leaves little doubt
that the person in question is the wife of— well, of a pseudonym? 'Multatuli'
is evidently a pen-name (Latin: 'multa tuli'), which the bearer himself trans­
lates in the text: T, Multatuli, who have borne much' ("ik, Multatuli, 'die veel
gedragen heb'", MH: 235) just as it is said several times of Havelaar that he
had suffered much. No, Tine/Everdine is evidently the wife of Eduard
Douwes Dekker, inventor of Multatuli.12 The chain of identification thus
stretches beyond the pseudonym to the Biographical Author, exploding the
fictional frame and revealing the book as autobiography, since Dekker =
Multatuli = Sjaalman = Havelaar, and the Author = Narrator = Character
(Genette 1991: 83).
It is also possible, of course, to read the claim to factual truth in the novel
as being made within the limits of the fictional world presented by the
narrative. The buck then stops with Multatuli rather than with Eduard Douwes
Dekker. In that case the dedication is presumably also part of the fictional
universe, and Everdine/E.H.v.W. is then indeed the wife of Multatuli. But this
does not affect the identification of Havelaar's wife with Multatuli's dedica­
tee.
40 THEO HERMANS

Now, at the novel's deepest narrative level, the story of Max Havelaar,
we are at one point presented with a conversation between Havelaar, his wife
Tine and some Dutch friends; they discuss the fact that Mrs Slotering, a native
woman who is the widow of Havelaar's predecessor and lives in the same
compound as the Havelaars, prefers to keep to herself. Tine remarks she can
very well understand Mrs Slotering's keenness to run her own household (we
learn the real reason later), and asks her husband:
"Weet je nog hoe je myn naam vertaald hebt?"
to which Havelaar replies:
"E.H.V.W.: eigen haard veel waard. (MH: 121)
In the translations the exchange is rendered as follows:
"Do you remember how you once translated my initials?"
"E.H.V.W. Eigen haard veel waard."
Translator's Note: Own hearth great worth (One's own hearth is worth a
good deal) (Siebenhaar: 159)
"Do you remember how you once translated my initials: E.H. v. W.?"
"Yes. Eigen haard veel waard"
Translator's Note: "Literally 'one's own hearth is worth much'. Cf.
There's no place like home'" (Edwards: 170)
"?Te acuerdas de cômo interpretaste mis iniciales como si fueran siglas de
una máxima: E.H.V.W.?"
"— Ya lo creo: 'Eigen Haard, Veel Waard."
Translator's Note: "Literalmente: 'El propio es el hogar que mucho
vale', pero aquí se refiere a que vale mucho lo que uno hace y la casa
propia gana si la administra y cuida la misma duena. Las iniciales
corresponden a: Everdina Huberta van Wijnbergen". (Carrasquer: 199)
[Omitted] (Garros: 179)

The extent to which the initials are 'contextually overdetermined' will be


clear by now. Both the immediate pragmatic context of the conversation and
the complex chain of identification linking the fictional character Tine with
the book's dedicatee bear directly on the passage. This severely reduces the
Biographical Translator's room for manoeuvre, as the task consists of com­
bining the initials (ultimately, those of a historical person, and therefore not
open to manipulation) with an appropriate four-word proverb in which each
word begins with a certain letter. Even the Spanish translation, which on other
THE TRANSLATOR'S VOICE IN TRANSLATED NARRATIVE 41

occasions coped with metalinguistic statements within the terms of Spanish, is


at a loss here. The French version omits the exchange, sacrificing one of the
strongest links in the identification chain, which in turn directly affects the
book's status as representing fact or fiction (irrespective of whether the claim
to factual truth is seen as being made within the fictional world only).
In so doing, however, the translated French text as it stands shows no
trace of the Translator's discursive presence in this particular instance. In the
Siebenhaar, Edwards and Carrasquer versions a sudden abyss opens up, even
if we disregard the Translator's Note. As in some of the other examples listed
above, the translated discourse operates under the convention that although
Havelaar and company are Dutch-speakers, we read their words, and the
presentation of their actions, in English or Spanish. This convention routinely
spans the credibility gap by putting in place a suspension bridge, the willing
suspension of disbelief. This convention collapses here. Since the Havelaar
circle in this scene, as in most other scenes in the book, is monolingual, the
sudden shift from English or Spanish to Dutch and back again creates a severe
discursive anomaly, a crevice. Oddly enough, the Narrator appears not to
register the anomaly: the voice leaps out of and back into its language of
narration as if it were speaking independently of its linguistic medium. It is
the other voice, the one which co-produces the text, the Translator's, which
picks up on the disparity and provides an explanation in the form of a
paratextual Note.
The Note in each case also translates the untranslatable. In the source
text, Tine's question regarding Havelaar's having once 'translated' ('ver-
taald') her initials into a proverb rather than a proper name, prompts an
intralingual response: the initials transposed into a proverb. In the English and
Spanish versions, we get what we expect of translation, an interlingual shift.
But instead of transforming the unfamiliar into the familiar, the shift (from
English/Spanish to Dutch in the text) only produces an incomprehensible
string ("eigen haard veel waard"), the opposite of translation. Since the
discourse itself does not remark on this incongruity, the assumption must be
that this is Dutch, Havelaar's 'real' language, which we knew was there but
were meant to gloss over — but also, coincidentally, the 'original' language
of the book as a whole. The Translator's Note confirms this by translating the
Dutch proverb (though not the initials, for that transposition, barring a linguis­
tic fluke, can only take place in Dutch) and showing the intractability of the
problem. In invoking, explicitly or not, the book's dedication and thus going
42 THEO HERMANS

beyond the immediate Havelaar scene, the Note brings the translated nature
of the entire discourse into the picture, self-referentially and self-reflexively.
The combination, then, of the non-translation of the initials, the reversion to
Dutch and the deferred translation of the Dutch proverb defines the discursive
position of the Translator's voice.

6.

The discussion in the preceding paragraphs was designed to show the pres­
ence of the Translator's voice, as a discursive presence, in translations of Max
Havelaar. If the points made have any validity, it is worth considering some
broader and more general claims. Indeed, while we may have been concerned
with the detail of one particular novel, and a particularly complex one at that,
the issue would appear to be of some wider theoretical import.

6.1

What the discussion of Max Havelaar in translation will have shown is that
the Translator's voice is always present as co-producer of the discourse. The
Translator's voice may remain hidden behind the voice(s) of the Narrator(s)
for long stretches. In some narratives it many never become clearly discern­
ible at all, and in those cases we have to fall back on positing an Implied
Translator as the source of the translated text's invention and intent. How­
ever, the inference from the Havelaar case must be that it is not only
reasonable but necessary to postulate the presence of the Translator's discur­
sive presence in translated fiction, because it is possible to cite specific cases
which clearly bring that 'other' voice to the surface, e.g. in instances where its
intervention is seen to cater for the needs of the Target Text reader (as a
consequence of the cultural and pragmatic embedding of texts and the
displacement resulting from translation), or in cases where the discourse
short-circuits itself through linguistic self-referentiality or contextual over-
determination.
Hence even in those instances where the Translator's discursive pres­
ence is not directly traceable, it is a presence that must be posited, just as we
must also posit a Target-Culture Implied Reader superimposed on the Source-
Culture Implied Reader. It also follows that, if a theoretical model of narrative
THE TRANSLATOR'S VOICE IN TRANSLATED NARRATIVE 43

communication is to be comprehensive, it must create room for instances like


those highlighted here.
This makes untenable the model of translated narrative which simply
assumes that every enunciation has a source (every text implies the presence
of a voice producing it) and that the translator's presence does not interfere
with that discursive situation, being located outside it as the reporting voice
which repeats, verbatim, the words spoken by the first voice. That model,
simplified, looks like this:
(text)
[I say; (text)]
{I translate: [I say: (text)]}
What is needed, rather, is a model of translated narrative which accounts
for the way in which the Translator's voice insinuates itself into the discourse
and adjusts to the displacement which translation brings about. The model,
that is, needs to incorporate the Translator as constantly co-producing the
discourse, shadowing, mimicking and, as it were, counterfeiting the Nar­
rator's words, but occasionally — caught in the text's disparities and inter­
stices; and paratextually — emerging into the open as a separate discursive
presence. In Giuliana Schiavi's essay (in this issue) the theoretical case for an
Implied Translator is argued and an alternative model proposed. In practice,
many translated narratives indeed do not allow us to go further than the terms
of that model, because what has been called here the Translator's voice or the
Translator's discursive presence in the text is wholly assimilated into the
Narrator's voice. It is only in specific cases that the 'other' voice becomes
dissociated from the one it mimics.

6.2

The further question arises: Why do current approaches to narrative have this
blind spot when it comes to the Translator's voice? Why do we, as readers,
prefer to ignore this 'other' discursive presence?
The reason, it seems to me, lies in the cultural and therefore also the
ideological construct which is translation. This takes us straight back to the
dominant concept of translation in our culture: translation as transparency and
duplicate, as not only consonant but coincident and hence to all intents and
44 THEO HERMANS

purposes identical with its source text; the view of translation as reproduction,
in which the translation is meant to reproduce the original, the whole original
and nothing but the original; the image of a translation being 'as good as' its
original, except in regard of status. A translation is a 'good' or a 'proper' or a
'real' translation, we tend to say, if there are no loose ends, no foreign bodies;
it should not contain anything that might affect the integrity of the original.
Translators are good translators if and when they have become transparent,
invisible, when they have spirited themselves away. Only a Translator who
speaks 'under erasure' can be trusted not to violate the original. The loyal
absence of the one guarantees the primacy and aura of the other.
This hierarchy governing the relation between original and translation is
nothing new. Historically it has been construed in a number of ways, mostly
around oppositions such as those between creative versus derivative work,
primary versus secondary, art versus craft, authority versus obedience, free­
dom versus constraint, speaking in one's own name versus speaking for
someone else. In each pair it is translation which is circumscribed, hemmed
in, controlled, subordinated. And in case we think these are after all natural
and necessary hierarchies, it may be useful to remind ourselves of the fact that
in our culture the male/female distinction, too, has been constructed in terms
of very similar oppositions of creative versus reproductive, original versus
derivative, active versus passive, dominant versus subservient.
Deep down, of course, we know — as soon as we stop to think about it —
that the oppositions in which we have placed translation are cultural and
ideological constructs. But they still structure our way of thinking about these
issues, because they are ingrained in our culture and in our mental habits and
projections. To abandon them, and to abandon the control mechanisms which
they keep in place, would be to upset established hierarchies, to deny the
primacy and inviolability of the original, to stress the intertextual transforma­
tive streak in all writing, to assert the plurivocality of discourse. And to let in
plural voices means destabilizing and decentring the speaking subject, and
creates the prospect of a runaway inflation of voices and meanings.13
Translation therefore needs to be controlled, to be kept under the lid,
because it is recognized as an activity that continually risks producing this
kind of proliferation and dissemination. Translation is controlled through the
ideology of transparency, identity, reproduction, the translator's absence
from the translated text. This allows us to suppress the loose ends, the
hybridity of translation: we pretend they do not exist — or at least that they
THE TRANSLATOR'S VOICE IN TRANSLATED NARRATIVE 45

ought not to exist, so as not to endanger the notion of univocal speech, the
single voice issuing from an identifiable source. And so we say that we read
Dostoyevsky, that we hear Yeltsin: we do not want to hear the Translator's
voice. However, as this essay has tried to demonstrate, that discursive pres­
ence is there — not only in exceptional circumstances in unusually complex
books, but always. For in the final analysis it is impossible not to ask how far
the plurivocal nature of translation extends. A translation as a rule addresses
an audience different from that addressed by the original. If this adjustment
calls for the recognition of a Target-Culture Implied Reader, for positing an
Implied Translator and for the possibility of discerning the Translator's
discursive presence in certain cases and under certain circumstances, then
there is nothing to prevent extending this principle from translated narrative
to translated texts in general. Translation is irreducible: it always leaves loose
ends, is always hybrid, plural, and different.
Author's address:
Theo Hermans • Department of Dutch • University College London • Gower
Street • LONDON WC1E 6BT • England

Notes

1. The notion of translation as re-enunciation, and its relation to quotation and reported
speech, is taken from Barbara Folkart (1991).
2. Different theorists use different terms for the various elements in the diagram. The triad
'text', 'story' and 'fabula' is from Mieke Bal (1985). I follow Rimmon-Kenan (1983),
Chatman (1978, 1990) and others in assuming an Implied Author and an Implied Reader,
despite Genette's reservations (1983).
3. This makes the present essay complementary to Giuliana Schiavi's "There is Always a
Teller in a Tale" (in this issue), which argues the case for positing an Implied Translator
in the translated text. My focus is on those instances where another discursive presence,
another voice, becomes discernible in the text itself. Both essays posit an Implied Reader
specific to the translated text.
4. In the case of a paratextual note there is a difference between a Translator's Note and
other possible Notes. In the context of narrative fiction Notes would normally be
attributed to a Narrator, and signal the intervention of a different Narrator or a switch to a
different narrative mode (as in, say, John Fowles' The French Lieutenant's Woman). In
principle, Translator's Notes could be presented in exactly the same way and be indistin­
guishable from other Notes (i.e. be detectable only through comparison with the original).
However, in our translational culture the translator's professional ethos and other institu-
46 THEO HERMANS

tional rules and conventions forbid this option: Translator's Notes are normally identified
as such. This gives them a different status, which raises a theoretical problem. The main
point, however, is the observation that Translator's Notes break through the narrative
discourse in a way different from other Notes; the voice which produces the Translator's
Notes is clearly a different voice, with an identity of its own.
5. In Ulysse gramophone (1987) and in Du droit à la philosophie (1990); see Segers 1994.
6. I will henceforth refer to the Dutch text (volume 1 of the 1992 edition) as MH. The
translations will be referred to by the name of the translator (Siebenhaar, Edwards,
Carrasquer, Garros) immediately followed by the page number.
7. In the fifth edition (1881) the dedication is "To the revered memory of Everdine Huberte,
Baroness van Wijnbergen, loyal wife, heroic, loving mother, noble woman" (Dutch: MH:
1). The manuscript (printed edition: Multatuli 1949) bears only the dedicatee's initials:
"To E.H.v.W." (Dutch: Multatuli 1949: 1). The importance of the dedication will become
clear later.
8. 'N.d.T.' stands for 'Note du Traducteur', i.e. Translator's Note. Interestingly, the abbre­
viation makes it clear that the producer of the Note is the translator as a textual presence,
a voice, not to be confused with the Biographical Translator. How else to explain the use
of the masculine (generic?) Traducteur' when it was Mme Roland Garros who translated
the book?
9. In view of what was said in Note 4 above, this obviously raises the question of how
Multatuli's Notes (from the 1875 edition on) relate to Translator's Notes. Both are para-
textual, and as such interrupt the narrative flow. Since 'Multatuli' is both a Narrator in the
text and the name on the cover and title page of the book (which, as we shall see, plays on
the distinction between fiction and non-fiction), and since the introduction to the Notes
(MH-.239-246) clearly indicates that he is their provider, the linguistic subject of these
Notes is beyond doubt. They appear as a kind of editorial intervention, associated with the
name on the cover and title page rather with the T, Multatuli' who enters as a Narrator in
the book's final pages. This is partly because Multatuli-as-Narrator does not appear until
the very end, and partly because many of the comments go well beyond the novel's
narrative world, e.g. in commenting on the editor of the first edition of 1860.

10. The term 'differential voice' is Barbara Folkart's, who in the section in question develops
an argument similar to mine: " . . . le ré-énonciateur ne se manifestera dans l'énoncé qu'il
produit que sous forme de déviances, tant pragmatiques et référentielles que
sémiologiques". She goes on to speak of "un ensemble hétéroclite de déviances (dont la
saisie requiert une analyse plus ou moins poussée, du moins une confrontation avec le
texte de départ . . .)" (Folkart 1991: 394-395). As I hope to have shown, the Translator's
'differential voice' can be seized in the translated text itself, without confronting it with
the source text.
11. The following discussion, like all such discussions, is indebted to Sotemann 1972, the
classic and detailed study of the structure of Max Havelaar. Since Sötemann's study
refers to the printed edition of the manuscript (Multatuli 1949) it does not address issues
like those raised in Note 9 regarding the 'editorial' status of the Notes in the 1875 and
subsequent editions.
12. Everdine van Wijnbergen died on 13 September 1874, hence the extended dedication in
the fourth edition, which came out in October 1875 (Multatuli 1992, II:538). Although in
THE TRANSLATOR'S VOICE IN TRANSLATED NARRATIVE 47

the manuscript and the first three editions of Max Havelaar the initials E.H.v.W. do not
explicitly identify the dedicatee as the author's wife, the conclusion seems inescapable,
as indeed Sötemann (1972: 23, 275ff.) has shown.
13. This is an idea which Karin Littau developed in several seminars and conference papers,
with reference to Foucault, Barthes and Derrida. See Littau 1993.

References

Bal, Mieke. 1985. Narratology: Introduction to the Theory of Narrative, tr. Christine van
Boheemen. Toronto: University of Toronto Press.
Chatman, Seymour. 1978. Story and Discourse: Narrative Structure in Fiction and Film.
Ithaca-London: Cornell University Press.
Chatman, Seymour. 1990. Coming to Terms: The Rhetoric of Narrative in Fiction and
Film. Ithaca-London: Cornell University Press.
Derrida, Jacques. 1991. Between the Blinds: A Derrida Reader, ed. Peggy Kamuf. New
York etc.: Harvester Wheatsheaf.
Derrida, Jacques. 1992. Acts of Literature, ed. Derek Attridge. New York-London:
Routledge.
Descartes, René. 1968. Discourse on Method and the Meditations, tr. F.E. Sutcliffe.
Harmondsworth: Penguin.
Folkart, Barbara. 1991. Le conflit des inondations: Traduction et discours rapporté.
Candiac: Les Éditions Balzac.
Genette, Gérard. 1972. "Discours du récit". Gérard Genette. Figures III. Paris: Seuil,
1972. 65-281.
Genette, Gérard. 1982. Palimpsestes: La littérature au second degré. Paris: Seuil.
Genette, Gérard. 1983. Nouveau discours du récit. Paris: Seuil.
Genette, Gérard. 1991. Fiction et diction. Paris: Seuil.
Harris, Brian. 1990. "Norms in Interpretation". Target 2:1. 115-119.
Hermans, Theo. 1988. "On Translating Proper Names, with Reference to De Witte and
Max Havelaar". Modern Dutch Studies: Essays in Honour of Peter King, ed. M.
Wintle. London: Athlone, 1988. 11-24.
Littau, Karin. 1993. "Intertextuality and Translation: The Waste Land in French and
German". Translation — the Vital Link, ed. Catriona Picken. London: Chamelon
Press, 1993. 63-69.
Multatuli. 1927. Max Havelaar or the Coffee Sales of the Netherlands Trading Company,
tr. W. Siebenhaar, intro. D.H. Lawrence. New York.
Multatuli. 1949. Max Havelaar of de kofflj-veilingen der Nederlandsche Handel-
maatschappij, ed. G. Stuiveling. Amsterdam: Van Oorschot.
Multatuli. 1968. Max Havelaar, tr. Mme Roland Garros. Paris: Editions universitaires.
Multatuli. 1975. Max Havelaar o las subastas de café de la Companía Comercial
Holandesa, tr. Francisco Carrasquer. Barcelona: Frontera.
Multatuli. 1987. Max Havelaar or the Coffee Auctions of the Dutch Trading Company, tr.
Roy Edwards. Harmonds worth: Penguin.
48 THEO HERMANS

Multatuli. 1992. Max Havelaar of de koffiveilingen der Nederlandsche Handelmaat-


schappy (2 vols.), ed. A. Kets-Vree. Assen-Maastricht: Van Gorcum.
Pym, Anthony. 1992. Translation and Text Transfer: An Essay on the Principles of
Intercultural Communication. Frankfurt am Main etc.: Peter Lang.
Rimmon-Kenan, Shlomith. 1983. Narrative Fiction: Contemporary Poetics. London-
New York: Methuen.
Schiavi, Giuliana. 1996. "There Is Always a Teller in a Tale". Target 8:1. 1-21.
Segers, Winibert. 1994. "Derrida's onvertaalbaarheidsweb". Bouwen aan Babel: Zes
opstellen over onvertaalbaarheid, ed. Raymond van den Broeck. Antwerpen-
Harmelen: Fantom, 1994. 89-100.
Sötemann, A.L. 1973. De structuur van de 'Max Havelaar'. Groningen: Wolters-Noord-
hoff.

You might also like