Professional Documents
Culture Documents
The Benjamins Translation Library aims to stimulate research and training in transla-
tion and interpreting studies. The Library provides a forum for a variety of approaches
(which may sometimes be conflicting) in a socio-cultural, historical, theoretical,
applied and pedagogical context. The Library includes scholarly works, reference
works, post-graduate text books and readers in the English language.
GENERAL EDITOR
Gideon Toury (Tel Aviv University)
ADVISORY BOARD
Marilyn Gaddis Rose (Binghamton University)
Yves Gambier (Turku University)
Daniel Gile (Université Lumière Lyon 2 and ISIT, Paris)
Ulrich Heid (University of Stuttgart)
Eva Hung (Chinese University of Hong Kong)
W. John Hutchins (University of East Anglia)
Zuzana Jettmarová (Charles University of Prague)
Werner Koller (Bergen University)
Alet Kruger (UNISA)
José Lambert (Catholic University of Leuven)
Franz Pöchhacker (University of Vienna)
Rosa Rabadán (University of León)
Roda Roberts (University of Ottawa)
Juan C. Sager (UMIST, Manchester)
Miriam Shlesinger (Bar-Ilan University, Israel)
Mary Snell-Hornby (University of Vienna)
Sonja Tirkkonen-Condit (University of Joensuu)
Lawrence Venuti (Temple University)
Wolfram Wilss (University of Saarbrücken)
Judith Woodsworth (Mt. Saint Vincent University, Halifax)
Sue Ellen Wright (Kent State University)
Volume 39
Andrew Chesterman, Natividad Gallardo San Salvador and Yves Gambier (eds.)
Translation in Context
Selected contributions from the EST Congress, Granada 1998
Edited by
ANDREW CHESTERMAN
University of Helsinki
NATIVIDAD GALLARDO SAN SALVADOR
University of Granada
YVES GAMBIER
University of Turku
Introduction ix
Part I: Conceptual analysis
Sandra Halverson
Prototype effects in the "translation" category 3
Cay Dollerup
"Relay" and "support" translations 17
Brian Mossop
The workplace procedures of professional translators 39
Hanna Risku & Roland Freihoff
Kooperative Textgestaltung im translatorischen Handlungsrahmen 49
Candace Séguinot
Knowledge, expertise, and theory in translation 87
Astrid Jensen & Arnt Lykke Jakobsen
Translating under time pressure 105
Paul Kussmaul
Types of creative translating 117
M T. Bajo, F. Padilla & P. Padilla
Comprehension processes in simultaneous interpreting 127
Nedret Kuran-Burço lu
At the crossroads of translation studies andimagology 143
Ritva Leppihalme
Foreignizing strategies in drama translation 153
Alexandra Lianeri
Translation and the shaping of modern democracy 163
Tiina Puurtinen
Translating linguistic markers of ideology 177
Kristiina Taivalkoski
Les simplifications narratives dans une traduction française
de Joseph Andrews 187
Jeremy Munday
Seeking translation equivalents: a corpus-based approach 201
Anja Schwärzt
Machine translation for translators? 211
Martin Kaltenbacher
Eliana Franco
Documentary film translation: a specific practice? 233
K. Malmkjær, J. Milton & V. Smith
Translation and mass culture 243
Masa Nomura
Text, image and translation 261
Christina Valdés
Reception factors in translated advertisements 271
Rainer Kohlmayer
Frank Wedekind's sex tragedy Lulu in English and French versions 293
Leticia Herrero
Sobre la traducibilidad de los marcadores culturales 307
Åse Johnsen
El mundo de [la filo]sofïa 317
Anne Martin
Institutional references from the English-speaking world
in the Spanish press 327
Stefania Taviano
Paul Bandia
Towards a history of translation in a (post)-colonial context: an African
perspective 353
Emine Bogenç Demirel & Hülya Yilmaz
La critique de la traduction littéraire en Turquie (1940-1992) 363
Index 391
Introduction
Translation Studies itself has been through many transformations since James
Holmes first approached the issue of naming the discipline and drew up a map
of Translation Studies which still seems surprisingly modern thirty years later.
Over the past two decades, the context of Translation Studies has been shifting
and expanding at an increasing rate.
Texts are now more specialised than in the past and translation is
inconceivable without documentation techniques, both factual and
terminological. Translation practices have undergone a radical change with the
incorporation of computer-based tools. Translation is no longer an isolated
activity; it has come into contact with, and often overlaps, other activities such
as documentation, publishing, technical and professional writing, and editing.
Analysis of the translation process has integrated new psycholinguistic and
cognitive aspects. Cultural, sociological, political, and situational factors have
been introduced that place translation within a broad social context. Translator
X Introduction
Translation in Context
Andrew Chesterman
Natividad Gallardo
Yves Gambier
PART I
Conceptual analysis
PROTOTYPE EFFECTS IN THE "TRANSLATION"
CATEGORY
Sandra Halverson
Department of English
University of Bergen, Norway
Zusammenfassung
Diese Untersuchung sucht empirisch zu bestätigen, daß die interlinguale Übersetzung
den zentralsten Bereich der Prototypenkategorie darstellt, die wiederum mit
verwandten Kategorien nach festgelegten Prinzipien verbunden ist. Studenten werden
um Abgabe von Werturteilen über eine Reihe von Übersetzungspaaren gebeten. Die
Zusammenstellung der Textreihe diente folgendem Ziel: Prüfung der Zentralität der
Sprache als ein System unter anderen und der Relevanz der Bewegung zwischen
semiotischen Systemen oder innerhalb derer. Die Haupthypothese beinhaltete die
Annahme über die Betrachtung der Beispiele mit intersystemischer Bewegung als das
beste/die besten Beispiel(e), insbesondere die Beispiele mit Bewegung zwischen zwei
deutlich distinkten Sprachen. Die Ergebnisse bestätigten die Hypothese.
Résumé
Dans cette contribution, nous avons cherché une base empirique à l'idée que c'est la
traduction interlinguistique qui est centrale dans une catégorisation prototypique. Nous
avons demandé à des étudiants d'évaluer des paires de traduction, notre objectif étant
de tester à la fois la place essentielle de la langue parmi les systèmes sémiotiques et
l'importance du mouvement entre ou à l'intérieur de ces systèmes. Notre hypothèse
principale a été que les meilleurs exemples seraient ceux portant sur le transfert
intersystémique, plus précisement entre deux systèmes linguistiques bien distincts. Les
résultats confirment l'hypothèse.
Resumen
Este estudio tiene como objetivo demostrar empiricamente que el concepto de
traducción interlingual es el elemento principal de una categoria prototipo que a su vez
esta muy ligada a distintas categorias relacionadas. Se pidió a un grupo de estudiantes
que seleccionaran los mejores ejemplos entre una serie de pares de traducción
disenados para comprobar la centralidad de los sistemas semióticos. La hipótesis
partia de la base de que los mejores ejemplos estarian relacionados con el movimiento
4 Sandra Halverson
1. Introduction
2. Theoretical background
3. Methodology
It is important to note that, in the references to Toury's typology, not all of the
labels are at the same level of generality. Thus, while the seven text pairs on
the questionnaire are all at the lowest level of generality in Toury's typology,
the superordinate labels given to pairs one, two, three, and seven in Table 1 are
merely an artifact of the labeling system employed by Toury, and the lack of
conventional (or natural language?) designations for these particular forms.
A further note is necessary on text pair five, as it is an exception to the
overall system. The two texts in pair five are both linguistic, and thus are
assigned a value of 1 on that particular dimension. The two are the same type
Prototype effects 7
4.1 Results
The mean scores and standard deviations for the seven translation pairs are
given in Table 2. Readers may recall that the seven-point scale ranged from 1
(yes, completely sure) to 7 (no, completely sure). A response of four indicated
"do not know".
8 Sandra Halverson
Table 2. Mean scores and standard deviation for each text pair
It is clear from Table 2 that only one of the pairs was clearly excluded from the
category, i.e. text pair seven, intersemiotic (non-linguistic) translation
(translation from a visual (painting) system to an aural one (musical)). All
other text pairs were identified as translations, though the mean scores indicate
that the degree of certainty varied. Furthermore, the means clearly indicate
graded membership; text pair number four (interlingual translation) was found
to be the most central member, while text pair number one (intrasystemic) was
particularly close to the "do not know" score. The remaining pairs also showed
varying degrees of membership.
Interestingly, the most central member, text pair four, also showed the
lowest standard deviation, which indicates that the group of respondents was
most agreed on this particular pair. It is also interesting that as the mean
approaches 4.0 ("do not know"), the standard deviation also increases, and
there is a direct correlation. The same pattern is shown for both yes and no
responses; in other words, the standard deviation is lower for text pair number
seven, the only pair excluded from the category, than it is for text pair six, i.e.
intralingual translation, with respective means of 5.04 and 3.54. Thus, the
pattern is maintained. The conclusion that can be drawn is that there is greatest
intragroup certainty about the most central member, and that as average
uncertainty grows, so does the range of responses (a finding which is
intuitively appealing).
On the basis of the above, an additional test was carried out to test for
the overall significance in group responses. A one-way ANOVA (see Butler
1985: 132) indicated that both between-group and within-group variance was
significant (P = .0326 and P = .0001 respectively). A scheffé test was then used
to check the significance of differences between means, followed by a z-test.
(Full statistical details are available on request from the author.)
Results showed the score for text pair seven is significantly different
than those for all of the other pairs. In addition, the group of text pairs three,
four and five does not show significantly different ratings. Similarly, scores on
pairs one, two and six are not significantly different. Indeed, the six text pairs
Prototype effects 9
which the respondents included in the category seem to show a rather ripple-
like pattern around the center, with the most central member most immediately
surrounded by the (not significantly different) text pairs three and five, and, in
the next "ring" the texts which were significantly different from three four and
five, i.e. text pairs one, two and six, which in turn were not significantly
different from each other.
This may be illustrated in Figure 1, below, in which the dotted circles
are intended to represent permeable boundaries:
4.2 Discussion
The results presented in Section 4.1 provide clear evidence of prototype effects
in the "translation" category. The most obvious and immediate evidence is that
indicating graded membership, as indicated by the means and corroborated in
the statistical analysis of subject ratings. As pointed out in the introduction,
graded membership is one of the characteristics of a non-classical category.
The second characteristic, the absence of a clear category boundary constituted
by necessary and sufficient conditions, is more difficult given the methodology
employed and requires a somewhat more detailed look at the results presented.
A clear category boundary is demonstrated if any of the dimensions
tested constitutes a necessary and sufficient condition by means of which
instances may be included or excluded. The only text pair that was excluded
from the category was number seven, which exemplified intersemiotic, non-
linguistic translation. It had values of 1 on type relationship, 1 on token
relationship, and 0 on linguality. In other words, there was a difference in
semiotic system at both type and token level and there was no language
10 Sandra Halverson
In Section 4.2, it was shown that none of the three dimensions constituted
necessary conditions for membership, though the involvement of language as
one of the semiotic systems seemed to be a sufficient condition. The presence
of a boundary at either type or token level was not found to be necessary or
sufficient. It seems obvious, however, that the various combinations on these
dimensions are playing a role in the judgments that the subjects made.
Figure 1 illustrated the fact that text pairs three, four and five were not
significantly different, though text pair four (interlingual translation) was
clearly the most central member (as indicated by the means). All of these pairs
involved linguistic systems, and all of them (with the possible exception of
number five) demonstrated a boundary at the token level, i.e. an intersystemic
relationship between the pairs. In addition, pair number three showed a
boundary at the type level. Thus, the most central members all seem to involve
Prototype effects 11
at least one linguistic system and at least one boundary, more specifically, a
lower (token) level one.
In the next ring in Figure 1 are text pairs one, two, and six.
Interestingly, pair one does not demonstrate a boundary at any level; nor does it
involve language. This pair was probably close to being excluded from the
category, as it was closest to a score of 4 (don't know), and also showed the
largest standard deviation (2.634). Furthermore, two different statistical tests
gave different outcomes as to the significance of the difference between pairs
one and two. In other words, pair number one, at least relative to one other
member in the outer "ring", seems to be even more peripheral to the category.
At the very least, this suggests that the status of pair number one is less clear
than that of the other pairs within that group. It may provide further indication
of the overall borderline status of this particular pair with respect to the
category as a whole. Additional testing could illuminate this question further.
Text pairs two and six diverge from the central members along the language
and boundary dimensions, respectively. In other words, pair two has a
boundary at token level (but not at type level) and no language, while pair six
involves language, but no boundaries at either level. Thus, the category shows
typical radial structure, with extensions along the token-level boundary
dimension and the language dimension.
The results presented in Section 4.1 also provide information about the
relative salience of the main dimensions in various areas. In other words, the
involvement of a boundary and/or language is not equally important for each
pair judged. This is particularly clear with respect to the existence of a system-
level boundary. In order to investigate this issue in more detail, it is useful to
consider various combinations of text pairs which differ only along that
particular dimension. Text pairs one and two, for example, are similar on
dimensions one and three, while they differ only in that pair number two
involves two different systems, while pair number one does not. These two
pairs were not significantly different from one another according to the scheffé
test, though they were significantly different from each other according to the
z-test. This would suggest that the salience of a boundary in this area is
questionable.
If, on the other hand, one compares pairs four, five and six, the result is
slightly different. Also these three differed only along the token-level
dimension, and the presence of a clear boundary was decreasing from four to
six. Four had a clear boundary (between two distinct languages), five had a less
clear (or no) boundary (between a dialect and a standard language), while in six
there was no boundary. Pairs four and five, it will be remembered, were among
the most central members, while six was not. Pairs four and five were not
significantly different from each other, while both were significantly different
from pair six. Thus here, it seems that the presence of a boundary was
influential in the subjects' category judgments.
12 Sandra Halverson
The difference between text pairs one and two and the group of four,
five and six lies in the involvement of language. In pairs which are not
linguistic (the former), the involvement of a boundary seems to play less of a
role, if any. In pairs which are linguistic (the latter), on the other hand, the
clarity of the boundary does seem to be relevant. What this seems to indicate is
that the two dimensions, i.e. the presence of a boundary and the involvement of
language, are interacting, at least close to the center of the category.
In the interest of deriving the cognitive model, the relevant information
is that the involvement of language and the crossing of at least one, preferably
a token-level, semiotic boundary seem to be shared by all of the most central
category members. Furthermore, these dimensions, separately, seem to be used
productively in the process of extending category membership. Finally,
dimensions two and three seem to be interacting in a significant way, at least at
the center of the category. Thus, the cognitive model posited for this category
must account for the interplay of these two dimensions in the center, along the
periphery and outside the category (i.e. the exclusion of text pair seven).
5. Conclusion
APPENDIX - QUESTIONNAIRE
1 1 2 3 4 5 6 7
2 $10 69,-kr. 1 2 3 4 5 6 7
3 vikeplikt 1 2 3 4 5 6 7
Translation/explanation of Questionnaire
1. Translation of instructions :
Please indicate whether you believe that B is a translation of A, and circle the
number which best represents both your response and how certain you are. By
circling one of the numbers from 1-3, you are answering yes, that B is a
translation of A. By circling one of the numbers from 5-7, you are answering no,
that B is not a translation of A. If you are not able to answer either yes or no,
circle the number 4.
* The painter Victor Hartmann's paintings and drawings form the basis for the
composer Mussorsky's piece 'Pictures at an Exhibition', which has been described as
Mussorsky's review of his favorite pictures.
16 Sandra Halverson
References
BUTLER, C. 1985. Statistics in linguistics. Oxford: Basil Blackwell.
COLEMAN, L. & P. KAY. 1981. Prototype Semantics: The English word lie. Language 57(1),
26-44.
DERRIDA, J. 1985. Des Tours de Babel. Difference in Translation ed. by J. Graham, 165-207.
Ithaca and London: Cornell University Press.
HALVERSON, S. 1997. The concept of equivalence in translation studies: Much ado about
something. Target 9(2), 207-233.
HALVERSON, S. 1999. Conceptual work and the 'translation' concept. Target 11(1), 1-31.
KOLLER, W. 1995. The concept of equivalence and the object of translation studies. Target
7(2), 191-222.
LAKOFF, G. 1987. Women, Fire and Dangerous Things. Chicago: University of Chicago
Press.
MYERS, D. 1994. The Chinese morpheme gong. Cognitive Linguistics 5(3), 261-280.
PYM, A. 1995. European translation studies, Une science qui dérange, and why equivalence
needn't be a dirty word. TTR - Traduction, Terminologie, Rédaction 8(1), 153-176.
PYM, A. 1997. Koller's 'Äquivalenz' revisited: Review of Werner Koller's Einfüring in die
bersetzungswissenschaft'. The Translator 3(1), 71-79.
Cay Dollerup
Centre for Translation Studies,
University of Copenhagen, Denmark
Zusammenfassung
Im Fokus dieses Artikels steht die Sachlage, daß viele Übersetzungen nicht die Sprache
des Originals als Ausgangssprache aufweisen, sondern Übersetzungen aus einer
intermediären Drittsprache (Viert- usw.) darstellen. Da die meisten intermediären
Realisationen für wirkliche Rezipienten gemacht worden sind, wird vorgeschlagen,
diese Realisationen aus Gründen der Exaktheit als 'relays' (Relais) für andere Texte im
Unterschied zu 'indirect translation' (indirekte Übersetzung) zu bezeichnen. Der Term
'indirect translation' bezieht sich auf intermediäre Realisationen mit original keinen
Konsumenten. Dem Verfahren der Übersetzer, manchmal Übersetzungen in andere
Sprachen als in die eigene Zielsprache zu konsultieren, wird ebenfalls im Artikel
Aufmerksamkeit geschenkt. Die Bezeichnung 'support translation' würde sich auf
diese Fälle beziehen.
Résumé
Les traductions ne sont pas toujours faites directement de la langue de l'original vers la
langue d'arrivée mais passent par une troisième (quatrième, etc.) langue intermédiaire.
Comme ces produits intermédiaires ont été réalisés pour des récepteurs bien précis, on
pourrait les appeler "relais", servant pour d'autres textes d'arrivée, tandis que l'expression
"traduction indirecte" pourrait désigner les cas où les textes intermédiaires n'ont pas
d'audience réelle. L'auteur parle aussi de "traduction-soutien", en constatant que les
traducteurs s'appuient parfois sur des traductions produites en d'autres langues que les leurs.
Resumen
Este articulo se centra en el hecho de que a menudo la traducción no se realiza
directamente entre la lengua origen y la lengua término, sino que pasa por una tercera
(o cuarta, etc.) lengua intermediaria. Puesto que la mayoria de las actuaciones
intermedias se han realizado pensando en un publico real, se argumenta que seria más
preciso llamar a estas actuaciones "relevos" para otros textos término y reservar el
término "traducción indirecta" para los casos en los cuales no existen unos receptores
18 Cay Dollerup
reales para las actuaciones intermedias. El articulo senala, además, que a veces los
traductores consultan traducciones a otras lenguas diferentes de su propia lengua
término, utilizando asi lo que se podria denominar una "traducción de apoyo".
1. Translation as communication
Discussions focus on the relationship between the source-text original and the
target-language version. The terms employed about this relationship describe a
reality in which, even when there are intermediary realisations, the focus is on the
relationship between the original sender and the end consumer:
'Original sender ' > message > intermediary decoding and (near) simultaneous
encoding by transmitter > message > decoding by end receiver.
2. 'Indirect' translation
However, much interpreting, subtitling, and translation is not based on the text of
the original in the source language, but on realisations of the original in yet other
languages.
An interest confined exclusively to the relationship between the original
sender and the end consumer may occasionally be justified, provided the aim of the
communication is to convey the exact meaning between only two parties, two end
interlocutors. This was the case in a Danish murder trial, where a Thai witness made
a deposition in court as follows:
The woman's deposition reached the jury in circuitous ways, since it was first
rendered from Thai into English by one interpreter and then into Danish by yet
another interpreter. (Politiken. 12 September 1998. 1:4, cc. 5-6. My translation.)
3. Relay
Relay often also involves a delay in the arrival of the message in interlingual
transmission as opposed to an instantaneous - but not necessarily simultaneous -
20 Cay Dollerup
3. J Conference interpreting
The Common Market (now the European Union) had four official languages,
namely Dutch, French, German, and Italian when originally founded. The entry of
the UK, Ireland, and Denmark in 1973, added two more languages, namely English,
covering both Ireland and the UK, and Danish. (For background information see
Dollerup and Ceelen 1995.) At the European institutions, it is (in principle) the
prerogative of national delegates to speak their own language. Danish members of
the European Parliament exercised this right, and also listened to Danish inter
preters. Other nationalities noted that Danes were slow on the uptake, indeed
dimwitted. Clever journalists then noticed that there seemed to be degrees of dim-
wittedness among the members of parliament: if, for instance, a Dutch delegate
cracked a joke, the Dutch would laugh; after 5-10 seconds, the French and Germans
would get the point; after some more seconds, the English, the Irish, and the Italians
would catch on, and, then,finally,the Danes would join in the general merriment.
The figure of 5-10 seconds is traditionally cited in research on simultaneous
interpreting as the 'time lag' between the original sender's utterance and the inter
preter's rendition. This is the explanation for the ripple response described above:
• The Dutch speaker is understood by his countrymen, who laugh right away.
• The German and the French interpreters render the joke in German and French,
respectively. Consequently, after 5-10 seconds, the German and the French
listeners appreciate the joke.
• The English and the Italian interpreters do not know Dutch, so, understanding
German and French, they use these languages as their source texts and their
delegates laugh after 10-20 seconds. In this example, the Danish interpreters use
English as their source-text: 'dim-wittedness' is due to relay in which all
listeners use interpreted renditions.
Bureaucrats, politicians, and interpreters who have lost touch with the
profession argue that relay leads to confusion, since there must be numerous errors
in every transfer from any one language into another. Working interpreters tend to
agree with my views which are based on fairly systematic listening to relays:
indisputable errors may arise, but there are few (Dollerup 1987).
It is true that there are errors in the product of relay: however, these have
rarely anything to do with the phenomenon of relay, but are due to the first in
terpreter's misunderstanding of the original sender. Normally, these errors are
caused by the original speakers: delegates mumble, hit the microphone, turn their
"Relay" and "support" translations 21
heads, speak dialect, use slang, quote figures in incomprehensible ways, and,
without warning, use prepared manuscripts, thus shifting delivery into a written
mode without informing anybody, least of all the interpreters (for an interpreter's
views on such occurrences, see Pearl 1995).
Similarly, the use of relay may lead to errors in the product. Let us assume
that Danish speeches are relayed by way of Dutch to English and German, from
these languages to French and Italian, and so on. This gives a time lag of some 30
seconds before the Danish speech reaches Portuguese. Let us then assume - and it is
a fair assumption - that the chair of the meeting understands Dutch. When the
Dutch interpreter concludes the brilliant summing-up of the last Danish speech, the
chair logically assumes that this is the end and gives the floor to the next delegate,
say, a Greek. The Greek delegate opens with a rousing statement. Here is the snag:
the Greek original begins before the entire original (Danish) message has made it
through all relays; at the same time, the interpreters have to reorganise their chains
of relay. This sometimes forces the interpreters either to cut short the previous
speech, or omit the beginning of the new speech. The delegates notice that
something has gone wrong and assume it to be the interpreters' fault. We see that it
is not due to relay as such, but to the management of relay: it is the fault of the chair.
Overall, then, at the institutions where relay is most often seen by the public
(the European institutions, the UN), the process of relay does not lead to many
misunderstandings at the linguistic level, for the interpreters are professionals who
are trained to deliver well-phrased and easily understood speech.
On the other hand, relay interpreting is undoubtedly a risky undertaking
when there are no professional middlemen around: in my experience, it usually
means that communication takes place by means of sign language and gestures.
3.2 Translation
Relay and concomitant delay are widespread in translation, indeed so common that,
in literary studies, for instance, it is hardly noted at all. We do not consider it strange
that literature takes time to cross cultural barriers. We accept that Shakespeare was
not translated into Spanish until the 18th century.
Sometimes the original no longer exits. The Jesus of the New Testament
must have spoken Aramaic, but no Aramaic text has been preserved. Modern Bibles
are made from relay editions using different languages. The English Bible overseen
by John Wyclif (c. 1385) used the Latin Vulgate as its source text. The Vulgate
derived from St. Jerome's Bible (c. 400), itself a Latin translation of Greek sources
which St. Jerome subsequently revised by means of Hebrew texts. William
Tyndale's English Bible of 1526 was based on Greek texts. Knowing no Greek,
Miles Coverdale (1535) used Tyndale's version of the New Testament and
translated the Old Testament from Hebrew and from Martin Luther's Bible, thus
also involving German. (See Delisle and Woodsworth 1995.) The main point is that
22 Cay Dollerup
some original source texts of the Bible no longer exist, so that translations are based
on relay translations in languages where, like the renditions of present-day
conference interpreters, they have had audiences of their own: in Greek, in Latin,
and in German.
Today, most translation activity is concerned with texts that are not intended
to last for generations, but are accepted as transitory and short-lived, such as letters,
instructions, manuals, used in industry, in trade, and in tourism. In these contexts,
translation is carried out soon after the appearance of the original, if at all: there is
no point in translating old manuals, instructions for obsolete tools, out-of-date
tourist brochures. Most translation in the modern world is therefore comparable to
consecutive interpreting: once the message is complete in the source language, it is
followed by a translation. In principle, it is immaterial whether an interval of five
minutes or one year elapses, for the point is that we are not talking about a century
or more between the appearance of the original and its translation(s).
Literature, however, offers many examples of relay in which the original is
still extant and where translation is both relayed and delayed.
The Tales of the brothers Grimm were initially published in 1812-1815 in
German. In 1823 an English translation by Edgar Taylor of selected tales achieved
success among English audiences. This collection - and its successors in English -
also provided source texts, relays, for numerous translations into other languages
(Sutton 1996, Dollerup 1999). Similarly German translations of fairytales first
published by the Dane Hans Christian Andersen, in 1835, were made for German
audiences as early as 1837 and enjoyed great success. Such German texts served as
the source texts for the first translations of Andersen into English and later into
central and southern European languages, for instance Slovene (personal
communication by Hans Vermeer and Silvana Orel). In subsequent periods direct
translations from Danish into English served as source texts for yet other target
translations into, for instance, Chinese (see e.g. Xu 1998).
I have a reason for singling out tales: they are popular and hence often
published. Accordingly, they offer a picture in miniature of relay, nowadays
especially so in European, indeed international, co-prints. In co-printing a publisher
in one country produces a book with illustrations and uses translations from national
publishers which fit these illustrations in terms of typography, that is, of length. The
point to note is that the text provided by the original publisher may already be a
translation and thus a relay of the original. This procedure is used extensively in the
publication and translation of children's literature and of illustrated textbooks. It is -
to the best of my knowledge - largely overlooked in comparative philological
translation studies.
The term relay highlights the dynamics in the interlingual movements of translation.
"Relay" and "support" translations 23
The end product of translation is static, but this static translation may be used for
other dynamic translation or interpreting processes, leading to yet other static
translations (Dollerup 1988). There is, in principle, no finality to this process,
translations of the same text can continue indefinitely, not only within the same
binary language pair, but also between languages and cultures. This is why relay is a
better term than indirect translation, which implies not only that intermediary
realisations are ephemeral, but also that the translation in hand is the only end
product imaginable.
It is, furthermore, a fact of life that literary translation is not highly
remunerated, and that translators commit errors, ranging from mistakes involving
false friends to complete distortions. Of course, most readers, including reviewers,
do not notice deviations because they do not collate the books they read for fun with
the originals; indeed there is no reason why they should read a translation at all if
they are well versed in the language of the original.
No matter how we define errors or deviations, it is clear that in written
public translations, each translator using relay will normally add new deviations to
those made by predecessors in the chain. In relation to the original, there is thus a
cumulation of deviations every time a work is relayed. In this respect, then, public
written translation differs radically from public professional simultaneous conference
interpreting, for thanks to the professionalism involved in delivery in relay
interpreting, nearly all errors occur between the original speaker and the first
interpreter and not in the chain of relay. Conversely, relay is indeed a major source
of deviations in written translation.
Cultural differences account for some deviations. Others are explicable in
terms of the presence or absence of the sender: in conference interpreting, the
presence of both senders and receivers in the same room, the body language of
senders, audiences, and other contextual features constitute strong corrective
elements. (See Gottlieb 1997 for errors in relay subtitling.) In literary translation,
relay translation (as well as delay) implies that the sender, the original author,
recedes into the background. The communicational chain is not complete. Fidelity
and loyalty to the author become weakened, not out of ill will, but for practical
reasons - the translator will not always be in a position to have the author elucidate
obscure points. In other words, the use of relay in translation shows that models of
the translation process which operate by means of unbroken chains from the original
sender to the final receiver are inapplicable to all the realities of translation work.
4. 'Support translation'
Relay translation should be distinguished from what I call support translation. This
is the strategy in which, translating a given source text, translators check translations
into languages other than their own target language in order to see whether
24 Cay Dollerup
5. Concluding remarks
References
ANDERSON, Kristine. 1997. "Karen Blixen's bilingual oeuvre: the role of her English editors".
Perspectives: Studies in Translatogy 5(2). 171-189.
BUSH, Peter. 1997. "Strawberry Flowers in Realms of Chocolate: The Training of Literary
Translators". The Changing Scene in World Languages. Issues and Challenges (ATA Scholarly
Monograph Series IX) ed. by Marian B. LABRUM, 109-117. Amsterdam & Philadelphia: John
Benjamins..
DELISLE, Jean & Judith WOODSWORTH. 1995. Translators through history. Amsterdam &
Philadelphia: John Benjamins.
DOLLERUP, Cay. 1988. "An ontological approach to translation and intranslatability". Semantik,
Kognition und Äquivalenz ed. by Gert JÄGER & Albrecht NEUBERT, 138-147. Leipzig: VEB
Verlag Enzyklopädie.
DOLLERUP, Cay. 1987. "Control of interlingual mediation in practice: Denmark as a case study".
Multilingual Journal of cross-cultural and interlanguage communication. 6(2), 169-190.
DOLLERUP, Cay. 1996. "Language work at the European Union". Translation Horizons Beyond
the Boundaries of Translation Spectrum ed. by Marilyn Gaddis ROSE, 297-314. Binghamton:
Center for Research in Translation.
DOLLERUP, Cay & CEELEN, Leo. 1996. A corpus of consecutive interpreting comprising
Danish, Dutch, English, French, German and Italian. Copenhagen: Centre for Translation Studies.
DOLLERUP, Cay. 1999. Tales and Translation: the Grimm TalesfromPan-Germanic narratives
to shared international fairytales. Amsterdam & Philadelphia: John Benjamins.
GOTTLIEB, Henrik. 1997. "Subtitling: diagonal translation". Subtitles, Translation and Idioms ed.
by Henrik GOTTLIEB, 107-134. Copenhagen: Centre for Translation Studies.
26 Cay Dollerup
PEARL, Stephen 1995. "Lacuna, myth and shibboleth in the teaching of simultaneous
interpreting". Perspectives: Studies in Translatology 3(2), 163-190.
SNELL-HORNBY, Mary, HÖNIG, Hans, KUSSMAUL, Paul & SCHMITT, Peter A. (eds). 1998.
Handbuch Translation. Tübingen: Stauffenburg Verlag.
SUTTON, Martin. 1996. The Sin-complex. A Critical Study of English Versions of the Grimms'
Kinder- und Hausmärchen in the Nineteenth Century. Kassel: Brüder Grimm Gesellschaft.
TOURY, Gideon. 1995. Descriptive Translation Studies and beyond. Amsterdam & Philadelphia:
Benjamins.
XU Yanhong. 1998. "The routes of translation". Perspectives: Studies in Translatology 6(1), 9-22.
EL CONCEPTO DE NUMERO DIFUSO APLICADO
AL PERFIL DEL TRADUCTOR
Abstract
The concept of fuzzy number as applied to the translator's profile.
This paper uses the mathematical concept of fuzzy number to quantify the
communicabihty degree of a given message (nothing, little, much, all) according to
the gap, both in time and space, existing between the transmitter and the receiver, and
the degree to which they share the same code. If communicabihty is not complete, it
can be retrieved by introducing a new element, the translator himself, who will have to
intervene to a greater or lesser degree in order to ensure a degrate communication.
Résumé
La notion de nombre flou appliquée au profil du traducteur.
Notre travail utilise la notion mathématique de nombre flou pour mesurer la
communicabilité d'un message donné (rien, peu, beaucoup, tout), en fonction de la
distance espace-temps qui existe entre l'émetteur et le récepteur et le degré avec lequel
ils partagent un même code. Cette capacité de communiquer, au cas où elle ne serait
pas totale, peut être rétablie au moyen de l'introduction d'un élément nouveau, le
traducteur, qui devra intervenir plus ou moins afin de rééquilibrer l'éventuel
déséquilibre communicatif.
Zusammmenfassung
Der Begriff der diffusen Zahl mit Anwendung auf das Profil des Übersetzers
In diesem Artikel wird der mathematische Begriff der diffusen Zahl verwendet, um
den Kommunikationsgrad einer bestimmten Mitteilung (nichts/wenig/viel/alles) im
Verhältnis zur existierenden räumlichen und zeitlichen Entfernung zwischen Emittent
und Rezeptor sowie dem Grad, in dem beide die gleichen Kode teilen, zu
quantifizieren.
Im Falle eines Mangels kann der gewünschte Kommunikationsgrad durch die
Einführung eines neuen Elements erzielt werden: durch den Übersetzer, der in mehr
oder weniger hohem Maße eingreifen muss, um eine potentielle.
28 Lourdes Pérez González
esta no resuite perturbada. Ahora bien, lograr este equilibrio puede conseguirse
sin intervención o puede requerir la intervención de un elemento nuevo.
Y es este nuevo elemento, esta conexión a veces ajena pero necesaria
(el Traductor), lo que restablece el equilibrio necesario para que un mensaje
comunique. Pero la necesidad del traductor será mayor o menor en función del
desequilirio existente en el acto comunicativo. Y para poder determinar esa
mayor o menor necesidad de traduction utilizaremos el concepto matemâtico
de difusividad que permitirâ analizar la falta de precisián, la indefinición, la
vaguedad con la que se presentan estas situaciones.
Este concepto de difusividad fue propuesto por L.A. Zadeh en 1965
(Zadeh 1965:338-353) como nueva via para simplificar el estudio de una serie
de problemas nacidos de la creciente aplicación de la matemática a numerosos
campos de la ciencia y de la técnica.
En la teoria clâsica todos los resultados con los que se trabaja son
exactos. Sin embargo, la experiencia nos demuestra que esto no es siempre
posible en la realidad. Ni son tan claras nuestras preferencias personales a la
hora de tomar una decision, ni son tan precisos los resultados de un
experimento.
Supongamos que queremos considerar personas altas y bajas y, para
ello, dividiremos el conjunto de las estaturas en dos subconjuntos A y B.
Diremos que una persona es alta si su estatura está en el conjunto B y
diremos que es baja si está en el conjunto A. Estos dos conjuntos son conjuntos
ordinarios. Sin embargo la frontera entre estaturas bajas y altas no es tan clara
como parecen indicar los conjuntos anteriores, pues hay una serie de valores
que tanto podrian estar en un conjunto como en otro. No obstante, aunque no
nos decidamos claramente a introducirlos en uno de los dos conjuntos si nos
aventuramos a decir que una estatura es mâs bien alta que baja o viceversa. La
introducción de este juicio subjetivo en la construcción de los conjuntos se va a
reflejar en la función de pertenencia y darâ lugar a subconjuntos difusos.
La diferencia fundamental con los conjuntos clâsicos es que mientras en
éstos los elementos pertenecen o no a ese conjunto (si o no, 0, 1, blanco,
negro), en los conjuntos difusos los elementos pueden pertenecer en mayor o
menor medida al subconjunto difuso (entre 0 y 1 hay infinitas posibilidades de
pertenencia), pertenencia que se expresa mediante una función que permite
evaluar (cuantificar) la presencia de un elemento dado en dicho subconjunto, lo
que esta mâs de acuerdo con la mayoria de las situaciones de la vida real y de
las aplicaciones cientificas y técnicas, donde las clasificaciones excluyentes no
ofrecen un modelo adecuado de la mayor parte de los fenómenos en estudio.
El numero difuso permite una semântica del tipo "aproximadamente
entre a y b" por lo que va a permitir modelizar la capacidad comunicadora de
los mensajes, representada ésta por el numero difuso obtenido al multiplicar los
numeros difusos que describen, respectivamente, la posición relativa entre
emisor y receptor y el grado de compartición mutua del código utilizado. Como
30 Lourdes Pérez Gonzalez
Las cuatro posibilidades del código: nada, poco, mucho, todo compartido, no
presentan dificultades de comprensión, ya que de lo que se trata es, en
definitiva, de los grados de "dominio" de una lengua o de un lenguaje, que un
receptor dado poseerá en mayor o menor medida.
Pero quizâ debamos insistir un poco más en las posibilidades de
emisor/receptor que cuantificâbamos como: presente, cercano, lejano e
inexistente, ya que incorpora las variables "en el tiempo y/o el espacio" y que
influirâ en el grado de comunicabilidad de los mensajes, similar a los niveles
de comprensión del mensaje definidos por Strawson (Strawson 1970) y que de
menor a mayor son: primer nivel (significacion lingüistica), segundo nivel
(significación referencial) y tercer nivel (significación completa).
Emisor/receptor presentes: se trata de presencia fïsica -que permite la
interpelación- por tanto de un código oral, siendo los contextos que movilizan
ambos netamente iguales. También podria tratarse de un texto escrito cuyo
contexto compartieran ambos.
Emisor y receptor cercanos: no existe presencia fïsica, ni posibilidad de
interpelación, se trata, pues, de un código escrito. Los contextos que movilizan
son diferentes - distancia espacio/temporal pequena - pero no sustancialmente
diferentes. Contextos culturales de paises cercanos - diferentes, pero similares-
o contextos históricos de momentos cercanos.
Emisor y receptor lejanos: no existe presencia fïsica ni posibilidad de
interpelación; se trata, también, de un código escrito. Los contextos que
movilizan son sustancialmente diferentes - distancia espacio/temporal grande.
Contextos culturales de paises remotos o contextos históricos de momentos
lejanos.
Emisor y receptor inexistentes. Se trata de casos inviables ya que al no
existir emisor/receptor no existe posibilidad de emisión de mensaje.
Las posibles combinaciones (16) nos va a permitir representar cuánto
comunica cada uno de los mensajes mediante el numero difuso producto de los
numeros que cuantifican los estados de las variables código y emisor/receptor
(Bonissone 1982:332). De modo que, en forma de cuadro, quedaria del
siguiente modo:
32 Lourdes Pérez González
Emisor/ código
Receptor a b C d
A 1 toda <0,5 mucha >0,5 poca 0 nula
B <0,5 mucha <0,5 mucha >0,5 poca 0 nula
C >0,5 poca >0,5 poca >0,5 poca 0 nula
D 0 nula 0 nula 0 nula 0 nula
Antes de entrar en el detalle de este cuadro, hemos de recordar que hay una
serie de casos inviables: aquellos en los que la variable emisor/receptor tiene el
valor cero, es decir uno de ellos o ambos no existen, ya que en la medida en
que no hay quien emita es imposible que haya mensaje, pero a efectos de una
mejor "visualización" seguiremos incorporândolos a los cuadros. Como
podemos apreciar, solo en un caso (emisor y receptor presentes, código
totalmente compartido) el mensaje comunica totalmente, en el resto de las
situaciones hay un desequilibrio, falta "algo" en mayor o menor medida para
que se logre la comunicación. Y, para compensar ese desequilibrio hemos de
introducir el concepto matemâtico de complementario respecto de la unidad,
que se obtiene mediante la diferencia entre los numeros difusos (Bonissone
1982:332) y que, en forma de cuadro, quedaria del siguiente modo:
Emisor/ código
receptor a b C d
A 0 nula >0,5 poca <0,5 mucha 1 toda
B >0,5 poca >0,5 poca <0,5 mucha 1 toda
C <0,5 mucha <0,5 mucha <0,5 mucha 1 toda
D 1 toda 1 toda 1 toda 1 toda
Este complementario, que equivale a "lo que falta" para que exista un
equilibrio entre los elementos que constituyen el acto comunicativo, cuantifica
la necesidad (mayor, menor o total) de descodificación-recodificación y de
acercamiento entre emisor y receptor - de traducción - para cada tipo de
mensaje aqui analizado. Analizaremos los cuatro casos que se nos presentan.
En el primer caso los valores que cuantifican los conjuntos
emisor/receptor y código son mâximos, es decir se trataría de un código
totalmente compartido y de la presencia de emisor y receptor.
Seria una comunicación nitida, transparente.
Puede tratarse de mensajes elementales, esencialmente informativos,
que no requieren ninguna competencia especial por parte del receptor, porque
El concepto de número difuso 33
los contextos del emisor y del receptor estân reducidos a cero. Imaginemos un
mensaje tipo: "esto es una mesa" o "son las tres y media".
Pero también puede tratarse de contextos activos pero iguales,
imaginemos una fórmula matemâtica comunicada entre dos expertos de igual
rango. En cualquiera de los casos, se trata de situaciones comunicativas nitidas,
que no requieren en absoluto traducción, ya que, como vimos, el
complementario es cero.
El segundo caso es la otra situación extrema, la contraria de la anterior,
es decir aquella en la que todos los elementos son cero. Situación en la que la
comunicación no es posible, ya que el grado de comunicabilidad es nulo, no
solo porque el código no es compartido en absoluto, sino porque emisor y
receptor son inexistentes.
En el tercer caso hay que incorporarle algo para que cumpla su función,
tiene que intervenir un elemento en principio ajeno pero absolutamente
necesario, (el Traductor), para que la comunicación exista. Este elemento
servirá para "compensar" la inicial no comunicabilidad, el inicial desequilibrio
existente entre los elementos de la comunicación. Existen varios tipos dentro
de este tercer caso según la mayor o menor necesidad de traducción requerida
por ciertos mensajes para que comuniquen. Y esta mayor o menor necesidad de
traductor, que estarâ más o menos cerca de los contextos del emisor y/o del
receptor y que tendrá que recodificar total o parcialmente es la que determinará
los distintos tipos de "ayuda" a la comunicación.
PRIMER TIPO
Que identificamos con los casos en que la necesidad de traducción es poca,
representados cuantitativamente por >0,5
Emisor/ código
receptor a b c d
A >0,5 poca
B >0,5 poca >0,5 poca
C
D
SEGUNDO TIPO
Que identificamos con las situaciones en las que el grado de comunicabilidad
es pequeno por tanto la necesidad de traducción es grande
Emisor/ código
receptor a b c d
A <0,5 mucha
B <0,5 mucha
C <0,5 mucha <0,5 mucha <0,5 mucha
D
En los tres primeros casos se trata de las mismas situaciones que en el epigrafe
anterior, incluyendo el caso de emisor/receptor lejanos, pero el código está
poco compartido, por tanto la necesidad de traducción es mayor. Imaginemos
las situaciones anteriormente mencionadas pero con mensajes emitidos en una
lengua solo parcialmente conocida por el receptor. Evidentemente se impone la
necesidad de traducción aunque, al no ser esta necesidad total, la traducción
puede ser ejercida por un traductor no experto. Imaginemos un mensaje
informativo emitido en un idioma que el destinatario sólo conoce parcialmente,
éste no captarâ el mensaje en su totalidad, pero la intervención de algün
espontáneo - en el caso de emisor/receptor presentes - que conozca la lengua
El concepto de número difuso 35
TERCER TIPO
Emisor/ Código
receptor a b c d
A 1 toda
B 1 toda
C 1 toda
D
Referendas bibliograficas
BONISSONE, P.P. 1982 "A fuzzy sets based linguistic approach: Theory and applications"
Approximate Reasoning in Decisions Analysis. 329-339. Amsterdam, New York, Oxford:
North-Holland Publishing Company
Brian Mossop
Government of Canada Translation Bureau and York University
School of Translation, Canada
Zusammenfassung
Eine Untersuchung zur Klärung der von Übersetzen bei ihrer Arbeit verwendeten
Prozeduren, wird vorgeschlagen. Wie ist die Abfolge der unterschiedlichen Aufgaben
beim Übersetzen (Interpretieren, Konstruieren, Recherchieren, Kontrollieren)?
Welche Kontroll- und Recherchiermethoden finden Anwendung? Sind die Prozeduren
anders bei einem kurzem, gutgeschriebenen Text als bei einem langen, mangelhaft
verfaßten Text? In diesem Artikel werden Schwächen des ÜberSetzungsunterrichts
wegen Nicht-Vorhandenseins einer Untersuchung dieser Art angesprochen. Behandelt
werden auch die möglichen praktischen und theoretischen Vorteile der Untersuchung
neben der interessanten Frage, warum keine Untersuchungen über diese Prozeduren
vorliegen.
Résumé
Nous proposons une enquête destinée à mettre à jour la méthode de travail employée
par différents traducteurs en milieu de travail. Dans quel ordre les différentes tâches
(interprétation, rédaction, recherche, vérification) sont-elles executées? Quelles sont
les méthodes de recherche et de vérification employées? La méthode varie-t-elle selon
que le texte à traduire est court ou long, bien ou mal rédigé? Nous indiquons les
faiblesses de tout conseil prodigué aux apprenants en l'absence d'une telle enquête,
les avantages pratiques et théoriques qu'on pourrait tirer de l'enquête, et les raisons
pour lesquelles la méthode de travail n'a pas été étudiée de façon systématique.
Resumen
Se propone la realización de un estudio para revelar los procedimientos seguidos por
varios traductores en el lugar de trabajo. Se investiga el orden seguido en la
realización de las distintas etapas (interpretación, composición, investigación,
revisión), cuáles son los métodos utilizados en la revisión e investigación y si se sigue
el mismo proceso para un texto corto bien escrito que para un texto largo mal escrito.
40 Brian Mossop
Se estudian los fallos que la ausencia de un estudio semejante puede producir en los
consejos ofrecidos a los principiantes. También se investigan los beneficios teóricos y
prâcticos que este trabajo puede aportar a la teoria y la práctica de la traducción, asi
como las razones por las que nunca se han estudiado estos procedimientos.
1. A Research Proposal
What we do not know, and need to know, is how translators distribute the
tasks over the phases. Each task can be performed during any phase (with the
exception of task 4, which obviously cannot occur during phase 1). One can try
to interpret a passage (task 1) either before or during drafting, or one can leave
a blank in the draft and interpret during the post-drafting phase. Similarly with
composing (task 2): one can jot down possible translations during pre-drafting;
or one can leave blanks during drafting and fill them in during the
post-drafting phase. Thus phases and tasks are not to be confused: checking
(task 4) is not a synonym of post-drafting (phase 3); composing (task 2) is not
a synonym of drafting (phase 2). 1
The purpose of this paper is to suggest why it would be worth carrying
out a large-scale study of work procedures in terms of tasks and phases. The
proposed study would consist of:
(1) systematically observing the procedures used by a large number of
The workplace procedures 41
The study might reveal - though I think this highly unlikely - that one
particular way of distributing the tasks over the phases, or one way of
organizing the research or revision work, makes it possible to produce higher
quality work within a given amount of time, or to complete jobs faster at a
given quality level, or even to achieve both higher quality and greater speed.
Of course, the procedure in question would not necessarily be adoptable by all
translators, since it may not be compatible with their personal psychologies.
A more likely outcome of the study would be that there are two or three
'best' procedures, all of which lead to similar results with respect to quality
and speed. And even more likely, I would guess, is that the best procedure will
vary with the type of project: short or long text, urgent request or not, familiar
or unfamiliar topic, well written or poorly written source text, translation for
publication or for information, and so on. It may become apparent early on in
the study that valid comparisons among translators can only be made for texts
having a given set of features (for example: short, familiar topic, poorly
written, urgent request).
It is of course also possible that almost any work procedure can yield
acceptable results. In other words, the secret of producing good translations
quickly may lie elsewhere - not in procedure at all.
It is unclear whether there would be any practical benefits for translation
students since the procedures used by experienced professionals may not be
conducive to learning translational skills. A simple example: professionals will
use either a unilingual or a bilingual dictionary to obtain information,
whichever is fastest, whereas students should be encouraged to use a unilingual
dictionary simply because they tend to be overreliant on bilingual ones.
In the absence of an empirical basis for advising professionals that this or that
procedure is better, advice given during professional development workshops,
or in the workplace to junior translators, will inevitably be based on some
combination of received prescriptions, experience and logic.
Professional development trainers use 'received prescriptions' when they
pass on concepts of how one ought to proceed - concepts remembered from
translation school or gleaned from books they have read or lectures they have
heard (for example: "start by reading the source text in its entirety, underlining
The workplace procedures 43
(1) The reader of the translation will be getting the meaning from the
translation, not from the source text. Therefore, during comparative checking,
read a sentence of the translation first, then the corresponding passage of the
source text in order to emulate the reader's experience.
The idea here is that if you read the source text first, you will already have the
intended meaning in your mind, and you may then erroneously read this
meaning into your translation. This certainly seems logical, but a study of the
kind advocated here might well reveal that the order of reading translation and
source text is completely unrelated to final quality.
Here are some more examples of logical advice about checking:
(3) Changes to the draft take time and there is a risk of introducing error while
making the changes. Therefore minimize changes.
The results of the advocated study may well prove all this advice wholly or
partly invalid. With regard to (2), it may turn out that with, say, half the
translators observed, one read-through was just as good as two: the extra time
required for two read-throughs did not lead to higher quality. With regard to
(3), it could turn out that this advice is valid, but only for a quarter of the
observed subjects, who do tend to introduce errors and would be best off
44 Brian Mossop
staying with their original drafts; others improve their translations significantly
by making large numbers of changes. With regard to (4), the study might show
that the extra time spent checking a printout rather than checking on-screen is
well worth the while because far more errors are caught; the best procedure
might prove to be checking on a printout but then saving time by inputting any
changes directly on screen rather than handwriting them first on the printout.
Of course, time permitting, an individual translator could conduct an
empirical study on herself in order to validate such logical advice. She could
save a batch of translations checked on screen and another batch checked on a
printout. If possible, she would allow for comparability between the batches in
terms of topic, urgency, length and writing quality. Then, after a few months,
she or a colleague would go over a sample of each of the two batches of
translations to see whether more errors were missed when checking was done
on screen. Such individual diagnoses would probably be necessary even if the
advocated large-scale study were to be carried out. For it is unlikely that, on all
matters for which advice is required, the study would point to one particular
approach (say, checking on a printout) as the best one for all translators.
A study of the sort advocated here would be very expensive. It would take
many researchers and much time to agree on the study methods, find subjects,
make the observations and analyse the results. For a study aimed at universal
validity, there would also be problems of international coordination. The best
starting point would perhaps be a small-scale study of work procedures in a
single large translating organization. This would make it possible to verify the
research methodology and it would eliminate the problem of trying to compare
subjects who are working to different quality standards.
However I do not think the main obstacle to the proposed study is
financial or logistical. Rather the main obstacle lies in the discipline of
46 Brian Mossop
Translation Studies (TS) itself. There are at least four features of TS which
discourage observation of translators' work procedures.
First, with the exception of think-aloud studies, TS is still strongly
oriented to the study of product rather than process. The translator is still not at
the centre of TS, even though some progress is being made in this regard.
Second, work in the area of translation pedagogy is almost entirely
oriented toward the teaching of students. Professional development - a regular
theme at congresses of practising translators - is practically never discussed at
TS conferences. As a result, the idea of observing professionals at work in
order to help them improve has not arisen within TS.
Third, the practitioners of TS are almost all university-based, but the
idea of studying workplace procedures would most likely seem attractive to a
practising professional. Unfortunately, few of us are active in TS.
Finally, TS remains strongly oriented to its source disciplines -
linguistics and literary/philosophical studies. The proposed study does not fall
within the scope of these fields. As a study of workplace procedures, it is most
closely related to the field variously known as methods engineering, work
design, and work measurement. Approaches that have been developed in these
areas might well be worth considering when designing the proposed study.
But what about the procedures, in the sense discussed in this article?
The idea of a standardized translation procedure is not nonsensical. However a
highly likely outcome of the proposed study is that while certain procedures
are indeed inefficient in terms of quality/speed, and should perhaps be
discouraged, the range of efficient procedures is quite large, so that there
would be no point in proclaiming one of them as a standard.2 Better to leave
the choice to individual translators.
Notes
1. Sager (1994:151-242) presents four phases (he treats interpretation of the commission as a
phase preceding the pre-drafting phase), each phase being divided into several stages,
such as the 'reading comprehension' stage and the 'dictionary look-up' stage (15 stages in
all). However he makes no distinction between a temporal phase and a task.
2. The German Standards Institute's proposed standard for the conduct of translation projects
(Deutsches Institut für Normung 1996) contains no specifications for procedures in the
sense discussed in this article. Most of the 12-page document is devoted to matters like
preparing the contract and billing. There is a very brief discussion of checking, but only to
list the things that must be checked (completeness etc.), not the procedure for doing so.
Note: The italicized phrases started reading and when draft completed mark the beginning of
the drafting and post-drafting phases respectively. Words in parentheses refer to
unobservables, which might be elicited through interviews.
Case A
1 Glanced very briefly through the ST (found topic was familiar, writing quality poor).
2 Looked at commission: 1000 words by 4 hours from now, for one person's information.
3 Started reading the ST and composing sentence-by-sentence, fixing up poor writing
while translating.
4 Did only a little research (when the concept was unclear). Instead left blanks, or
guessed and inserted ?-marks.
5 (Where the concept was clear but no TL word came to mind), sometimes briefly
consulted a bilingual wordbook or thesaurus.
6 Consulted an electronic database for abbreviations, as encountered.
7 (When a passage clarified a previous passage), went back to fill in blank or correct.
8 Rarely edited a sentence after composing.
9 When draft completed, consulted an SL colleague about some of the ?-marks and made
appropriate changes.
10 Called the author (about the meaning of remaining ?-marked passages) - author did not
call back - left ?-marks in the translation.
11 Did a full proofreading of the TT on screen for gross language errors but made no style
corrections. Looked at the ST on two occasions (when the draft seemed nonsensical)
and reworked the relevant passage of the translation.
12 Ran electronic spell check.
48 Brian Mossop
13 Made a printout and checked it for formatting errors. Counted paragraphs in ST and TT
(to make sure none omitted).
Case B
1 Paged through ST (found topic unfamiliar, but text well written).
2 Looked at commission: 9,000 words for 1 week from now; translation to be published
and used by specialized readership for many years; author unavailable.
3 Scanned the text (looking for keywords) and searched an electronic term bank for them.
4 Searched the Internet for parallel TL texts.
5 Searched an in-house library for parallel TL texts.
6 Read passages of the parallel texts and wrote a few notes in the margins of the ST.
7 Phoned to clarify the commission: there would be a subject-matter editor.
8 Started reading the ST and composing sentence-by-sentence.
9 (Where concept unclear), stopped to do research, but occasionally aborted the research
effort and inserted a ?-mark.
10 (Where concept clear but no TL word came to mind), often consulted a thesaurus or
collocation dictionary or bilingual wordbook.
11 Phoned for a copy of the intertext (because part of the ST proved to be a detailed
response to another text).
12 (When a passage clarified a previous passage), went back to fill in blank or correct.
13 Frequently edited a sentence after composing.
14 When draft completed, went to a university library to read some items in the article's
bibliography, and made changes in the translation accordingly.
15 Telephoned a TL subject-matter expert with questions and made appropriate changes.
16 Ran electronic spell check (to eliminate distracting typographical errors).
17 Printed out a copy, and did a full comparative re-reading for accuracy and
completeness, handwriting changes on the printout.
18 Input all changes, made a fresh printout and did a full language & style editing.
19 Separately checked transcription of numbers (because important to message).
20 Revised the title (in light of the actual content of the text).
21 Created the table of contents.
22 Ran a final spell check.
23 Made a printout and checked it for formatting errors.
References
CHANDLER, Daniel. 1993. "Writing Strategies and Writers' Tools". English Today 9(2), 32-
38.
Abstract
Today's complex translation projects require the co-operation of experts from
different fields. In this paper, we use the approaches of constructivism and situated-
ness to develop a theoretical basis for understanding co-operation in translation. We
apply this framework to authentic examples of co-operation between a freelance
translator and his co-operation partners. The two focal points of the analysis are as
follows:
1. Co-operation partners among such parties as the initiator of the process, the
commissioner, designers and developers of graphics or text, the external providers of
information, and the user or recipient, and
2. Fields of co-operation such as co-ordination, planning, developing, and evaluating.
Résumé
La complexité des projets de traduction actuels exige une coopération entre
spécialistes de différents domaines. Nous référerons au constructivisme et à l'analyse
situationnelle pour développer une base théorique utile à la compréhension de cette
coopération. Un tel cadre sera appliqué ensuite à des exemples réels de coopération
entre un traducteur indépendant et ses partenaires experts. Deux points surtout vont
retenir notre attention:
1. Les partenaires de la coopération, comme par exemple l'initiateur de la traduction,
le donneur d'ouvrage, les éditeurs chargés de la mise en page, les fournisseurs
d'information, le récepteur ou usager du document traduit;
2. Les moments de coopération, comme la coordination, la planification, le
développement et l'évaluation.
Resumen
En la actualidad, los proyectos de traducción complejos requieren la cooperación de
expertos en diversas areas. En este trabajo utilizamos el enfoque del constructivismo y
del análisis situacional para desarrollar una base teórica para estudiar el fenómeno de
la cooperación en la traducción. Aplicamos este marco a ejemplos reales de
50 Hanna Risku & Roland Freihoff
cooperación entre un traductor autónomo y los socios con los que coopera. Los dos
puntos esenciales del análisis son los siguientes:
1. La cooperación entre socios tales como el iniciador del proceso, el que encarga la
traducción, los disenadores gráficos o del texto, los proveedores externos de la
información y el usuario o destinatario y
2. Las areas de cooperación tales como la coordinación, planificación, desarrollo y
evaluación.
Als das menschliche Handeln und so auch das Übersetzen noch vor allem als
Befolgen gelernter Schemata und Umsetzung vorhandenen Wissens betrachtet
wurde, war die Sache einfach: die Beschreibungsformeln lieferten eine Theorie
und damit auch eine Methode. Konventionen und Methoden dienten als fertige
Handlungsanleitungen. Abweichungen von der üblichen Praxis und Sprach
verwendung wurden als situationsbedingte Sonderfälle behandelt. Diese Sicht
ist typisch für Sprachtheorien, die sich in der Auflistung von Merkmalen,
Scripts und anderer Wissenseinheiten erschöpfen (propositionale Theorien,
Symbolmanipulation, semantische Netze, frühe konnektionistische Ansätze)
und genauso typisch für Translationstheorien, die sich auf solche Sprach
theorien stützen bzw. Translation als die Anwendung erlernter Handlungs-
abfolgen beschreiben (z.B. äquivalenztheoretische und fertigkeitsbasierte
Translationstheorien).
Kooperative Textgestaltung 51
Die konkrete jeweilige Situation ist jedoch nicht etwas, womit wir
unsere vorgefassten Schemata auffüllen, sondern gerade sie ist die fundamen
tale Quelle und unverzichtbare Ressource für die Sinnbildung und damit für die
Handlungsfähigkeit. Schemata und Konzepte sind der Handlung vor- und nach
gelagert, sie bestimmen nicht die Handlung selbst, sondern dienen lediglich als
vage Pläne und nachträglich erstellte Erklärungen. Handlungen und Bedeu
tungen dagegen bauen sich immer im Hier und Jetzt auf. Die Aufgabe der
Schemata ist nur, uns in eine Position zu bringen, aus der heraus wir die
momentanen Ressourcen nutzen und eine sinnvolle Interpretation der Situation
samt ihrer verbalisierten Teile (der darin situierten Texte) entwickeln können.
Deshalb sind Situationen der Schlüssel zum Erlernen komplexer Aufgaben. Die
Expertenkompetenz liegt nicht im Vollziehen eines Plans, sondern im kontinu
ierlichen Erstellen neuer Hypothesen (Suchman 1987:188; Hendriks-Jansen
1996).
Wenn wir also jetzt das Konstruktive und Situative, das Kreative, an der Trans
lation zutage fördern, fragen sich viele, wie die daraus resultierende Unsicher
heit und Unvorhersehbarkeit zu bewältigen ist. Werden hier nicht Willkür und
Unkontrollierbarkeit gezüchtet? Schon bahnt sich eine Gegenreaktion an: die
Überlegung, ob wir nicht durch eine graduelle Vorfixierung der Texte, der
Übersetzungsprozesse und -aufgaben eine Stabilität erzwingen könnten. Dies
wäre jedoch eine totale Verkennung des Grundproblems: derartige Fixierungen
führen in eine erstarrte Scheinwelt, in der wir mit quasi-objektiven, letztlich
aber willkürlich festgelegten Prozeduren und zwischenkulturellen Äquivalenz
setzungen arbeiteten.
Der vorliegende Beitrag zeigt eine andere Lösung: die kooperative
Textgestaltung. Wenn alle Beteiligten in den Übersetzungsprozess eingebun
den werden, lassen sich Flexibilität, Kreativität und Verantwortung voll aus
schöpfen - ohne Realitätsverlust. Das ist durch die kognitiven und kommuni
kativen Grundlagen der Translation zu erklären: erfolgreiche Kommunikation
ist nicht auf Erlernung und Beherrschung vorgefasster Bedeutungen und
Konventionen zurückzuführen, sondern im Gegenteil auf unsere Fähigkeit, im
Interpretationsprozess ständig unsere Hypothesen zu prüfen, Missverständnisse
aufzudecken und zu bewältigen. Bei diesem Prozess der Kohärenzbildung grei
fen wir auf vielfältige situative Hinweise zurück: auf den Anlass, den Raum,
das bisher Gesagte, die beteiligten Personen. Wir leben nicht nur mit und in der
Unsicherheit, sondern wir benötigen sie sogar für komplexe intelligente Tätig
keiten. Darin besteht die eigentliche Leistung, und nicht etwa in der Beseiti
gung der Ambivalenz. Wir sollten nicht versuchen, diese fundamentale Un
sicherheit und Unvorhersehbarkeit wegzuerklären oder wegzuregulieren, son-
52 Hanna Risku & Roland Freihoff
dem uns auf die Ressourcen konzentrieren, die uns die Bewältigung der Fall
spezifik eröffnen (Suchman 1987: 69). Wesentlich sind also die Strategien des
Krisenmanagements. - Die durch Verallgemeinerungen und Vereinfachungen
gewonnenen, oft nur vermeintlichen „Standardfälle", in denen sich erlernte
Routineprozeduren oder Automatismen zur Texterstellung anbieten, werden
immer mehr mit Hilfe von Textbaustein- und Übersetzungsspeichersystemen
am Computer erledigt (Freihoff 1998).
Wenn die Forschung die Kontakte der Übersetzenden zu den anderen Beteilig
ten ignoriert, so verzichtet sie auf wesentliche Teile, die dieser Arbeit als
Hilfeleistung, als Ermöglichung von Verständigung für andere, erst einen Sinn
geben. Denn letztlich sind es die gegenseitigen Erwartungen der Beteiligten,
die die jeweiligen Ziele und den Handlungsrahmen für die translatorische
Produktentwicklung bestimmen. - Wir wollen die Diskussion auf den Boden
der Tatsachen holen und verwenden dazu das Rollenmodell von Holz-Mänttäri
(1984:106). Anhand von praktischen Erfahrungen erläutern wir die impliziten
und expliziten Erwartungen der Beteiligten.
EIN HANDLUNGSRAHMEN
Textge stalter, Textproduzent und Textdesigner sind Oberbegriffe für Verfasser und
Gestalter. Verfasser/innen schreiben Texte im Hinblick auf die situative Einbettung.
Gestalter/innen entwerfen (und verwirklichen) die drucktechnische Gestaltung
(Layout, Typographie).
3.2 Vertextung
Nun gilt es prospektiv zu denken und zu zeigen, wie für den Texter die Rah
menbedingungen sozusagen das Wurzelgeflecht bilden, auf dem die Texte
wachsen. Eine Grafik mit den verschiedenen Ebenen von Zielbildung und
Maßnahmen kann uns dabei helfen (s. Abb. 2).
2. Translator-Skopos
3. Vorgehensweise
4. Form
zer muss sich von der Welt der Endrezipienten durch Zeitschriften und Rückfragen
beim Bedarfsträger/Besteller ein Bild machen.
Als „Stamm-Freelancer" kennt der Übersetzer „Tradition" und Usancen der Firma und
der Branche. Für Recherchen nutzt er sein Paralleltextarchiv, Übersetzungsarchiv
(teils elektronisch), gedruckte Übersetzungen (Prospekte, Broschüren etc.), Glossare
(offizielle und eigene), Internet-Quellen und Ausgaben aktueller Fachzeitschriften, die
das Unternehmen ausleiht.
Experten (Marketing, Produktion, Forschung und Entwicklung) sind Partner für Sach
wissen und Terminologie. Sie bestätigen die Vorschläge, geben selbst den Terminus
vor, den sie unbedingt im Text haben wollen, oder sie erläutern das „Konzept", damit
der Übersetzer selbst weitersuchen kann. Oft vermittelt die Kommunikationsabteilung
den Kontakt und leitet Fragen des Translators weiter.
Sonstige Informanten sind Partner z.B. für Formulierungsfragen oder Sachwissen. Sie
haben mit dem Auftrag weder direkt noch indirekt zu tun. Es sind z.B. Bekannte mit
Spezialkenntnissen und Muttersprachler, die eine Formulierung beurteilen können.
AT-Gestalter als Partner sind Experten, die einen Fachartikel schreiben: Produkt
entwickler, ein Werbebüro bzw. dessen Texter, zuweilen auch ein Regisseur für Ton-
aufhahmen, wenn der Übersetzer Texte für Präsentationen selbst übersetzt, über
arbeitet und auch auf Band spricht. Die Gestaltung wird auch vom AT-Autorenteam
beeinflusst bei Termindruck und Absprachen. Auch die Niederlassungen als ZT-Appli
katoren sind Partner für Überprüfung und Feedback. -
Kooperative Textgestaltung 57
4. Ausblick
Literatur
CLARK, Andy 1997. Being there. Putting Brain, Body, and World Together Again.
Cambridge: The MIT Press.
HENDRIKS-JANSEN, Horst 1996. Catching Ourselves in the Act. Situated Activity, Inter
active Emergence, Evolution, and Human Thought. Cambridge: The MIT Press.
RISKU, Hanna 1998. Translatorische Kompetenz. Kognitive Grundlagen des Übersetzens als
Expertentätigkeit. Tübingen: Stauffenburg.
SUCHMAN, Lucy 1987. Plans and Situated Actions. Cambridge University Press.
TRANSLATION INTO A NON-MOTHER TONGUE
IN TRANSLATION THEORY: DECONSTRUCTION
OF THE TRADITIONAL
Zusammenfassung
Nur wenige Übersetzungswissenschaftler haben ernsthafte Überlegungen über die
Unterschiede zwischen Übersetzungen in die Muttersprache und aus der
Muttersprache des Übersetzers angestellt. Trotzdem wird oft die Annahme vertreten,
daß die Übersetzer nur in ihre eigene Muttersprache übersetzen sollten, um sprachlich
und kulturell akzeptable Übersetzungen zu produzieren. Auf Grund einer genauen
Lektüre von einigen grundlegenden Werken zur Übersetzung läßt sich diese
traditionelle Annahme jedoch auf aprioristische Auffassungen zurückführen. Es fehlen
wissenschaftliche Beweise für die tatsächliche sprachliche und kulturelle
Überlegenheit der Übersetzungen in die Muttersprache. Die Annahme ist entgegen
weitverbreitetem Glauben nicht traditionell oder universal akzeptiert - weder in der
Praxis noch in der Theorie des Übersetzens.
Résumé
Peu de chercheurs en traductologie ont analysé sérieusement les différences entre les
traductions faites vers la langue maternelle du traducteur et celles faites à partir
d'elles. Malgré tout, on affirme souvent que les traducteurs devraient travailler
exclusivement vers une langue, de façon à produire des versions acceptables, à la fois
linguistiquement et culturellement. Un regard attentif sur des ouvrages fondamentaux
en traduction révèle néanmoins que cette affirmation "traditionnelle" s'appuie plus sur
des conditions a priori que sur des preuves scientifiques. En dépit de l'opinion
courante, l'affirmation est loin d'être universellement acceptée, que ce soit du côté des
praticiens ou du côté des théoriciens.
Resumen
En los Estudios de Traducción, pocos estudiosos han considerado seriamente la
diferencia entre las traducciones directas e inversas realizadas por los traductores. A
pesar de esto, suelen pensar que los traductores sólo deberian traducir hacia su propia
62 Nike Kocijancic Pokorn
The assumption that translators can master only their mother tongue and must
therefore translate only in that direction developed rather late in the Western
world. In fact, it seems to have been Martin Luther who for the first time
assumed that one translated satisfactorily only into one's own language (Kelly
1979:110). His conviction was taken over and strengthened by the first and the
second, nationalist generation of Romantic authors, who also made a great
contribution to the rise of national philologies, especially because of the almost
obsessional identification of language and nation (Lefevere 1982:49). It was
perhaps Wilhelm von Humboldt who expressed most powerfully the
mystification of the native speaker and the special status the mother tongue had
at that period. Humboldt claimed that the nation is deeply connected with its
language, in fact that the nation's language is the spirit of that nation,1 which
means that only those who speak the language of a particular community can
access the hidden essence of the nation. That is also why the translator can
never write the way the author of the original would have written in the
language of the translator (Humboldt 1977:42), because the complete passage
from one language to another is impossible. Every language has its own way of
expression, which remains inaccessible to everyone who does not speak that
particular language from birth. Translation should therefore always proceed
from foreign languages to one's mother tongue and never vice-versa, since the
hidden essence of the target language is not attainable by any foreign speaker.
The roots of the conviction that we can grasp the ungraspable only in our
mother tongue, and consequently create a convincing translation only in our
native language, thus stem from this Romantic identification of the
transcendental nature of the nation and its language.
The claim that one should always translate into one's mother tongue
can also be found in contemporary writings on translation. Thus for example
Peter Newmark is short but blatantly direct in regard to this same problem; he
too states that translators cannot translate into a non-mother tongue and so
repeats and strengthens the "traditional" position.
Translation into a non-mother tongue 63
Translators should thus translate only into their mother tongues; even if a
person lives in a TL culture for years, his or her writing will be, according to
Newmark, "unnatural and non-native", full of "unacceptable or improbable
collocations" (ibid.). In spite of these statements, however this "indisputable"
stand reveals some considerable cracks. In examining Newmark's assertions
more closely, it seems difficult to avoid the following questions: Who is the
native speaker Newmark is talking about? The speaker of English, Scottish,
Canadian, American English or also the speaker of Singapore, Indian,
Pakistani, Nigerian English? Are the children of immigrants who are born in an
English-speaking country also native speakers of English or not? Is a person
who moved to an English-speaking environment in his/her childhood a native
speaker of English, and if so, when does childhood stop? What about pairs of
translators consisting of a native and a non-native speaker of English - which
language is their mother tongue?
Newmark does not answer these questions, and in fact, even if he had
wanted to, he would not have been able to provide final answers. The notions
of'native speaker" and "mother tongue" are far from being well defined in
linguistics; in fact, there is a considerable variation in the connotations
attributed to those terms, which seem to depend on the ideological position of
the person providing the definition, or at least on the motives hidden behind
his/her need to determine them (see Skutnabb-Kangas & Phillipson 1989:450-
477; Phillipson 1992:39).
Moreover, linguists have no answer to the question of when the critical
age for acquiring a particular language occurs (see Krashen 1981:72-76; Stern
1983:362-63; Littlewood 1984:66-67; Singleton 1992:47; Cook 1996:108); nor
do they, in spite of some attempts (e.g. Coppieters 1987, Davies 1991), have
any acid tests which would once and for all determine the difference between
native speakers and highly proficient non-native speakers of a particular
language (Paikeday 1985). Although we must agree that the majority of foreign
learners never achieve native-like competence, it remains unproven whether
some individuals nevertheless eventually succeed in being accepted by a
particular foreign community as native speakers of that language.
Newmark's categorical claims thus seem more than suspect. He seems
to ethnocentrically defend the notion of the superiority of the "natural native
64 Nike Kocijancic Pokorn
speaker", the innate state that can never be acquired, and thus rejects the
marginal and peripheral (i.e. translators from immigrant communities and the
practice of team translation) as necessarily inferior.2
It is the vagueness of the basic terms "mother tongue" and "native speaker"
that has most probably led many translation theoreticians to avoid the question
whether translators should be native speakers of the source or of the target
language; instead, they have idealised the subject involved in the process of
translation. In other words, translators are assumed by some scholars to be
perfect bilingual speakers of the source and target languages, translating from
one mother tongue to another.
We find the claim that translators should be bilingual for instance in J.
C. Catford (1965:27): "The discovery of textual equivalents is based on the
competent bilingual informant or translator", where the vagueness and
evasiveness of the notion "equivalence" is coupled with a similarly vague and
abstract notion of "a bilingual informant or translator". The notion of
translators as perfect bilingual or even bicultural persons can also be found in
theoretical works by Mary Snell-Hornby, e.g. where she writes that the aim of
all translation teaching is to create "not only a bilingual but also a bicultural (if
not multicultural) specialist working with and within an infinite variety of areas
of technical expertise" (Snell-Hornby 1992:11). Linguists argue that "real", i.e.
perfect, bilinguals are rare, if they exist at all:
The notion of proficiency raises some very complex issues. Again, the
"obvious" answer is to say that people are bilingual when they achieve native*-
like fluency in each language. But this criterion is far too strong. People who
have "perfect" fluency in two languages do exist, but they are the exception,
not the rule (Crystal 1987:362).
translation into one's mother tongue guarantees a good translation. This opaque
discourse on the translators' mother tongue reflecting the "traditional" view
can thus be found in Victor Hugo's writing on translation (Hugo 1992:18)6
through Friedrich Schleiermacher's "Über die verschiedenen Methoden des
Übersetzens" (Schleiermacher 1985:322)7 to Walter Benjamin (1982), where
he writes:
It is the task of the translator to release in his own language that pure language
which is under the spell of another, to liberate the language imprisoned in a
work in his re-creation of that work (Benjamin 1982:80).
Benjamin demands that a translator should not only master his own mother
tongue but also allow the foreign language to transform the target language by
means of translation and thus liberate the power of the pure, original language
- and this extremely important and difficult task seems to be only possible if
one is translating into one's mother tongue. By accepting the assumption that
translators always translate, or at least should translate, into their mother
tongue, Benjamin's work seems to reveal one of its sources - German
Romantic thought.
In addition to traditional writings, the hidden assumption that one
always translates into one's mother tongue can also be found in contemporary
theories - thus we can find it in Barbara Johnson's deconstructive thoughts
(Johnson 1985:142)8, as well as in Sherry Simon's feminist work (Simon 1996
94)9. But this hidden discourse is especially interesting and contradictory in
George Steiner's hermeneutic work on translation theory.
Steiner never openly discusses the translators' choice of the target
language, despite the fact that he repeatedly fails to determine his own native
language, supposedly possessing equal fluency in English, French, and
German. In fact, he even claims that he experiences his first three tongues as
perfectly equivalent centres of himself (Steiner 1992:120). Although he
recognises the difficulties in defining the notion of the mother tongue itself, he
indicates repeatedly that the TL of the translator should also be his mother
tongue. For example: "The translator labours to secure a natural habitat for the
alien presence which he has imported into his own tongue and natural cultural
setting" (ibid.:365). Or: "He [the translator] will import from abroad
conventions, models of sensibility, expressive genres which his own language
and culture have not yet reached" (ibid.:370). And finally: "[...] it is logically
conceivable that the translator, having gained great mastery over a source-
language, will conclude 'I understand this text but find no way of restating it in
my own native tongue "'(ibid. :372).
Although he cannot define his native language, and although he has
done some translation himself, Steiner never opens the question of choosing
one's target language in translation. The principle that the translator is allowed
66 Nike Kocijancic Pokorn
to translate only into his native language seems so deeply rooted in his thought
that he never challenges it. He also never disputes another principle, connected
to the first one and also typical of the canonised translational norms in the
English-speaking world - the principle of fluency and naturalness. And indeed,
the norm of fluency and naturalness has prevailed over other translational
strategies in English-speaking cultures and shaped the canon of foreign
literatures in English (see Venuti 1995). And since it was also agreed that
"perfect" fluency in the TL and the mastery of its different styles could only be
achieved in one's mother tongue, the norm that the translator (of at least
literary texts) should be a native speaker of the TL became widely accepted
too; indeed, it seems even more deeply grounded in English-speaking cultures
than the fluency principle.
But norms can change. Thus Lawrence Venuti challenges the absolute
validity of the norm of fluency and tries to get his readers to reflect on the
ethnocentric violence of a transparent, fluent translation, i.e. of a translation
that does not appear to be a translation but imposes itself as the "original"
(Venuti 1995:41) - he even pleads for the production of translations that reveal
"the linguistic and cultural difference of foreign texts" (ibid). But in spite of
his openness towards the foreign, and sensitivity to ethnocentric violence,
Venuti never touches the problem of the translator's TL, and thus accepts,
perhaps not consciously though, the prevailing and ethnocentric norm that
proclaims the superiority of TL translators.
Venuti seems to completely ignore the prevailing practice in peripheral
language communities, where many translators work into a language that is
non-native to them, and the deplorable fact in Western societies that many
translations are praised despite the fact that their "translators" did not
understand the SL10. In fact, he does not find this to be an issue of discussion;
thus he quotes Goethe translated by André Lefevere, a Belgian translating from
German into English (ibid. :99), and he explains the translational practice of
Ezra Pound, without mentioning his "Cathay" (1915), despite the fact that this,
probably the most praised of Pound's translations of Chinese poems, is also
famous for the fact that Pound did not understand Chinese when he translated
from E. F. Fenollosa's transcription of and commentary on the source text.
Moreover, Venuti emphasises that the TL should be the translators'
mother tongue, since he qualifies the language and culture the translator is
supposed to translate into as "domestic", and the SL culture as "foreign": "[...]
the translator's interpretive choices answer to a domestic cultural situation and
so always exceed the foreign text" (ibid.:37; emphasis added). Or: "Because
translation is a double writing, a rewriting of the foreign text according to
domestic cultural values, any translation requires a double reading - as both
communication and inscription" (ibid. :312). According to Venuti then,
translators choose a "foreign" text and translate it according to a "domestic"
Translation into a non-mother tongue 67
cultural situation, which implies that they never work away from their native
language but always into their mother tongue.
By not discussing the possibility of translations into a non-native
language, by undertheorizing the basic notions of "foreign" and "domestic",
Venuti thus accepts the prevailing practice in major-language communities,
where, indeed, translation is usually done into the translator's mother tongue.
But since he not only accepts this practice but also generalises it, he thus
imposes yet another ethnocentric norm on the rest of the peripheral world by
suggesting the necessity and thus superiority of translations into the translator's
mother tongue, i.e. he covertly imposes the superiority of the TL culture.
5 Conclusion
Notes
1. "Die Sprache ist gleichsam die äußerliche Erscheinung des Geistes der Völker; ihre
Sprache ist ihr Geist und ihr Geist ihre Sprache, man kann sich beide nicht identisch genug
denken" (Humboldt in Stolze 1994:24).
2. Newmark's influence is particularly felt in books on translation teaching. See also Alan
Duff (1981:111, 125 and 1989:11) and Geoffrey Samuelsson-Brown (1995:16).
3. "[ ... ] the only complication is that the source language communicator does not master the
receptor language. Therefore he needs the help of a bilingual person to produce a receptor
language stimulus that will communicate his informative intention" (Gutt 1990:143).
4. "[...] the translator must possess linguistic competence in both languages and
communicative competence in both cultures [...]" (Bell 1991:42).
Nike Kocijancic Pokorn
70
5. Lefevere and Bassnett are even more demanding: "Since languages express cultures,
translators should be bicultural, not bilingual" (Lefevere & Bassnett 1990:11).
6. "To translate a foreign writer is to add to your national poetry [...]" (Hugo 1992:18;
emphasis added).
7. "Seine [a translator's] erste Regel muß sein, sich wegen des Verhältnisses, in dem seine
Arbeit zu einer fremden Sprache steht, nichts zu erlauben was nicht auch jeder
ursprünglichen Schrift gleicher Gattung in der heimlichen Sprache erlaubt wird"
(Schleiermacher 1985:322; emphasis added)
8. "Through the foreign language we renew our love-hate intimacy with our mother tongue"
(Johnson 1985:142).
9. "Translation plays out in the open the 'everyday frustrations' of writing, projecting them
into an external form. We transfer our frustrations to the mother tongue [...]" (Simon
1996:94).
10. "Some of the most persuasive translations in the history of the metier have been made by
writers ignorant of the language from which they were translating (this would be so
notably where rare, "exotic" languages are involved)" (Steiner 1992:375).
References
BENJAMIN, W. 1982. "The Task of the Translator". Walter Benjamin, Illuminations ed. by
H. ARENDT, 69-82. London: Fontana.
COOK, V. 1996. Second Language Learning and Language Teaching. London, New York &
Sydney, Aukland: Arnold.
DUFF, A. 1981. The Third Language: Recurrent Problems of Translation into English.
Oxford: Pergamon Press.
GUTT, E.-A. 1990. "A Theoretical Account of Translation - Without a Translation Theory".
Target 2(2), 135-164.
KELLY, L. G. 1979. The True Interpreter: A History of Translation Theory and Practice in
the West. Oxford: Basil Blackwell.
LEFEVERE, A. 1995(b). "Is this the same text?: A Consumer's Approach to Translation".
Unpublished lecture at Cetra translation seminar in Leuven.
LEFEVERE, A. & S. BASSNETT (eds) 1990. Translation, History and Culture. London:
Pinter.
McALESTER, G. 1992. "Teaching Translation into a Foreign Language: Status, Scope and
Aims". Teaching Translation and Interpreting: Training, Talent and Experience ed. by C.
DOLLERUP & A. LODDEGAARD, 291-297. Amsterdam & Philadelphia: John Benjamins.
NIDA, E. 1964. Toward a Science of Translating: With Special Reference to Principles and
Procedures Involved in Bible Translating. Leiden: E. J. Brill.
NIDA, E. & TABER C. R. 1982. The Theory and Practice of Translation. Leiden: The
United Bible Society, E. J. Brill.
QUIRK, R. 1990. "Language Varieties and Standard Language". English Today 21(6/1).3-10.
SIMON, S. 1996. Gender in Translation: Cultural Identity and the Politics of Transmission.
London & New York: Routledge.
SINGLETON, D. 1992. "Second Language Instruction: the When and How". In Language
Teaching in the Twenty-first Century: Problems and Prospects. AILA Review (9).46-54.
STEINER, G. [1975] 1992. After Babel: Aspects of Language and Translation. New York,
Oxford: Oxford University Press.
VENUTI, L. 1995. The Translator's Invisibility: A History of Translation. London & New
York: Routledge.
TERMINOLOGY POLICIES, DIVERSITY, AND
MINORITISED LANGUAGES*
Zusammenfassung
Nach der Beschreibung der Unterschiede zwischen Minoritätensprache und
minorisierter Sprache, skizzieren die Verfasser die heutige Situation der Terminologie
in Katalonien, und zwar angesichts des Sprachenkontakts, der bis hin zur
Vermischung besteht. Diese Situation macht eine Erörterung der Funktion der
Terminologie notwendig: Ist sie auf die Standardisierung der Verwendung
ausgerichtet Anwendungen oder auf die Entwicklung von Besonderheiten der
jeweiligen linguistischen Gemeinschaft? Verschiedene Argumente sind vorgebracht
worden, aber offensichtlich lenkt der Status der betreffenden Sprachen die
Stellungnahmen. Einige Beispiele schliessen die Studie ab.
Résumé
Après avoir différencié langue minoritaire et langue minorée, les deux auteurs
esquissent la place de la terminologie aujourd'hui en Catalogne, quand il y a contact
de langues, parfois jusqu'au mixage. Cette situation oblige alors à s'interroger sur les
fonctions de la terminologie: vise-t-elle à standardiser les usages ou à développer les
spécificités de chaque communauté linguistique? Divers arguments sont avancés mais
d'évidence le statut des langues concernées infléchit les prises de position. Quelques
principes concluent la démonstration.
Resumen
Las autoras analizan la situación del catalan, lengua que hasta hace poco se
encuadraba dentro del grupo de lenguas minoritarias y que en la actualidad se
denominan lenguas menos favorecidas; las politicas terminológicas; la normalización;
la discriminación frente a otras lenguas en la investigación terminológica y la
necesidad de que las lenguas minoritarias del Estado se incluyan en los planes de
estudio de las Facultades de Traducción e Interpretación.
74 Aguilar-Amat & Santamaria
The concept of "minority" with respect to language is dynamic rather than static.
"Minority" is the expression of a relation not an essence. The relations can
assume two forms: diachronic and spatial. (p. 86)
languages); and we shall also distinguish between the terms majority language
and majoritised language. Viewed from this perspective, "all languages are
potentially minority languages" (Cronin, 1995:87-88). These concepts are also
relative, since a language can be majoritised in one context, for example,
Castilian in Spain, and minoritised in another, such as the position of Castilian in
Europe - only around 1% of texts produced in the European Union are written in
Castilian. In comparison with other labels such as expanding languages and
receding languages, the terms we propose here have the advantage of not
implying any given "direction".
In favour of the first option, Arntz & Picht (1989:171) defend the
standardising role of terminology as a means of ensuring unequivocal
communication among specialists. Efforts to clarify the linguistic resources of
specialised fields are part of a long tradition. Versalius (1514-1564), the
founder of modern anatomy, can be considered one of the precursors.
However, it was not until the second half of the 19th century that
terminological work became systematic. This was mainly due to the
technological revolution and the disordered coexistence of a whole range of
technological discoveries. At first, in Germany there were 15 different units of
measurement for electrical resistance, and eight different units for motor
power. Inevitably, this situation gave rise to expensive mistakes. As a result,
the industrialised states began to see the need to standardise objects, procedures
and their associated terminology.
However, although these authors also use the word "standardisation",
the objectives of this process on an international scale are totally opposed to
those of the standardisation of a minoritised language. In the former case,
certain lexical options are imposed on other languages, whereas in the latter
case, the aim is to preserve and strengthen the autochthonous terms. Therefore,
we shall refer to this international level as "homogenisation" or "international
standardisation", as opposed to simple "standardisation", as it was translated by
Amelia de Irazazâbal.
In the area of homogenisation, the International Electrotechnical
Commission (IEC) was founded in 1906 and set about the task of clarifying
electronic terminology. The DNA (Deutscher Normenausschluss - German
Standardisation Commission) appeared in 1919, followed by the ISA
(International Standards Association). Wüster (1898-1977) published his work
Internationale Sprachnormung in der Technik, besonders in Elecktrotechnik
(International Linguistic Standardisation in Technology, especially in
Electronics) in 1931, which was based on de Saussure's structuralism. The
main objections raised against this work were the decisive role assigned to the
idea of system and the belief that language cannot be standardised. Wüster's
reply was that he was not referring to everyday language, but only to
specialised language. Wüster assumed that international standardisation cannot
be conceived as static, that it is not equally meaningful in all specialised areas
and that it cannot be applied with equal intensity in all cases.
The controversy aroused by Wüster was interrupted by the outbreak of
the Second World War. In 1946 the ISO (International Standards Organisation)
was founded to replace the ISA. In 1950 a technical committee was formed to
deal with terminological matters (ISO/TC 37 Terminology - principles and co-
ordination). The committee completed the drafting of the basic standards in
1973, and these standards have been undergoing a process of revision since
1974. Other bodies such as the IEC (International Electric Committee) and
Technical Committee 7, also work in conjunction with the ISO.
Terminology policies 79
to replace an already established term with a new one. In many cases, the new
term and the old term continue to be used alongside each other as synonyms,
and this of course can lead to ambiguity.
Neologism is a valid resource for certain new or emerging fields of
specialisation, but it stands little chance of success if it used in an effort to
replace the traditional conceptual and lexical system. This point must be firmly
borne in mind when dealing with minoritised languages.
5. Conclusions
Notes
1. Tejo: a piece of roof-tile which is used as a toy. Tejuelo: a piece of metal which serves as a
support for a machine, a metal blank which will be made into a coin, or a piece of paper
attached to a book's spine for cataloguing purposes. Tirar/echar los tejos: to make verbal
attempts to seduce someone. A teja vana: covered in roof-tiles but without any interior
covering or ceiling. A toca teja: to pay in cash. De tejas abajo or de tejas arriba: of this
world or of a supernatural world.
84 Aguilar-Amat & Santamaria
2. United into networks and forums like Unión Latina, RITERM (Red Iberoamericana de
Terminologia), REALITER (Red Panlatina de Terminologia).
References
CABRÉ, M.T. 1992. La terminologia. La teoria, els mètodes, les aplicacions. Barcelon: Empúries.
PUEYO, M. 1992. Tres escoles per al català. Lleida: Pagès, coll. Argent Viu.
Psychological/cognitive aspects
KNOWLEDGE, EXPERTISE, AND T H E O R Y IN
TRANSLATION
Candace Séguinot
York University, Toronto, Canada
Zusammenfassung
Der Artikel beschäftigt sich mit der Übersetzung im Licht der Erforschung von
Entwicklung, Anwendung und Institutionalisierung von Wissen. Es wird hier dem
Verständnis von der Natur der Probleme von der kognitiven zu einer mehr ökologischen
Sichtweise nachgegangen und vorgeschlagen, daß die Unterschiede zwischen
Übersetzungstheorien von einer prozeßorientierten versus wissensorienterten
Herangehensweise an Probleme herrühren. Die Termini expert und professional sind
manchmal synonym verwendet worden, aber die Unterschiede zwischen diesen beiden
Termini sprechen für den Bedarf einer eher soziologisch orientierten Untersuchung.
Résumé
Le point de vue sur la traduction adopté ici est celui du développement, de l'utilisation
et de l'institutionnalisation des savoirs. Le traitement cognitif de la compréhension des
problèmes cède alors le pas à une approche plus écologique. Une ligne de partage est
proposée entre les différentes théories de la traduction, selon que la résolution des
problèmes y est perçue comme processus ou comme savoir. Parfois considérées comme
négligeables, les différences entre savoir-expert et savoir professionnel incitent à
donner plus d'assise sociologique à la recherche.
Resumen
Este trabajo enfoca la traducción desde el punto de vista del desarrollo, uso e
institucionalización del conocimiento. Parte de una comprensión cognitiva de los
problemas hasta el desarrollo de una perspectiva más ecológica. Concluye que las
diferencias entre las distintas teorias de la traducción radican en si se aproximan a los
problemas basándose en el proceso o en el conocimiento. En ocasiones se han utilizado
los términos experto y profesional como sinónimos pero existe una diferencia entre
ambos que apunta a la necesidad de una investigación basada en un enfoque más
sociológico.
88 Candace Séguinot
1. Introduction
The impulse behind this paper, written and delivered as a talk, was to look at
translation in the light of research into institutionalized forms of knowledge.
How do people learn to translate? How do people learn to translate well? You
might think that we could find the answer by studying the people who have
made a career of translating, but it is difficult to use professionals as subjects.
Because so many of their processes have become automatic, they lose
themselves in their work, forget to verbalize their thoughts, and their expertise
is recorded only in the text they have created. In a testimony to the subliminal
flow state described by Douglas Robinson (1997b:95), a professional translator
once told me that he is completely unaware of what he will say until he sees his
own words on the page. One kind of information that is available for the study
of the translation process is what happens when things break down, when there
are difficulties that interrupt the process. It is then that we learn about
knowledge and can identify signs of expertise.
What's that honey? It says to cut the butter into the flour? You need two knives
to do that . . . wait a minute, I think I've got a special tool...where did I put
it...it looks like a little rocker with blades...actually, we could have done this
part in the Cuisinart. No, you can't use a blender. No matter, we can just use
the knives, see...like this. That's right. Now blend in the rest of the ingredients,
and . . .
And the first recipe is done, and we're going to start on a second. This is my
grandmother's recipe for Coconut Peaks. You won't believe the secret
ingredient...it's made from mashed potatoes. Uh oh . . . I don't have any
leftover potatoes. What do you want to do, sweetheart, we can make something
else or. . You really want to use Nana's recipe? O.K. We'll just go to the store
and get some potatoes. Hey, while we're here, let's pick up some ice cream.
I've got a 20% off coupon so we might as well use it. Wait a minute, that can't
be right. You only charged me 50 cents for the ice cream...it should be at least
a couple of dollars. Well yes, I realize there's a long line in back of me. No, I
Knowledge, expertise, and theory in translation 89
don't want to cause trouble. Well, you're the one who's going to be in trouble
because you're losing money on that ice cream . . . All right, all right.
So I pay the cashier, and we can leave, and return to the reality of the written
page. No problem. Now what do I mean when I say that? That in all likelihood,
you already know that you are sitting in front of a written page. You do not
have difficulty imagining yourself in that situation.
But now, let's return to my niece and her problem. The difference
between the problem in 'no problem' and my niece's situation in the kitchen is
that her 'problem' was a task to perform. She had to bring a dessert to school,
and the culture dictates, home made. Had she been older, and in her own home,
and I in hers, there would still have been a task, but no problem. However in
this version, she was not in her home, she was in mine: the problem was that she
needed my permission, perhaps even my supervision and certainly my guidance,
to use my kitchen and work with my recipes. 'The' problem turned out to have
several different component problems, each of which either inhibited or
interrupted the task.
There are two routes to understanding the nature of task-oriented
behaviour, what strategies lead to success or failure, what kinds of experience
best contribute to the development of expertise, and so on. The first is to start
with an understanding of particular situations, such as this one, and generalize
from there. The second is to look at what has been written about problems and
tasks.
Let's start with the second approach and look at interrupted tasks from the
perspective of a cognitive psychologist. Richard Mayer (1989:39-41), for
example, wrote:
A problem solver has a problem when a situation is in one state, the problem
solver wants the situation to be in a different state, and the problem solver does
not know an obvious way to eliminate obstacles between the two states. In
short, a problem consists of three components: the given state, the goal state,
and obstacles that block movement from the given to the goal state . . .
From the perspective of translation, problems have been defined with much the
same flavour in the empirical literature. Wolfgang Lörscher (1991:79-80), for
example, cited a different source defining problems in essentially the same way.
The programming model of human thought from which it stems is obvious;
obvious, too, its influence in shaping process models of translation and
interpretation. And not surprisingly, as the behavioural sciences have been
greatly influenced by the claim that all human thinking is information processing.
90 Candace Séguinot
For many cognitive psychologists a decade ago, the actual processes needed in
artificial intelligence were a source of information about human intelligence.
Artificial intelligence does not have an intelligensor, so information
processing could be studied separate from the processor taken as a whole. In
this view of the world, hypotheses can be generated about any processes, the
unobservable as well as the observable. This meant that the mind was once
again allowed to be an object of speculation, and from there it became central.
Knowledge became key. In fact, from the point of view of meme theory, people
have been described as incidental to the ideas that are housed in them, mere
libraries. The ideas live on long after their unsuspecting hosts pass on. A
problem, from the cognitive perspective, equates to a need for knowledge. Not
knowing how to get from where you are to where you want to be explains why
processes are difficult or unsuccessful.
Not knowing implies a view from the inside, speculating on what goes on
in my niece's or the translator' s head. What is outside is modeled in our minds
in a variety of ways, through networks or hierarchies of associations, frames,
schema, etc. Problems, for the cognitive psychologist, do not actually exist "out
there". It is our perception that identifies something as a problem. In other
words, it is the construct of an individual.
In contrast to this psycholinguistic approach, there are a number of
schools of Translation Studies that can be categorized as focussing on the
knowledge side of the equation, i.e. that try to identify what a competent
translator needs to know. They look at problems in a different way. The
understanding of what constitutes a translation problem seems to take one of
three forms. The first, for linguistically-oriented scholars like Vinay and
Darbelnet (1958) and Catford (1965), is in the structural differences between
languages. The second is an intellectual construct: you compare a source to a
target text, you match up bits that appear to be related, and you label what
appeared in the source text a problem and in the target text a solution
(Chesterman 1997). The third knowledge-based approach is the one exemplified
by Mary Snell-Hornby (1988:52 and 96), who spoke of problematic situations,
as when words have different meanings in context than in isolation or when a
linguistic expression is multidimensional. Christiane Nord (1997:64) has most
recently articulated the difference in perspective between the internal and
external views. Outlining the benefits of source-text analysis, she rejects the use
of the word 'problem' for what she calls the subjective difficulties encountered
during a translation session.
So on the one hand we have the process-oriented researchers, like
Lörscher, Tirkkonen-Condit, Jääskeläinen, and several others, including myself,
who study the individual and work toward finding objectively verifiable ways of
identifying problems that are realized in the course of translating sessions; and
on the other, linguistic, text, and culturally oriented studies whose authors'
Knowledge, expertise, and theory in translation 91
One explanation for the differences is, as Chesterman (1997:42-46) points out,
our different starting points. Looking at a particular question will lead to one
answer, starting with a different question will lead to another. For as Karl
Popper describes the world of scientific enquiry (1989:47),
Yet to point out the lack of coherence is to miss the progress on the descriptive
front where it was formerly assumed that linguistic difficulties could predict
difficulties in the translation process. The pivotal change from the view of the
translation problem as external to the individual to the data-driven re-orientation
of that understanding can be seen in a study published by Hans Krings in 1987.
In the section on methodology Krings began (1987:162),
On analyzing the data, however, he realized that only one-fifth of the problems
were common to all his subjects, and over half the problems were idiosyncratic
(1987:167-169). Furthermore, the fact that 90% of the problems were lexical
spurred other researchers to ask whether that figure wasn't more a function of
the subject population (he was studying second language learners) than an
insight into what was truly problematic for translators. This was the beginning
of an empirical focus on the question of whether translation as practised by
professionals is different from the translating of the more inexperienced, which
was an important step in understanding the nature of translation competence.
For as Gideon Toury (1995:238) has said,
92 Candace Séguinot
This latter point, the question of whether socio-cultural factors contribute to the
development of translators who will be successful as professionals, leads to two
questions which will be raised again later in the paper. The first is whether there
is a clear understanding of the difference between translator competence, or
expertise, and professional competence, meaning the ability to assume a position
in the translation marketplace. The second is whether current empirical research
methods which depend on psycholinguistic methods can be enhanced or
complemented by other avenues of research to explore these socio-cultural
factors.
translator has to face on a regular basis. How someone comes to learn to cope
in the face of the unexpected is one of the hallmarks of expertise. The paradox
is that in order to help someone develop their expertise, we need to make clear
the opposite, the expected. Problems come in two types: the routine, and the
unfamiliar. Routine problems are those for which the solutions are ready-made.
That does not mean that the novice necessarily knows the solution, but that the
procedure for arriving at a solution is well-known.
What role does academic instruction play in establishing those routine
connections? We know we contribute to the development of good translating
habits by showing students how to use documentation correctly, we know we
encourage automatic correspondences through exercises in comparative
stylistics. These are activities that have built-in repetitions. What is less clear is
how material we present in the form of facts, for example types of approaches
for different text types, gets transferred into procedural knowledge.
What distinguishes sensible from silly operations is not their parentage but their
procedure, and this holds no less for intellectual than for practical
performances. "Intelligent" cannot be defined in terms of "intellectual" or
"knowing how" in terms of "knowing that"; "thinking what I am doing" does
Knowledge, expertise, and theory in translation 95
You might think the answer would be to study professionals in action to see
their strategies first-hand. Yet even as I carry out this kind of research, I know
that you cannot have the desired effect - the integration of a description of
practices into the practice itself - because of the developmental stages involved
in acquiring expertise. For reasons that will become clear later, the nature of the
expert's knowledge, as well as the processes themselves, is different from that
of less experienced learners.
The point of looking at the problem of the unexpected is to see that experts
handle such situations in ways that distinguish them from the inexperienced, and
they do so in ways that may have little to do with what they are capable of
explaining. There is a second developmental issue in my kitchen example that I
would like to turn to now. As an adult, I feel the pressure of time. It would have
been more productive to switch to another recipe than to go to the store if my
only goal had been to make cookies. But real life, as you know, is messy. As
adults, we have to learn to deal with multiple goals. In this case, I was making
cookies, but I saw this as a way to spend quality time with my niece. That, for
me, was the more important of the goals, and so I let my niece decide what we
would do. Productivity is not a child's goal, and children do not readily
compromise, so we stuck with our original plan and went to buy the missing
ingredient.
Productivity is not necessarily a byproduct of expertise, though in the
field of translation we tend to conflate the characteristics of the professional
with those of expertise. Anthony Pym (1994:2), for example, includes a
reference to expert status in his explanations of the use of the word 'expert' in
discourse about translation. What does characterize the behaviour of the expert?
To return to my everyday world example, my expertise in the kitchen, at least
relative to my niece, showed itself in two ways. I had an arsenal of recipes at my
disposal; my niece did not. I also had more than one strategy for accomplishing
a part of the procedure. If we go back to the bit where she questioned me about
how one went about cutting butter into flour, my answer contained more
information than a simple instruction. I told her indirectly that there were three
ways we could go about cutting in butter: using two knives, using a special tool,
or using a kitchen machine. So in addition to knowledge, my experience has
taught me that there are often alternatives.
In the literature on procedures, a distinction is made between what are
called toolbox skills and algorithmic skills. In the case of a toolbox skill, you
96 Candace Séguinot
have a problem so you reach into your toolbox and pull out a tool, one of many.
In the case of an algorithmic skill, there is only one formula: learn it, and you
can arrive at your solution. Translation is a toolbox skill, and the learner needs
to be at a given level to take advantage of the offering of alternatives. Because
the beginner tends to focus on the lower level problems, however, there is an
opening up that comes from seeing how others have handled the same material.
In an article on creativity in translation, Paul Kußmaul (1991:98)
stresses the need to ensure that students always preserve a critical attitude
toward the ideas that come into their heads. During the incubation and
illumination phases "they must be able to stand back, as it were, and observe
what they have been doing." My store of experience leads to my facing novel
experiences with a set of expectations. When I went to the grocery store, for
example, I had an idea of how much my groceries should cost. No matter that
the total was significantly less than I expected, that it was to my advantage from
a purely monetary perspective. I felt moved to say something. I put aside the
task of buying groceries for a moment, and in so doing I created more problems
for myself by opening myself up to hostility on the part of the cashier and the
people in the line behind me. One of the ways that we differ from computer
programs is that our monitoring includes a scanning for ethical and aesthetic
implications. To return to the definition of a problem, it seems unlikely that we
store our responses to moral dilemmas as individual instances that we recognize
on subsequent encounters. How, then, do we account for the ways in which
ethical and aesthetic norms affect the performance of a task?
One answer lies in the theories of behaviour that have developed from a
recognition of the inadequacies of the cognitive view. No one would deny if we
have a task to perform, whether to imagine ourselves here or elsewhere, to bake
cookies or to translate a text, the 'how to' takes place in our head. We rely on
our previous knowledge, we think about alternatives, and we come to decisions.
This is the arena that Mayer (1989:42-43) is addressing when he explains the
nature of problem solving :
Problem solving can be analyzed into two major phases: representation and
solution. Representation involves moving from a statement or presentation of
the problem in the world to an internal encoding of the problem in memory.
More specifically, representation involves mentally encoding the given state,
goal state, and legal operators for a problem. Solution involves filling in the
gap between the given and goal states. More specifically, solution involves
devising and carrying out a plan for operating on the representation of the
problem...The representation and solution processes may be analyzed,
respectively, as building a problem space and using procedures for searching
Knowledge, expertise, and theory in translation 97
for a path through the problem space.The input to the solution process is a
cognitive description of the problem (including the given state, goal state,
intermediate states, legal operators) and the output is a search path (i.e.,
application of a series of operators) between the given and goal states. The
search through the problem space (i.e., the connections among the given, goal,
and all intervening states) is controlled by a set of procedures, such as a means-
ends analysis strategy.
But neither the assignment of the task nor the attempts to deal with it take place
in a vacuum of mental processes. To carry out her task, my niece looked to me
for help; I asked for her opinion. In the first instance, there was an obstacle, and
it was a lack of knowledge. My niece looked outside herself, to another person,
for the information that she could not understand. In taking her wishes into
account I was acknowledging our relationship and the importance of that
relationship in the determination of how I was going to act.
One could, of course, build a model that allows for an interpersonal
component. As Donald Norman points out (1985:310), however, there is a
fundamental objection to the mind-as-machine approach. This is its "lack of
consideration of other aspects of human behaviour, of interaction with other
people and the environment, of the influence of the history of the person, or
even the culture". The search for models that more accurately conform to our
common sense understanding of how people operate, referred to as the search
for ecological validity, has meant a movement away from mechanistic
explanations involving chains of cause and effect and a move toward more
complex, decentralized explanations (Resnick:1994). Mitchell Resnick offers the
analogy of a flock of birds. They fly in formation, changing direction as if to a
hidden signal, but there is no leader to give the signal.
translators. She looked at what the translators said about how they were
translating the source text and concluded that the differences could be attributed
in part to their degree of involvement with the target culture: one translator was
involved in the legal system, another a volunteer with a refugee association.
But what about the difference between expert knowledge and professional
knowledge? For that we need to change from a psychological to a sociological
perspective; we need to look at the development of the professions. Expertise in
the form of a commodity is a relatively recent notion. Prior to the industrial
revolution, people relied on authority, and their own common sense. The
division of labour and the densification of population with the growth of cities
made it possible for people to sell their advice and services, and from there it
became possible for people to support themselves through this form of
"commodification". The expectation of payment became combined with a
degree of autonomy. There was a recognition within society that a certain form
of learning was necessary for the successful performance of services, and
mechanisms were developed to ensure exclusivity to rights of practice, notably
through some form of certification. The combination of these factors served to
define a group of practitioners as professionals. Today when we think of
professional translators we tend to think of these outward aspects still: people
who support themselves by translating.
Meanwhile the learning that was necessary for the exercise of the
professions had to be passed on, and in this century it became the job of the
universities. In return for the rewards of being absorbed into institutions of
higher learning, the professions had to accept the positivist ideology that is the
hallmark of the modern university: science leads to progress, and a better world.
But there was a problem when it came to practical knowledge, because
100 Candace Séguinot
(1) The types of deadlines that are associated with translation assignments and
translation tests.
(2) The types of workplace simulations that are built into the programme.
(3) The weight given to the surface manifestations of professionalism:
requirements that assignments be word processed, be in on time, comply with
particular standards, etc.
(4) The reinforcement of routines through relevant forms of practice (e.g.
building and managing a personal term file).
(5) The regulating of curriculum offerings on academic or professional grounds.
(6) Whether Schools of Translation and Interpretation regulate themselves with
a governing body or charter that imposes particular kinds of courses.
These are just some of the ways that we define the academic and the
professional in setting out our interpretation of translation goals. This says
nothing about the preparation of students for their role as translators, though
classes play a part in this aspect as well. We could say that translation as a
discipline is, to the extent that it is directed toward the professional, concerned
with the transmittal rather than the construction of knowledge. To that extent it
reflects the reality of the institutionalized practice. The individual translator
working in a commercial setting has very little power (see Pym 1997:100 and
Robinson 1997b). The norms with which today's translators operate emanate
from the institutions in which or with which they work, and not only do those
institutions differ in their expectations, their expectations are changing to
incorporate the less obviously translation-specific activities required to produce
camera-ready copy and text suited to both print and screen. The importance of
contextualizing translation in order to predict the norms that will surface makes
it difficult for empirical studies to isolate particular variables for study. We
know instinctively that experimenting with short, artificial corpuses as is done in
psychology would not reflect the reality of the translation workplace (Séguinot
1996b). So while process studies appear to be a promising window into the
nature of professional expertise, they are constrained both by the rigor of
scientific methodology and by the notion of ecological validity.
The question of how we can help our students become problem setters, as well
as problem solvers, brings us back to the topic with which I began, namely the
problem of problems. The more we structure classroom practices around what
commercial interests see as problems - the need for specialized vocabulary,
familiarity with technology, and productivity - the more we need to introduce
and therefore reward activities that tend to conform. With this in mind I would
like to close with some words from David Bohm, a physicist who wrote about
the relationship between art and science, and who thought a great deal about the
distinction between the mechanical and the creative. He wrote that (Bohm
1998:4)
The irony is that research and education that are tied to producing
specific kinds of knowledge, to meet needs specified by the profession, may not
serve the profession as well in the long run as less instrumental approaches that
bring both the profession and the discipline back to question the nature of
Knowledge, expertise, and theory in translation 103
translation expertise.
References
BOHM, D., ed. by NICHOL, L. 1998. On Creativity. London & New York: Routledge.
CATFORD, T.C. 1965. A Linguistic Theory of Translation. London: Oxford University Press.
DREYFUS, H. L. & DREYFUS, S.E. 1986. Mind over Machine. Oxford: Blackwell & New
York: The Free Press.
FRASER, J. 1996. "Mapping the process of translation ". Meta 41(1), 84-96.
MAYER, R. E. 1989 "Human nonadversary problem solving". Human and Machine Problem
Solving ed. by K. J. GILHOOLY, 39-56. New York and London: Plenum Press.
NORMAN, D. A. 1985. "Twelve issues for cognitive science". Issues in Cognitive Modeling
ed. by A. M. AITKENHEAD & J. M. SLACK, 309-336. Hillsdale, N. J.: Lawrence Erlbaum
104 Candace Ségumot
POPPER, K. 1989. Conjectures and Refutations: the Growth of Scientific Knowledge. 5 rev.
ed. London: Routledge.
PYM, A. 1994. "Ideologies of the expert in discourses on translator training". Koiné 4.139-
149.
RESNICK, M. 1994. Turtles, Termites, and Traffic Jams. Cambridge, Mass.: The MIT Press.
ROBINSON, D. 1997a. "The invisible hands that control translation". Paper posted in the
On-Line Colloquium on translation hosted over the internet by the Facultat de Traducció i
d'lnterpretació of the Universitat Autônoma de Barcelona.
SIMEONI, D. 1998. "The pivotal status of the translator's habitus". Target 10 (1), 1-39.
TOURY, G. 1995. Descriptive Translation Studies and Beyond. Amsterdam & Philadelphia:
John Benjamins.
Zusammenfassung
Dieser Artikel berichtet von einem Übersetzungsexperiment, bei dem die Methode des
Computerlogging von real-time writing-Daten in Kombination mit der Methode des
Lauten Denkens angewendet wurde. Drei verschiedene Texte wurden von sechs
Personen unter drei unterschiedlichen zeitlichen Bedingungen übersetzt. Alle
Tastatureingaben, d.h. alle Cursorbewegungen und Korrekturen bei der Produktion
und Revision des Zieltextes, wurden mit Hilfe der Software Translog registriert, und
die Daten des Lauten Denkens wurden auf Tonband festgehalten. Obwohl nicht alle
anfänglichen Hypothesen durch das Experiment bestätigt wurden, half die
Kombination quantitativer und qualitativer Daten bei der Ermittlung wichtiger
Informationen über den Einfluss des Zeitdrucks auf die Revision des Zieltextes, über
die Benutzung von Wörterbüchern und die Anwendung von wörtlicher Übersetzung
oder Paraphrase.
Résumé
Dans l'expérience rapportée, on a eu recours à la fois à l'écriture saisie
informatiquement en temps réel et à la verbalisation à haute voix. On a demandé à six
sujets de traduire trois textes de presse différents, selon trois délais différents. Toutes
les frappes utilisées pour écrire et éditer les traductions ont été saisies grâce au
programme Translog; tout ce qui a été verbalisé a été enregistré. Toutes les hypothèses
initiales n'ont pu être étayées par notre expérience mais on a pu souligner fortement,
grâce aux données quantitatives et qualitatives, combien la finalisation des textes
d'arrivée est dépendante des délais, de l'utilisation de dictionnaires et de la stratégie
de traduction (littérale ou paraphrastique) choisie par chacun des groupes testés.
106 Jensen & Jakobsen
Resumen
El experimento de traducción objeto de este reporte tue llevado a cabo combinando la
escritura informatica en tiempo real y el razonamiento en voz alta. El experimento
consistió en la traducción realizada por seis sujetos de tres textos diferentes extraidos
de la prensa bajo tres restricciones de tiempo distintas. Todas las pulsaciones
utilizadas al escribir y editar los textos de llegada se registraron por medio del
programa Translog, mientras que la producción verbalizada fue grabada en cinta
magnética. A pesar de no poder corroborar mediante el experimento todas las
hipótesis iniciales, merced a la combinación de datos cuantitativos y cualitativos se ha
podido poner claramente de manifiesto hasta qué punto la calidad de los textos de
llegada está en relación con el periodo de tiempo disponible, la posibilidad o no de
utilizar diccionarios, y el tipo de estrategia traslativa elegida por cada uno de los
grupos sometidos al experimento.
1. Introduction
Since the middle of the 1980s, research into the process of translation has
tended to focus on methods of eliciting and analysing qualitative data. In
particular, the method of thinking aloud has been widely used. In the present
paper, we report on an experiment in which quantitative data logged by a
computer program were used in combination with qualitative think-aloud data.
We believe a quantitative process-oriented approach is an interesting
alternative or supplement to the various approaches hitherto imported from
psycholinguistics. Logging keystrokes in real time is a method which allows us
to track and study the writing process involved in translation across time and
through all revisions (Jakobsen 1998; 1999). Furthermore, this can be done in a
controlled but ecologically valid environment while escaping some of the
criticism that has been levelled against qualitative approaches. It is also
possible to combine a quantitative approach with more invasive, qualitative
methods such as observation, interviewing, or thinking-aloud protocolling.
Combining a quantitative and a qualitative approach makes it possible to
synchronise and compare data elicited by different methods but pertaining to
the same phenomenon. Finally, the logging tool that we use for collecting data
also permits us to work with retrospection in a new way that seems to enhance
both the accuracy and the quantity of retrospective information obtained from
subjects.
The key-log program used in the experiment was Translog, which was
originally designed by Arnt Lykke Jakobsen in 1995. The program has a
facility for editing and displaying source text in various ways and a facility for
Translating under time pressure 107
In addition to supplying information about the file and text used, the total
number of keystrokes, (electronic) dictionary lookups and the total duration of
a writing event, the View facility represents a writing event as a linear
succession of keystrokes (characters, cursor movements, deletions, dictionary
lookups, etc.) with an indication of time in relevant places. Time delay is
indicated by asterisks. In the first representation, the time unit has been set at
four seconds. This means that for each time there was a delay lasting four
seconds an asterisk is printed. However, if a delay occurred during the writing
event which would require ten or more asterisks, the View facility writes the
absolute time delay instead, e.g. [*:56.41].
A time-unit value of four (or more) seconds gives a very rough
representation of time delay. What can be seen in this first representation of the
writing event is that, initially, there was a long orientation pause lasting 56.41
seconds, followed by scrolling of the source text [i] and [↓] to bring the entire
source text into view. This was followed by an even longer pause lasting
106.88 seconds. If the Translog data were our only source of information, we
would assume that this pause was spent processing the headline of the source
text. Synchronised audiotape data might supply additional information, e.g.
that the subject was looking up a word in a manual dictionary, or the subject
might supply think-aloud data at this point. At the end of this brief extract,
there is another long pause lasting almost a minute (59.70 seconds). This could
be interpreted either as a monitoring pause during which the target text so far
produced (or a part of it) was monitored, or as a forward-looking processing
pause in which subsequent target text was being processed.
A representation with a time-unit value of 4 seconds (or more) allows
us to observe only the major delays, which tend to appear at text, paragraph or
sentence boundaries. If the main focus of interest is on such major pauses, or if
it is on, say, text revision rather than time delay, a representation with a high
time-unit value probably brings out the interesting data most clearly. A high
Translating under time pressure 109
time-unit value has the advantage of representing the target text in a way that is
usually easy to read. The target text appears to have been typed very regularly
with occasional minor delays appearing between sentences (after the headline
line break, after the sentence ending with 'mod klart sprog'), in connection
with the (proposed but abandoned) revision of 'flyder', or between words
(som*** smør). (The centred dot represents a touch of the space bar.)
Translog's default time-unit value (i.e. the value of one asterisk) is 0.50
seconds. With this value, the identical text segment will be represented in the
following manner:
[*:56.41][l]*****[^]*********[^]*******[t]**[t][*:106.88]investeringsforening*
**.*** sreklame****:*-*pakket-*ind-i-*weasel-**words** 5[*:8.24]Det-ermulig
t- **at- **ikke-noget-ande*t-dokument-på* ****i verden- * **har-beg*ået-så-inange-
syndermod*klar*t**sprog.[*:7.96],**Sproget,****flyder*********fi:******
********r*som-[*:12.20]smør[*:59.70]
As can be seen, the only difference in the way the View facility displays logfile
information is in the way time delays are represented. With a time-unit value of
half a second, more detailed (and more accurate) information about time delay
is made to appear. We can see that time elapsed between all of the initial scroll
movements of the source text, and that sometimes there were delays not just
between words and sentences, but also within words (ande*t, beg*ået, klar*t).
However, delays appear on the whole very clearly to coincide with
grammatical (word and sentence) boundaries. Finally, we notice that all delays
lasting 5 seconds or more are now represented with the absolute delay time,
e.g. [*:8.24]. If we are interested in studying time delay in even greater detail,
we can ask the program to recalculate the logfile information with a smaller
time-unit value.
For translation within the news media, time and deadlines are key issues.
Another serious challenge in newspaper translation is the multitude of topics
dealt with., In order to cope with these two main constraints, the translator of
110 Jensen & Jakobsen
texts for the news media must have efficient methods of information retrieval,
must be able to solve comprehension and translation problems quickly, must be
capable of making fast decisions, and must have efficient translation strategies.
Some problems are no doubt solved in a flash and many decisions made in an
instant. However, a lot of problem solving takes measurable time and therefore
can be more easily investigated.
One purpose of the experiment carried out by Astrid Jensen was to see
what problem-solving activity could be identified by comparing think-aloud
and logfile data and to see how such activity was distributed across eighteen
translation events. The effect of time pressure (if any) on the subjects' use of
translation strategies was also measured. Translog was used to help identify
problem-solving activity and to ascertain both what translation strategies were
used by non-translators and professional translators and what changes in the
use of translation strategies would correlate with different time conditions. It
was assumed that non-translators and professional translators would behave
quite differently. This assumption was based on a number of studies within the
past ten years comparing non-translators with professional translators using
introspective methods (Börsch 1986; Gerloff 1986; Krings 1986a and 1986b;
Jääskeläinen 1989a and 1989b; Lörscher 1991; Jääskeläinen & Tirkkonen-
Condit 1991; Fraser 1996). Translation strategies were operationally restricted
to direct translation and paraphrase. The extent to which dictionaries were
resorted to was also observed. (On time pressure, see also Jensen 1999.)
Six translators, two non-translators, two recently graduated
professionals, and two experienced professional translators were asked to
translate three different journalistic texts under three different time constraints
using Translog. The dictionaries consulted were traditional (non-electronic)
dictionaries. While doing the translation, subjects were asked to think aloud.
What they said was audiotaped, transcribed and synchronised with the process
information recorded by Translog. The two non-translators, or 'educated
laymen' (Jääskeläinen 1993: 99) had an M.Sc. degree in engineering, and were
accustomed to using English as a working language but had no experience with
translating. The time conditions set were 15, 20 and 30 minutes (with an initial
warm-up exercise). Each subject translated texts under all three time
conditions. Thus a total of 18 translations were produced.
The texts chosen comprised three topics from different sources: the
Economist, the Financial Times, and the Observer. The texts were translated
from English into Danish, the mother tongue of all the subjects, and subjects
were instructed to translate as if the articles were to appear in the Danish
newspaper Berlingske Tidende, a well-known quality national newspaper. For
the professional translators, this set-up was very similar to a routine translation
task. The average length of the texts was 120 words. Subjects were allowed to
use dictionaries of their own choice, and the experiment took place in their own
office, uninterrupted by the researcher.
Translating under time pressure 111
For the purpose of analysing the translation, the source text was segmented into
portions consisting of individual sentences (S1, S2, etc.) and the relevant
verbalisations from the subjects' protocols were combined with the data from
the Translog print-outs in a table (see Table 1). In that table, the duration of
time delays was represented in absolute figures, and relevant editing was
inserted in the text segments where it belonged. Further, a distinction was made
according to whether the editing took place during translation or during
revision.
Table 1: Source language text units synchronised with logfile and think-aloud
data
5. Problem-solving activity
An initial, quantitative survey was made from the Translog records of pauses
that could be associated with problem solving. By problem-solving activity we
understand information processing aimed at solving a specific and identifiable
translation problem. This may be reflected in verbalisation, but not necessarily.
In the experiment, there were very considerable differences in verbalisation.
Some subjects produced very few comments, while others, perhaps more eager
112 Jensen & Jakobsen
to please the researcher, spoke volumes. There was no correlation between the
amount of verbalisation and the speed (or quality) of target-text production.
Therefore, no correlation was found between the quantity of verbalisation and
the amount or frequency of problem-solving activity.
On the other hand, verbalisation was clearly linked to problem solving.
After replaying the recorded translation event in Translog and listening to the
synchronised audio-tape recording several times and thus comparing logged
text-production data with think-aloud data, it was intuitively clear what
problems were being processed in what delays. Some delays were caused by
dictionary search, others by meta-comments, memory search, or word
association. A coding scheme for six identifiable problem-solving activities
was applied, and though not all delays could be linked to a specific problem-
solving activity, a correlation emerged between delays lasting four seconds (or
more) and think-aloud evidence of problem-solving activity.
A subsequent count of all delays lasting more than four seconds did not
reveal significant differences across the six subjects. Nor were there any clear
differences in the number of delays devoted to each of the three texts used. All
subjects had an average of very close to 2 problem-solving delays per minute,
regardless of text and time condition. There was a clear increase in the number
of problem-solving delays as more time was granted for a task (32% more with
20 minutes than with 15; 57% more with 30 minutes), the increase
corresponding to the increase in translation time available.
However, by differentiating between delays occurring during the initial
editing phase and during revision, two significant tendencies emerged. With
more time, the number of problem-solving delays during the initial editing of
the target text increased much less (20% and 32%) than the number of
problem-solving delays during revision (83% and 137%). Also, there was a
markedly higher incidence of problem-solving delays during revision in the
translations by the two most experienced professionals than in the other
translations.
These findings were interpreted as suggesting that raw target-text
production has its own rhythm and speed, which responds to the time factor but
is fairly independent of it (within the time ranges tested in the experiment). The
increase in the incidence of problem-solving delays during revision suggests
that for the majority of subjects in the experiment 15 minutes did not allow
much time for revision. In other words, the number of translation problems
negotiated was relatively independent of the three time conditions in the
experiment. What happened when more time was available was that time was
sometimes spent in two (or more) separate rounds of attempting to solve
problems to which problem-solving time had already been devoted.
Translating under time pressure 113
6. Dictionary lookups
There was no significant variation in the number of dictionary lookups per time
condition or per text. The most striking differences in the use of dictionaries
were (again) between three groups of subjects. Non-translators used
dictionaries far more frequently (45 times in all) than the young professionals
(with 1-3 years of professional experience), who in their turn used dictionaries
far more often (32 times) than the experienced professionals (with more than 8
years of professional experience; 18 dictionary look-ups).
7. Translation strategies
No significant correlation was found between the three time conditions or the
three texts and the use of translation strategies.
The preferred translation strategy by the group of non-professional
translators was literal translation, whereas professional translators all preferred
paraphrase. There were only half as many instances of direct translation in the
114 Jensen & Jakobsen
The only significant effect of time pressure that was identified in the
experiment was the decrease in problem-solving activity during revision as less
time was given for a translation task. However, it appears that only the non
professionals came under serious time pressure to the extent that in at least one
case they were unable to complete a first draft of the translation. Therefore, the
assumption that different translation strategies would be used, given sufficient
time pressure, could neither be confirmed nor dismissed. In retrospect, it can be
seen that there ought to have been a shorter time condition than 15 minutes so
that all subjects would have come under more acutely felt time pressure.
The main findings of the experiment were in unexpected areas, in the
correlation of dictionary use and professional translational experience and in
the correlation of paraphrase and professional translational experience. Both
findings confirmed familiar assumptions. The results corresponded rather
closely with observations on 'developmental stages' made e.g. by Jääskeläinen
(1996:67) but seemed to suggest that the translational behaviour of young
professionals may be closer to that of semi-professionals (graduate students)
than to that of professionals with eight or more years of experience.
An initial assumption that time pressure would force all translators to
use some of the coping tactics observed in interpreting (cf. Gile 1995:207),
because of limited processing capacity or because of gaps in the subjects'
knowledge base, was only partially supported by the data. The findings
confirmed that all groups made use of coping tactics, in fact almost a third of
all translation strategies employed could be analysed as coping tactics.
However, at this point it would be premature to conclude that the use of coping
tactics was caused by time pressure only. It seems more probable that the
combination of (some) time pressure and certain types of translation difficulty
motivated the use of coping tactics.
Translating under time pressure 115
References
BELL, R.T. 1991. Translation and Translating, Theory and Practice. London: Longman.
GERLOFF, P. 1986. "Second language learners' reports on the interpretive process: talk-aloud
protocols of translation". HOUSE, J. & S. BLUM-KULKA (eds), 243-261.
GILE, D. 1995. Basic Concepts and Models for Interpreter and Translator Training.
Amsterdam & Philadelphia: Benjamins.
HANSEN, G. (ed.) 1999. Probing the Process in Translation: Methods and Results.
Copenhagen Studies in Language 24. Copenhagen: Samfundslitteratur.
JAKOBSEN, A.L. 1998. "Logging time delay in translation". Copenhagen Working Papers in
LSP 7, 73-101. Copenhagen: Copenhagen Business School.
JAKOBSEN, A.L. 1999. "Logging target text production with Translog". HANSEN, G. (ed.),
9-20.
JÄÄSKELÄINEN, R. 1996. "Hard work will bear beautiful fruit? A comparison of two think-
aloud protocol studies". Meta 41(1), 60-74.
KRINGS H.P. 1986a. Was in den Köpfen von Übersetzern vorgeht. Tübingen: Narr.
Paul Kussmaul
University ofMainz, Germersheim, Germany
Zusammenfassung
Um etwas über kreative Prozesse bei englisch-deutschen Übersetzungen (die meisten
stammen aus T S . Eliot's Old Possum 's Book of Practical Cats) zu erfahren, wird von
den Begriffen Perspektive, Fokus, Figur/Grund Relation, Szene, Rahmen, Kern,
unscharfe Ränder und Verkettungen als heuristische Mittel Gebrauch gemacht.
Dadurch lassen sich eine Reihe kreativer Techniken unterscheiden: 1. Verkettung von
Kategorien 2. Auswahl von Szenenelementen 3. Szenenerweiterung 4. Einrahmung
einer Szene 5. Neurahmung einer Szene.
Résumé
Jusqu'où y a-t-il créativité dans les traductions anglais-allemand, notamment dans les
traductions de T.S. Eliot? Pour l'évaluer, nous avons fait usage heuristique des notions
comme celles de point de perspective, focus (ou saillance), figure, fond, scénario, frame
(ou cadre cognitif), éléments saillants, frontières floues et liant ou enchainement. On a
distingué finalement 5 types de modes créatifs: l'enchainement de catégories, la
sélection d'éléments du scénario (ou script), l'élargissement du scénario, la mise en
cadre d'un scénario et la mise en place d'un nouveau frame.
Resumen
Con el fin de estudiar la creatividad en las traducciones del inglés al alemân (la mayoria
tomadas del Old Possum's Book of Practical Cats de T.S. Eliot), se han utilizado como
medio heuristico los conceptos perspectiva, enfoque, alineación flgura/suelo, escena,
marco, nücleo, bordes difusos y encadenamiento. Se pueden distinguir varias técnicas
creativas: 1) Encadenar categorias 2) Seleccionar elementos de la escena 3) Ampliar la
escena 4) Enmarcar la escena 5) Imaginar un marco nuevo.
118 Paul Kussmaul
1. Heuristic models
"He can't think straight for a minute!" This usually is not exactly a laudatory
comment. Creative people, however, very often do exactly this; they jump from
one idea to another. This does not mean they cannot think straight, but when it
comes to having a bright new idea, thinking straight is replaced by thinking
around the corner, or, in the terms of creativity research, by thinking laterally
(de Bono 1970) or divergently (Guilford 1975). Lateral thinking, a term coined
by Edward de Bono, has indeed become one of the basic notions in creativity
research. Lateral thinking leads to new ideas, and so it does in translation.
As to the products that result from new ideas, it is generally recognised
that they should be new in one way or other and also that they should be
adequate to the task (cf. Preiser 1976:6ff). These two features should be kept
in mind when evaluating creative translations. I am not interested here in
translations that are not good enough or even bad. I am specifically studying
good translations. They can be taken as models of how to translate successfully,
especially if the cognitive categories behind them can be revealed.
There are two further notions closely connected with lateral thinking:
perspective and focus (cf. Brodbeck 1995:40-41, de Bono 1990:154-168 uses
the synonym "attention area"). Changing the perspective and changing the focus
are ways of seeing familiar things in a new light and thus of coming up with new
ideas. When we deal with linguistic categories and with the use of language
these notions are, of course, metaphorical. It may be helpful, therefore, to use
pictures as a starting point in order to illustrate what goes on in our minds when
we are creative.
Figure 1
When we look at the cube in Figure 1 from the usual point of view, its front side
is on the lower left and we look upward to the right. But we can change the
point of view and look at it so that the front side is on the upper right and we
look down to the left.
When we make a drawing of a man on a bike we usually look at him
from the side, but we may change the perspective and look at him from the top,
and when he is a Mexican he will look like Figure 2:
Types of creative translating 119
Figure 2
These are amusing games, but they also reveal some basic features or creative
thinking. Creative thinking makes us see details we may not have noticed
before: in the drawing of the Mexican on a bike the unusual size and the
sunshade function of his hat.
It is interesting to see that the notions of perspective and focus are also
used in cognitive semantics. Ronald W. Langacker uses the notion of
figure/ground alignment to explain perspective and focus (1987:120ff.). Here
again, a picture may help to illustrate what he means. (Langacker only gives a
verbal description.) In Figure 3 we usually see a white dot on a black field. The
focus is on the white dot.
Figure 3
Figure 4
here of the notion "canonical viewpoint", which is the usual and natural way of
looking at things (Langacker 1987:123). Thus, when looking at the cube we
may find it hard to preserve the new point of view; after a few seconds we may
relapse into the old and familiar perspective.
Although we may find it difficult to do so we can structure things in a
different way. The first figure we can see as a black piece of cardboard with a
little hole in it, and the second figure we can see as a white plate on a black
table mat. This is a more unusual focus, and it is very similar to the change of
focus used as a starting point for divergent thinking. In creative translating, as
we shall see, changing the perspective and focus plays a very important part.
Being guided by the principle of fidelity, we will try to preserve the notions and
concepts of the source text in our translation, but we may change the point of
view from which we see a concept and/or we may focus on different elements of
a concept.
There is another frequently used term closely linked with prototypes.
When referring to larger notions, use is made of scene or even scenario (cf.
Fillmore 1977). Scenes may consist of a large number of elements. It will be
seen that the notion of a scene and focusing on scene elements are helpful tools
for explaining what happens in creative translating.
Focus and perspective, it seems to me, are also at work in prototype
semantics and in what George Lakoff calls chaining, a notion based on
prototype semantics. Chaining takes place via core elements (which are
focused) of scenarios (Lakoff 1987:95). When translating creatively we
sometimes do this. If we do not want to throw the principle of fidelity
overboard altogether there will, after all, have to be some link between the
source and target texts.
Scenes are linguistically represented by frames (cf. Fillmore 1977). In
creative translating, it will be seen, scenic details mentioned in the source text
are sometimes replaced by a frame in the target text. For practical purposes we
can distinguish the frame from the scene because usually a frame is more
abstract and consists of fewer words than a description of a scene.
2. 1 Chaining of categories
In the Asterix comics the Druid's name in the French original is Panoramix. The
name refers to the far-sightedness and visionary powers of the wise old man. In
the English version he is called Get-a-fix, an allusion to the distribution of the
magic potion, and in the German version his name is Miraculix which refers to
the miraculous effects of the potion. These translations can be regarded as
Types of creative translating 121
creative because they are new in the sense that they differ from the source text.
If, following Lakoff, we think of the names as categories, we might find it
difficult to see the chaining or linking element between the original name and its
translations: what have visionary powers to do with the magic potion? Are these
not two completely different categories? The connection becomes more
obvious, though, if we see the categories within an overall scenario: the druid
with all his various functions and capabilities. They are the linking elements
between the various names for him. We may regard the different names as a
result of a different perspective. For someone interested in planning and looking
ahead a proper name for the Druid may well be Panoramix, for someone
interested in the effects of a special type of booze he may adequately be called
Get-a-fix, and for those who take him along to fight against the Romans he is a
proper Miraculix. The different names can also be explained by focus; the
different names in the French, English and German version put the emphasis on
different features of the character. In Langacker's terms we may call these
features the figure against the ground of the overall scenario, and what was part
of the ground in the French source text (remember the restructuring of
Langacker's figure-ground alignment!) becomes the figure in the translations.
The fact that categories are parts of larger scenarios can be used as a
hypothesis to explain creative translations. These embeddings are not dissimilar
to the hierarchical structure of semantic fields, but in contrast to them they are
not caused by the language system but by our comprehension of the text and by
our experiences. This is the psychological aspect of the model, and this is what
enables us to chain categories.
In translations we can often observe that a relatively abstract word of the source
text is replaced by a more detailed and concrete expression. This can be
analysed from a cognitive point of view.
In T. S. Eliot's Old Possum's Book of Practical Cats (which was used as
the libretto for the musical Cats) one of the poems is about Gus, the Theatre
Cat, who as an old cat tells everyone how famous he was in the old days:
For he isn't the cat that he was in his prime;
Though his name was very famous, he says, in its time.
(T.S.Eliot 1961:72)
The words name, famous, in its time are the frames which are filled in, as it
were, by the reader's mental pictures. These pictures are part of the
comprehension process, which is closely linked, as I have been able to observe
in many instances, with the actual translation process. Translators can be
122 Paul Kussmaul
creative at the comprehension stage, and when they verbalise what they have
comprehended they produce a creative translation. Thus, imaginative reading,
one might say, is be a prerequisite for creative translating. Carl Zuckmayer, who
translated this poem, appears to have been an imaginative reader. He wrote:
The abstract frame famous gives rise to a scenario which potentially consists of
many elements, and one of them is picked out and verbalised by Zuckmayer.
Being based on our experiences, scenarios are historically determined. Today
the celebrity scenario would include the fact that we know famous actors from
talk shows and television serials and commercials, and if Carl Zuckmayer still
lived and had to translate the poem today, he would perhaps make use of one of
these elements.
I propose to call the creative strategy underlying this translation: picking
out scene elements within a frame. Here again we can observe the phenomenon
of'focusing. One element is mentioned in the translation, and we thus view one
particular element of the scene, which may be said to stick out as the figure
against the ground of the rest of the scene. It might have been possible to view
a different element, for instance the well-known theatres or the famous plays in
which he acted or the high fees he earned.
We may ask if reading about an actor in the newspapers is a core element
of the notion fame. I think it is. With fame we associate publicity and with
publicity we associate the media, and newspapers are part of them. Is the
translation adequate? It seems we are on the safe side if we pick out core
elements as translations, because they are prototypical. Television commercials,
although perhaps part of the celebrity scenario, are not core elements really but
rather positioned at the fuzzy edges. (See Rosch 1973.) Thus a translation such
as the following would not be so good:
The central elements of the source-text scene are "old", "weak", "invalid".
These are the elements which link (or "chain") source and target scenes. The
target scene, however, is more intensified and thus adds a new aspect to the
source scene. The "old-invalid-man scene" is enlarged to the "old-man-close-to-
death scene". Is this enlargement adequate? One could turn things round and
say that the new scene includes the old one. Death implies - as prototypical
notions - sickness and old age. In other words, the larger frame would include
the smaller one. Logically, the translation would thus be adequate, but we have
only looked at two lines of the poem. We would, of course, have to make a
detailed analysis of the complete poem in order to find out if the scene
enlargement, with the mentioning of its details, is justified. There is no room to
do this here: we have just pointed out some aspects of evaluation.
In the example mentioned above about Gus, the theatre cat, as a famous actor, a
frame was replaced by an element of the corresponding scene. The converse
method is the replacement of a scene element by a frame. Mr. Mistoffelees, the
conjuring cat, knows many tricks, for instance
He can pick any card from a pack. (Eliot 1961:59)
Card tricks is a phrase that covers many things. It is a frame for a scene that
consists of many elements. One of them is the element chosen by T.S. Eliot. But
other parts of the scene are also the person who shows the tricks and the
audience who is baffled by them. There can be different kinds of tricks, too. The
magician may guess which card a person from the audience has chosen, or a
person may tell the magician - as in our example - which card to draw, and
there are many more varieties. In Kartentricks the focus is not on a particular
one of these elements, but the word includes them all. We zoom out, as it were,
until we reach the frame so that the whole scene is in the picture.
In the translation the source text has been altered, that is, to some extent
it fulfils the criterion of novelty so essential for creative products. But is it really
creative? It was mentioned earlier that a creative product must not only be new
but also adequate. Semantically speaking the translation is adequate insofar as a
frame always includes the elements of the corresponding scene. But is it also
adequate from a poetical point of view? One might argue that card tricks is less
vivid than picking any card from a pack, and that vividness, after a careful
analysis of T.S. Eliot's Book of Practical Cats, might turn out to be a central
feature of the poems. Thus the translation by Friedrich Podszus in the English-
German version of the Suhrkamp edition, although more literal, would be more
creative because he preserved the vividness of the original by writing:
Er zieht auf Wunsch jede Karte aus dem Päckchen heraus. (Eliot 1961.61)
'At the audience's wish he draws any card from a pack.'
version is more adequate, although less creative, and as a libretto for the
musical the translation by Michael Kunze is an adequate and more creative
version.
Kartentrick is an already existing frame for the scene suggested by the source
text. This is the frame people use for the kinds of mental pictures suggested by
the source text. A new frame, however, would be a tailor-made frame, as it
were; in lexical terms, a neologism.
I asked my students to translate a satirical text about British motorway
restaurants (cf. Kussmaul 1995:27). The assignment was: "Preserve the
humorous tone in order to add to the enjoyment of reading the text! The text
will appear in a booklet on British everyday culture." The second part of the
sentence "Starch is cheap and filling, so sausages composed chiefly of bread
seem somehow right" was translated by a student by "...Hot dogs mit viel Brot
und Würstchen mit Baguette" ('hot dogs with a lot of bread and sausages with
baguette'). In a seminar we discussed this translation, and said that top-down
processes had here probably blotted out the bottom-up material. We tried to
remember what English sausages looked and tasted like, namely that they
consisted of about 50% bread, 40 % fat and 10% meat. One student remarked
that in the German translation the bread was round the sausage and in the
English text the bread was in the sausage. After having agreed that with regard
to the assignment the translation was not adequate one of the students
suggested: "Let us use Brotwurst ('bread sausage')!"
This is a creative translation. It is adequate because it refers to the
British culture and it is new, because the word does not exist in German. From a
cognitive point of view it comprises, i.e. frames, by implication, the elements
mentioned in the source text: cheap, filling, composed chiefly of bread, but it
focuses on the last one. In addition, it is funny, because it is a play on words
{Bratwurst 'sausage for frying' - Brotwurst) and thus preserves the humour of
the source text.
3. Conclusion
The five types of creative translation I have presented are chiefly based on four
types of mental processes: (a) moving from frame to frame (Asterix examples)
(b) moving from frame to scene (the theatre cat) and (c) moving from scene to
scene (again the theatre cat) and (d) moving from scene to frame (the conjuring
cat and the British motorway text). They all involve some degree of focusing,
perspective change or zooming.
126 Paul Kussmaul
References
DE BONO, Edward 1970. Lateral Thinking. A Textbook of Creativity. London: Ward Lock
Educational.
ELIOT, T. S. 1961. Old Possums Katzenbuch. Englisch und Deutsch. Frankfurt am Main:
Suhrkamp.
KUSSMAUL, Paul 1995. Training the Translator. Amsterdam & Philadelphia: John
Benjamins.
LAKOFF, George 1987. Women, Fire and Dangerous Things. What Categories Reveal about
the Mind. Chicago: University of Chicago Press.
Zusammenfassung
In zwei Versuchsreihen haben qualifizierte Dolmetscher, Dolmetschstudenten,
zweisprachige Versuchspersonen und Fachkräfte aus anderen Berufszweigen
Aufgaben gelöst im Zusammenhang mit dem Arbeitsgedächtnis, dem Textverständnis
und dem Zugang zur Semantik und dem Lexikon. In der ersten Serie haben wir
versucht, das Textverständnis unserer Versuchspersonen zu definieren, sowie auch
andere Aspekte dieser Kompetenz (Aktivierung des Lexikons und der Semantik). In
der zweiten Versuchsreihe haben wir versucht die Fähigkeit und Effizienz des
Arbeitsgedächtnisses zu überprüfen. Die Ergebnisse erweisen eine relative
Überlegenheit der Dolmetscher im Bereich der linguistischen Fähigkeiten und in der
Benutzung des Arbeitsgedächtnisses und sind ein Hinweis dafür, dass diese besondere
Strategien benutzen zur Erweiterung ihrer Fähigkeiten und Effizienz und damit ihr
Verständnis verbessern.
Résumé
Au cours de deux séries d'expérience, des interprètes professionnels, des étudiants en
interprétation, des sujets bilingues et des praticiens provenant d'autres domaines du
savoir ont réalisé des tâches incluant la mémoire opérationnelle, la compréhension et
l'accessibilité lexicale et sémantique. Lors de la première série, nous avons cherché à
établir la capacité de compréhension de nos sujets, mais aussi à cerner certains aspects
relatifs à cette activité (activation lexicale et sémantique). La deuxième série de tests
visait à établir la capacité et l'efficacité de la mémoire opérationnelle des interprètes.
Les résultats ont montré la supériorité des interprètes quant aux compétences
linguistiques et l'usage de leur mémoire opérationnelle, car il a été prouvé que ceux-ci
acquièrent des stratégies spécifiques dans le but d'améliorer leurs capacités et leur
efficacité, renforçant ainsi leurs processus de compréhension.
128 Bajo & Padilla
Resumen
En dos series experimentales, intérpretes profesionales, estudiantes de interpretation,
bilingües y profesionales de otros campos del saber realizaron tareas relacionadas con
la memoria de trabajo, comprensión, acceso léxico y semántico. En la primera,
intentamos definir la habilidad de comprensión de nuestros sujetos, asi como aspectos
diferentes relacionados con esta habilidad (activación léxica y semântica). En la
segunda serie, intentamos définir la capacidad y eficacia de la memoria de trabajo de
los intérpretes. Los resultados mostraron la superioridad de los intérpretes en destrezas
lingüisticas y en el uso de su memoria de trabajo, al indicar que desarrollan estrategias
peculiares para ampliar su capacidad y eficacia, y asi potenciar sus procesos de
comprensión.
1. Introduction
2. Hypothesis
The first question that we explored was whether the comprehension processes
involved in interpreting are different in nature from those involved in normal
reading or listening and if so, how. According to psychohnguistic theories of
discourse comprehension (Kintsch 1988; Gernsbacher 1990) two groups of
processes play an essential role in comprehension, especially when it takes
place in conditions of simultaneity (Gile 1997). First, linguistic processes such
as lexical/semantic access, propositional analysis, syntactic processing etc.
need to be performed efficiently to construct a mental representation of the
discourse. Second, working memory capacity and resources need to be
distributed among the different processes involved. In our experiments we have
focused our attention on these linguistic and memory processes, and we have
attempted to determine whether they are performed differently in interpreting.
Dillinger (1994) has recently suggested that interpreters do not carry out
comprehension processes in ways that differ qualitatively from the ways in
which other people carry them out. The ability to interpret, that is, to
simultaneously understand, translate and produce discourse, is only due to the
greater linguistic skills that these people possess because of their degree of
bilingualism. That is, knowledge and use of two or more languages enhance the
linguistic skills of the interpreters and these enhanced abilities make it possible
for them to simultaneously understand and produce language. Alternatively,
other proposals (Daro & Fabro 1994; Padilla, Bajo, Carlas & Padilla 1995)
emphasise the role of working memory and linguistic training as determinants
of the interpreting skill. This debate has important pedagogical implications,
because if the first hypothesis is proved, much of the training given to potential
interpreters in cognitive skills would be irrelevant, and it should be oriented
only to increase knowledge of their second or third language. Therefore, an
important question in our experiments was to determine whether the
interpreters' training and experience had an effect over and above the possible
effect of knowledge and use of two languages.
130 Bajo & Padilla
3. Subjects
However, there were significant differences in reading times among the groups
(F(3,34) = 2.77, p = 0.05), in the analysis of variance. Comparisons between
the different groups showed significant differences between the interpreters
132 Bajo & Padilla
and non-interpreters (p = 0.01) and between the interpreters and the bilinguals
(p = 0.05). Figure 2 shows mean reading times obtained for each group.
Results seem to indicate that the interpreters show greater efficiency in
processing the information present in the text. They were capable of reading at
greater speed with no decrease in accuracy in their understanding and recall of
the text. Our first experiment thus indicates that the superiority shown by
interpreters is not only due to the linguistic skills derived from their knowledge
of two languages, but possibly to the training in some other skills involved in
comprehension. In our second and third experiments we attempt to clarify
whether other linguistic processes normally trained in interpretation schools are
also enhanced in the interpreters.
The categorization task also showed the superiority of the interpreters. In this
task we presented pairs of words and the subjects had to decide if the concept
denoted by the second word was a member of the category represented by the
first. An important variable in categorization is the typicality of the exemplars
of the category (Rosch 1975). Many categorization experiments have shown
that response times depend on typicality, with more typical exemplars being
responded to faster than less typical ones (Rosch & Mervis 1975). For this
134 Bajo & Padilla
These results seem to indicate again that the linguistic skills of the interpreters
are superior to those of the bilinguals, and that, therefore, different processes
from those involved in second language competence may be responsible for
this superiority.
apart from knowledge of the second language (as Dillinger suggests, 1994),
specific training in other linguistic skills leads to better use of them.
With this purpose in mind, at the end of the academic year we asked
subjects in the student group to take part in the second phase of the study. Our
aim was to compare their performance in the second phase with that of the first
phase. Thus, the students performed the comprehension, categorization and
lexical decision tasks once at the beginning of the academic year, before
receiving training in linguistic skills, and then at the end of the year, after the
intensive training received during this period (8 subjects participated in both
phases). To take into account the possible effect of practice on the tasks, a
group of 10 university students (non-interpreters) also participated in the
experiment as controls. They participated at the beginning of the academic year
(phase 1) and at the end of it (phase 2). If, in our previous experiments, training
was the cause of the interpreters' superior performance, we expected that
interpreting students would show improved performances in the second phase.
Students Controls
Figure 5 shows mean reading times for interpreting students and controls as a
function of phase. As can be seen, only the group of interpreting students
experienced a reduction in the response times from the first phase to the second
(t(15) = 2.08, p = 0.06). The slight reduction in time from phase 1 to phase 2
experienced by the control group was not significant (t(16) = 0.89, p = 0.38).
The categorization task showed the same pattern of results (Figure 6).
The control group showed no significant improvements in any of the three
conditions of the task. Thus both typical and non-typical members were
categorized equally fast in the first and second phase of the experiment, t(16) =
- 0.63, p =,0.53 for typical exemplars and t(16) = 0.61, p = 0.55 for non-typical
136 Bajo & Padilla
Finally, Figure 7 shows the results of the lexical decision task. As can be seen,
there is a tendency for the interpreting students to respond faster to words in
the second phase (t(15) = 1.86, p = 0.09). This improvement was significant in
the non-word condition (t(15) = 3.37, p = 0.04). In contrast, response times for
the control group remained identical in both phases and conditions (t(15) = -
0.17, p = 0.86 and t(16) = 0.82, p = 0.42).
Our results seem to indicate that linguistic skills such as access to
semantic and lexical information that are particularly developed in interpreters
are not only due to the greater linguistic skills that knowledge of two languages
gives them. This superiority did not show in the bilingual group, which also
had linguistic competence in two languages. Our data show that these skills are
susceptible to training, as our interpreting students performed the tasks much
more effectively after training.
Comprehension processes 137
capacity will be more capable of carrying out these processes and will show
greater reading skills.
Daró and Fabbro's research (1994) has also shown the importance of
this memory when the task is performed in the conditions of simultaneity
imposed by interpreting. For this reason we thought that the greater skill of
interpreters in the comprehension task could be due to their greater short-term
memory capacity. Therefore our next objective was to obtain data on the
memory capacity of our interpreter subjects. In this second series of
experiments, we omitted the group of bilinguals, since there was no particular
reason to suppose that they might be superior and our previous experiments
had showed that their linguistic skills were the same as those of the students.
In order to measure working memory capacity we used two tasks that
have been widely used for this purpose: the digit span task and the phrase span
task. In the digit span task, three sequences of four digits were presented first.
Then the number of digits was gradually increased until it was impossible for
the subjects to recall them in order. In the phrase span task, subjects were
presented with a group of phrases and asked to recall the last word of each
phrase. The number of phrases in the group was gradually increased. The size
of the largest group of phrases in which all the words were recalled represented
the memory span of that subject. This task was also included because it
simultaneously measures the processing and storage capacities of working
memory, while the digit span test is more related to the storage capacity.
Daneman and Carpenter (1980) have shown that this task can predict
individual differences in reading skills.
Figure 8 shows the memory span obtained by each of the groups. As
can be seen, the interpreters' memory span as measured by the two tasks was
higher than that of the rest. The analysis of variance computed on the digit span
data showed the differences to be significant (F(2, 27) = 3.26, p = 0.05). The
analysis on the phrase span data also showed the differences to be significant
(F(2,27) = 5.32, p < 0.01).
In addition we calculated the correlation between the scores obtained by
our subjects in these memory tasks and the reading times obtained in the
previous experiments. The Kendall correlation between phrase span and
reading times was significant (K(25) = 0.24, p = 0.05). These results suggest
that the memory capacity of the interpreters is larger than that of non-
interpreters or interpreting students, and that this larger capacity may be the
basis for their efficient reading skills.
Daneman and Carpenter (1980) suggest that the greatest source of
differences in reading skills is not in the passive storage of the information but
in the most active part of the processing of the working memory (the central
executive, in Baddeley's term). People differ in their functional capacity, that is
to say, in the processes they perform in order to use the limited resources they
Comprehension processes 139
D Digits
□ Phrase
It thus seems that the interpreters' superior comprehension is not only due to
their greater speed in accessing semantic information, but also to their greater
short-term memory capacity and to their efficient use of this memory.
8. Conclusions
quicker when accessing lexical and semantic information stored in their long-
term memory. Second, they are able to perform a series of mental tasks without
the support of a rehearsal component in their working memory (the articulatory
loop). Contrary to Dillinger's results, interpreters' superiority in
comprehension processes is not due to the linguistic abilities acquired by
knowledge of a second language, as our bilingual subjects did not show that
superiority.
In agreement with other theoretical approaches in interpreting (Daró &
Fabbro, 1994), our data support the key role of working memory in
simultaneous interpreting. Efficiency in comprehension seem to be due to
better management of working memory resources. Our data show that better
reading abilities go together with better performance in tasks implying the use
of working memory. An important component in improving comprehension is
the capacity to compute syntactic and semantic relationships between
successive words and phrases in order to build a coherent representation of the
discourse. For new information to be integrated with previously processed
information one must have access to the results of previous processes. Storing
this information temporarily is of crucial importance to understanding.
Therefore, people with a larger memory capacity will be more capable of
storing information and, consequently, will be better able to execute a series of
sub-processes needed for comprehension (looking for referents, making
inferences, solving ambiguities etc.).
According to our conclusions, the training of future interpreters must be
focused, to a large extent, on an efficient use of their memory, in such a way
that intensive practice in related tasks has immediate repercussions on the
acquisition of the technique of interpreting.
1. Preparation of this paper was supported by Grant PB95-1180 of the DGICYT (Ministerio
de Educación y Cultura) of the Spanish Government to the first author. The authors are
indebted to José Canas for his help in preparing and discussing many of the experiments
presented in the paper.
References
BADDELEY, A. 1990 Human Memory: Theory and Practice. Hove: Lawrence Erlbaum
Assoc.
DANEMAN, M & CARPENTER, P.A. 1980. "Individual differences in working memory and
reading''. Journal of Verbal Learning and Verbal Behavior 19, 450-466.
DANKS, J.H.; SHREVE, G.M.; FONTAN, S.B. & MCBEATH, M.K., eds. 1997. Cognitive
Processes in Translation and Interpreting. London: Sage Publications.
142 Bajo & Padilla
DARÓ , V. & FABBRO, F. 1994. "Verbal memory during simultaneous interpretation: effects
of phonological interference". Applied Linguistics 15 (4).
GILE, D. 1995. Basic Concepts and Models for Interpreter and Translator Training.
Amsterdam & Philadelphia: John Benjamins.
KINTSCH, W. & VON DIJK, T.A. 1978. "Toward a Model of Text Comprehension and
Production". Psychological Review, 85 (5) 363-394.
PADILLA P., BAJO M.T., CANAS J.J. & PADILLA F. 1995. "Cognitive processes of
memory in simultaneous interpretation". Topics in interpreting research ed. by J. Tommola.
61-71. Turku: Centre for Translation and Interpreting.
ROSCH, E. & MERVIS, C.B. 1975. "Family resemblances. studies in the internal structure of
categories". Cognitive Psychology 7, 575-605.
AT THE CROSSROADS OF TRANSLATION
STUDIES AND IMAGOLOGY
Nedret Kuran-Burçoglu
Bogaziçi University, Istanbul, Turkey
Zusammenfassung
Der Artikel befaßt sich mit dem Einfluß von "image of the other" (Bild vom Anderen)
auf den Übersetzungsprozeß. Der Einfluß kann an drei Stellen wirksam werden: (1) vor
dem Übersetzen, (2) während des Übersetzens und (3) während des
Rezeptionsprozesses. Diese drei Stellen repräsentieren Bereiche, in denen die zwei
Disziplinen Übersetzungswissenschaft (Translation Studies) und Imagologie
(Imagology) einander fruchtbar beeinflussen und dadurch ihre Perspektive erweitern
und der Forschung neue Wege weisen können.
Résumé
Quel est l'impact de "l'image de l'autre" sur le processus de traduction? Cet impact
peut avoir lieu à trois moments différents: 1) avant la traduction, 2) pendant, puis 3)
lors de la réception. La traductologie et l'imagologie pourraient s'influencer avec profit
en abordant ces impacts, de façon à élargir leurs perspectives et à ouvrir aussi de
nouveaux champs de recherche.
Resumen
Este articulo estudia el impacto de las imágenes sobre el fenómeno traductor. Antes de
centrarse en ejemplos, distingue tres posibles ámbitos de influencia entre las dos
disciplinas. Éstos son 1) antes de la traducción, 2) durante el proceso traductor y 3) en
el proceso de recepción. Finalmente, ilustrando cada ámbito con ejemplos, pretende
demostrar lo importante que resulta tanto para los Estudios de Traducción como para el
Estudio de la Imagen considerar las influencias reciprocas para asi ampliar sus
perspectivas y abrir nuevas areas de investigación.
144 Nedret Kuran-Burçoglu
1. Introduction
Roland Barthes states in his book Mythologies that images are not God-given,
but created by men; he goes on to say that
Indeed there is much evidence demonstrating that images are created by man. If
the processes used in creating the images are properly assessed, several
constraints will be observed that are significant in shaping the final product.
These are temporal and spatial, as well as socio-cultural and political
constraints, i.e. value judgements, prejudices, conventions, habits, traditions,
political and power relationships. Within the signification process, these are
concretized as representations, the corpus of which will eventually create the
image. I use "concretized" in the sense of Iser (1972, 1976). "This term is
sometimes used synonymously with realization, but it carries the strong
implication of making actual what was actually virtual" (Mario Valdes in
Makaryk 1994:527).
A critical approach to this creation process challenges the reliability of
the images. Roland Barthes' view that the meanings attached to the images may
well appear "natural" and "immanent" deepens this scepticism and concern even
more, as it suggests that images may be misleading and thus insidious because
they seem "innocent" and "naive" at first sight.
This conclusion justifies Edward Said's criticism (1978, 1993) of the so-
called "orientalist gaze," as well as Frantz Fanon's (1986) and Homi Bhabha's
(1994) cautious approach to images that fall within "postcolonial discourse".
Another example can be cited from the Image Studies carried out by
imagologists in the analysis of "Nazi discourse". Feminist, gay and lesbian
studies are also greatly concerned about the attitude of the public towards their
subject matter. Such examples indicate how powerful images can be in shaping
human relationships by various means and on various levels. Imagology, which
investigates the genesis of "the image of the other," as well as the socio-cultural
constraints influential upon its creation and transformation processes, tries to
highlight these complex phenomena.
Within the scope of Translation Studies, "the image of the other
(culture)" and its imagological analysis should, in my mind, not be
underestimated, as this image very often plays a formative role in the translation
phenomenon, and the translations in turn may have an initiating, formative or
Translation studies and imagology 145
The stages of the translation phenomenon where the image of the other may
have an impact are the following. (See also Kuran-Burçoglu 1995, 1998,
forthcoming (a).)
(1) Prior to the translation process. The translator's choice of the text s/he is
going to translate may be directly or indirectly guided or influenced by the
image of the other. This decision (among others) is designated by Gideon Toury
as a "preliminary norm" in Translation Studies (Toury 1980, 1985).
(2) During the translation process. During the translation process the translator
has to make a number of decisions which are reflected in the product, i.e. the
target text, in terms of omissions, additions or lexical choices. During this stage,
designated the "encoding process" in Semiotics and Communication Studies,
the image of the other, in the mind of the translator, can influence the target
text.
(3) During the reception process of the target text. The reader's choice as to
which translated text to read, as well as the so-called reception process of the
target text, designated as such by Communication and Reception Studies, may
also be influenced by the image of the other.
After having pinpointed the potential intersecting areas of the two disciplines, I
would like to cite examples for each case.
146 Nedret Kuran-Burçoglu
The first of two examples illustrating the first case is taken from the history of
Turkish translation, which has two significant eras: the Tanzimat and the
Cumhuriyet. The former is the so-called "Renovation era," which started with
the Declaration of Tanzimat in 1839 and lasted into the 1850s, while the latter
began in the years following the Declaration of the Turkish Republic in 1923,
and lasted through the 1940s. These eras mark the two periods of radical
transition in Turkish culture from Eastern to Western cultural norms and values.
It can be justifiably claimed that the Tanzimat era had paved the way for the
Cumhuriyet era. During these two periods of modernization in Turkish culture,
which were naturally accompanied by a series of reforms, translation gained
impetus. Special institutions such as the Encümen-i Danis, and the Tercüme
Bürosu, in the 1840s and 1940s respectively, were set up to select works that
should be translated within the scope of a program encompassing the translation
of world masterpieces into the Turkish language. Among them, works from
Western literature occupied a significant place. In the Cumhuriyet era they were
thought to serve as models for a new nation that was undergoing a cultural
restructuring. Translators were encouraged by competitions, as well as by
commissions from such state institutions as the Ministry of Education and the
Ministry of Culture. During the Tanzimat era, as well as during the Cumhuriyet
era, "Western culture" was associated with concepts such as developments in
technology, rationalism, positivism, secularism and modernism by the Turkish
elite, comprised principally of medical doctors and military men, who naturally
linked them to the Enlightenment. These concepts emerged as a positive image
of the other as opposed to a slightly negative image of Eastern culture, which
was then associated with ideas of backwardness, superstition and
fundamentalism.
The second example illustrating influence prior to the translation itself is
taken from German cultural history during the German Enlightenment period
starting in the 1720s, when the society was undergoing significant change.
German intellectuals emphasized the effectiveness of translations in reshaping
the culture of the German community according to models taken from particular
European countries whose cultures were highly esteemed. These countries were
France and England. While Johann Christopher Gottsched - German
philosopher, theologian and literary scholar - was in favor of French neo-
classical models, German poets like Bodmer, Breitinger and Gotthold Ephraim
Lessing were in favor of English models, as they thought that the English values
would be more appropriate for them to adopt than any others (Kuran-Burçoglu
1995). These judgements, diverse though they were, were actually both based
on favorable images of those particular cultures, and the priorities of these
intellectuals, quite influential men in their country, played a significant role in
Translation studies and imagology 147
the choice of the literary works to be translated into German in that period of
German history.
These two examples show how the image of the other was instrumental
in shaping period norms in particular countries at turning points of their cultural
histories; it also affected the choice of translators, either directly through
commission by the state and its institutions, or indirectly through public opinion
shaped by policy makers and/or influential people, such as the intellectuals of
the country.
For the second intersecting stage of the two disciplines, the translation process
itself, I will cite two examples based on translations of the Flemish poets
Timmermanns and Karel van de Vosteijne into the German language. The
examples are taken from Johan Soenen's article "Imagology and Translation"
(1997). In this article Soenen also emphasizes the significant impact of the
image of the other on translation. He draws attention to particular stylistic
changes, mitigations and intensifications in content, omissions and additions
made by the translator during the translation process, which transform the
original text wittingly or unwittingly according to "the stereotyped image."
Soenen considers those as constraints that stem from "image-bound
interpretations" on the part of the translator (Soenen 1997:128-129)
He first elaborates on the "stereotyped Flemish image" of the period in
Germany, a mixture of sensuality and mysticism, and then points to the
popularity of German translations of the Flemish author Timmermanns' works
into German, which confirm and justify "the existing Flemish Image" in the
minds of the German public.
Soenen then discusses the case of Karel van de Vostaijne, whose works
have also been translated into German in such a way that they reflect "the
existing Flemish image" in the minds of the target public, which actually
represents a distortion of the source text. In order to satisfy the receptor
audiences and justify "their existing stereotyped image" the translator has
transformed the innocent, almost platonic love scenes of the original poems into
erotic scenes through using stylistic elements and emotive words that reflect
sensuality and mysticism (ibid.).
These first two intersecting areas are called "the linking points of
Comparative Literature and the Translation phenomenon" by the imagologist
Hugo Dyserinck (1981). The impact of the image of the other in these two areas
inevitably leads to the third potential influence area, the reception stage.
148 Nedret Kuran-Burçoglu
At this third stage, under certain conditions, the following kinds of results can
be observed.
If the target reader is unfamiliar with the source culture, i.e. if translation
is for her/him the first means of getting to know the foreign culture, and hence
the image of the other is not developed yet, translation may strongly influence
the newly emerging image. In this case the image of the other which is in the
minds of the commissioner and the translator will, via the skopos (Vermeer
1990) and its implementation, have a particularly strong impact on the creation
of the image of the other in the mind of the reader. Translations done from the
languages of remote cultures for the first time represent channels through which
these foreign cultures are introduced into the polysystem, in Itamar Even-
Zohar's (1979) terms, of the target culture. Examples of this situation from
former centuries are abundant, but there are also recent cases.
Here I will cite an example that is taken from another recent study done
by Johan Soenen on the image of the Turkish literature in Flanders. The list of
translated works from Turkish literature into Flemish shows that up to recent
times Dutch and Belgian readers have not been familiar with this literature, but
that during the 1980s and 90s translations from this literature have flooded the
book markets. The reception research Johan Soenen did on Dutch and Belgian
readers has revealed that the Turkish literature is received as "the reflection of a
world of harsh realities, brutal current affairs, as well as the existing opposition
to it" by the reading audience (Soenen, forthcoming). The author of this study
explains this "one-sided," "stereotypical," "caricatural," "incomplete," and
"subjective" image (ibid.) as a natural result of the works that have been
selected for translation from the inventory of current Turkish literature by the
publishing houses, i.e. the commissioners, as well as by the translators. These
were books written mainly by authors who for socio-political and economic
reasons have migrated to Belgium or to other European countries. It is the
ideological content of the works - the violation of human rights, problems of
democracy, high inflation rate, the Kurdish problem, as well as religious
fundamentalism and the disadvantaged position of women - rather than their
literary quality that seems to have attracted the commissioners and the
translators to select them for translation, as this would justify the "stereotyped
image" of the Turk in the minds of European readers in the 1990s (ibid.) and
thus guarantee a success. The sudden increase in the number of such books that
were translated from Turkish to various European languages after the military
coup of 1980 in Turkey indicates that translation has certainly played a
significant formative or transformative role in the image of the Turk in Europe
today.
Translation studies and imagology 149
4. Conclusion
References
BARTHES, Roland. 1957. Mythologies. Paris: Le Seuil. Translated by Annette Lavers, 1972.
Mythologies. New York: Hill and Wang.
BHABHA, Homi K. 1994. The Location of Culture. London & New York: Routledge.
FANON, F. 1952 Peau noire, masques blancs. Paris: Le Seuil. Translated in 1986. Black
Skin, White Masks. London: Pluto.
KUHN, Anette. (1985). 1987. The Power of the Image. London, New York: Routledge and
Kegan Paul.
150 Nedret Kuran-Burçoglu
KURAN-BURÇOGLU, Nedret (ed.) (Forthcoming, b). The Image of the Turk in Europe from
1923 to 1990s.
SAID, Edward. 1993. Culture and Imperialism. New York: Vintage Books.
TOURY, Gideon. 1980. In Search of a Theory of Translation. Tel Aviv: The Porter Institute
for Poetics and Semiotics.
TOURY, Gideon. 1985."A Rationale for Descriptive Translation Studies". The Manipulation
of Literature ed. by Theo HERMANS, 16-41. London & Sydney: Croom Helm.
Translation effects
FOREIGNIZING STRATEGIES IN DRAMA
TRANSLATION
THE CASE OF THE FINNISH OLEANNA
Ritva Leppihalme
University of Helsinki, Finland
Zusammenfassung
Eine exzessive Treue zu ausgangstextuellen linguistischen Anweisungen kann bei
Dramenübersetzung Probleme durch Erschwerung der Relativierung der adressierten
Inhalte beim Zielrezipienten hervorrufen. Das Stück Oleanna (1992) von David
Mamet war Anlaß für beträchtliche Debatten in der englischsprachigen Welt, während
die finnische Produktion in Helsinki (1994) nur geringen Einfluß hatte. Dieser Artikel
beinhaltet die Behauptung, daß die finnische Rezeption wenigstens teilweise durch die
bewußte Übersetzer-Entscheidung zur sprachennahen Imitation des Dialogs von Mamet
hervorgerufen war.
Résumé
Au théâtre, une traduction à la fidélité excessive peut mettre le public-cible en difficulté
face aux questions soulevées par l'oeuvre originale. Si la pièce de David Mamet,
Oleanna (1992), suscita de multiples controverses dans les pays anglophones, en
Finlande en revanche, sa traduction (1994) ne souleva guère de polémique. Cette tiède
réception peut être attribuée en partie à la décision du traducteur d'imiter, verbalement,
le dialogue de Mamet.
Resumen
El publico receptor de una traducción teatral puede experimentar dificultades para
identificarse con el tema si la traducción ha mantenido una fidelidad excesiva a los
recursos lingüisticos del texto original. La obra de David Mamet, Oleanna (1992),
suscitó una gran polémica en el mundo angloparlante mientras que una producción
fmlandesa puesta en escena en Helsinki (1994) tuvo poca repercusión. El presente
articulo sugiere que la recepción que tuvo en Finlandia se debió, al menos en parte, a la
decision consciente del traductor de seguir literalmente a nivel verbal los diálogos de
Mamet.
154 Ritva Leppihalme
1. Introduction
Example (1)
Source text:
Target text:
JOHN: . . . mitä?
\ . . what?'
(Mamet 1993b:2)
Word for word, the target-text lines certainly replicate the fits and starts of
Mamet's dialogue, but without its basis of familiar domestic speech patterns.
Thus the translation deprives the audience of a disturbing feeling of recognition:
that this is how many of their own conversations go, too.
Such pragmatic differences between American and Finnish conversation
styles are particularly noticeable on the stage, where the presence of actors
speaking the language of the audience creates the illusion that the characters
belong to the same language culture as the audience. While dramatic dialogue is
not intended to represent authentic conversation with the same degree of
accuracy as do transcripts made for conversation analysis, and Mamet's
dialogue is clearly stylized, there is still a need for interaction that makes sense
in pragmatic terms. Mechanical repetition of the interlocutor's words does not
always work very well in the translation as Finnish speakers would sometimes
choose to repeat other words in the lines: in the above extract, for example, the
repetition of puhua 'to talk' suggests wonder at the idea that with two people in
a room, one perhaps wishes to talk about something to the other! What is more,
many conversational strategies are different in Finnish. The pace of Finnish
speech is usually slower than that of (American) English and the toleration of
silences between speeches much higher (Kukkonen 1993:100-111). In Finnish
conversation there would likely be less of a scramble for the turn to speak and
less of a desire to interrupt one's interlocutor; in fact, Finnish speakers
commonly use silences to indicate non-comprehension and disagreement. In an
alternative Finnish translation of Oleanna, the techniques used by Mamet to
indicate lack of communication could perhaps at times be replaced by other
strategies.
Any representation of dialogue in Finnish needs to be in one of two
alternative registers. The dialogue of the play can be written in formal literary
language — for a long time the only possible style on the stage except the
country dialects spoken mostly by comic characters. Or the characters could be
made to speak colloquial urban spoken Finnish, the language of real-life Finnish
students and professors. The latter is a choice often made when contemporary
plays, either domestic or translated, are produced, and has been used
successfully in Finnish productions of Mamet's plays both before and after the
production of Oleanna. The use of the formal register almost inevitably results
in a distancing effect, as this is the language variety that tends to be spoken in
public, in front of audiences, but not in face-to-face encounters. The translator's
choice of the formal register in the dialogue of John and Carol may well be
symptomatic of the distance between the two speakers; but it also, at the same
Drama translation 157
time, distances the speakers from the audience. If this distance is increased
further by lines that sound un-Finnish, this is also likely to have an effect on the
responses of the audience. I will return to this point below (section 3).
A further difference between the translated dialogue and real-life Finnish
speech is the use of question tags and pragmatic particles. If English usage is
reproduced, as in this translation, so that nearly every well, all right, and yes is
retained and rendered by their dictionary equivalents, their communicative
significance is overlooked. As Östman (1982:144) points out, "the manner of
expressing pragmatic information varies from one language and culture to
another, and [...] the need for expressing this information via explicit
manifestations or implicit suggestions will also vary across languages". The
literal approach adopted by the translator also leads to a striking absence of the
Finnish pragmatic particles that would be common in precisely the sort of
situationsMamet'ss text represents, where speakers find neither understanding
nor agreement.
One relevant difference between English and Finnish is the opacity of learned
English vocabulary compared to the transparency of corresponding words in
Finnish. Towards the end of Act I, Carol is uncertain about John's word
predilection (Mamet 1993a: 31), checking whether it means "liking". In Act III
she requests that John explain to her the word indictment (63). A little later
(66), she erupts in fury at the word transpire and demands that he say happen
instead. While Finnish also has its share of Greek and Latin loan words, these
are generally either fully domesticated everyday loans (such as hotelli 'hotel',
televisio 'television'), alternatives to more common transparent words (for
example traditio 'tradition' = perinne) or strictly technical terms in various
special fields (diskurssi 'discourse', metafora 'metaphor', elektroni-
mikroskooppi 'electron microscope'). Finnish semantic equivalents of vast
numbers of learned loanwords in English are by contrast of vernacular origin.
With such difference in lexis, is it possible to convey the implications of
John's choice of opaque words — partly perhaps the reason why he is labelled
elitist in Carol's report to the tenure committee? Can a distinction be made
between Carol's preferred everyday vocabulary and the academic vocabulary
which is part of the professorial mask behind which John tries to hide? If we
look at the three instances singled out above, we find indictment rendered with
a technical legal term, kanne 'action, suit' (Mamet 1993b:37); and predilection
with lukkarinrakkaus (18), a jocular and somewhat dated expression of roughly
the required meaning. Both renditions work reasonably well even though they
are derived from mainly domestic roots: it is conceivable that Carol might be
uncertain about the precise meaning of kanne and unfamiliar with the other
158 Ritva Leppihalme
Example (2)
CAROL: "Happen"?
JOHN: Yes.
CAROL: Then say it. For God's sake. Who the hell do you think that you are?
(Mamet 1993a: 66)
the rendition into Finnish is weak. John is made to say Ja mitä seuraisi 'And
what would transpire?' (Mamet 1993b:39). Carol violently demands that he use
the word tapahtua 'happen' instead. Unhappily, both words are equally
standard items of Finnish vocabulary with no difference in distribution or
register. This deprives Carol's outburst of all logic in the framework of the play.
From the functional point of view and as an indication of the different
vocabularies it might be a useful translation strategy to pepper John's lines with
a number of scholarly words in Finnish by way of compensation, even where
such words do not occur in precisely the same lines in English. The occurrences
in the source text of words like paradigm, which Carol also queries (Mamet
1993a:45), are helpful in that such words are part of Finnish academic
vocabulary as well and hence can be retained (nearly) unchanged. An item of
restricted register, even with a different meaning, could be substituted for And
what would transpire?, making Carol's sudden rage more intelligible. Such a
strategy would be justified on the grounds that functional equivalence takes
precedence here over semantic equivalence.
3. Discussion
This paper has established that the global translation strategy chosen by the
Finnish translator of Oleanna ignores several pragmatic differences between
American English and Finnish, and does not always result in appropriate
semantic renderings for crucial words in the dialogue. The dialogue thus sounds
unnatural and distant, with the occasional rendering even puzzling for the
audience. Thus the literal translation weakens the dramatic effect of the play.
Mamet's dialogue is known to be carefully crafted, and in Oleanna it
emphasizes the miscommunication caused by the different speech styles of the
two characters. Their interruptions, repetitions and evasions try to give an effect
of authentic conversation and hence to bring the spectators to a realization that
despite the torrent of words there is little true contact between the two
Drama translation 159
characters — and that this might be true of their own lives as well. The
translator's decision to imitate Mamet's dialogue as closely as possible on the
verbal level is doubtless based on respect for the author's use of language, but it
fails to take into account the differences in source and target language cultures:
it does not enable the target audience to recognize familiar speech patterns and
so to become aware of the lack of communication. Though the words of the
characters are in Finnish, the conversation is not Finnish but foreign.
Judging by a short paper by the translator (Baran 1998:36-39), it is a
deliberate decision on his part to translate plays as literally as possible. "I would
rather be too literal than provide creative approximations", he says when
discussing his principles, though without reference to any particular play
(translation RL). Baran also addresses the choice of the formal "book language"
register when he asserts that the informal variety he rejects aims at "making the
translated discourse sound tougher, more fluent, more modern" (38), and that
such "fluency strives for a feeling of familiarity rather than for the truth" (38).
He thus rejects familiarity as a goal, preferring to foreignize.
Foreignizing strategies may of course be legitimate and even
commendable under certain circumstances. Lawrence Venuti (1995) argues that
fluency in translation is the result of domesticating strategies and that a fluent,
"invisible" translation often expresses repression of cultural differences,
ethnocentrism, and even xenophobia. In his view foreignizing strategies "open
up a space for something that's less ethnocentric" (47). But Venuti admits that
his argument mainly applies to a situation where the target culture knows very
little about the source culture (48), in other words where the source text
represents a little-known and peripheral culture in the eyes of the target
readership; and that is clearly not the case where a contemporary American play
by an acclaimed and well-known author is translated for theatregoers in what
has (rightly or wrongly) been called the most Americanized country in Europe.
The reception of the play in Finland can be characterized as indifferent.
While the play's issues of sexual harassment2 and political correctness may well
be less explosive in Finland, with a tradition of considerable equality between
the sexes and a relatively homogeneous population, surely miscommunication,
lack of contact and abuse of power are equally thorny problems in both
language cultures. And yet, the Finnish production failed to provoke much
discussion or debate. When I saw the translated version of the play in Helsinki it
was evident that during the intermission and after the performance there was
neither excitement nor heated discussion among members of the audience,
contrary to what had been reported to be the case in American theatres where
"loud arguments broke out" (Ebert 1995), and the play provoked "vigorous
after-show discussion and debate" (Gist 1996). What commentary there was in
Helsinki seemed to focus on Carol as a student presumed to be neurotic. The
larger issues were thus overlooked. Of course, there are cultural differences in
160 Ritva Leppihalme
4. Conclusion
The lack of interest in the play in Finland may well, as argued in this paper, be
related to its translation. In Christiane Nord's words,4 cultural distance is
incompatible with the appelative function. While faithful on the verbal level, the
Oleanna translation disregards pragmatic considerations and differences in
source and target semantic-lexical resources, and thus deprives the play of much
of its dramatic tension as it fails to create human contact between the characters
and the audience. Some familiarity with conversational analysis and with
intercultural differences in conversational styles might have contributed to a less
literal but more effective version, closer to what Mamet was aiming at. A
misguided attempt to respect the language of a famous dramatist thus led to a
translation that did less than justice to his text.
Drama translation 161
Notes
1. Baran is the theatre's dramaturge. A dramaturge in a Finnish theatre adapts texts for use
in the theatre, and can also write and translate such texts (Aaltonen 1996:28, 213, 219).
While it would be fruitless to speculate on circumstances which may have had an effect
on this particular translation (such as the amount of time at the translator's disposal or
the expectations of the director, Kurt Nuotio), it can be noted that a dramaturge is
generally thought to hold more power in a theatre than do actors or those translators who
are not directly connected with production (Aaltonen 1996:54).
2. In 1995, an inquiry at Helsinki University showed that 2.7% of students and 6.8% of staff
had encountered some sexual harassment in the past two years. C. 75% of the
respondents were female. I would like to thank Teija Mankkinen for providing these data.
3. Kindly provided by the Finnish National Theatre. All translations from Finnish are mine.
References
BASSNETT, S. 1998. "Still Trapped in the Labyrinth: Further Reflections on Translation and
Theatre". Constructing Cultures. Essays on Literary Translation ed. by S. BASSNETT & A.
LEFEVERE, 90-108. Clevedon: Multilingual Matters.
EBERT, R. 1995. "'Oleanna' film doesn't match play's intensity". Available on Internet:
http://www.kentuckyconnect.com/heraldleader/movies/mq/oleanna.html.
KAJAVA, J. 1994. "Kun hiiri saa kissan hampaisiinsa". Helsingin Sanomat, 14 Feb. 1994.
KERNS, W. 1997. "Mamet's Oleanna challenges cast, prompts audience debate". Available
on Internet: http://www.lubbockonline.com/ news/ 061397/mamets.html
162 Ritva Leppihalme
KNOWLES, R.P. 1997. "Focus, Faithfulness, Shakespeare, and The Shrew: Directing as
Translation as Resistance". Essays in Theatre 16(1), 33-52.
Zusammenfassung
Der Artikel untersucht die Rolle der Übersetzung bei Formung der Konzeptionen von
Demokratie im 19. Jh. Es wird davon ausgegangen, daß die Ideen der Demokratie
weitgehend als Produkt von Übersetzung konstruiert waren. Sie zielte auf die
Einbürgerung der Transzendenz von historisch spezifischen Interpretationen durch
Argumente für deren "Treue" zu einem a priori definierten Original. Sowohl Rhetorik
als auch Praxis der Übersetzung dienten als Mittel zur semantischen Transformation
von Demokratie als eine inhärent ambige Konzeption, die gleichzeitig Glorifikation
von autonomem politischem Gedanken und Gehorsam gegen autoritativen Prinzipien
ausdrückte.
Résumé
Quel est le rôle de la traduction dans la mise en place des conceptions de la démocratie
en Angleterre au 19ème siècle? Selon notre recherche, la notion de démocratie est
largement le fruit de traductions qui ont tenté de dépasser des interprétations
historiquement marquées, en prétendant rester "fidèles" à un original défini à priori.
La pratique traductionnelle, y compris dans sa rhétorique, a donc servi de moyen pour
transformer le sens de démocratie comme concept intrinséquement ambigu - servant à
la fois à la glorification de la pensée politique per se et à la défense de principes
autoritaires.
Resumen
Este ensayo examina el papel de la traducción en la conceptualización de democracia
en Inglaterra durante el siglo XIX. Propone que las ideas democráticas constituyen en
gran medida un producto de la traducción que intentaba establecer la transcendencia
de interpretaciones históricamente especificas al alegar su "fidelidad" a un original
defmido a priori. Tanto la retórica como la práctica traductora sirvieron como medio
para la transformación semántica de la democracia como un concepto intrinsecamente
164 Alexandra Lianeri
After the first decades of the 19th century, translation was explicitly described
by Victorian thinkers as a discourse which pointed outside its immediate
historical context, to a realm of knowledge that was perceived as detached
from the ideological conflicts and cultural contingencies of English society.
Conceived as a means of conveying intact the remoteness of the original
(Bassnett 1991:68), translation had "no pretension to any other merit" as John
Stuart Mill put it, "than that of fidelity" (Mill 1978: 42); a "merit" which not
only implied the assertion of textual transparency, but also located translation
beyond the boundaries of a transient cultural present, in a conceptual space
whose commensurability with this present was also the proof of its
participation in a universal, essentially ahistorical - rather than national or
culturally specific - Weltanschauung. In a typical articulation of this creed
Matthew Arnold, in his famous essay on the translation of Homer, drew a
parallel between the "union of the translator with his original" and the "union
of the human soul with the divine essence", arguing that such a union alone is
the presupposition of a good translation, which can take place "when the mist
which stands between them [the translator and the original] - the mist of alien
modes of thinking, speaking, and feeling on the translator's part - 'defecates to
a pure transparency', and disappears" (Arnold 1960:103). The religious
allusions of this metaphor do not merely set the presuppositions of "good"
translations. By describing faithfulness as the expression of unity between the
human and the divine, Arnold seeks further to set the criteria for the validation
of certain translations as direct realisations of the universal meaning inscribed
in canonical source texts, such as the Homeric epics to which he refers. The
claim to "transparency", which implies, as Lawrence Venuti has argued, the
self-concealment and illusory naturalisation of translation (Venuti 1995:1-42),
is simultaneously the key to the sublimation of the translated text, which
becomes capable of approaching the "truth" at the moment it is deprived of
authenticity and autonomy, and becomes subsumed to the authority of the
original.
Paradoxically, this conception of translation was employed in the 19th
century in order to dispute religious justifications of political truths and values,
challenge the God-given authority of kings and aristocracy, and substitute the
unquestionable supremacy of divine knowledge for the fallible sensibility of
political subjects. A rhetoric which strictly determined the boundary between
"faithfulness" and "betrayal", and defined transparency as the proof of the
translation's validity, was evoked in order to shape and justify the disparate
rewritings and significations of democracy, which later brought about the
notorious semantic ambiguity and contestability of the concept.
166 Alexandra Lianeri
The refutation of this model of political thought, which had long been
established as authoritative and accurate, was actualised through the
"realisation" of the mistakes, limitations and often political prejudices of
previous translators. This is precisely the point J. S. Mill emphasised when he
decided to publish a series of translations of Platonic Dialogues, which began
with one of the most significant ancient political debates on democracy: that
between Socrates and the Sophist Protagoras, included in Plato's homonymous
Dialogue. For Mill, the truth of Plato's political thought was obscured by the
faults of conservative translations and the political biases of interpreters and
commentators who exaggerated and transformed aspects of the source text "for
the base purpose of discrediting free institutions and freedom of inquiry" (Mill
1978:42-43). In his attempt to oppose these interpretive models, Mill was all
too eager to stress the fidelity and transparency of his translation: "it is the
object of these papers" he asserts, "not to explain or criticise Plato but to allow
him to speak for himself' (Mill 1978:60). The rhetoric which highlights Mill's
"invisibility" was a premise of no passing significance. Fully acknowledged as
a dominant norm in the 19th century, the claim to the self-annihilation of the
translator in the face of canonical source texts was the necessary
presupposition for the recognition of the value of Mill's work and the
shortcomings of its predecessors.
Mill's conception of faithfulness evoked a double standard of
evaluation, on the basis of which he could reject all "negative" appraisals of
classical democracy as "mistranslations", while legitimising as accurate the
identification of Greek political thought with his contemporary liberal ideals.
His claim, which would result in a radical semantic transformation of the
concept of democracy, did not merely consist in the use of translation as a
means of political propaganda, which employed the same ideological weapon -
the notion of transparency - as its predecessors. Liberal thought did constitute
a historically "faithful" interpretation of classical democracy during the 19th
The shaping of modern democracy 167
subject: the political principles which were for him actualised in the freedom
and equality of the Athenian democracy. This aim determined the translator's
position towards the two main figures in the dialogue, i.e. Socrates and
Protagoras, whose opinions on the nature of political juxtapose two political
philosophies. The first, expressed by the ancient Sophist, is a moderate support
of relativist thought, a determination of political virtue as primarily existential
rather than essential, in the sense that this virtue can only be defined by
reference to historical human knowledge and evaluation rather than some
absolute and immanent standards. For Protagoras truth can only be human, and
justice, politics and ethics have only man as their measure. The opposing
argument, developed by Socrates, echoes Plato's hostility to the insecure nature
of human, temporal truths and seeks to identify the indelible features of
political knowledge which stand beyond their specific conceptions and
transformations by different men and cultures.
Within a context of strife over the political rights and abilities of
individual citizens, the Protagorean idea that human knowledge should be
considered as the only legitimate measure of political action could justify, for
Mill, not only the acquisition of freedom and the potential for political
participation, but also the renunciation of unquestioned obedience to political
authority lying outside human standards for critique and debate. Perceived as
an early support of liberal democratic ideals, Protagoras's translation found a
discursive location in a broader ideological attempt to restitute a widespread
Victorian image of the Sophistic movement as mischievous and immoral while
(re)constructing its unity with nineteenth-century social struggles and political
debates on democratisation. This move took, in the translation, the form of a
search for the "true" meaning of the ancient notion of 'Sophist', which had
"misleadingly", as Mill argues, become "significative of quibbling and deceit"
(Mill 1978:43). Mill suggested the retranslation of the notion as "a teacher of
wisdom" and explicitly endorsed Protagoras's political insight and
contemporary relevance. For, as he argued in the introduction to his translation,
main components of political virtue, justice and prudence can be expected from
all men reared in democratic communities and constitute an indispensable
presupposition of social coherence and reproduction. Based on the Athenian
example he states precisely this point:
Mill follows the passage relatively closely in his translation, changing only the
notion of political into social virtue:
The Athenians and others ... are ready, when the subject is social virtue, which
depends wholly upon justice and prudence, to listen to all advisers; because of
this virtue all should be partakers, or states cannot exist (Mill 1978:49, my
italics).
We will return to the significance of this change later. The idea expressed in
Protagoras's argument that all citizens should be partakers of "justice" and
"prudence", and therefore all are worthy advisors when it comes to public
issues, was the core principle of the Athenian democracy and provided strong
intellectual support to Mill's political creeds and democratic principles. This
idea, which appears recurrently in Protagoras's speech, is meticulously
rendered by the translator, in opposition to a significant part of Socrates's
thought which is either condensed or entirely omitted. In particular Mill
summarises a speech of several pages (332a-334c) into a mere description of it,
in which he informs his readers that "Socrates forces Protagoras ... to admit
that [prudence] is the same thing with wisdom, that is the same
thing with justice, or at least inseparable from it" (Mill 1978:54). This speech
constitutes a significant part of Socrates's argument that all components of
political virtue are ultimately reducible to "knowledge" and wisdom, in the
sense that knowledge of the "good" and the "evil" is the only presupposition
and legitimisation of human virtue. Mill was not initially negative to the
privileged position knowledge occupied in Socrates's thought. What he
rejected in his translation was the totalising conception of knowledge,
expressed in the source text, as a system of absolute moral and political truths,
and the implied devaluation of critical political judgement and endorsement of
political heteronomy and submission. Against this idea, which dissociated
political virtue from shared social experience and knowledge, Mill juxtaposed
Protagoras's glorification of human political capacity, the belief expressed in
his translation that "it is thus with virtue: all men can teach it" (Mill 1978: 51)
170 Alexandra Lianeri
Despite his faith in human political potential Mill was not willing to endorse a
total relativisation of political principles, which could easily follow the
recognition of equal validity in diversified conceptions of morality and politics.
His attempt to demonstrate the possibility of a self-instituted, rational
democracy, and to sustain a secularised model of political thought, was
versatile and complex enough to accommodate the notion of authority,
however differentiated this notion was from Socratic idealism. Following
Protagoras' thought he sough to suggest that no system of morality, and indeed
no democratic society, could be maintained when devoid of some shared
principles and creeds capable of justifying the very necessity of political
justice, equality, and the self-conscious, responsible conduct of human life.
Hence while Protagoras's speech ostensibly highlights the significance
of personal political opinion, it also points out the existence of universal
principles and historical laws as an indispensable feature of the Athenian
democracy. Justice, prudence, and respect, which are for him the predominant
presuppositions of political virtue, are neither disputable nor historically
relative. As he claims in an important deviation from rational argumentation
which employs the Greek mythical tradition, these values were given by Zeus
to all people when societies were initially formed, and it was Zeus who ordered
men that whoever fails to partake of these values should be killed, as a public
pest (322c-d). Articulated in a time in which myths had certainly lost their
literal value, Protagoras's narrative had more a symbolic than a truly religious
meaning and function. It served to express a system of historically stable
standards of Athenian democratic thought, on the grounds of which individuals
could be educated to become politically capable and responsible citizens. As
Protagoras quickly observes after narrating the myth, the Athenians do not
consider Zeus's gifts as natural and spontaneous, but as the result of careful
education. It is precisely this education, which takes place in the context of
democratic institutions and is guarded collectively by the community, which
creates citizens able to articulate and implement political judgement; as well as
to subsume their selfish interests, when necessary, to the common good. In the
Protagorean conception of the Athenian democracy, political autonomy and
liberty are the indispensable rights of citizens rather than of individuals, and
authority lies with those principles which have been capable of nourishing the
moral and political responsibility of those citizens among themselves as well as
towards the entire city (cf. 323d-328d).
On the treatment of this issue Mill's translation departs from the source text at
two main points: it replaces the notion of the citizen by the notion of the
The shaping of modern democracy 171
All civilised men, even the most unjust, if compared with men among whom
there is no training, no tribunals, no laws, with the wild men [of whom poets
tell us,] would appear a perfect master in virtue (Mill 1978:51).
The translation introduces two significant transformations into the source text.
Mill uses the notion of "civilised men" in order to render the Protagorean
description of men reared within laws and society, and he further substitutes
the idea that these men are educated to be themselves just, as well as "creators"
of justice, in the phrase "perfect master in virtue". The ideal of political
creation expressed in the source text constituted a notion of man not as
individual but qua citizen, whose education derives precisely from his
participation in the political processes of the community and who is able to
express - as part of a social body - moral and political judgements. The
necessity of active political participation is absent from Mill's thought. For
him, as for the majority of 19th-century liberal thinkers, democratic
government requires the agreement rather the participation of the governed in
politics. Democracy itself should be based, Mill argued, not on the often
mistaken and vulgar judgement of the mass of the people, but on the
consensual establishment of an educated and progressive élite which is capable
of political creation: "The idea of a rational democracy is not that the people
themselves govern, but that they have security for good government [...] the
best government (need it be said?) must be the government of the wisest, and
these must always be a few" (quoted in Burns 1969:294). Not surprisingly, the
translation excludes the people from this potentiality of political judgement by
emphasising their "mastery in virtue" rather than their capacity for political
creation acquired in a democratic society. Precisely as in the first extract from
the translation examined in this essay, the virtue of the majority is presented to
be "social" rather than "political" in nature. This substitution accords with
Mill's life-long dedication to a fully extended participatory democracy, which
172 Alexandra Lianeri
does not nevertheless recognise the responsibility of all citizens to express and
implement a collective ethics and politics, that is, the responsibility of a self-
instituting society expressed in the source text.
What is worth some further attention here is the notion of the "civilised
man", introduced in the translation as the subject of social virtue. While it is
beyond the scope of this essay to examine the full range of meanings which
constituted the concept of "civilisation" in modern civil societies, it is
important to note at least one aspect of these meanings, which becomes evident
in Mill's approach to the Platonic dialogue: the identification of civil virtue
with the pursuit of personal interests and the wilful legitimisation of egotistic
utilitarianism and hedonism. Mill's conception of the "principle of utility" -
initially articulated by his father, James Mill, and Jeremy Bentham - as the
quest for individual pleasure and fulfilment of personal desires, is expressed in
the translation of the following Socratic statement:
... pleasure is the same thing with good, and pain with evil: and if a pleasure is
bad, then it is because it prevents a greater pleasure, or causes a pain which
exceeds the pleasure: if a pain is good, it is because it prevents a greater pain,
or leads to a greater pleasure (1978:58).
The translation of the passage stays fairly close to the source text. Yet the idea
expressed by Socrates at this point is part of a broader argument developed in
the dialogue, which ultimately seeks to refute such a relativistic conception of
virtue and maintain that political qualities are intrinsically "good", as
expressions of an a priori political truth, rather than judged as good by
temporal standards of pleasure. It is only by omitting a significant part of this
argument, as mentioned above, that Mill is able to declare, when commenting
on Socrates's thought at the end of his translation, that "the principle of utility,
- the doctrine that all things are good or evil, by virtue solely of the pleasure or
the pain which they produce, - is as broadly stated, and as emphatically
maintained against Protagoras by Socrates, as it ever was by Epicurus or
Bentham" (Mill 1978:61). The doctrine of utilitarianism, which Mill reads in
the Platonic text, illustrates the value of individuality and selfhood, and
advocates the indisputable democratic right of freedom to fulfil one's personal
goals and desires. Yet the conception of people as individuals who are devoted
to the maximisation of their own satisfaction involves an implicit deification of
this satisfaction, a fatalistic submission to self-centred impulses and
inclinations which are presented as the standard for the judgement of one's life
and conduct. One cannot help noting that this standard of civil societies, which
The shaping of modern democracy 173
the state compels them [the children] to learn the laws ... [and] the city outlines
the laws ... and compels us to exercise and obey authority in accordance with
those ... and he who transgresses [the laws] is to be corrected, or, in other
words called into account, which is a term used ... seeing that justice calls men
to account (Jowett, 1875:151, my italics).
4. Conclusion
References
ARNOLD, Matthew. 1960. "On Translating Homer". On The Classical Tradition ed. by R.H.
SUPER. Ann Arbor: The University of Michigan Press.
BASSNETT-MC GUIRE, Susan. [1980] 1991. Translation Studies, revised edition, London &
New York: Routledge.
BURNS, J. H., 1969. "J. S. Mill and Democracy 1829-1861". Mill A Collection of critical
Essays, 280-328, ed. by J. B. SCHNEEWIND. London: Macmillan. First published in Political
Studies 5,1957.
ELYOT, Thomas. [1531] 1883. The Boke Named The Governour, edited by Henry HERBERT.
& Stephen CROFT, 2 vols. London: Kegan Paul.
JOWETT, Benjamin. 1875. The Dialogues of Plato. Translated into English with Analyses and
Introductions. Vol. 1. Humphrey Milford: Oxford University Press.
JOWETT, Benjamin. 1892. The Dialogues of Plato. Translated into English with Analyses and
Introductions, 5 vols. Humphrey Milford. Oxford University Press.
LEFEVERE, André, 1992. Translation, Rewriting and the Manipulation of Literary Fame,
Routledge: London and New York.
MILL, John Stuart. [1834] 1978. Notes on Some of the More Popular Dialogues of Plato: The
Protagoras. Essays on the Philosophy and the Classics (Collected Works of John Stuart Mill.
Vol. xi) ed. by J. M. ROBSON. University of Toronto Press
QUENNELL, Peter. (ed.) 1950. Byron. A Self Portrait. Letters and Diaries 1789 to 1824.
London: John Murray.
VENUTI, Lawrence. 1995. The Translator's Invisibility. A History of Translation. London &
New York: Routledge.
WILLIAMS, Raymond. [1976] 1983. Keywords: a Vocabulary of Culture and Society. Revised
edition. London: Fontana Press.
TRANSLATING LINGUISTIC MARKERS OF
IDEOLOGY
Tiina Puurtinen
University ofJoensuu, Savonlinna School of Translation Studies,
Finland
Zusammenfassung
Dieser Artikel ist ein Versuch, Ideen und Ergebnisse der kritischen Diskurs analyse auf das
Übersetzen anzuwenden. Das Ziel ist, einige Sprachstrukturen - und zwar Passivisierung,
Nominalisierung und premodifizierte Attributkonstruktionen - zu besprechen, die Ideologie
in englischen und/oder in finnischen Texten reflektieren können, sowie auch die
ideologischen Folgen von bestimmten sprachlichen Entscheidungen des Übersetzers zu
illustrieren..
Résumé
Nous chercherons ici à appliquer à la traduction certaines idées et certains acquis de l'analyse
de discours (AD) critique. Notre objectif est double: d'une part, discuter de quelques
structures linguistiques, à savoir la passivisation, la nominalisation et les constructions
participiales attributs prémodifiées, qui peuvent refléter une idéologie dans les textes anglais
et/ou finnois; d'autre part, illustrer les conséquences idéologiques de certains choix
linguistiques opérés par le traducteur.
Resumen
Este trabajo aplica a la traducción los conceptos y resultados presentados por analistas del
discurso criticos. El objetivo es el estudio de algunas estructuras lingüisticas, en particular
la construcción de la pasiva, la nominalización y las construcciones atributivas de participio
que pueden reflejar una ideologia en los textos en inglés y en finlandés. También se propone
ilustrar las consecuencias ideológicas de determinadas decisiones lingüisticas del traductor.
178 Tiina Puurtinen
1. Introduction
Use of language always reflects the user's attitudes, beliefs and viewpoints more
or less inconspicuously. Language use also embodies values and power relations.
Ways in which such ideological meanings are encoded in linguistic expression have
for the past two decades been the subject of critical discourse analysis (CDA; see
e.g. Fairclough 1990, 1995). CDA aims at revealing how ideology affects linguistic
choices made by a text producer and how language can be used to maintain,
reinforce or challenge ideologies. The concept of ideology itself is ambiguous and
has been variously defined (for criticism of the definitions, see Solin 1995,
Thompson 1984); in this paper, following Simpson (1993:5), ideology refers to
taken-for-granted assumptions, beliefs and value systems shared collectively by
social groups. Every text has an ideology, because it must be based on some values
and ways of thinking. The embedding of ideology in a text is not necessarily
intentional; neither the writer nor the reader need to be conscious of it, especially
if the values and beliefs are those that we have internalised as members of a
particular society. Reflection of such unconscious ideologies in a text can,
however, be particularly effective, because it reinforces generally accepted notions
and attitudes (or so called "common sense", see Fairclough 1990:77-108, Fowler
1987:490) instead of challenging them.
Embedded ideological meanings are interesting for Translation Studies,
firstly because ideologies of societies and cultures are different, and secondly
because the lexico-grammatical realisation of ideology is likely to vary in different
languages. However, very few CDA-oriented studies of translation have been
carried out so far (Hatim & Mason 1997, Knowles & Malmkjær 1989). This paper
discusses some linguistic structures (passivisation, nominalisation, premodified
participial constructions) which can reflect ideology in English and/or Finnish texts,
and illustrates the ideological consequences (often unintentional) of certain
linguistic choices made in translation.
and thereby reinforce, values which are unquestioningly accepted by a society, and
by the writer as a member of that society. Of course a text producer can also
deliberately challenge such generally held beliefs by making the text implicitly (or
explicitly) reflect an opposing ideology. For example an author of a children's
detective story might portray female characters as more active than male ones, and
thus more visible linguistically. However, the reverse is likely to be more common
in children's literature. (See Knowles & Malmkjær 1996 for related examples; and
also Hollindale 1988, summarised in Knowles & Malmkjær 1996:65-68.)
Thompson (1990:59-67, summarised in Knowles & Malmkjær 1996: 46-
59), describes five modes of operation of ideology: legitimation refers to the
representation of power relations as legitimate; in dissimulation power relations
are hidden, denied or obscured; through unification diverse social groups are
brought together e.g. by using such symbols of unity as national flags and anthems,
whereas fragmentation is a way of differentiating between groups; through
reification power relations are represented as if they were natural and permanent.
Operation of ideology is served by a number of strategies, which can be realised
in text by micro-linguistic structures (involving single words or phrases) or macro-
linguistic structures (involving a clause or more). The strategies can either operate
implicitly in a text, or they can rely on explicit verbalisations. In the following, I
will concentrate on reification, and three associated strategies, passivisation,
nominalisation, and premodified (participial) constructions (the last one is my
addition to Thompson's list). The reason for focusing on these particular strategies
is that, first, they are realised by specific linguistic forms, which can be easily
identified in a text, and second, both their internal structure and their usage in
various text types is different in English and Finnish. It must be emphasised,
however, that none of the linguistic forms discussed in the following is always an
indicator of ideology in a text and should not automatically be interpreted as such.
CDA-based research of translated and non-translated texts must be supported by
a good knowledge of the values held by the surrounding culture.
3.1 Passivisation
The source text's passive forms seem to imply that in addition to the writer, there
are other people and groups (The Ecologist magazine, environmental groups and
organisations) behind the demand, although none of them is explicitly mentioned.
The Finnish translation, on the other hand, does not give the same impression but
is likely to be interpreted simply as the writer's personal opinion that consumers
and big companies have a moral obligation to act in the recommended manner. (Of
course, this may be just my way of reading between the lines, and other
interpretations are possible.) An alternative translation, which retains the
Markers of ideology 181
(2) Thanks to these new technologies, it's now possible for countries to develop
sound, broad-based economies without industrializing, without draining people from
rural areas and concentrating low-paid workers in company towns or urban ghettos,
and without degrading and exhausting their land, air, and water. (Mayur & Daviss
1998; 46-47)
The use of the passive vs. active voice and the related issue of role
assignment can also be a method of supporting characterisation in fiction. Knowles
& Malmkjær (1996:79-80) show that in The Secret Garden by F. Hodgson
Burnett, role assignment correlates with the increased activity of one of the main
characters. At the beginning of the story Mary Lennox is less active and less often
in the role of actor, i.e. things happen or are done to her rather than by her, while
towards the end she becomes more active and tends to occupy the actor role in
182 Tiina Puurtinen
3.2 Nominalisation
has been turned into the nominal form pyynnöt kontrolloimisesta ('requests for
controlling'), which obscures agency; who should control the clear-cutting and
logging practices: industrialized or developing countries?
(3) Developing countries, which control 60 percent of the world's forests, have
resisted calls from the industrialized nations to control their clear-cutting and
logging practices. (Gupte 1997)
Premodified nominal phrases purport to tell the readers what they already know;
the information in modifiers is taken as known, and thus there is no need to
foreground it in a way which invites the reader to object to it (cf. e.g. President
Clinton's outrageous behaviour angered the Americans vs. President Clinton
behaved outrageously and that angered the Americans - the first, premodified
construction implies that Clinton's behaviour is generally considered outrageous,
whereas the second version leaves this view open to discussion and disagreement).
Premodified constructions can be seen as a method of classification (see Fowler &
Kress 1979): premodified nominal phrases seem to be lexical items, unitary
elements, which cannot be split up. Finnish premodified participial attribute
constructions are similar to nominalisations in that they can turn a process into a
mere modifier by embedding the verb as a participial attribute. In the following
Finnish translation of a sentence in Rudyard Kipling's The Jungle Book (1987) the
verb nukkua ('to sleep') appears as a premodifier to those who sleep, and seems
to classify gipsies and woodcutters as a group of people who always tend to or
have to sleep outside. The premodified construction is like a lexical unit, a natural,
fixed entity. The source text's postmodification creates no such impression but
seems to be just an explanatory addition referring to incidental behaviour. (See also
Viertiö 1998 on various functions and interpretations of participial attribute
constructions in sports news in the Finnish press.)
(4) It was the noise that bewilders woodcutters and gipsies sleeping in the open, and
makes them run sometimes into the very mouth of the tiger. (Kipling 1987:11)
184 Tiina Puurtinen
4. Conclusion
Sources of examples
KIPLING, Rudyard. [1894] 1987. The Jungle Book. New York: Penguin. Transl. into Finnish by Eila
Pennanen & Juhani Jaskari 1965. Viidakkokirjat. Porvoo: WSOY.
KLEINER, Kurt. 1995. "Language Deaths 'Bad for Us A l l " . New Scientist 4 March, 15.
MANSER-FONDS, Bruno. 1998. "Why Are the Forests Burning?". The Ecologist 28(1), 8.
MAYUR, Rashmi & Bennett DAVISS. 1998. "The Technology of Hope". The Futurist, October,
46-51.
References
EGGINS, Suzanne, Peter WIGNELL & J.R. MARTIN. 1993. "The Discourse of History: Distancing
the Recoverable Past". Register Analysis. Theory and Practice ed. by Mohsen GHADESSY, 75-
109. London & New York: Pinter.
FAIRCLOUGH, Norman. [1989] 1990. Language and Power. London & New York: Routledge.
FAIRCLOUGH, Norman. 1995. Critical Discourse Analysis: The Critical Study of Language.
London & New York: Longman.
FOWLER, Roger. 1987. "Notes on Critical Linguistics". Essays in Honour of Michael Halliday ed.
by T. THREADGOLD & R. STEELE, 481-492. Amsterdam & Philadelphia: John Benjamins.
FOWLER, Roger. 1991. Language in the News. Discourse and Ldeology in the Press. London &
New York: Routledge.
FOWLER, Roger & Gunther KRESS. 1979. "Critical Linguistics". Language and Control ed. by
Roger FOWLER, Bob HODGE, Gunther KRESS & Tony TREW, 185-213. London: Routledge &
Kegan Paul.
186 Tiina Puurtinen
HATIM, Basil & Ian MASON. 1997. The Translator as Communicator. London & New York:
Routledge.
HOLLINDALE, P. 1988. "Ideology and the Children's Book". Signal 55.3-22. Reprinted in 1992,
Literature for Children: Contemporary Criticism ed. by P. HUNT, 19-40. London & New York:
Routledge.
KNOWLES, Murray & Kirsten MALMKJÆR. 1989. "Translating Ideology: Language, Power and
the World of the Tin Soldier". ELR Journal 3: Language and Ideology. 205-241. Birmingham:
University of Birmingham.
KNOWLES, Murray & Kirsten MALMKJÆR. 1996. Language and Control in Children's
Literature. London & New York: Routledge.
NG, Sik Hung & James J. BRANAC. 1993. Power in Language. Verbal Communication and
Social Influence. Newbury Park, London & New Delhi: Sage Publications.
RAVELLI, L.J. 1988. "Grammatical Metaphor: An Initial Analysis". Pragmatics, Discourse and
Text. Some Systematically-Inspired Approaches ed. by Erich H. STEINER & Robert VELTMAN.
London: Pinter, 133-147.
SIMPSON, Paul. 1993. Language, Ideology and Point of View. London & New York: Longman.
SOLIN, Anna. 1995. "Ideology, Power, Interpretation - An Assessment of the Social Theory of
Critical Linguistics". The New Courant 4 ed. by Eija VENTOLA & Anna SOLIN, 162-180. Helsinki:
University of Helsinki.
THOMPSON, J.B. 1984. Studies in the Theory of Ideology. Cambridge: Polity Press.
THOMPSON, J.B. 1990. Ideology and Modern Culture: Critical and Social Theory in the Era
of Mass Communication. Cambridge: Polity Press.
TREW, Tony. 1979. "Theory and Ideology at Work". Language and Control ed. by R. FOWLER
et al. 94-116. London: Routledge & Kegan Paul
Kristiina Taivalkoski
Université de Helsinki, Finland
Abstract
The article deals with shifts in reported discourse, on the basis of an analysis of the
French translation (1743) of Fielding's Joseph Andrews. Types of reported discourse
are classified according to the model proposed by McHale (1978). The analysis shows
that the translator transformed the rarer types into more common types, a result that
corroborates Toury's law of growing standardisation (1995). The author suggests that
the reasons for these shifts are to be found in the norms of the dominant literary
system in the target culture. The shifts have repercussions that are not only stylistic
but also thematic.
Zusammenfassung
Der Artikel befaßt sich mit Verschiebungen (shifts) in erlebter Rede, anhand einer
Analyse der französischen Übersetzung (1743) von Fieldings Werk Joseph Andrews.
Die Klassifizierung von Typen erlebter Rede erfolgt nach dem Modell von McHale
(1978). Die Analyse zeigt, daß der Übersetzer die selteneren Typen in generellere
Typen transferiert hat. Dieses Ergebnis bestätigt Tourys Gesetz der wachsenden
Standardisierung (1995). Der Verfasserin zufolge liegt der Grund für diese
Verschiebungen in den Normen des dominierenden literarischen Systems in der
Zielkultur. Die Verschiebungen weisen Nachwirkungen auf, die nicht nur stilistisch
sondern auch thematisch sind.
Resumen
El articulo aborda las desviaciones en el estilo indirecto basándose en el análisis de
una traducción al francés (1743) de la obra de Fielding, Joseph Andrews. Se clasifican
los tipos de estilo indirecto según el modelo propuesto por McHale (1978). El anâlisis
muestra que el traductor transformé los tipos menos frecuentes a tipos mâs corrientes,
un resultado que confirma la ley de estandarización creciente de Toury (1995). El
autor propone que estas desviaciones se deben a las normas impuestas por el sistema
literario dominante en la cultura término. Las desviaciones presentan repercusiones
tanto estilisticas como temáticas.
188 Kristiina Taivalkoski
1. Introduction
Par textème, Toury entend le rôle particulier que joue une unité sémiotique
Simplifications narratives 189
Je fais une distinction entre les types non mixtes et les types mixtes du discours
rapporté. Les types non mixtes ne contiennent que du discours du narrateur
(résumé diégétique, compte rendu diégétique, discours indirect) ou du
personnage (discours direct, discours direct libre). Les types mixtes, par
contre, contiennent aussi bien du discours du narrateur que du personnage. Tels
sont le discours indirect mimétique et le discours indirect libre. Le tableau (1)
récapitule mon classement:
J'ai classé les types mixtes du discours rapporté comme textèmes et les types
non mixtes comme répertorèmes, parce que les formes mixtes ont été rares
dans la littérature occidentale du XVIIIe siècle. Selon Ann Banfield (1982:225-
235), elles ont certes fait des apparitions sporadiques dans l'histoire de la
littérature. Cependant, les premiers auteurs pour lesquels l'usage de ces formes
est généralement reconnu sont La Fontaine en France, Henry Fielding et Jane
Austen en Angleterre, et Goethe en Allemagne. Le premier auteur qui en
développa un art systématique ne fut pourtant que Flaubert au XIXe siècle.
Comme j'ai testé la loi de la standardisation croissante dans mon étude,
il m'a fallu définir ce que je considère comme standardisation: dans ce
contexte, elle est le glissement d'un type mixte du discours rapporté vers un
type non mixte. Le tableau (2) illustre tous les glissements possibles - en
théorie - entre les types mixtes et non mixtes du discours rapporté et leur
valeur vis-à-vis de la loi de la standardisation croissante. Les glissements entre
les types non mixtes sont neutres, les glissements de types mixtes vers les types
non mixtes sont des standardisations (en caractères gras) et les glissements de
types non mixtes vers les types mixtes et les glissements entre les types mixtes
sont des anti-standardisations (en italiques):
TA: qu'il leur parla ainsi: "Les Voleurs de grand chemin se sont
tant multipliés depuis quelque tems, que les gens ne peuvent
dormir en sureté dans leur lit.
TD: then he began to revile them saying, 'that Robberies on the
Highway were now grown so frequent, that People could not
sleep safely in their Beds, (--).' (AV, II, xi, 257; JA, II, xi, 173,
c'est moi qui souligne)
Je peux conclure que la narration, exprimée par le discours rapporté, est moins
polyvalente dans le texte d'arrivée que dans le texte de départ de mon corpus,
ce qui corrobore l'hypothèse de la standardisation croissante de Toury. Les
raisons de cette simplification narrative ne résident pas dans l'incompétence du
traducteur mais dans son opinion sur ce qui est acceptable dans la culture
d'arrivée, ce que le traducteur dit de façon explicite dans son Dictionaire
Néologique. Il est à noter que les glissements narratifs dans le discours rapporté
sont toujours possibles dans les traductions littéraires actuelles. Ils peuvent
passer inaperçus par le traducteur dans des cas de glissement entre des types du
discours rapporté apparentés comme le discours indirect et le discours indirect
mimétique. La disparition du plurilinguisme - de sociolectes, d'idiolectes, de
jargons professionnels - a des répercussions également sur la narration d'un
texte.
Les répercussions de la standardisation du discours rapporté dans la
traduction ne sont pas insignifiantes. C'est que les différents types du discours
rapporté jouent souvent un rôle particulier dans l'ensemble d'un ouvrage
littéraire. Ce rôle n'est pas uniquement stylistique. Les types du discours
Simplifications narratives 197
Notes
1. Le texte d'arrivée de mon corpus comprend 2513 mots, comptés par l'ordinateur, tandis
que le texte de départ en contient 3342. Cela s'explique par le fait qu'il y a plusieurs
omissions dont la plus grande ne comprend pas moins de 532 mots. En outre, le texte
d'arrivée contient une grande addition de 201 mots.
4. Par l'indication «non mixte 2», je veux dire autre type non mixte du discours rapporté.
5. Par l'indication «mixte 2», je veux dire autre type mixte du discours rapporté.
7. Cette réplique est suivie par l'addition suivante qui rend le style du texte d'arrivée encore
plus pathétique: «Ce seroit une injustice criante, contraire à l'administration de la Police
en Angleterre.» {AV, II, xi, 258)
8. Voir aussi Lefevere (1992: 87-92). Le même motif - les exigences de la poétique française
du XVIIIe siècle - expliquerait les changements que fit Houdart de La Motte dans sa
traduction de l'Iliade d'Homère en 1714. Selon Lefevere, l'épopée n'avait pas bonne
presse dans la France du classicisme. Par conséquent, La Motte jugea l'Iliade selon les
critères de la tragédie, genre prédominant à l'époque. Comme on le sait, sa traduction
déclencha la seconde phase de la Querelle des Anciens et des Modernes en France.
9. Selon Brian McHale (1978: 256-257) reconstruire «l'énoncé originel» est une illusion dans
la littérature parce qu'il n'y a pas d'énoncé réel derrière l'énoncé du texte. Le seul énoncé
qui existe est dans le texte.
10. Le pathos de cette réplique est augmenté par l'addition suivante dans le texte d'arrivée:
«Son poids m'accable; en le révélant, je trouverai peut-être quelque secours.» {AV, IV, xiv,
289)
198 Kristiina Taivalkoski
Références
FIELDING, H. 1743. The History of the Adventures of Joseph Andrews, and of his Friend Mr.
Abraham Adams. Written in Imitation of the Manner of Cervantes, Author of Don Quixote.
London: A. Millar.
LEFEVERE, A. 1992. Translation, Rewriting and the Manipulation of'Literary Fame. London
& New York: Routledge.
MAY, R. 1994. The Translator in the Text. On Reading Russian Literature in English.
Evanston: Northwestern University Press.
McHALE, B. 1978. "Free Indirect Discourse: a survey of recent accounts." Poetics and Theory
of Literature 3, 249-287.
TOURY, G. 1995. Descriptive Translation Studies- and beyond. Amsterdam & Philadelphia:
John Benjamins.
Jeremy Munday
University of Bradford, U.K.
Zusammenfassung
In diesem Artikel wird ein praktischer Ansatz beschrieben hinsichtlich der
Verwendung umfangreicher Computer-Korpora sowohl in zweisprachiger
Lexikographie als auch im Bereich des Übersetzens. Folgendes wird dargelegt:
(1) Methoden der Ermittlung von Übersetzungsäquivalenten für Einträge in
zweisprachigen Wörterbüchern, Spanisch-Englisch, wo der Lexikograph sich
zwischen Quasi-Synonymen zu entscheiden hatte;
(2) mögliche Anwendungen für professionelle Übersetzer. Aus bereits veröffentlichten
Literatur-Übersetzungen werden Beispiele herangezogen und die jeweiligen
Entscheidungen der Übersetzer anhand großer "KontrolP'-Korpora überprüft.
Der Artikel diskutiert auch mögliche negative Implikationen der
Korpuslinguistik, legt aber auch nahe, daß die immer schneller zunehmende
Verfügbarkeit von Computer-Korpora für Forschungszwecke spannende
Möglichkeiten sowohl für Übersetzer als auch für die Wissenschaftskritik wird
bereitstellen können.
Résumé
Cet article décrit une approche pratique pour l'usage des grands corpus informatisés
en lexicographie bilingue et en traduction. Il présente:
(1) des méthodes afin de déterminer les équivalences pour les entrées de dictionnaires
bilingues espagnol-anglais, où le lexicographe a dû choisir entre plusieurs quasi-
synonymes;
(2) des applications possibles pour le traducteur professionnel. Des exemples sont tirés
de traductions littéraires publiées, pour comparer les choix des traducteurs par rapport
à de larges corpus de contrôle.
L'article traite des répercussions négatives possibles de la linguistique de
corpus, mais il suggère aussi que la disponibilité de plus en plus grande de corpus
informatisés pour la recherche fournira d'énormes possibilités non seulement au
traducteur professionnel mais aussi au chercheur critique.
202 Jeremy Munday
Resumen
Este articulo describe un enfoque prâctico para el uso de los grandes corpus
informatizados tanto en la lexicografia bilingüe como en la traducción. En él se
plantean:
(1) métodos para determinar equivalencias para palabras cabeza de articulo en
diccionarios bilingües español-inglés, cuando el lexicógrafo ha tenido que elegir entre
cuasi-sinónimos;
(2) posibles aplicaciones para el traductor profesional. Se sacan ejemplos de
traducciones literarias ya publicadas, comprobando las selecciones hechas por los
traductores y comparândolos con ejemplos extraidos de grandes corpus "control".
Aunque el articulo baraja las posibles repercusiones negativas de la lingüistica
del corpus, acaba concluyendo que la disponibilidad cada vez más grande de los
corpus informatizados para la investigacion proporcionarâ posibilidades de enorme
interés no solo para el traductor profesional sino también para el critico académico.
1. Introduction
The first large computerized corpus of texts designed for specific large-scale
lexicographic use was put together for the COBUILD English Dictionary
project at Birmingham, U.K. This has now developed into the Bank of English
(BoE). Other British publishers have now followed this line, notably with the
British National Corpus (BNC). The BNC comprises a variety of texts
(newspapers, serious and light fiction, non-fiction, and some spoken
transcriptions) totalling, in January 2000, around 110 million words.
In-house, the large dictionary publishing companies have bigger SUN
A corpus-base approach 203
systems for the analysis of their corpora. These corpora are now increasingly
available for researchers: the British National Corpus, requiring 4-6 Gb of hard
disk space, is currently available in CD-Rom form to research institutions for
around $400 or via a single-user Internet subscription for around $110 per year.
Smaller-scale commercially-produced research tools for the linguistic analysis
of researchers' own corpora are also now more widely available. Most of these
programs can be easily used by non-IT specialists. A very good example is
Oxford University Press's user-friendly and very effective WordSmith (Scott
1996), which contains a range of useful tools for the analysis of texts and can
be used on most PCs and with little training. The researchers' corpora can be
their own academic texts or books, newspaper collections purchased on CD-
Rom, or literary texts for which the researcher has sought copyright to hold on
disk, as was the case with my own analysis of Garcia Márquez stories in
translation (see 5.2 below).
The great strength of computers is their accuracy and speed in counting and
ordering stored information; and the easiest linguistic phenomena for the
computer to count and order are individual words. The most straightforward
definition of words, which is the one that computers find easiest to assimilate
(see Baker 1993) and which will be used here, is that of an orthographic word-
form, that is, a group of typographical characters with a space on either side
(e.g. an, orthographic and word-form are three examples of words). WordSmith
is able to count the word-forms, discounting hyphens as required, to provide
basic statistics for a text (text length in words, sentences, paragraphs, etc.), to
give a frequency list of the word-forms either in alphabetical order or in
descending order of frequency, and to provide concordances, basic collocation
statistics and type-token ratio (the number of distinct word-forms as a
percentage of the total word-forms in a text). Some of these measurements are
potentially extremely useful in computational stylistics to describe a writer's
"fingerprint", and in descriptive translation studies, where the texture and
structure of STs and TTs can be compared (see, for example, Holmes 1994 on
authorship attribution and Munday 1997, 1998 on translation). The present
paper, however, focusing on seeking small-phrase translation equivalents, will
limit itself to two specific and related areas, those of concordances and
collocations.
4.1 Concordances
was an energetic literary hostess but also a fond aunt to her nephews and nieces.
e family fortunes. They had all bade him a fond farewell, then settled down
had seen the look on Jimmy's face. These fond farewells were too damn facile,
hing for the other person. I've told you how fond I am of Lena. Have you so little
"the food, however, remains a glowingly fond memory for both. Back at
ves of French rosarians or the daughters of fond nurserymen. Sometimes,
Atlantic and married a British barrister fond of amateur theatrics, Cecil
are the right person to know. You are very fond of Anna, rightly so. She is also
rica's out-of-touch marketing man. He was fond of bragging that his relations
Then perhaps he should, Angela. I'm very fond of Gerald." "But then everyone
ou, if I may. I do have a pet dog. I am very fond of her, and it would grieve me
e another to put matters right." "I got very fond of him. And since he was uniq
a had called him Uncle Mark and had been fond of him, unlike most of her other
ham may be our Judas, but he's too fly, too fond of himself, too ambitious to
und a honey-pot. Don't get me wrong. I'm fond of my wife and my boys are
kept trying for a French twist, Michael was fond of recalling, but it kept Connie
but that was a long time ago as he was fond of reminding her. And what can
a good dose of asphalt, starting, as he was fond of saying, with his own back
All his, and no one else's. Pearle wasn't fond of the rabbits, which was just as
our German cousins and was particularly fond of them, though the most loyal
right behind the house. Mrs Roosevelt was fond of this room, with its lovely
Almost any search term can be called up. Thus, a lexicographer may search for
the particular term which forms the entry word being compiled, while
translation analysts may use word frequency lists or their own initial reading of
the text to decide which search terms to examine. It is important to realize,
however, that the intellectual work and interpretation of the concordances is
provided by the user, the computer simply being a very useful tool in
facilitating the analysis.
together (in the case of fond, as will be discussed in the analysis below, it
would be of and words such as farewell and memory). It shows the
paradigmatic axis of language. The most obvious adjacent collocates may be
visible from a concordance, but WordSmith also has a basic collocation
program, which calculates and presents in tabular form the most frequent
collocates of any search term. Much more sophisticated analytical programs
are employed by in-house lexicographers. Yet, if our goal is to use the corpus
as a kind of interactive thesaurus to seek out typical or appropriate translation
equivalents, then either of the packages is useful, since their main function in
our case would be to suggest ideas to the translator which might otherwise
have been overlooked. Thus, statistical results for collocates of rain (verb and
noun) point out the most common uses to be acid rain, heavy rain, wind...rain,
pouring rain, rain forest, rain... down, torrential rain, driving rain, rain...
snow, rain... falling, through... rain, began... rain, cold... rain. Such a list can
be useful for a translator seeking the translation of a collocate within the co-
text of rain (see the "oblique, wind-driven rain" example in 5.2 below).
only one is possible, then the lexicographer will mostly choose the one that
covers the widest lexical field, or, as Manley suggested, "is insertable in a wide
range of contexts". In our case, this may end up being attached to since it does
fit all three example sentences and is the most common of the three words,
even if it may not be the first-choice translation in all cases.
The description of previous work in this field has not only isolated different
A corpus-base approach 209
Note
1. My thanks to Oxford University Press ELT Dictionaries and COBUILD for allowing me
to access their electronic corpora and to use examples from them. Thanks also to Jonathan
Cape (Random House) for allowing me to hold the Garcia Marquez stories on disk for the
period of my research.
References
BAKER, M. 1995. "Corpora in Translation Studies: An Overview and Suggestions for Future
Research". Target 7(2), 223-43.
BURROWS, J. F. 1987. Computation into Criticism: A study of Jane Austen 's novels and an
experiment in method. Oxford: Oxford University Press.
HOLMES, D. I. 1994. "Authorship Attribution". Computers and the Humanities 28(2): 87-106.
MUNDAY, J. 1998. "A computer-assisted approach to the analysis of translation shifts". Meta
43(4), 542-56.
SINCLAIR, J.M. 1991. Corpus, Concordance, Collocation. Oxford: Oxford University Press.
TOURY, G. 1995. Descriptive Translation Studies and beyond. Amsterdam & Philadelphia:
John Benjamins.
Illustrative texts
(DC) GARCIA MÁRQUEZ, Gabriel. 1992. Doce cuentos peregrinos. Madrid: Mondadori.
(SP) GARCIA MÁRQUEZ, Gabriel. 1994. Strange Pilgrims. (Translated by Edith Grossman).
London: Penguin UK.
MACHINE TRANSLATION FOR TRANSLATORS?
Anja Schwarzl
University of Salzburg, Austria
Zusammenfassung
In den letzten Jahren kamen vermehrt Übersetzungsprogramme auf den Markt, die auf
handelsüblichen Personalcomputern laufen. In diesem Aufsatz geht es um die
Übersetzungsqualität dieser sogenannten semi-professionellen Systeme. Kriterien
werden aufgelistet, die bei der Entscheidung, ob sich ein Text für maschinelle
Übersetzung (MÜ) eignet oder nicht, hilfreich sein können. Beispiele für MÜ werden im
Rahmen eines allgemeinsprachlichen Tests und eines Vorbearbeitungstests angeführt.
Ein weiterer Schwerpunkt ist die Rolle der MÜ in der modernen Kommunikation wie
WWW und e-mail.
Résumé
Ces dernières années, on a mis sur le marché des systèmes de traduction automatique
(TA) qui peuvent fonctionner sur des ordinateurs personnels standard. Notre
contribution s'interrogera sur les performances de ces systèmes semi-professionnels.
Nous chercherons à définir des critères pour savoir si on peut recommander la TA pour
certains textes. Nos exemples de réalisation en TA sont de trois sortes: des phrases de
langue générale, un texte technique pré-édité et des documents disponibles en ligne.
Resumen
Desde hace unos afios, existen sistemas automâticos de traducción compatibles con un
ordenador estándar. Este trabajo se centra en el funcionamiento de estos sistemas
denominados semiprofesionales. Este trabajo procurará establecer unos criterios para
determinar si se puede recomendar el uso de la traducción automâtica para ciertos
textos. Se muestran ejemplos del funcionamiento de traducción automâtica en una
prueba de lenguaje general y en otra de preedición. Un tercer tema de este trabajo se
centra en la importancia de la traducción automâtica en los nuevos medios de
comunicación tales como Internet y el correo electrónico.
212 Anja Schwarzl
1. What is MT?
2. Semiprofessional systems
3. MT for me?
4. Examples of MT output
This section reports on some small-scale tests carried out on the three MT
programs. Two kinds of test were done: (1) a test on general language using the
BNC, and (2) a kind of pre-editing experiment with a technical text.
4.1 Test 1
Sample sentences for general language were acquired from queries in the British
National Corpus (BNC), which is a 100-million word collection of samples of
written and spoken language from a wide range of sources, designed to
represent a wide cross-section of current British English, both spoken and
written (URL: http://info.ox.ac.uk/bnc/).
The query was done for English word light, because it is an everyday
word which nonetheless exhibits a lot of ambiguities. First it shows category
ambiguity: it can be a noun, a verb or an adjective. The adjective itself is
lexically ambiguous, as the opposite of both heavy and dark. The first 100
random sentences of the BNC query's result were used for translation. As three
of them showed too many characteristics of spoken English, i.e. ellipsis, fillers,
incomplete sentences, they were excluded, which left a total of 97 sentences.
It is remarkable that with the same program some categories that were
correct in one language were wrong in another, which may be regarded as
evidence of the use of a transfer and not an interlingua system. In all examples
here an asterisk (*) in front of a translation marks an incorrect translation.
In other cases, again, some versions were correct and others were not:
Table 1 shows the outcome of this test, regarding just the category of the word
light, regardless of the quality of the rest of the output. The first column lists
the programs and the respective translation directions; the second and third
column give the absolute number of sentences, where light was translated
correctly or wrongly respectively; the last column gives the percentage of
translation mistakes.
4.2 Test 2
This test was basically a pre-editing experiment. Its aim was to find out how
sentence length effects the quality of the translation outcome. I took a
paragraph from a technical text in German (Eppinger & Herter 1993:84) and
had it rated by the Word97 ™ spelling and grammar check. Then I rewrote the
paragraph with the main aim of shortening the sentences while keeping the same
level of vocabulary and information. Below are the original and the pre-edited
text, both of them with the readability statistics from Word97 ™:
Text 2a (ORIGINAL):
1 Es existieren viele Verfahren, die diese Datenreduktion (von analog nach digital)
auf verschiedene Weise zu erreichen versuchen: Verfahren der Signalformcodierung
quantisieren und komprimieren ausschließlich das Sprachsignal, ohne auf die
charakteristischen Merkmale der Spracherzeugung, wie die unterschiedliche
Machine translation 217
Generierung von Konsonanten und vokalen oder den Einfluß des Vokaltrakts,
einzugehen.
2 Verfahren, die diese Merkmale der Spracherzeugung mit einbeziehen, werden
Quellencodierungsverfahren genannt, und es lassen sich mit ihnen wesentlich niedrigere
Bitraten erzeugen, da das Codierverfahren dem Wesen des Sprachsignals angepaßt ist
und dieses somit besser erfaßt wird.
3 Eine Mischung beider Verfahren, die Verfahren der hybriden Codierung, versuchen
die Vorteile beider Verfahren miteinander zu verknüpfen.
Text 2b (PRE-EDITED):
1 Eine Datenreduktion ist erforderlich, wenn ein analoges Signal in ein digitales
Signal umgewandelt wird. Viele Verfahren versuchen, diese Datenreduktion zu
erreichen. Sie tun dies auf unterschiedliche Weise. Die Signalformcodierung codiert nur
das Sprachsignal. Sie geht nicht auf die typischen Merkmale des Signals ein. Diese
Merkmale sind z.B. die Generierung von Konsonanten und Vokalen. Ein weiteres
Merkmal ist z.B. der Einfluß des Vokaltrakts.
2 Es existieren auch Verfahren, die diese Merkmale mit einbeziehen. Diese werden
Quellencodierungsverfahren genannt. Damit können niedrigere Bitraten erzeugt werden.
Dieses Codierverfahren ist dem Wesen des Sprachsignals angepaßt. Das Sprachsignal
wird daher besser erfaßt.
3 Es gibt auch eine Mischung der beiden Verfahren. Dies nennt man hybride
Codierung. Sie verknüpft die Vorteile beider Verfahren miteinander.
The Flesch Reading Ease Score has a scale ranging from 0 (very complex) to
100 (very easy). It can be seen that the Flesch Reading Score and the Syntax
complexity changed considerably in the second text, whereas the complexity of
vocabulary remains almost the same. The first aim was to shorten the sentence
length while keeping the vocabulary standard. The second aim was to reduce
the frequency of deictic expressions (e.g. pronouns). The three underlined
words in each text were not found in the Word97™ dictionary. To illustrate the
result it will be sufficient to present just the respective translations of two
programs (Langenscheidt, Pons):
218 Anja Schwarzl
Langenscheidt: Text 2a
1 Many procedures exist, that this data reduction (of analogous according to digital)
in different way, attempt to achieve: Procedures of the waveform codification quantize
and pack exclusively the voice signal, without generation of consonants different to the
characteristic features of the voice generation as those and vokalen or the influence of
the vowel tract to come in.
2 Dealt, the these features of the voice generation with may include, are source
coding techniques mentioned, and it can be made itself bit rates lower with them
considerably since the enciphering procedure is adapted to the being of the voice signal
and consequently this is registered better.
3 A compound of both procedures, the procedures of hybrid codification, do the
advantages of both procedures attempt with each other to associate.
Pons: Text 2a
1 It many methods, the this data reduction ( exist of analogous after digital ) try to
reach in different way: Or the influence of the vowel section spends the signal form
coding of Quantisieren in travelling and exclusively compresses the speech signal,
without on the characteristic features of the speech generation, as the various generation
of consonants and vocal, come in.
2 methods which include these features of the speech generation are called methods of
source encoding and have fundamentally lower bitrates it produced with them for
themselves since the method of encoding is adapted to the nature of the speech signal
and this is included therefore better.
3 Methods, the methods of the hybrid coding, more twoly try to connect the
advantages of both methods together for a mixture.
the signal. E.g. these features are the generation of consonants and vowels. E.g. a
further feature is the influence of the vowel section.
2 Methods which include these features also exist. These are called methods of source
encoding. Lower bitrates can be produced with that. This method of encoding is adapted
to the nature of the speech signal. The speech signal is included therefore better.
3 This one also gives both methods a mixture. One calls this hybrid coding. She
connects the advantages of both methods together.
It is very obvious that in both cases the translation of the second text (2b) is
much better, though still far from perfect.
of each document is stored, but delivery takes considerably more time because
of MT. This would be the only option for e-mail communication.
In both cases, however, texts and documents have to be translated in a
way which is 'as effective as possible'. The most obvious difficulties (apart from
MT problems) are deciding which languages should be made available;
differences of character set (even among European languages, let alone others);
representation conventions (e.g. date, time, certain symbols and images); and
the necessity of multilingual aligned hypertext. At least today, the solution via
MT is certainly out of the question, or only a temporary alternative. As 'written
proof you might like to check the much-discussed Starr Report on the
AltaVista Homepage (http://starr-report.altavista.com). A warning notifies the
user that the translations of the document have been done without human
intervention and that therefore the quality of these translations may not fully
capture the precision of the original legal text. Additionally, MT would only be
worth it - and thus only be provided - for large language groups, in Europe e.g.
for English, French, Spanish and perhaps German. But CMC will probably be an
(economic) pusher for MT research and development.
References
EPPINGER, B. & HERTER, E. 1993. Sprachverarbeitung, p. 84. Wien: Carl Hanser Verlag.
HUTCHINS, W.J. & SOMERS H.L. 1992. An Introduction to Machine Translation. London:
Academic Press.
LAUER, A. et al. (eds) 1996. Übersetzungswissenschaft im Umbruch: Festschrift für
Wolfgang Wilss zum 70. Geburtstag. Tübingen: Gunter Narr.
MELBY, A. 1992. "The Translator Workstation." Computers in Translation. A practical
appraisal ed. by John NEWTON, 147-165. London: Routledge.
MELBY, A. & WARNER, T. 1995. The Possibility of Language. A Discussion of the Nature
of Language, with implications for Human and Machine Translation. Amsterdam &
Philadelphia: John Benjamins.
MELBY, A. 1997. "Should I use Machine Translation?" Web-Document; URL:
http: //www.ttt.org/theory/mt4me. html)
O'HAGAN, M. 1996. The Coming Industry of Telecommunication. Topics in Translation 4.
Clevedon: Multilingual Matters.
SCHMITT, P.A. 1996. "Computereinsatz in der Translation". LAUER, A. et al., 187-196.
ASPECTS OF UNIVERSAL GRAMMAR IN HUMAN
VERSUS MACHINE TRANSLATION
Martin Kaltenbacher
University of Salzburg, Austria
Zusammenfassung
Viele Unterschiede zwischen Sprachen lassen sich auf eine konträre Auswahl von
binären universalgrammatikalischen Optionen zurückfuhren. Die strukturellen
Unterschiede zwischen Deutsch und Englisch, die sich durch drei unterschiedliche
Parameter ergeben, werden beschrieben, und vier Gruppen von Englisch als
Zweitsprache-Lernern sowie zwei Übersetzungprogramme werden auf korrekte
Übersetzung hin getestet. Universalgrammatikalisches Wissen ist nicht in
Softwarealgorithmen integriert, was in maschineller Übersetzung zu einer Reihe von
Fehlern führt.
Résumé
Nombre de différences entre les langues sont dues aux sélections différenciées dans les
options binaires disponibles en grammaire universelle. On décrit ici des différences
structurales entre l'allemand et l'anglais, selon trois paramètres distincts. Les
traductions de quatre groupes d'apprenants en anglais seront comparées à celles
produites par deux logiciels de TA. La non-intégration du savoir de grammaire
universelle dans les calculs informatisés mènent à divers types d'erreur en TA.
Resumen
Muchas de las diferencias entre las distintas lenguas se basan en contrastar selecciones
entre las opciones binarias distintas que se encuentran en la gramática universal. En
este articulo se describen las diferencias estructurales entre el alemán y el inglés que
tienen su origen en tres parâmetros distintos. También se han realizado y evaluado
pruebas de traducción correcta con cuatro grupos de estudiantes de inglés como
segunda lengua y con dos programas de traducción automática. Se producen errores de
distinta indole en la traducción automâtica ya que los conocimientos de gramâtica
universal no se integran en los algoritmos del software.
222 Martin Kaltenbacher
(17) Ich kann nicht zu deiner Party kommen, weil ich mir den Fuß gebrochen
habe.
I cannot come to your party because I have broken my foot.
GL: I cannot come to your party because I have broken myself the foot.
LS: I can not come to your party because I burst the foot for myself.
(18) Weißt du, wann er morgen kommen wird?
Do you know when he will come tomorrow?
GL: Do you whiten, when he/it will come tomorrow?
LS: Do you know when he will come tomorrow?
(19) Er sagt der Polizei, was er über den Mord weiß.
He tells the police what he knows about the murder.
GL: He/it tells the police, what he/it knows about the murder.
LS: He tells to the police what he knows via the murder.
(20) Es tut mir leid, daß du einen Unfall hattest.
I am sorry that you had an accident.
GL: I am sorry that you had an accident.
LS: It does me suffer that you had an accident.
Universal grammar 227
The learner groups as well as the computer programs make the necessary
adjustments to the verb position. Among the total number of learner translations
there is only one case of TL-violation, in which the learner produced *I am
sorry that you an accident had for (20). Surprisingly, the number of missing
translations is quite high even in the more advanced test-groups. A possible
explanation for this avoidance strategy could be the greater length and higher
structural complexity of hypotactic sentences. At first sight translation software
again seems to be able to reset the parameter value appropriately. But the two
programs get into trouble when translating sentences that involve NP-movement
such as wh-extraction, as in (21):
4. Pro-drop parameter
The pro-drop parameter setting for German and English is identical: [- pro-
drop]. Among other phenomena (Chomsky 1981: 140), this implies that
sentences in both languages require an overt subject either in the form of a
referential subject or in the form of an expletive pronoun {Es regnet 'It is
raining'). (Haider 1993; Quirk et al. 1985.)
228 Martin Kaltenbacher
In test-sentence (22) the human translators, even those from the more advanced
test-groups, quite often fail to insert the necessary expletive there. Instead, they
produce Last year were many accidents, which simply reflects the German
sentence structure. This violates the English subject before verb requirement,
which in this case demands the expletive there. The computer programs apply
an SV word order in (22), but their output is not grammatical either. Sentence
(23) exhibits a similar problem area for L2-learners, although the human
translators performed better here than in (22), probably because school curricula
provide for particular training on the German phrase es gibt and its English
counterpart there is. For MT the phrase es gibt seems to be particularly difficult
to translate. The two programs produce contradictory results. Neither of them
manages to produce 100%-correct translations of the phrase in varying
contexts. Test-sentence (23) is translated correctly by GL, whereas LS produces
a word-by-word translation of the German sentence structure. In other contexts,
however, the two programs give different results. More subtle tests have shown
that MT solutions can be affected by the presence of morphologically
ambiguous forms, as in test-sentence (24), where GL parses der Straße as a
dative:
Tests with other sentences of this kind, using acronyms instead of nouns, also
revealed inconsistencies in the programs.
Universal grammar 229
5. Conclusion
The mastery of universal rules and the options they impose on different
languages is one of the keys to correct translation. One of the reasons for MT's
limited efficiency in coping with universal grammar, or with syntactic structure
in general, can be identified in the limited significance that was assigned to the
importance of abstract universal principles in the early days of MT. Schwarzl
(1998: 116) emphasizes that "for a long time linguistics has been left out in the
general MT research ... MT researchers consider[ed] themselves engineers and
the often abstract linguistic theories were only of limited practical use". Human
translators are innately endowed with knowledge about universal principles, and
gradually reset their L1-parameter setting to the appropriate L2-setting along
the route of foreign language acquisition. In MT-software such knowledge is
only partially integrated in syntactic analysis and generation modules. One of the
main deficiencies of MT is the inability of the software to assign the proper
syntactic functions to different grammatical forms. Human translators can easily
determine the syntactic functions of phrases within a sentence owing to their
understanding of context and their knowledge of the world. Hubmayer (1998:
133) points out that human translators are subconscioulsy aware of many facts
that for them 'go without saying', whereas MT must have recourse to actual
words implanted in a lexical database. MT seems to master some of the
processes of syntactic shift required by diverging parameter settings between
two individual languages. When an error is produced its origin is not always of a
purely syntactic nature. Very often it is the ambiguity of morphological or
grammatical forms that prevents MT from correctly analysing the syntax of the
source sentence. These shortcomings will have to be reconsidered and rectified
by both linguists and software engineers in co-operation. How this can be
achieved, however has to be left on the agenda.
References
ARNOLD, D., BALKAN, L., LEE HUMPHREYS, R., MEJER, S. & SADLER, L. 1994.
Machine Translation. An Introductory Guide. Oxford: Blackwell Ltd.
HAIDER, H. 1993. Deutsche Syntax - generativ. Vorstudien zur Theorie einer projektiven
Grammatik. Tübingen: Gunter Narr Verlag.
POLLOCK, J.Y. 1989. "Verb Movement, Universal Grammar, and the Structure of IP".
Linguistic Inquiry 20. 365-424.
WHITE, L. 1989. Universal Grammar and Second Language Acquisition. Amsterdam &
Philadelphia: John Benjamins.
PART VI
Zusammenfassung
In der Forschung audiovisueller Übersetzung kommt generell der Übersetzung von
Dokumentarfilmen eine untergeordnete Rolle zu. Dieser Artikel stellt fest, daß der
Kategorie von Dokumentarfilmübersetzung eine eigene Übersetzungsspezifik
zugrundeliegt, der mehr Beachtung zu schenken ist. Den Ausgangspunkt für die
Betrachtung der Dokumentarfilmübersetzung bildet das Genre, das wichtige
übersetzungsaktbezogene Fragen aufwirft, welche sich von Fragen zur Übersetzung von
fiktiven Texten unterscheiden. Die Untersuchung der Praxis vom
Dokumentarfilmübersetzen zeigt, daß diese Übersetzungsaktivität nicht in dem Ausmaß
von objektiver Handlung geprägt ist, wie oft angenommen. Als Erklärung dafür dient,
daß das übersetzerische Verhalten primär kulturgebunden definiert wird.
Résumé
Les documentaires n'ont jamais eu une grande place dans les travaux touchant la
traduction audiovisuelle. Pourtant, selon l'auteur, ils mériteraient une attention
particulière. D'abord comme genre - ce qui soulève nombre de questions quand on en
vient à leur traduction. Ensuite comme genre appelant un type particulier de transfert,
loin d'être aussi objectif qu'on le pense mais toute traduction du fait même qu'elle
prend place dans une culture, n'exige-t-elle pas interprétation?
Resumen
La traducción de documentales suele ocupar un lugar poco privilegiado dentro de la
investigación en traducción audiovisual. En esta presentación se pretende demostrar que
esta actividad constituye una práctica especifica que merece una atención especial. En
primer lugar, se analiza el tema desde una perspectiva de género, lo cual implica varios
aspectos importantes a tener en cuenta durante el proceso traductor que difieren de los
aspectos encontrados en los productos de ficción. Después, se analiza el tema desde una
perspectiva practica, revelando asi que la traducción de documentales no es tan objetiva
234 Eliana Franco
1. Introduction
Before answering the question addressed in the title of the paper, the corpus
upon which my answer is based should be defined and justified. The corpus
consists of documentaries about Brazil, produced by West Europeans, and
broadcast by cable television channels in Belgium. The decision to investigate
documentaries focusing on my home country/culture (Brazil) came about as a
result of my observation that what Belgians - and other western and Central
Europeans - seem to be interested in, and what they know, about Brazil is no
longer "expected" topics such as football and carnival, but has to do with a very
specific issue: Brazilian street children.
Two years of videotaping (from November 1995 until December 1997)
led to some interesting findings about programmes in general and documentaries
in particular, such as: (a) programmes about Brazil were frequently broadcast (a
total of 80 programmes were recorded over two years); (b) in terms of
production, the vast majority of these programmes were produced by West
Europeans (85% of the total); (c) in terms of topics discussed, the most
frequent one (27% of the programmes) was the issue of "the street children", a
trend that justifies Europeans' current interest in Brazil and also confirms the
social relevance of such programmes (this issue featured in about 27% of the
documentaries recorded on Brazil); and last but not least, (d) most of the
documentaries (66%) made use of a hardly explored audiovisual mode of
transfer, namely "voice-over".
A preliminary investigation of the 22 documentaries which constitute the
corpus indicated that, in general terms, translating documentaries can be viewed
as a specific practice mainly because of aspects considered typical of the genre,
which differ from those involved in the translation of feature films (either
subtitling or dubbing). Many of the translation decisions, however, seem to be
determined not by genre-specific factors but by the host culture. The following
brief account of some of the aspects involved in documentary film translation
suggests that the activity remains to be explored by researchers.
Documentary film translation 235
Unlike scripts for the translation of fiction, documentary scripts are very often
post-production texts, a fact which is a consequence of the regular procedure
adopted by documentary filmmakers in which shooting precedes scriptwriting.
Scriptwriting of factual material, therefore, may mean either the transcription of
the original from the videotape before actual translating or straightforward
translation from the tape.
Thus, the important point to be recognized is that, contrary to fiction,
the videotaped material acquires a pivotal position in the translation of
documentaries, being the source text from which original transcriptions and/or
first-translated versions derive. So, in the case of a German documentary where
interviews with Brazilians are translated directly from the videotape, when this
documentary is imported by France the available models for the French
translator are the original tape and the German written version. If a transcription
of the original is also available, the number of candidates - i.e. written source
models at the disposal of the translator of the French version - consequently
increases.
4. Modes of transfer
5.7 Orality
instrument, and then to the woman who starts playing her instrument) it all
starts...(now all play and sing together)'
(lb) Die Musik kommt aus meinem Innern. Ich kam nach Candeal und machte
Musik. Die Musik lebte schon immer in mir. Für mich ist alles Musik -
Trommeln, Bongos, das Klatschen und das Schlagen der Stöcke. Das
Klatschen ist der Rhythmus damit fängt alles an. Und dann setzen die
Trommeln ein.
'The music comes from my inner self. I came to Candeal and made music. The
music has always lived in me. For me everything is music - drums, bongos,
clapping and the beating of the drum sticks. Clapping is the rhythm from which
everything takes off. And then come the drums'.
5.2 Objectivity
Surprisingly enough, the German versions did not prove very objective as far as
the meaning of foreign/exotic information was concerned. The German texts
contained some semantic shifts which seem to have been motivated by two
distinct factors. One is the cultural/linguistic gaps which exist between German
and Brazilian Portuguese. For example, in Surfavela, a Portuguese production
about a Brazilian social project which aims at integrating street children into
society through surfing, the surfers' jargon becomes the language of the
documentary. It is characterized by specific vocabulary, slang terms (many of
them derived from English) and idiomatic expressions. Owing to the absence of
a "surfing culture" in Germany (where the word Surfing generally refers to
windsurfing) and the consequent lack of vocabulary and expressions to support
the translation, the transfer of meaning became partly impaired and the German
version of Surfavela resulted in a text that was difficult to understand in the
recipient culture.
Secondly, some semantic shifts in the German translations seem to have
resulted from deliberate attempts to deviate from the original intended message.
One example can be found in Carlinhos Brown, and the comparison of the two
language versions (with French reproducing the main idea of the original) makes
my point clear:
240 Eliana Franco
(2a) Quand je crée un groupe comme Lactomia avec des enfants, je pense à
quoi ? A cette arnaque autour des enfants de rue du Brésil.
'When I create a group like Lactomia with the children, I think of what? Of this
farce about the street children of Brazil.'
(2b) Ich habe die Gruppe Lactomia für die Kinder Brasiliens gegründet. Ich
möchte ihnen helfen, denn die Stras senkinder haben hier ein sehr hartes
Leben.
'I created the Lactomia group for the children of Brazil. I wanted to help them
because the street children have a very hard life here.'
By deviating from the original message, the German translation provides it with
a negative perspective about the source culture which reveals the opposite of
what the speaker is trying to convey to his foreign audience. It is fair to suggest
that such a behaviour was ideologically motivated.
Interestingly enough, the few instances of semantic shifts observed in the
French versions seem to have resulted from a clear attempt to eliminate bias.
These are mainly found in the German production Hundert Morgen Paradies,
where commentary and questions put to interviewees often take on an
evaluative tone throughout the documentary, which aims to emphasize the poor
living conditions of landless families, especially the children. Let us have a look
at an example:
(3 a) "Ihr schlaft doch nicht alle auf diesem schmalen Bett? "
'You don't all sleep in this small bed?'
Attitude is conveyed here through the emphatic negative (doch nicht) in the
German interviewer's original question (3a). The utterance then becomes more
an expression of the interviewer's indignation about the interviewee's living
situation than an actual question that needs to be answered. By omitting the
negative, the French version completely eliminates the evaluative tone of the
utterance.
6. Conclusion
References
LUYKEN, G. M. & al. 1991. Overcoming Language Barriers in Television. Dubbing and
subtitling for the European audience (Media Monographs no. 13). Manchester: The European
Institute for the Media.
MAILHAC, J.P. 1998. "Optimising the linguistic transfer in the case of commercial videos".
Translating for the Media (Papers from the International Conference, Berlin, November 22-
23, 1996) ed. by Yves Gambier. Turku: Centre for Translation and Interpreting, 207-223.
WINSTON, B. 1995. Claiming the Real The documentary film revisited. London: British
Film Institute.
Documentaries
Hundert Morgen paradies. Brasiliens landlose nehmen sich ihr Recht (1996), a German
production (1996) by Lourdes Picareta and Philip Siegel, broadcast by WDR on 14/05/96 and
by ARTE on 10/05/96.
Carlinhos Brown (1996), a French production by Claude Santiago, broadcast in French and
German by ARTE on 26/10/96.
Surfavela (1996), a Portuguese production by Joaquim Pinto and Nuno Leonel, broadcast in
French and German by ARTE on 05/11/96.
TRANSLATION AND MASS CULTURE
Zusammenfassung
Übersetzungen fur den Massenmarkt stellen ein Problem für die
Übersetzungswissenschaft dar. In diesem Artikel wird eine Analyse der
Gegenwartsszene der Massenmarkt vorgenommen und betrachtet, wie Übersetzung in
der Realität "gemacht" wird: Angefangen wird mit einem allgemeinen Überblick über
die Übersetzung für Massenmarkt anhand von Buchklub-Übersetzungen in Brasilien
(JM). Nach dem Überblick folgt eine Untersuchung darüber, wie ein dänischer
Erfolgsautor eine Massenleserschaft auf dem englischsprachigen Markt erreichte (KM).
Anschlieend wird eine Untersuchung über die Naturalisierung der Kinder- und
Jugendliteratur für den deutschsprachigen Markt vorgestellt (VS).
Résumé
Les traductions pour la culture de masse pose un certain problème à la traductologie.
Nous nous proposons de voir comment est réalisée la traduction pour le marché de
masse, en étudiant trois exemples distincts: les traductions pour un club de livres au
Brésil (JM), la trajectoire de l'auteur danois P. Høeg devenu célèbre dans le monde
anglophone (KM), et la façon dont la littérature pour enfants est adaptée, naturalisée
pour les marchés germanophones (VS).
Resumen
Las traducciones para los mercados de masa constituyen un cierto problema para los
Estudios de Traducción. Este articulo hace un balance de la situación actual del
mercado de masas para ver cómo se realiza realmente la traducción. Se utilizará para
ello un estudio general de la traducción para el mercado de masas, utilizando como
ejemplo traducciones realizadas para un club de libros en Brasil (J.M.). Después se
hizo un seguimiento con estudios de cómo un conocido escritor danés de grandes éxitos
consiguió llegar a un amplio publico en el mercado angloparlante (K.M.), y cómo se
naturaliza la literatura infantil para el mercado alemân (V.S.).
244 Malmkjær, Milton & Smith
The illustrations in this section are taken from studies into the translations of the
Clube do Livro book club and the the translation of classic fiction in Brazil
(Milton 1995).
1.2 Standardization
of art and destroy the ritualistic and magical elements surrounding the original.
With film and photography there is never a single original. The very nature of
both arts is essentially reproductive: films and photographs cannot be collected
as paintings can. Moreover, the cost of a film is so high that it must be made
available to the highest number of people possible. Indeed, the nature of film is
highly democratic as it will make all myths and great authors and artists
available to cinema-goers.
The fragmentation of a film is very different to the unity of a painting or
a poem. The director, responsible for the final product, will depend on the
expertise of many other different areas, sound, photography, etc., of which he
will have only a limited knowledge. He will be more of a manager of the final
product. Compare this with the painter or novelist, who will have all elements
of the work of art under their control. The final result of the film or photograph
will come about as the result of the combination of a number of very different
elements. Edgar Morin points out that in spite of this standardization, the
culture industry does not sell soap powder, and that within a certain framework,
there is a demand for originality. New products must always be invented. This,
for Morin, is the "dynamic contradiction" of the culture industry (Morin
1977:28).
Though Benjamin's essay concentrates on the reproducible arts of film
and photography, his essay also helps us to understand commercial translation,
where the editor will coordinate the work of original author, translator,
copydesk, illustrator and finance department.
do Livro study, very simple spelling mistakes were found, especially in the
names of authors: Virginia Wolff, Charlotte Bronté, George Elliot, and Kunt
(sic!) Hamsun.
Commercialism is not beyond a few tricks — the reader may not be treated with
all that much respect. Translations were generally introduced as special
translations, a euphemism for "contains many cuts".
There has been a steady stream of Danish literature into English since the early
19th century, but relatively few authors have reached the British (or American)
best seller lists. The most recent writer to do so is Peter Høeg (b. 1957).
When Høeg published Forestilling om det tyvende Àrhundrede
('Vision/performance of the 20th century', 1988), it was likened by reviewers to
the magical realist work of Allende and Garcia Marques, and its ironical tone to
that of the older, socially critical Danish novelist, Hans Scherfig (1905-1979).
It was first published in the UK in 1996 as The History of Danish Dreams,
translated by Barbara Haveland.
Høeg's second publication was a collection of short stories, Fortællinger
om natten ('Tales about the/at night', 1990; translated as Tales of the Night).
Critics note the similarity to Karen Blixen and the shock effect in the short
stories which is certainly retained in De måske egnede ('The perhaps suitable',
translated as Borderliners) published in Denmark in 1993 and in the US in
November 1994. This book accuses an array of well-known, contemporary
educationalists, many of whom are still living, of having ruined the lives of very
many children in their care, but it is also a more general attack on the Danish
establishment's way of dealing with difference and failure. It is not, however, the
book that initially made Høeg's name in English.
That book was Frøken Smillas fornæmmelse for sne (1992), published
in the United States, translated by Tiina Nunnally as Smilla's Sense of Snow, in
1993. Høeg wanted alterations made in the translation before publication in
Britain, but Nunnally refused to let the altered translation go out under her
name. Consequently, the name F. David was invented for the version published
in Britain as Miss Smilla 's Feeling for Snow (Guido Waldman, personal
communication). Smilla has been described as a thriller, as the Danish answer to
Jules Verne, as a crime/science fiction novel, and as a socially critical fable of
suspense. It has now been made into a film.
Kvinden og aben ('The Woman and the Ape') was published on 1 April
1996. This book has been seen as reflecting the animalism of our times, a
critique of modern civilisation. Critics have noted echoes of The Jungle Book,
King Kong, and Planet of the Apes.
The Danish editor at Farrar Straus and Giroux, Høeg's American publisher, is
reported (Korse 1994) to have said that what is special about Høeg is that his
style is at once Danish and international; and it is possible that the combination
of this mixture of Danishness and international-ness with Høeg's subject matters
is a major factor in making him a best seller in English. This raises some
Translation and mass culture 249
questions about the troubled notion of the translator's invisibility (Venuti 1995).
This notion is not simple, but part of what Venuti seems to mean by it is a kind
of fluency. Venuti also wants to link the question of visibility versus invisibility
to the question of foreignising versus domesticating (1995:21): "Anglo-
American culture ... has long been dominated by domesticating theories that
recommend fluent translating." He claims that in the postwar period, when
modernism was marginalised in English-language translation, "Most translators
chose a fluent, domesticating method that reduced the foreign text to dominant
cultural values in English" (1995:225), and he seems to hold that this trend is
still in force.
However, Høeg's texts in English are not inscribed with dominant
English values, at least not obviously so. They are made fluent in English to
some degree, but there are plenty of markers of Danishness in the language, in
particular in the names of persons and places. Other markers of linguistic
foreignness are transferred from the originals, with respect to which they are
also foreign (though perhaps less so in Danish than in English in the case of
names from Greenland). But the fact that cultural difference is marked up in the
translations, which are therefore (and for other reasons) visible as translations,
may actually be more a domesticating feature, than the opposite (see below). So
there is some difficulty in placing Høeg with respect to the oppositions Venuti
sets up; and it may also be the case that Venuti's concepts are rather too broad
and imprecise to be used for analytic purposes.
Better, perhaps, to approach the work with something like Lefevere's
notion of a typology of changes (1992: 61) in mind, and try to determine which
adjustments to the receptor culture have been made in the translations, and
which foreign elements have been retained, and then try to decide whether these
can be said to serve domestication or foreignisation, and if they can, how and in
what sense. Such a typology might include the translation-oriented categories of
Additions, Omissions and Changes, because it is by adding, omitting and
making changes over and above those demanded by the basic translation
process that translators domesticate or foreignise. There are, however, no
significant omissions, so far as I am aware, in the Høeg texts.
The typology should probably also include Setting as a category because
of its obvious connection with culture, and Innovation because it often amounts
to the type of generic change which, as Lefevere (1992: 23) points out, is
desirable when there is a perceived need for it "in the environment of a literary
system for that system to be or remain functional" in the sense of having "an
impact on the environment by means of the works it produces, or the rewritings
thereof'. But it is obvious from what was said above about the reception of
Høeg's work that he has been considered innovative, so no further use will be
made of this category of analysis.
250 Malmkjær, Milton & Smith
2.2 Additions
2.2.1 Extratextual
The Høeg translations, but not the originals, offer some information about the
author. This highlights the foreignness of the writer and may therefore be called
foreignising; but it constitutes domestication insofar as it conforms to a British
norm for book presentation. The translation of Smilla includes, on the page
immediately before the narration begins, a map of Copenhagen which is
presumably intended to help the reader follow Smilla around the town and
which obviously highlights the foreign setting for the story. The History of
Danish Dreams has a historical setting and includes historical characters. Many
of these are listed at the end of the translation in a "Biographical List of Real-
Life Characters". Obviously, the list highlights the foreignness of the setting,
and so does the paragraph which introduces the list, which mentions Danes and
Denmark repeatedly.
2.2.2 Intratextual
F. David, in Smilla, inserts explanatory phrases like "the children's book" in
front of "Rasmus Klump on the Ice Cap"(p. 12), the title of a book which every
Dane would recognise as a children's book. It could be argued that the mere
fact of having to categorise a book is foreignising, but in general this strategy as
employed by David promotes transparency and fluency, for example when she
inserts "his summer home" in front of "on Klitvej in Skagen" (Target Text p.
16; Source Text p. 24, my translation):
TT Here they haven't put in a tennis court. But not for lack of space. It's
probably because Loyn has a a couple of them in his back garden in Hellerup,
and two more at his summer home on Klitvej in Skagen.
ST In here they haven't made a tennis court. But not for lack of space. It is
probably because Loyn has a couple in his back garden in Hellerup, and two
more on Klitvej in Skagen.
2.3.1 Punctuation
Where Høeg uses punctuation sparingly, particularly in Borderliners and in The
History of Danish Dreams, to create a run-on, stream-of-consciousness effect,
Haveland's translations normalise the punctuation to a degree - tidy it up, to
make for a somewhat calmer, more controlled, conventional reading, and also
to make it conform more closely to English norms. This promotes transparency,
and is domesticating.
2.3.2 Names
Haveland retains almost all Danish names, though she translates some names of
institutions. These translations clearly provide some information about the
nature of the organisations, but it may seem foreignising, and so, of course, is
the retention of the Danish names, which is the primary strategy used in all the
translations. F. David adjusts the name of the child Esajas, who becomes Isaiah,
in accordance with English Biblical tradition, but not with Greenlandic Biblical
tradition, which is Danish based. Obviously, this is an instance of acculturation
towards English.
2.4 Setting
Why do translators of children's books feel licensed to change as they see fit
instead of providing the children with access to the original? [...] Is it possible
that children in their intelligent and imaginative capacities to learn and be
exposed to the strange world of the original are largely underrated?
Why should this be the case? And do the Enid Blyton stories to be discussed
here confirm these findings? A superficial reading of many of the children's
books available in translation might lead to the conclusion that the main stock-
in-trade of children's literature is the exciting story rather than the creative use
of language. This of course also applies to many mass-media translations where
translators are explicitly told to translate freely, as Smith and Klein-Braley
Translation and mass culture 253
Two main problems arise here in translation: firstly, how to transfer the
culturally-saturated setting of the English boarding school, and secondly, how
to update the text for contemporary readers. It is conceivable that the timing of
the translations into German, still relatively close to the end of the Second
World War, actually influenced the translation strategy. Otherwise it is hard to
imagine why the brown school uniform of Malory Towers should be changed to
dark blue! On the other hand there is evidence that the decisions taken were not
really based on any principled consideration of the setting; a consistent strategy
of domestication has not been adopted: the setting is naturalised. Blyton
located Malory Towers in a deserted part of Cornwall and has her chief
protagonist travel from London to get there. The translators omit references to
these real places but go on to translate details of the setting so literally that it
turns entirely into a fantasy landscape, since no German-speaking country offers
a similar topography.
An aspect which causes considerable confusion is the way sports are
dealt with. In Malory Towers one of the main sporting activities is the game of
lacrosse, played outdoors on a field with a stick. Despite the fact that this game
is not actually played by very many English native-speaker readers and so may
be as exotic for them as for German-speaking readers, the translators decide to
opt for a hybrid strategy of translating references to the game as Tennis or
Handball or Turnen (Gymnastics). (In the examples below, MT = Malory
Towers, SC = St. Clare's, HN = Hanni und Nanni, D = Dolly; numbers refer to
the volume in question.)
The problems this confused strategy leads to come to a crux in the fifth volume,
where a whole chapter is devoted to the description of a lacrosse game. From
the very start, the German version fails to mention which sport we are dealing
with (not even calling it handball), so we can only deduce from passing
references what is going on. But this is easier said than done. Tests done with
various readers show that during the reading process, football, handball and
basketball are all called to mind, only to be discarded when the sentence "wurde
ihr den Ball aus dem Netz geschlagen" ('the ball was forced out of her net') is
reached, leaving the readers completely in the dark. This passage shows what
can result when the ramifications of the options available are not followed
through consistently.
Translation and mass culture 255
"Shows what Malory Towers does to you!" said Darrell. "Still, I suppose any
good boarding-school does the same things - makes you stand on your own feet,
rubs off your corners, teaches you common sense, makes you accept
responsibility. (p. 110)
"Das zeigt nur, wie wohl uns Möwenfels tut!" meinte Dolly. (D5, p. 103).
Or in MT2:
"I'm going to the post with this for Daphne. But don't go and split on me, Gwen.
I know it's against the rules." Gwendoline stared at Mary-Lou in surprise. "You
breaking the rules!" she said. "I don't believe you ever did that before. You're
mad to think you can get to the post and back in time." (p. 126).
"Ich gehe für Diana zur Post, um das Päckchen aufzugeben", erwiderte Marlies.
Evelyn starrte sie ,berrascht an. "Jetzt noch?" fragte sie. "Du kannst niemals
pünktlich zum Abendessen zurück sein, wenn du jetzt noch zur Post gehst!'" (D2,
p. 119).
Again, in MT2, three whole pages in which the heroine Darrell confesses to the
headmistress that she had wrongly accused another pupil Ellen of stealing and
cheating at exams are missed out, while later the confession of the real thief is
dramatically curtailed. In SCI a chapter where Isabel (Nanni) accidentally finds a
set of exam questions which will give her an unfair advantage is left
untranslated. The moral lessons to be learned in these passages, which might
also be relevant for the new readership, i.e. that cheating, boasting, lying,
stealing and lack of consideration for others' feelings are not acceptable forms
of behaviour, do not figure in the German text.
Though these examples may give the impression that these stories are moral
treatises for the edification of young girls, it must be said that humour is an
important element in them. In both series, the French mistress is one of the main
comic characters. Mam'zelle is represented as a good-hearted but
temperamental figure, fond of theatrical gestures, whose command of English is
256 Malmkjær, Milton & Smith
somewhat shaky and who is the favoured target for tricks and practical jokes, as
she has herself a sense of humour. This character is partially reproduced in the
German text in a similarly stereotypical fashion. Some of the language errors are
also taken over and adapted to fit the German language, such as,
"We shall gobble together about it." "Gabble, Mam'zelle, you mean," said Sally,
with a giggle. (MT5, p. 20).
Nevertheless the German translator cannot sustain the effort of linguistic jokes
and leaves many of them simply untranslated. Elsewhere the language "errors"
ascribed to Mam'zelle are simply misunderstood by the translator, e.g. the word
"chest" appears to have been understood as "trunk" (= large suitcase), rather
than "heart":
"You will soon be their friends, n'est-ce pas? Tomorrow they will - what do you
call it - they will take you to their chests and you will be one of them." MTS, p.
20).
"Du wirst bald mit ihnen Freundschaft schliessen, n'est-ce pas? Morgen werden
sie dir ihre Sachen zeigen, und du wirst ihnen gehören." (D5, p. 27).
One of the highlights of MT5 is when, after years of having practical jokes
played on her, Mam'zelle takes revenge and plays a joke on the girls. Having
confiscated a catalogue of items for practical jokes, she sends for some fake
monster teeth and wears them to a sports event. She revels in the astounded
reactions of all she meets, including colleagues. The urge to laugh, however,
becomes too great and she decides to return indoors to remove the teeth. On
the way she meets various other people, including the headmistress with some
parents, and then in the company of the girls in her class she can hold out no
longer and explodes with laughter, dropping the teeth. Only then do the girls
realise that their teacher has played a "treek" on them, and her popularity is
reconfirmed. This whole chapter is omitted in German. It is not quite clear why:
perhaps the idea of a teacher resorting to such undignified behaviour is not
compatible with German school stories, perhaps the translation was just
growing too long. In any case, the readers are cheated of one of the most
hilarious incidents in the whole of Malory Towers.
This brief comparison of these English and German texts confirms some
of the tentative conclusions put forward by House:
Translation and mass culture 257
there may well be differences as regards children's books between the German
and English linguacultures: there is a German tendency towards depicting a
different role relationship between children and adults, there is [...] less humour,
a far greater explicitness and a greater need to impose edifying ideas and
ideology on the stories told in children's books. (1997:130)
Certainly the degree of humour found in the English originals of Blyton's stories
is not found in the German translations, and interestingly enough the degree of
explicitness in the German translations often does exceed that of the originals.
In view of the fact that much of the detail of the originals is omitted, this
conclusion may surprise, but in fact often the detail is replaced by a single
sentence which summarises and at the same time makes actions more explicit
than they would have been. The edification which House finds in her German
data is less present in these texts. This may be the result of an attempt to update
the books and situate them in a social context in which the boarding school and
the class considerations attached to it do not play an equivalent role.
In direct comparison, the German translations come across as less exciting from
a child's point of view. It is all the more surprising, therefore, that they enjoy
such enormous success with German-speaking readers. A critic from India
(Hindu Index, 26.04.97), documenting Blyton's success in India, claims that
Blyton has no real competitor among local authors, apart from those who have
chosen to clone her. Could it also be the case in the German-speaking world,
that, regardless of the dubious quality of the translations, Enid Blyton simply
fills a niche that no-one else has succeeded in filling?
Note
1
The guidelines advocate the following procedures: Sie müssen frei übersetzen. Übernehmen
Sie nicht die oft schwülstige, überladene Ausdrucksweise des Originals. [...] Wann immer Sie
im Original auf inhaltliche Fehler stoen, bitte ändern. [...] Gelegentlich reagieren die
Hauptpersonen unverständlich, oder ihre Gefühle sind nicht nachvollziehbar. Auch das
dürfen Sie ändern.[...] Sie dürfen nicht nur, sondern Sie müssen sogar sehr frei übersetzen. (A
free translation is essential. Do not reproduce the sometimes excessive and over-exaggerated
language of the original... Whenever you encounter content errors, correct them...
Occasionally the main characters act oddly or their emotional reactions do not seem natural...
This should also be changed. A free translation is not just recommended: it is a "must".)
References
FERREIRA, Jerusa Pires, 1993. O Livro de Sao Cipriano: uma Legenda de Massas. Säo
Paulo: Perspectiva.
KORSE, Jens. 1994. Review of De måske egrlede. Det Fri Aktuelt, 16 November.
LEFEVERE, André, 1992. Translation, Rewriting and the Manipulation of Literary Frame
London & New York: Routledge.
MILTON, John, 1995. "A Traduçao de Romances 'Classicos' do Ingles para o Português no
Brasil" ('The Translation of 'Classic' Novels from English to Portuguese in Brazil'),
Trabalhos em Lingüistica Aplicada, 24, Instituto de Estudos da Linguagem, UNICAMP,
Campinas, Brazil.
MILTON, John, 1996. "The Translations of O Clube do Livro". TradTerm 3, 47-65. Säo
Paulo: Universidade de Sao Paulo.
MORIN, Edgar, 1977. Cultura de Massas no Século XX: O Espfri to do Tempo 1- Neurose.
Rio de Janeiro: Editora Forense Universitaria.
VENUTI, Lawrence. 1995. The Translator's Invisibility: A History of Translation. London &
New York: Routledge.
St. Clare's, Vols. 1-6. Methuen, 1942-1945. German translations: Hanni und Nanni, 1-6.
Schneider Verlag, 1965-1970.
Malory Towers, Vols. 1-6. Methuen: 1946-1951. German translation: Dolly, 1-6. Schneider
Verlag, 1966-1967.
Translation and mass culture 259
Masa Nomura
University of Sao Paulo, Brazil
Zusammenfassung
Die vorliegende Arbeit behandelt die Art der Stereotypen, auf die die Werbung in
einem globalisierten Kontext zurückgreift. Eine Auswahl von Werbetexten im
Deutschen und im brasilianischen Portugiesisch soll dazu dienen, Fragen der Identität
und der Fremdheit anhand des Stereotypenbegriffs zu erläutern. Die in Werbetexten
gefundenen Stereotypen stimmen nur teilweise mit der allgemeinen Vorstellung
überein, was unter "Deutschsein" und "Brasilianersein" zu verstehen ist.
Résumé
Ce travail traite de quelques similitudes et différences entre les stéréotypes culturels
utilisés dans la publicité alors que la mondialisation est en cours. On a analysé une
sélection de textes publicitaires en allemand et en portugais brésilien dans la
perspective des questions d'identité et de différence, la notion de stéréotype éclairant
ces questions. Les stéréotypes rencontrés dans les textes étudiés confirment, en partie
seulement, ce qui est entendu par les expressions "être allemand" ou "être brésilien".
Resumen
Este trabajo habla de algunos puntos comunes y diferentes entre los estereotipos de
que se vale la publicidad en un contexto globalizado. Una selección de textos
publicitarios en alemân y en portugués brasileño fue analizado con la fmalidad de
discutir cuestiones relativas a identidad y diferencia y la noción de estereotipo que
ilumine estas cuestiones. Los estereotipos encontrados en los textos confirman
parcialmente la noción consensual de lo que sea "ser alemân" y "ser brasileño".
262 Masa Nomura
1. Introduction
When the Portuguese discovered Brazil in 1500, the first powerful image they
had of the new land was the colourful presence of parrots in a green landscape,
so they named it 'the land of parrots'. Reports by early European travellers
mention 'naked Indians eating human flesh'. These examples are mentioned
merely to illustrate the tendency that predominates in the field of advertising
for the tourist industry nowadays, especially the one directed at the European
public that has retained the emphasis on the exoticism of the 'Southern lands'.
Even presumably more informed depictions of what could be called the Third
World embody ideological barriers. As Lennox (1994:185) critically points
out, the lack of interest in the historical and cultural background of less
developed countries has led to a trend towards overemphasising negative
aspects, such as underdevelopment, poverty, social disarray, etc. Exoticism and
poverty are aspects that still mark the European view of Third World peoples.
In spite of the differences, the progressive intertwining of international
and economic relations imposes increasingly closer and more frequent contacts
with foreign cultures. This is confirmed by the expansion of markets beyond
geopolitical borders, a trend that requires companies to commercialise the same
products and services in different cultural environments simultaneously. One
should also consider issues normally studied in marketing and advertising
schools (Garcia 1990, Marcondes and Ramos 1995), such as the way the media
influence consumers, producing changes in deeply engraved habits or instilling
the compulsion to buy often superfluous products and services. Understanding
the discourse of international advertising thus involves questions that concern
culture-specific data and norms. Multinational companies seek to operate a
coherent production and marketing strategy on a global basis. Nevertheless, in
Europe and North America it is increasingly realised that translations of
avertising texts must be adapted, not translated (Schmitt 1985:2).
Advertising has now become a rich field of research also for translation
studies (cf. Reiss 1976, Snell-Hornby 1986, Azenha 1994, Azenha, Dornbusch
& Nomura 1997). The awareness of the confrontation between identities
sensitises the translator to the continuum that exists between reality and text:
"Advertising [...] is nothing but a part of a wider reality, namely, the set of
social relations gathered together at a specific historical moment" (Garcia
1990:9; translation M.N.). It is clear nowadays that whenever cultural barriers
are crossed advertising texts need to be rewritten or 'adapted'.
2. Stereotypes
3. Advertising texts
Samples of advertising texts were selected from Brazilian (100 texts) and
German magazines (another 100 texts). Both corpora have a similar journalistic
format (2 weekly magazines in each country: Veja and IstoE, Der Spiegel and
Focus), the same large circulation (approximately 1 250 000), the same type of
contents (news briefs, politics, sports, culture, new trends), and the same target
public (average working men/women, usually seeking to be well-informed
about all kinds of issues). Each one dominates the market over similar printed
media; they are opinion formers and known to be serious and reliable in their
treatment of the news. The advertising texts they carry are presumably
successful, thanks to the great number of potential consumers they reach.
A Brazilian advertisement for the Jeep Cherokee: "Motor 4.0L High Output,
suspensao Up Country, traçao Selec-Trac © 4X4 Full-Time com cambio
automático, ar-condicionado e duplo air-bag." (IstoE 1508, Aug. 98);
"MontBlanc. The Art of Writing." (Veja 36, Aug. 97)
Cerveja Brahma. A numero um. ('Brahma beer. The number one.', Veja 17,
April 99)
Das einzig wahre Warsteiner Spitzen-Pilsener der Premium-Klasse. (Der
Spiegel 17, April 99)
Se voce quer mais da vida - nada mais justo que voce tenha um cartao à
altura. - Faça algo de novo na sua vida. ('If you want more from life -
nothing is more important than a highly recommended credit card. - Do
something new in your life.') (IstoÉ 1499, June 98)
266 Masa Nomura
Por que nao mudar o seu ritmo? Por que nao ficar bonita e gostosa? Por que
nao mudar o seu dia-a-dia? ('Why don't you change your rhythm of life?
Why not be pretty and sexy? Why don't you change your routine?') (Nescafé,
Veja 36, Sept. 97)
Sie werden es mit allen sechs Sinnen erleben. Alle anderen nur mit einem.
Sehen Sie die Welt mit anderen Augen. Fühlen Sie die Sonne warm auf der
Haut. Atmen Sie den frischen Duft der Natur. Und gemessen Sie es, unterwegs
zu sein. Das neue Saab 9-3 Cabrio gibt Ihnen die souveräne Sicherheit, alles
perfekt unter Kontrolle zu haben. [...]. (Der Spiegel 8, Feb. 98)
3.3 Image
boy and a little girl; mother and children, a young father looking after his son;
many infants, preferably cute and blond with blue eyes, selling a large number
of different products. A Brazilian advertising agency was very successful in
launching a campaign for a multinational manufacturer of dairy products
(Parmalat) with children of several ethnic backgrounds dressed up as different
mammals. The appeal to many people's feelings of tenderness towards children
in their infancy paves the way for the sale of the advertised goods.
Sexuality (Br. 15%, Ger. 13%). The explicit or suggested resort to
sexuality is highly appreciated among advertisers in general. In Brazil, a
tropical country with a long coastline full of sunny beaches, images of young
men and women with perfect tanned bodies in the skimpiest swimsuits are
widely used in advertising. To appreciate naked bodies is therefore rather
natural in Brazil and has no special erotic effect on Brazilian onlookers.
However, it immediately catches one's attention if the advertising plays on
social taboos or prejudices. A black-and-white couple, both young, beautiful
and sexy, was chosen to sell a new brand of coffee; he is black and she is
blonde, or vice-versa, in six image versions. The verbal text by a cup filled
with hot coffee draws one's attention to the perfect match between the two:
Chegou o café Parmalat. O café à altura do nosso leite. ('Parmalat coffee has
arrived. The coffee to match our milk.')
Although miscegenation is (in theory) socially accepted in Brazil, the very fact
of presenting it so clearly in the black-and-white interplay caused a great
impact on the text recipients.
The automobile as a symbol of status (Br. 30%, Ger. 50%). Shining cars
in metallic colours, an image exploited both in Brazilian and in German
advertising, are used for the purpose of arousing the covetousness of the
potential buyer. In both cultures, cars are often depicted against a vast
landscape, as in many advertisements for car brands such as Cherokee, Toyota,
Mitsubishi and others. Such an arrangement emphasises the sense of conquest
of new spaces caused by the possession of that particular object of desire.
Characters belonging to folk fantasy and the world of arts (Br. 12%,
Ger. 15%)). A knowledge of history and mythology is to some extent
incorporated in German folk fantasy and often appears in German advertising:
characters from fairy tales, witches, goblins, giants, kings, queens, as well as
portrayals of classical paintings, famous sculptures, boroughs, castles and
churches. In Brazil, popular heroes and heroines come from the sports world
(Ayrton Senna, car racing champion; Ronaldinho, football star; Gustavo
Kuerten, tennis champion) or from the television entertainment world; the
mixture of fictional and real characters plays an important role in people's
fantasy. One of the most popular icons of Brazilian television show-business,
Text, image and translation 269
Xuxa, blonde, beautiful, young and rich, gains millions of dollars by serving as
a model for many advertising agencies.
Religious symbols (Br. 8%, Ger. 7%). In Brazil, as well as in Germany,
religious symbols are very common in advertising. The image used for a
German car manufacturer shows the interior of a Gothic cathedral to symbolise
the vast interior space of its vehicles. In Brazil, a country marked by
Catholicism and religious syncretism, images of Jesus Christ, the Pope and
popular Catholic saints, as well as gods and goddesses of the Afro-Christian
religion, may serve to attest the quality of some goods.
These examples show that the messages produced in the advertising
industry are structured, above all, around the interplay between image and
verbal text: the higher the number of visual and/or intertextual points of contact
around which such associations are built, the stronger the persuasive power of
the message. Owing to the richness of specific cultural references, the task of
the translator of such advertising texts on a global scale will be very complex.
The difficulty for translation is proportional to the extensiveness of these
multiple connections.
There are two main consequences that should be mentioned. One is the need
for adaptation (in the broadest sense of the term) with due appreciation of
culture-specific conditioning factors, stereotypes, and the peculiarities of the
advertising industry. The second is the way we are forced to reflect on the
notion of 'translating' itself, since, in the case of advertising, it is the appealing
element that will have to be translated, even if this means nearly recomposing
the entire text.
Secondly, the translation of advertisements highlights question of the
creativity involved in the process of rewriting texts, so that they can operate
with images related to intrinsic values of the target-culture, provided the
original message is preserved. It also underlines the importance of mastering
the expressive means geared to a specific translation job. The translation of
advertising reveals conditioning factors that are present, to a varied extent, in
the translation of any kind of text (political speeches, religious texts,
manifestos, legal documents, user's instructions). It therefore stresses the
hybrid character of texts. It also calls for reflection on the translator's role as a
re-creator, not just a simple mediator of the information exchanged between the
several parties involved in the communication process (source-text recipient,
target-text producer, mediator of the negotiation between advertiser,
advertising agency, publishers etc.). Finally, it invites the reconsideration of the
question of 'authorship' in translation (individual work, team work, the role of
the advertising agencies) and of ethics (the sanction imposed by the company
270 Masa Nomura
on the 'quality of the translation', the responsibility for the published content,
the method of diffusion etc.).
References
AZENHA, J. 1996. "Kulturelle Aspekte bei der Übersetzung technischer Texte - Deutsch-
(brasilianisches) Portugiesisch. Theorie und Praxis". Die Unterrichtspraxis/Teaching German,
73-76. New Jersey: AATG.
REISS, K. 1976. Texttyp und Übersetzungsmethode. Der operative Text. 3rd ed.: 1992.
Heidelberg: Niemeyer.
Cristina Valdés
University of Oviedo, Spain
Zusammenfassung
Der folgende Artikel beschäftigt sich mit der Rezeption von Werbeanzeigen, die für
den spanischen Markt übersetzt wurden, und schlägt eine grundlegende
Herangehensweise an die Beobachtung vor, dass einem Text bei der Rezeption durch
verschiedene Kulturen unterschiedliche Bedeutungen verliehen werden. Eine
deskriptive Analyse der ins Spanische übersetzten Anzeigen zeigt, dass die
Übersetzung fast immer eine Verändung der einzelnen Textelemente erfordet, da es
sich um multimediale Texte handelt, welche verbale und nonverbale Komponenten
enthalten. In allen untersuchten Beispielen seigt sich, dass verschiedene Faktoren und
die jeweilige Rezeption der Anzeigen zu unterschiedlichen Übersetzungsstrategien
fuhren.
Résumé
Cet article présente quelques-uns des facteurs touchant la réception de publicités
traduites pour le marché espagnol et propose une approche générale du fait que les
éventuelles cultures de réception donnent une signification différente à un même texte.
Les textes multimédia combinent des composants verbal et non-verbal: leur traduction
exige presque toujours une transformation de ces composants, comme le prouve
l'analyse de publicités traduites en espagnol. Dans tous ces cas, différents facteurs
déterminent les stratégies de traduction.
Resumen
El siguiente articulo presenta algunos de los factores que afectana la recepción de
anuncios traducidos para el mercado español y ofrece una aproximación general al
hecho de que distintas culturas receptoras asignan un significado diferente al mismo
texto. Ya que se trata de textos de naturaleza multimedia que combinan un
componente verbal y uno no-verbal, su traducción supone casi siempre una
transformación de dichos componentes, como demuestra el análisis descriptivo de un
numero de anuncios traducidos al espafiol. En todos estos casos, factores diferentes y
su recepción dan lugar a varias estrategias traductoras.
272 Cristina Valdés
1. Introduction
be, however, different versions of the same target text depending on the
medium employed (e.g. television or printed magazines).
In advertising, the feedback process is indirect and delayed; marketing
experts employ different methods to analyse consumers' responses in many
countries. This does not imply an individual reaction but a group response,
from particular segments of society which might react differently to the same
ad. Reactions can be found in letters sent to the editor in magazines, surveys
among television viewers, etc. Although the feedback effect is not immediate,
changes in the campaign may be required after its first reception in the target
market. A well-known example of this was a printed advertisement promoting
a new soap powder which showed a picture of dirty clothes on the left, a box of
soap in the middle and clean clothes on the right. The soap did not sell well in
the Middle East. Why? Obviously because the advertisers forgot that in that
part of the world, people usually read from right to left. The main obstacle here
was not the linguistic component but the placing of the pictures.
The first goal is to produce a text that will be considered 'acceptable' by the
target text receivers. The translator knows the target culture conventions for
that particular text type, as regards both verbal and non-verbal elements, and
works according to these conventions. Potential consumers have preconceived
ideas which affect the reception of ads. Once they identify the target text as an
ad, they relate it to their expectations about this type of text. They also bring to
it the knowledge they possess about the product and brand, and about the sort
of ads for that brand which they are used to seeing. If there is a dramatic
change in any one of these points, their reactions may be unexpected.
The translation of advertisements may entail the transformation of any
textual element, and these changes usually depend on receivers' expectations
as regards prototypes, that is, conventionalized advertisements for particular
products or brands, or on their lifestyle. Texts must be understood and must be
relevant to the target audience if they are to be convincing. Other decisions
made by the translator which involve a transformation in the target text are
related to changes in the non-verbal component (images and sounds) or to
restrictions imposed by the media. The examples discussed below show how
receivers influence the strategies adopted by translators.
Two advertisements for the airline Iberia which appeared in Spanish and in
English in 1997 share similarities as regards the colours, the picture, and the
content of the text; however, there is a clear difference between them, which
274 Cristina Valdés
has resulted from a change of focus in the translation: they were produced with
two different target audiences in mind. The Spanish source text was published
in the weekly magazine supplement El Semanal, which addresses middle-class
readers of both sexes. The slogan Lideres entre Europa y Latinoamérica
introduces the main theme of the text and the advantage the Spanish airline
company intends to promote: it acts as the link between Spain and the rest of
Europe and also with Latin America, as is explained in the first paragraph. The
translation into English, however, entails a change in the target of the ad since
the audience of the text is now those readers of the international publication
Time who might wish to fly to Spain and Latin America, as the slogan
indicates: Leading the way to Spain and Latin America. A different target has
meant different translation strategies: while in the Spanish text the verb used is
unimos (El Semanal, 26th October 1997), given the links between Spain and
Latin America, in the English version the translator simply "offers" (Time, 17th
November 1997) destinations in those countries.
It is the last part of the text that most clearly shows the difference in
focus between both advertisements: the Spanish text mentions the wish of the
company to become the best company in the world for the Spanish audience:
Alli donde tú vayas. Donde tu quieras llevarnos (El Semanal, 26th October
1997), so the target addressed is the Spanish audience for which Iberia is Tu
primera compania, emphasizing the familiarity of the audience with the
company. However, the English text has been written for an international
audience, so the translator has adopted the perspective of a foreigner flying to
Spain and Latin America: We 'd like to be the airline of your choice when flying
to Spain and Latin America. In this case the company is "No. 1 to Spain and
Latin America" and not from Spain to Latin America as it appears in the
Spanish text. A different recipient has thus given rise to a change of strategies.
2.2 A new product in the target market: Advanced Night Repair, by Estée
Lauder
in the accompanying picture of the bottle. This strategy is very often adopted
by translators when the product is new in the target market and its use is not
well known. The visual image of the product can help to indicate what the
promoted product is, but the advertisement can make more explicit its
advantage in the cosmetic market.
Although the Spanish translator keeps the core of the message of the
source text, s/he adds some extra information because of the skopos of the
target text, which is not only to sell but to introduce a product which is
unknown to the target audience. A long fragment has been added to emphasize
the benefits of the introduction of the product in the Spanish market:
In this fragment the translator urges potential Spanish customers to buy the
product as soon as possible in order to make up for lost time.
Translators also adopt this kind of explicitation strategy (see e.g.
Fawcett 1997:100) when the promoted product is closely related to a culture
which is distant from that of the receivers and therefore there may be more
information which needs to be made explicit. The translator fills these cultural
gaps in order to bring readers closer to the advertisement, but preserves the
foreign flavour of the product. An example of this strategy is provided by an
advertisement of Scotch Glenfiddich whisky translated into Spanish. The
central concept of the campaign highlights the origin of this whisky in the
Highlands and the importance that the water of a certain valley has in the
making of the liquor. The Spanish translator, presupposing the lack of
understanding among non-Gaelic speakers, added extra information about the
name of the product: Los nombres dificiles sugieren el misterio. Glenfiddich,
en gaélico, es el Valle del Ciervo'. If the target were a Gaelic-speaking
community, this addition would obviously be redundant. Compare also the
explanation of the English name in: The Highlands. Las Tierras Altas. Región
escocesa de extraordinaria belleza donde se elabora el mejor whisky del
mundo.
A similar strategy can be seen in the following advertisement for a
Finnish vodka:
The next example is a pair of advertisements for the liqueur Tia Maria. Both
the English source text and the Spanish target text present the same visuals but
with a different slogan. While the English source text invites readers to choose
between drinking Tia Maria "Neat, mixed or over ice?", the Spanish translator
has adapted his/her translation according to the drinking habits in the Spanish
culture, where such drinks are either consumed neat or with ice: ¿Solo o con
hielo? The strategy employed by the Spanish translator has thus been
influenced by the target culture norms: while the image follows the
globalization principle, the linguistic message has been adapted according to
the target.
Sometimes a different use of a product in the target culture entails a
change of the image as well, and a new message in translations. An example of
this can be found in the Kellogg's TV advertisements in England and the
United States on the one hand and in Spain on the other hand. While in the first
case, cereals are commonly consumed at breakfast by both adults and children,
this product was launched in Spain mostly as breakfast food for children, so
translators need to be aware of this usage when addressing their texts to the
different targets.
commercial. The text was produced, translated into English and dubbed by an
English native speaker in an agency in Madrid. However, when sent to the
London office it had to be translated and dubbed again. The reason was that the
actor responsible for the dubbing had to play the role of a lower-class London
shopkeeper but neither the refined style nor the intonation used was appropriate
for a shopkeeper. Once more, the multimedia nature of the commercial, the
combination of sound, image and text, imposed restrictions on the translation.
With globalization in mind, advertisers seem to rely nowadays less on
linguistic written forms in television commercials and more on music,
particularly on English pop songs. But still, unless the song is very popular,
hearers in contexts other than where the song originated will fail to recognize
the meaningful relationship established between the lyrics and the whole
message of the spot.
words related to shine. With that purpose in mind, the headline Cybershine is
translated into Italian as Cyberflash, plus a subheading in futuristic graphics,
La nuova moda conquista lo spazio, which has been added in the Italian
version. The cold colours employed and the main paragraph also retain the
connotations of words such as magnetica, futuro scintillante di tecnologia
spaziale, etc., and therefore they produce a similar effect on Italian viewers.
Another addition made in the translation shows a photograph in which some
lipsticks and nail-polish bottles appear as if they were astronauts floating in the
galaxy. This picture is closely related to the clothes worn by the model.
the blonde couple from the beginning; the camera makes the same movements
but this time framing the blonde girl. At the beginning and end of the
commercial the blonde couple is shown on the left side, occupying the position
the dark-haired couple had in the Spanish text. The message and the actors are
the same in both texts but the change of the target audience has meant some
transformations in the final presentation of the advertisement. The medium has
played a significant role in this example, since the use of different camera
movements is one of the advantages of television, for it allows the framing of
certain segments of the picture rather than others, to place more emphasis on
them.
The next example once more reveals the complexity involved in multimedia
translation and how translation strategies depend on the intended reception of
the text. The four advertisements I will describe are part of American Airlines'
international campaign "Smiles" in Europe and Latin America. The translation
of the original English text has not only meant a linguistic transfer into
Swedish, French and Spanish but also a transformation of the semiotic
interplay of non-verbal elements.
The text focuses on the picture of a smiling blonde American Airlines
air-hostess, Susan Kent, who is directly facing the viewer. The photograph is
the same for the English, Swedish and French advertisements, while in the text
for the Latin American market the air-hostess is called Isabel Philippi and
represents the stereotypical image of Latin women: dark skin and hair. The
smile is kept, as it serves as the persuasive central concept of the campaign.
The visual manipulation is required to achieve the viewer's identification with
the stereotyped image of the model, but this is done retaining the smile.
There are also some other changes in the smaller pictures related to the
skopos and the target audience of the advertisement. While in the Spanish
translation the promoted American Airlines service between Latin America and
United States is addressed to a general audience, the English source text, as
well as the Swedish and French versions, specifically target businessmen, as
we can infer from the references to work in these texts. This change of target
alters the structure of the Spanish translation, and the offer of five-course
menus and facilities for businessmen in two of the smaller pictures disappears
in the other versions. Other subtle transformations have to do with the different
conventions as regards clothes: for example, in the Latin culture men do not
usually wear braces but they do wear a necktie, as is depicted in the picture.
The marketing experts who mention this campaign state that it "ran in
Europe, basically changing only the language" (Wells et al. 1995:738), a
statement that a closer look at the four texts has proved untrue. The purpose
here has not been to make a detailed analysis of the four advertisements but to
280 Cristina Valdés
3. Conclusions
The study of these advertisements has shown how certain modifications in the
visual and oral components are required in order to convey the intended effect
of advertisements on the target audience. Multimedia texts offer a good
example of how cultural conventions and intended text reception affect
translation strategies in what marketing experts regard as "a globalized
context."
The idea of a "one-audience" world has prevailed in advertising so far.
However, evidence from translations shows that transformations often take
place. In spite of all the globalization attempts at making a one-culture world
by means of employing the same visual, oral and written elements, translators
still feel the need to adapt their messages to their target receivers and they thus
challenge the tendency of globalization, as the different strategies I have
mentioned have revealed.
References
BRIERLEY, S. 1995. The Advertising Handbook. London & New York: Routledge.
COOK, G. 1992. The Discourse of Advertising. London & New York: Routledge.
REISS, K. 1981. "Type, Kind and Individuality of Text. Decision Making in Translation".
Poetics Today 2(4), 121-131.
WELLS, W., J. BURNETT & S. MORIARTY. 1995. (3rd edition) Advertising. Principles and
Practice. New Jersey: Prentice Hall.
CREATIVITY IN LEGAL TRANSLATION: HOW
MUCH IS TOO MUCH?
Susan Sarcevic
University of Rijeka, Croatia
Zusammenfassung
Während Kreativität im juristischen Übersetzen im allgemeinen bestritten ist, vertritt die
Autorin den Standpunkt, gewisse Abweichungen vom Ausgangstext seien nicht nur
akzeptabel, sondern sogar nützlich, vor allem wenn dadurch Klarheit und richtige
Betonung gefördert und die durch die Paralleltexte angestrebten Folgen herbeigeführt
werden. Beispiele aus den zweisprachigen kanadischen Bundesgesetzen zeigen, wie die
französisch sprechenden Übersetzer ehemalige als sakrosankt angesehene Regeln
verletzen und neue Texte schaffen, die weder beim Lesen noch im Aussehen den
englischen Texten entsprechen.
Résumé
La créativité en traduction juridique prête plutôt à controverse. Notre point de vue est
que les modifications apportées à l'original sont acceptables, sinon même souhaitables,
pour arriver à plus de clarté et souligner ce qui est important, les textes parallèles
atteignant ainsi les objectifs recherchés. On citera des exemples de la législation
fédérale bilingue du Canada: les traducteurs (en français) transgressent désormais des
règles naguère considérées comme sacro-saintes, produisant de la sorte des textes
nouveaux mis à distance de leurs homologues en anglais.
Resumen
Aunque la creatividad en la traducción juridica suscita polémica, este trabajo propone
que cierta "desviación del original" es permisible y hasta deseable para lograr mayor
claridad y énfasis con el fin de conseguir que, en la practica, los textos paralelos
obtengan los resultados deseados. Los ejemplos citados de la legislación federal
canadiense bilingüe muestran cómo los traductores francófonos en la actualidad violan
unas reglas previamente consideradas sagradas, produciendo textos nuevos que no se
parecen a sus homólogos en inglés ni en su aspecto ni en su lectura.
282 Susan Sarcevic
1. Introduction
Fearing that any deviations from the original could pose a threat to uniform
interpretation and application, lawyers traditionally obliged translators to
preserve the letter of the law by reproducing the wording and syntax of the ST
as closely as possible. This view was finally challenged early in the twentieth
century by Virgile Rossel, a law professor who dared to translate the German
text of the Swiss Civil Code into "natural" French. Accused of heresy for
having altered the letter of the law, Rossel defended his "revolutionary"
translation by invoking the principle of language equality. In his opinion, the
French-speaking population of Switzerland had the right to have their laws
written in the genius of the French language (see Sarcevic 1997:36-40; also
Dullion 1997:379-381).
Legal translation 283
designations of "source" and "target" text no longer apply and the translator is
gradually converted into a co-drafter. The question to what extent co-drafting
can be regarded as translation is dealt with elsewhere (Sarcevic 1997:105-108).
Here it suffices to say that co-drafting is not traditional translation but involves
new forms of text production (cf. Vermeer 1996:34, where he now argues for a
broad definition of translation).
new French texts of Canadian federal legislation are "unusual" and contain an
"element of surprise" for all those accustomed to reading common law legislation.
From this point of view, Kussmaul's definition of creativity as a "shift from the
original" that is "functionally adequate" (1998:39) can also apply to these texts. In
my opinion, creativity in legal texts can be deemed "functionally adequate" only
when the "shifts" do not interfere with the uniform interpretation and application of
the parallel texts in question. On the contrary, they should promote uniform
interpretation and application by achieving clarity and proper emphasis, ensuring that
the texts will lead to the same results in practice. While creativity in legal translation
entails innovation, it should not be used for the sake of innovation itself Creativity is
at its best when legal translators use innovative means to compensate for differences
between legal systems and languages, with a view to promoting uniform
interpretation and application. Hence, creativity in legal translation is not simply free
translation.
Looking at the French translation of the above provision, also in the Revised
Statutes of Canada, 1970, we see that the translator faithfully reproduces the
wording and syntax of the ST, even preserving the capitalization of English
words. A notable exception, at least from the legal point of view, is the last part
of the statement of law that is formulated as a separate sentence in the French
text, thus breaking the common law drafting rule that each section must be
formulated as a single sentence:
At about the time co-drafting was introduced, the Labour Code was undergoing
a total revision, as a result of which the entire French translation was revised.
While the English text of the above provision remained practically unchanged
(now section 253(1)), major interventions occur in the French translation of the
provision published in the Revised Statutes of Canada, 1985. Daring to break
rules previously considered sacrosanct, the translator/co-drafter completely
reformulates and restructures the translation, making it read more like a French
provision:
The boldest innovation in the French text is undoubtedly the departure from the
basic conditional sentence, thus proving that a variety of surface structures can
be used to express a legal rule without disturbing the thought process. Instead of
expessing the conditions of the fact-situation in an introductory lorsque-clause as
earlier, the translator/co-drafter places them in an adverbial phrase (dans le cadre de
la présente partie ou de ses règlements) embedded in the statement of law
between the main verb and its modal (peut exiger), thereby qualifying the legal
action. Of course, there are restrictions to the translator's creativity. In particular, he
or she must ensure that the new surface structure expresses the deep structure of the
intended logical relations (here: if/then), and that the elements of the fact-situation
and statement of law are clearly formulated and easily recognizable.
Legal translation 287
Such interventions make it clear that the creative intentions of the Francophone
translators/co-drafters are not limited to language but also include the right to
create a new text with a distinctive style based on civil law drafting practices.
This is particularly evident when one compares the English and French texts of
section 253(1) of the Canada Labour Code, as revised and amended in the
Statutes of Canada, 1993:
This time the substance of the provision has been amended by inserting a time
specification into both texts. In English, the time specification and the existing
requirement are set off from the main part of the text by paragraphing, a
technique frequently used in common law legislation to enumerate a list of
contingencies, alternatives, requirements or conditions to be applied
cumulatively or alternatively. Although the device serves as a visual aid to
comprehension by indenting blocks of text, Francophone translators/co-drafters
contend that it is not always used to its best advantage. As a result, they reject
paragraphing whenever possible and incorporate the clauses into the main body
of the provision, as has been done in the French text of the amended provision.
As seen above, the rejection of paragraphing by Francophone translators/co-
drafters results in two very different looking texts, each with its own distinctive
style. Although such changes affect the composition of the French text, they are
usually tolerated as long as the text is clear and the substance unaltered.
As far as the substance is concerned, the question arises as to whether it
is unduly altered by acts of creativity resulting in semantic shifts. Whereas
common law provisions tend to be particular, civil law provisions are formulated
in general terms. Convinced that they can express the same idea in more general
terms, Francophone translators/co-drafters began to use a generic term to
express a series of subordinate terms in the English text. For example, in section
9 of The Official Languages Act (Statutes of Canada, 1988), the English
expression "all rules, orders and regulations governing the practice or
procedure..." is rendered in French as "les textes régissant la procédure et la
pratique. .."In this case, the generic term textes is acceptable because it includes
rules, orders and regulations. Such changes, however, involve a high degree of
risk as the translator must be absolutely certain that the generic term covers all
the subordinate terms, yet is not too broad so as to unintentionally widen the
scope of application, allowing the text to be applied in unintended situations. To
avoid possible misinterpretation, Francophone translators/co-drafters sometimes
identify the items covered by the generic term in an appositive phrase and use
the generic term in a subsequent subsection. For example, the subject of
subsection 1 of section 11 of The Official Languages Act, A notice,
advertisement or other matter is rendered as Les textes - notamment les avis et
annonces in the French version. Instead of repeating the subject and part of the
qualifier in a lengthy cross-reference as in subsection 2 of the English text, the
translator/co-drafter simply uses the generic term:
Legal translation 289
Where a notice, advertisement or other\ Il est donné dans ces textes égale
matter is printed in one or more\ importance aux deux langues officielles.
publications pursuant to section (1), it
shall be given equal prominence in each
official language
the entire territory of Canada; hence, two legal systems come into play:
common law and civil law. A system of federal courts has been established to
decide disputes in both English and French in all matters governed by federal
statute. Canadian federal law is common law; thus the sitting judges are trained
in the common law and apply common law rules of interpretation. The fears of
Canadian common law lawyers therefore appear to be justified: the increasingly
bold creativity by Francophone translators/co-drafters has reached its upper
limits of acceptability.
Multilingual communication in the law can be successful only if there is
interaction between text producers and receivers, i.e., those who interpret and
apply the parallel texts of a given instrument. Among other things, this means
that the drafting techniques used in each of the parallel texts should be
compatible with the rules of interpretation applied by the relevant courts. Since
common law rules of interpretation are applied by federal judges, including the
justices of the Supreme Court of Canada, it follows that drafting elements that
interfere with such rules pose a threat to uniform interpretation and application
and are thus unacceptable in federal legislation. In keeping with this reasoning,
common law lawyers have warned translators/co-drafters to refrain from using
elements of civil law drafting that are incompatible with the methods of common
law interpretation. In the interest of guaranteeing the harmonious co-existence
of the two texts, Francophone translators/co-drafters have apparently agreed to
strike a balance with the English text and the common law in federal statutes.
This, however, is not their last word.
References
COVACS, A. 1982. "La réalisation de la version française des lois fédérales du Canada".
Langage du droit et traduction ed. by J.Cl. GÉMAR, 83-100. Montréal: Linguatech & Conseil
de la langue française.
DRIEDGER, E. 1982. A Manuel of Instruction for Legislative and Legal Writing. Ottawa:
Ministry of Justice.
DULLION, V. 1997. "Lorsque traduire, c'est écrire une page d'histoire: La version française du
Code civil suisse dans l'unification juridique de la Confédération". L'histoire et les théories de
la traduction, 371-388. Berne: ASTTI & Genève: ETI.
GÉMAR, J.-CI. 1995. Traduire ou l'art d'interpréter, tome 2: Application. Sainte-Foy: Presses
de l'Université du Québec.
KUSSMAUL, P. 1995. Training the Translator. Amsterdam & Philadelphia: John Benjamins.
292 Susan Šarcevic
SAGER, J. 1998. "What Distinguishes Major Types of Translation?" The Translator 4(1),
69-89.
SARCEVIC, S. 1997. New Approach to Legal Translation, The Hague, London, Boston:
Kluwer Law International.
VOLMAN, Y., EIJSBOUTS, W.T. & VAN MONTFRANS, M. 1988. La cour et les langues.
Amsterdam: Universiteit van Amsterdam.
WILSS, W. 1996. Knowledge and Skills in Translator Behavior. Amsterdam & Philadelphia:
John Benjamins.
FRANK WEDEKIND'S SEX TRAGEDY LULU
IN ENGLISH AND FRENCH VERSIONS
THE CASE FOR AN INDIVIDUALISTIC VIEW OF
LITERARY TRANSLATION
Rainer Kohlmayer
University of Mainz, Germersheim, Germany
Zusammenfassung
Die Schlußszene von Wedekinds Lulu-Tragödie stellte die englischen, amerikanischen
und französischen Übersetzer vor eine schwere Aufgabe. Zum einen konnte die
schockierende Reihe von 'Freiern' religiöse, ethnische und kulturelle Normerwartungen
verletzen. Zum andern erwies sich die sprachliche Hybridität als ästhetisches
Rezeptionshindernis. Der Aufsatz gibt einen ersten Überblick über das breite Spektrum
der individuellen übersetzerischen Konzepte und plädiert für eine stärkere
Berücksichtigung der ästhetischen Autonomie der Übersetzer.
Résumé
La tragédie Lulu de Frank Wedekind (1864-1918) fut à l'origine très à l'avant-garde.
Le texte, qui mélangeait plusieurs langues, allait à l'encontre de bien des normes
morales et esthétiques de l'époque. Les traductions américaines, anglaises et françaises
présentent une multitude d'approches individuelles très différentes. Il est assez étonnant
que même les versions anglaises les plus récentes occultent complètement la polyphonie
linguistique de la pièce originale de Wedekind.
Resumen
El drama sexual de "Lulu" de Frank Wedekind (1894) fue el primer intento de los
escritores alemanes de crear una obra teatral realmente cosmopolita tanto por su tema
(el poder global del sexo) como por su polifonia lingüistica (al mezclar varios idiomas y
formas de expresión). Este articulo trata de las actitudes personales de los traductores
ingleses y franceses y de sus decisiones frente al texto rebelde de Wedekind.
294 Rainer Kohlmayer
The aim of my paper is mainly to describe how Wedekind's English and French
translators intervened linguistically, aesthetically, and ideologically in order to
make the Lulu-tragedy cross the linguistic and cultural border from Germany
into France, England, and the United States in the course of the 20th century.
My approach is, in certain respects, rather similar to that of the
Manipulation Group, especially to Lefevere's and Toury's widely accepted
thesis that translators tend to adapt their work to the norms of target cultures or
target cultural polysystems, i.e. mainly to the norms of the ruling aesthetics and
to the power structure of the ruling institutions. What I reject, however, at least
within the framework of literature, is the determinism that very often seems to
crop up in explanations of translators' and readers' behaviour. Changeable
attitudes, habits, and conventions are defined as norms, and it is frequently
taken for granted that the existence of a so-called norm is reason enough for a
translator to stick to it, or that the norm expectancy of theatre audiences or
reading publics is the most important basis of literary success or failure. Norms
are thought to govern literary translators' behaviour. Therefore the individual
translator's behaviour is thought to be of interest only insofar as he or she
follows or contradicts those pre-existing norms.
In my opinion, literature is definitely not a closed system with built-in
causalities. Literature is not made up of a universe of norms to be broken or
stuck to by writers and translators. Literary translators are, as a rule, not merely
functionaries of target norms or target groups. They have a considerable
amount of autonomy. Not even during the Nazi epoch in Germany, when the
German literary system was regulated as never before, did translators'
behaviour become entirely predictable. There were translations in favour of and
against the ruling powers of that time (see e.g. Kohlmayer 1994;1996a).
Taking into consideration the fact that the system approach cannot fully
explain, let alone predict a literary translator's behaviour, we ought to devote
more time and energy to finding out about literary translators as individual
persons and decision-makers. I do not deny the pressure of social norms, yet I
do not accept that literary translators are deconstructed into ego-less adherents
to, or mere fulfillers of, pre-existing norms. Literature and art are, as long as
they are original products, first and foremost something individual, and thus, by
definition, something un-systematic and even ab-normal.
I thus adhere to an "individualistic" theory of literary translation. For me,
all literary translations are, first and foremost, individual compromises. The
translator's basic maxim - a de facto Gricean maxim - is: "Be prepared to
compromise". The quality of the individual compromise depends, to put it in a
nutshell, on four characteristic features of the literary translator: on his or her
Lulu in English and French versions 295
individual knowledge and skill, conscience and courage. (See also Kohlmayer
1996b, 1988; and Gombrich 1993.)
First I discuss which particular features of Wedekind's earlier and later versions
presented the most obvious difficulties for translators.
Perhaps the most striking and surprising feature of Wedekind's original text of
1894 is its linguistic hybridity. The first three acts, set in Germany, are written in
296 Rainer Kohlmayer
German. The fourth act, however, set in Paris, makes abundant use of the
French language. Most of the nouveaux riches, pimps, journalists, prostitutes,
etc. assembled in Act Four speak French. I quote a few characteristic lines from
the beginning of Act Four:
The fifth act, set in London, and ending with Lulu's murder at the hands of Jack
the Ripper, contains passages in German, French, Swiss German, as well as
more than 250 dialogue passages in English.
In one of his notebooks of the time of writing, Wedekind commented on
his use of different languages by referring to Shakespeare:
Lulu. Why will you not stay all night with me, Sir?
Jack. I am married man.
Lulu. You say, you missed the last bus and that you have spend the night with
one of your friends...
Jack. Time is money. - How much do you want?
Lulu. Pound...
Jack. Good evening. - (geht zur Thüre)
Lulu (hält ihn zurück). Stay! - Stay! (Wedekind 1990:127).
Most of the time Wedekind's level of English is suitable for the characters he
portrays. In terms of aesthetics, the very lack of normal verbal communication
in the last act can be interpreted as an additional dimension of modernity. I
quote from Edward Bond's introduction:
Lulu barters with her customers in English. The play shows an African
speaking it. It is becoming the world language - English is the language of
capitalism. [...] The impression is strangely modern, like a multi-language film
with subtitles (Bond 1993:63).
A third feature creating problems for the linguistic transfer of the play into other
cultural systems was the heterogeneity of Lulu's four customers in Act Five and
the considerable critical potential inherent in their portrayal. The sequence with
Lulu's four customers in the last act can be seen as a kind of quick-motion
298 Rainer Kohlmayer
I shall discuss Lulu's first three customers in some detail, together with the
corresponding translations, before commenting on Bond's and Bentley's recent
English versions.
The first "gentleman" (Wedekind 1990:128) Lulu brings up from the street is
Mr. Hopkins, a pious Christian, judging from the book Schigolch and Alwa find
in the pocket of his coat: "Lessons for those - who are - and those who want to
be - Christian Workers" (Wedekind 1990:113). According to Lulu, he is "totaly
[sic] mad" (Wedekind 1990:128). Mr. Hopkins does not only refrain from
uttering a single word himself, he also forbids Lulu to speak by laying a finger
to his lips, putting his hand over her mouth, and so on. Mr. Hopkins'
pantomime is not only funny, it is also a Nietzschean attack against Christianity:
Mr. Hopkins' strange behaviour demonstrates the Platonic splitting up of man
into body and soul. Mr. Hopkins' mind does not want to know or hear about
what his body is doing.
In the 1913 version Wedekind slightly emphasised the ideological
content of the scene by changing the title of the book to "Ermahnungen für
fromme Pilger und solche, die es werden wollen" (Wedekind 1989:165;
"Exhortations to pious pilgrims and those anxious to become such" in Spender
1952:162), thereby evoking the well-known Christian tradition of the soul's
pilgrimage on earth.
Interestingly enough, the French version of 1969 cut out the passage
with the prayer-book so that, for the French public, Lulu's customer was no
longer identifiable as the archetypal Christian (Jouve 1969:172). And,
surprisingly, the English version by Beck and Barnes in 1971 also neutralised
the anti-Christian potential of the scene, by transforming the character into a
"deaf mute", who "puts his forefinger to his lips and ears and shakes his head
making strange rasping noises" (Beck & Barnes 1971:75). This character was
obviously physically handicapped, whereas Wedekind's Mr. Hopkins was, as it
were, mentally or spiritually handicapped, i.e. handicapped by Christianity.
Lulu in English and French versions 299
Lulu's third customer, Dr. Hilti, is the most complex character of the four
(Wedekind 1990:122-125). After the archetypal Christian and the extremely
barbaric variety of male sexuality, Wedekind presents us here with the capitalist
and calvinist variety of male sexuality. It is important to realize that Dr. Hilti
belongs to the uppermost layer of European society and culture. On the one
hand, he is a young professor of philosophy from Zurich, a "Darwinian", as he
says, coming from a wealthy aristocratic family, and about to get married to a
Swiss millionairess from Basel. On the other hand, he is absolutely illiterate in
emotional and erotic matters. For Dr. Hilti, sexuality is nothing but man's
instrument for the production of offspring. He comes upstairs with Lulu because
now, after getting engaged, he is in a hurry to find out how the thing works. He
is the quintessential image of what Max Weber, in his sociology of Calvinism's
Protestant ethic, termed "innerwordly asceticism" ("innerweltliche Askese"). All
these geographical, social, financial, and ideological features are symbolically
united in the strong Swiss German dialect he speaks.
A further difficulty for translators, on top of the cultural and linguistic
complexity of Dr. Hilti's character, is added by the multilingual conversation
between him and Lulu. I quote a short passage from Wedekind's original
manuscript in order to demonstrate Wedekind's stunningly avantgarde
technique of mixing pidgin English, French, Swiss German, and Standard
German:
In the 1913 version the scene was cut by more than half, the multilingual
passages disappeared, and the mixture of languages was reduced to German and
Swiss German (Wedekind 1989:174f). But even so Dr. Hilti contained
sufficient snags for translators. Here is a brief review of the translators'
individual solutions.
The first American translator, Eliot, reduced Dr. Hilti's social status
substantially, perhaps because of the mistaken belief that Dr. Hilti's strong
Swiss dialect might somehow be a sign of vulgarity, whereas the very opposite
is true: it signifies wealth. So for Eliot, Dr. Hilti is no longer the puritan,
provincial product of Switzerland's financial aristocracy, he simply comes "of a
very old country family", and speaks "in the broadest north-country vowels"
(Eliot 1914:74). Besides, he is no longer a Privatdozent, but merely a "tutor".
His fiancée undergoes the same fate of social relegation: she is no longer a
wealthy Basel aristocrat's daughter with two million to her credit but owns
merely "a hundred thousand" (75). Whereas Wedekind's Dr. Hilti is a
provocative caricature of the puritan ideal which combines sexual abstinence
with social and financial success, Eliot's Dr. Hilti is merely a boorish provincial.
Stephen Spender, Carl Richard Mueller, Beck and Barnes, and Eric Bentley
translate Hilti's Swiss dialect into perfectly normal English, with only Bentley
adding a footnote to inform the reader about Hilti's Swiss dialect, "a form of
speech Wedekind found amusing" (Bentley 1994:194). He does not inform the
reader about the social and financial connotations of the Basel dialect. Beck and
Barnes cut the Hilti scene by about half. The French translator leaves it out
completely, and so does, amazingly, Edward Bond, although in his preface he
had praised Wedekind's multilingual modernity.
It is surprising that none of the English versions managed to do justice
to Wedekind's Dr. Hilti, linguistically and ideologically the most interesting
character in the last act.
Lulu's last customer, Jack the Ripper, seems to have been less of a problem for
the English and French translators. The most obvious aesthetic and ideological
changes occurred in Beck's and Barnes' adaptation, where Jack the Ripper
becomes mythologized into something like a high priest of the theatre of cruelty.
The adapters add the following stage direction: "As in a dream he slowly raises
the bloodstained knife high like a chalice and licks it" (Beck & Barnes 1971:82).
302 Rainer Kohlmayer
Both translations are, in my view, important works of art; both are accompanied
by original poems by the translators themselves, thereby demonstrating the
extent to which they had become involved creatively in their work, and both
translators offer detailed interpretations of the play, with Bentley's introductory
essay (and translation) reacting critically to Bond's previously published
interpretation. Bond's interpretation is Marxist, Bentley's is, broadly speaking,
Feminist. He sees Wedekind as the "playwright of a sexual revolution" (Bentley
1994:26).
Both translators correct Wedekind's weak English. Most of the time they
rewrite Wedekind's sentences even when they are perfectly correct, and most of
the time they do so irrespective of the person speaking. Bond is somewhat more
restrained in this respect, anxious to preserve some of the "strange,
hallucinatory effect" (Bond 1993:XXVII) of Wedekind's English in the last act,
whereas Bentley turns Lulu into quite a proficient speaker of English in her
dealings with her customers. Both translators seem to reject the idea of
presenting the readers with speakers of pidgin English.
Lulu in English and French versions 303
Where Bentley keeps all of Wedekind's characters, Bond cuts out several
characters in Act Four (as well as Dr. Hilti in Act Five), in addition to changing
the end of Act Four. By cutting the complete story of a twelve-year-old girl
consenting to become a prostitute in Act Four, Bond reduces the sexual theme,
and, by adding a new ending to Act Four, he reinforces the anti-capitalist
content of the play, in keeping with his Marxist interpretation.
Each translation and each interpretation is a coherent piece of work, the result
of an individual aesthetic and ideological compromise and concept. Yet Bond's
Marxist and Bentley's Feminist interpretations of Wedekind's Lulu coincide in
one important point: both eliminate or reduce the linguistic hybridity as well as
the heterogeneity of the characters in Wedekind's play. If cultural hybridity is an
important feature of our time, Wedekind's original play from 1894 must be
considered more avantgarde than the English versions published a hundred years
later.
References
BECK, C. & BARNES, P. 1971. Frank Wedekind. Lulu. A sex tragedy. Adapted by Peter
Barnes from Frank Wedekind's Earth Spirit and Pandora's Box. Translation by Charlotte
Beck, with an Introduction by Martin Esslin. London: Heinemann
BENTLEY, E. 1994. Frank Wedekind's The First Lulu. English version by Eric Bentley. New
York/London: Applause Theatre Books.
BOND, E. 1993. Frank Wedekind. Plays: One. Lulu: A Monster Tragedy, translated and
introduced by Edward Bond and Elisabeth Bond-Pablé. London: Methuen Drama.
ELIOT, S.A. 1914. Frank Wedekind. Erdgeist (Earth-Spirit). /Pandora's Box. Translated by
Samuel A. Eliot, Jr. GLEBE 2 (3/4). New York: Albert and Charles Boni.
GOMBRICH, E. H. 1993. A Lifelong Interest. Conversations on Art and Science with Didier
Eribon. London: Thames & Hudson.
JOUVE, P.J. 1969. Frank Wedekind. Lulu. Version française et adaptation par Pierre Jean
Jouve. Héricourt: Editions L'Age d'Homme.
KOHLMAYER, R. 1988. "Der Literaturübersetzer zwischen Original und Markt. Eine Kritik
funktionalistischer Übersetzungstheorien". Lebende Sprachen 33(4), 145-156.
304 Rainer Kohlmayer
KOHLMAYER, R. 1994. "Oscar Wilde's Society Comedies and the National Socialist
Message". New Comparison 17(Spring), 11-22.
MUELLER, CR. 1967. Frank Wedekind. The Lulu Plays. Earth Spirit. Pandora's Box. Death
and the Devil. Newly Translated and with an Introduction by Carl Richard Mueller. New
York: Fawcett Publications.
SPENDER, S. 1952. Frank Wedekind. Earth-Spirit / Pandora's Box. Translated from the
German by Stephen Spender. London: Vision Press.
WEDEKIND, Frank. 1989 [=1913]. Lulu. Erdgeist. Die Büchse der Pandora. Herausgegeben
von Erhard Weidl. Stuttgart: Reclam.
WEDEKIND, Frank. 1990 [=1894]. Die Büchse der Pandora. Eine Monstretragoedie.
Historisch-kritische Ausgabe der Urfassung von 1894. Herausgegeben, kommentiert und mit
einem Essay von Hartmut Vinçon (Pharus III). Darmstadt: Jürgen Hausser.
PART VII
Culture-bound concepts
SOBRE LA TRADUCIBILIDAD DE LOS
MARCADORES CULTURALES
Leticia Herrero
Universidad de Alicante, Espana
Abstract
Culture markers (culture-bound terms) are elements that are so tied to a particular culture
that exact equivalents cannot usually be found in another language, without the concept
losing its cultural identity. Various translation strategies are nevertheless available, which
respect the cultural origin of such terms to different degrees. The translator's decisions
about what strategies to use naturally affect readers' understanding of the translation, but
they also affect the image of the source culture which is created in the target culture.
Zusammenfassung
Kulturmarker (kulturgebundene Termini) sind Elemente, die eine so enge Verbundenheit
mit einer speziellen Kultur aufweisen, daß ohne Verlust der kulturellen Identität dieser
Konzeptionen keine exakten Äquivalente für sie in einer anderen Sprache zu finden sind.
Zur Verfugung stehen jedoch verschiedene Übersetzungsstrategien, die die kulturelle
Herkunft dieser Termini unterschiedlich berücksichtigen. Es ist klar, daß die vom
Übersetzer getroffeneWahl der Strategie das Verständnis der Übersetzung beim Leser
beeinflußt. Darüber hinaus wirkt sich die gewählte Strategie auf das Image der
Ausgangskultur in der Zielkultur aus.
Résumé
Les marqueurs culturels (ou termes culturellement marqués) sont des éléments tellement
liés à une culture qu'il est presque impossible de leur trouver dans une autre langue des
équivalents exacts sans qu'ils perdent leur identité. Néanmoins, existent différentes
stratégies traductionnelles qui respectent, à des degrés divers, leur origine culturelle. Les
décisions du traducteur, quant à la stratégie à adopter, affectent la compréhension du
lecteur mais aussi l'image de la culture d'origine dans la culture réceptrice.
repentina atracción que parecen haber sentido las editoriales espanolas por las
literaturas de origen asiático, desde la india o hindu hasta la china y japonesa
(aunque la pasión por lo exótico demuestra llegar mås lejos y dar la vuelta al
mundo recuperando lo que, hasta el momento, habia sido apartado por diferente.
La literatura britanica actual ha descubierto entre sus escritores una
comunidad de artistas de origen indio que estân demostrando ser tan herederos de
Thackeray, Austen o Dickens como el más British de los escritores britânicos. El
enorme éxito obtenido en la propia Gran Bretana con libros firmados por Meera
Syal, Anita Desai, Vikram Seth o Rohinton Mistry, sin olvidar el reconocimiento
con el Premio Booker 1997 a la primera novela de Arundhati Roy, The God of
Small Things, son un aval más que suficiente para que las editoriales espanolas se
lancen a la búsqueda y captura de titulos escritos por autores angloindios.
Antes de que los nombres recién mencionados llegaran a Espafia, otros del
mismo origen ya lo habian hecho, pero los motivos que llevaron a traducir los
primeros titulos no coinciden con la politica actual que incita a su traduction, que
claramente responde a la demanda social por estos libros. Es ahora cuando
muchos traductores contratados por las editoriales comparten la ardua tarea de
transferir a nuestra lengua estos textos tan particulares cargados de referencias
culturales; y es esta coincidencia la que invita a retomar la eterna cuestión de los
limites de la traducción y reflexionar sobre los problemas que ineludiblemente
plantea la transferencia de elementos culturales entre lenguas.
Ciertamente, por la parte que les afecta, los Estudios de Traducción no
pueden permanecer ajenos a la creciente publicación de textos hïbridos que se
está produciendo en el mercado bibliográfico internacional. El término hïbrido se
aplica a la literatura de los escritores oriundos de las antiguas colonias para
enfatizar su vinculo con dos culturas, esto es, con una que les es propia por
nacimiento y con otra que lo es por adopción (en el caso que nos ocupa, la india y
la britanica respectivamente). Los rasgos definitorios de la novela angloindia son
radicalmente diferentes a los de un texto firmado por algunos ingleses
contemporâneos como Martin Amis o David Lodge, por lo que es lógico pensar
que la actividad traductora no planteará los mismos problemas en un caso y en
otro. Basândome en este razonamiento, las obras de Hanif Kureishi, Gita Mehta o
Amitav Ghosh constituyen mi punto de partida para reflexionar sobre la
traducibilidad de los elementos culturales.
Todos sabemos que el inglés, el idioma mâs difundido en todo el mundo,
es medio de comunicación incluso en lugares que originariamente hablaban otras
lenguas; el caso de India, donde la lengua de los britânicos permanece como
recuerdo de su pasado colonial, es paradigmâtico. Asi quedaba patente en una de
las multiples entrevistas que Arundhati Roy se viera obligada a atender por el
inesperado éxito de su novela, al describir su uso del inglés como una imposición:
"Después de 300 años de colonialismo" admite la escritora, "fue la lengua la que
me escogió a mi" ("Arundhati Roy. Reflejos de la India". El Pais semanal, 12-7-
98, pâg. 26).
Marcadores culturales 309
India ha visto cómo el uso del inglés le abre una puerta al exterior que, de
lo contrario, estaria herméticamente cerrada para un mundo que no conoce las
otras dos lenguas mâs importantes del pais. El inglés, en el pasado un elemento de
opresión, posibilita hoy al estado hindu un presente cosmopolita y una creciente
proyección al exterior si consideramos el ritmo de expansion de esta lengua.
Como dijera Salman Rushdie, la lengua inglesa ha dejado de ser exclusiva de los
ingleses (Rushdie 1991: 70) y ahora está en boca de chinos, indios, africanos y
caribeños. Este hecho que los hindúes mâs ortodoxos consideran una contrariedad
y una falta de identidad es el arma de trabajo de los escritores postcoloniales; la
lengua inglesa que una vez los colonizó es ahora dominada por los autores y
puesta al servicio de sus propósitos. El periodo postcolonial es el tiempo de la
recolonización de la lengua y la cultura.
Si vamos a hablar de los problemas que surgen al traducir uno de estos
textos, creo conveniente ofrecer unas nociones, aun muy sucintas, sobre la
literatura postcolonial. La novela postcolonial tiene un poco del género
autobiogrâfico, otro poco del histórico, algunas pinceladas fantásticas, y todo ello
dispuesto de un modo que la critica ha creido semejante al realismo mágico. De
acuerdo o no con tal comparación, si es cierto que la narración se convierte para
el autor postcolonial en un proceso de creación de su propia historicia, historia
entendida en su doble acepción de sucesión de acontecimientos reales a lo largo
de un periodo, y relation de los hechos que conforman un relato. (Esta
ambivalencia lingüística esta ingeniosamente representada en el juego de palabras
inglés [his]story).
El escritor postcolonial recrea la historia de un modo subjetivo, desde
contar su infancia hasta recapitular episodios nacionales. Hace con la historia lo
que el borgiano personaje Pierre Menard con El Quijote: escribirla de nuevo
tiempo después, dedicar sus escrùpulos y vigilias a repetir en un idioma ajeno —
recordemos que el inglés le es propio por imposición— algo ya contado (Borges
1939: 280). La historia final asemeja a una especie de palimpsesto en el que
viejos trazos de otras plumas son todavia perceptibles. La verdad histórica, decia
Borges que dijo Menard segün habia dicho Cervantes, no es lo que sucedió, es lo
que juzgamos que sucedió. Y eso es lo que hace el escritor postcolonial, juzgar y
escribir los acontecimientos segün su propia visión de los hechos, escribir la
historia para demostrar el poder mediático del lenguaje que durante siglos ha
silenciado a las culturas minoritarias y marginales.
Inevitablemente, como consecuencia de esta representación personal, el
lenguaje que la transmite tiene que ser asimismo recreado. Si el lenguaje concede
expresión a la realidad cultural, hay una necesidad imperiosa de inventar un
lenguaje diferente para contar una realidad diferente. El Ingles con I mayúscula,
el Inglés estândar, no esta cualificado para ello y, en su defecto, deben recurrir al
inglés con /' minúscula. Para contar su historia, el escritor juega con la lingüistica,
rompe el lenguaje, arranca las palabras de su contexto natural y las contextualiza
en un nuevo entorno cultural.
310 Leticia Herrero
(1) —He encontrado algunos sellos interesantes para Kety —dijo Soli
—Soli me ha traido un Scinde Dawke —balbució Kety con un
repentino estallido de animation.
(R. Lucas, Lafàbrica de hielo, pág. 73).
A lo largo de las novelas, son muchos los casos en que el lector puede
encontrarse perdido ante una referencia cultural, sin contar aquéllos que, debido
al perfil del lector modelo de estas traducciones —un lector de alto nivel
cultural—, son supuestamente reconocibles (por ejemplo: Old Vic, Whitehall,
Indira o Barbara Cartland).
Tampoco la traducción lingüistica del marcador es siempre conveniente
pues, en ocasiones, el resultado confonde al lector, quien reconoce
lingüisticamente el elemento meta, pero semânticamente se le escapa el referente,
como ocurre con los ejemplos n° 3, 4 y 5:
(5) [TO] the free radio scheme was a dead duck —» [TM] el plan de la
radio gratis habia fracasado
(S. Rushdie, Oriente, Occidente, págs. 26/29)
(8) [TO] (...) instructing Jockey to drive the sahib to the police-station →
[TM] le dio instrucciones a Jockey para que le llevase a la comisaria.
(R. Lucas, La fábrica de hielo, 102/128)
No es correcto decir que unas estrategias sean mejores que otras, pues, en
primer lugar, habria que especificar para qué o para quién son mejores. La
estrategia de traducción mas conveniente, mas adecuada o más funcional será
aquélla que el traductor asi lo estime en la situatión interlingüistica global. El
análisis de los marcadores culturales en la novela angloindia me ha permitido
observar la cantidad de condicionantes que determinan la labor del traductor. La
naturaleza del nombre, si es propio o comün, —generalmente, el grado de
conservatión es mayor con los nombres propios que con los comunes—, el
historial interlingüistico entre las culturas involucradas —las versiones prefijadas
son respetadas prácticamente siempre—, o el universo cultural de donde proviene
el marcador —los traductores tienden a ser más conservadores con los
marcadores hindúes y más sustitutorios con los británicos— son algunos de los
condicionantes de traduccián de los marcadores culturales, pero es posible
encontrar muchos otros factores que influyen en la decisión del traductor como el
grado de opacidad comunicativa o el de explicatión.
La traducibilidad dependerá, por tanto, del marcador, su contexto y la
situation de transferencia. Hay un ejemplo que me parece tremendamente
ilustrativo de la comodidad con la que se pueden resolver algunos problemas de
traducción; lo cita John T. Brewer en su articulo "The Role of 'Culture' in
Successful Translation" (1988: 24) y esta tornado de una novela alemana. La
escena narra la huida de un asesino del lugar del crimen tras borrar sus huellas y
eliminar todas las pruebas. La puerta de la habitation esta cerrada; el original
describe cómo el criminal la abre con el codo para evitar dejar marcas. Este
detalle, que en el entorno cultural origen de la novela resulta verosimil, para un
lector americano supondria una hazaña por parte del asesino, puesto que
prâcticamente todas las puertas en Estados Unidos tienen pornos redondos que
difïcilmente se prestan a ser abiertos con el codo.
En el caso anterior, la intervention del traductor en el texto supone un
cambio insignificante; el propio Brewer sugiere colocar un panuelo en el bolsillo
de la chaqueta del asesino con el que poder tocar el porno sin dejar impregnadas
las huellas - gesto, por otro lado, muy recurrente en el cine negro. Pero ¿qué
posibilidades encuentra el traductor cuando el problema de traducción tiene un
alcance mayor, cuando supera los limites de la novela y afecta a la forma de
pensar y de vivir de una comunidad cultural?
Marcadores culturales 315
"Kipling es siempre muy inglés al hablar de la India [...] y sin duda por eso lo
entendemos muy bien; no sólo habla un idioma muy accesible [...], sino que lo
ve todo a través de unas gafas inglesas [...]. Naipaul quiere presentarnos sus
316 Leticia Herrero
personajes tal cual son, no tal como los podemos comprender en Europa, se
identifica con ellos, entra maravillado en su comprensión intima, por lo común
enigmática para nosotros; señala, registra, deja hablar, reúne apasionadamente
materiales que no interpreta [...]. El escenario es tan vasto e intrincado que su
actitud desalienta un poco al lector, que echa de menos esa mania tan occidental
de simplificar las cosas para hacernos la ilusión de que las dominamos.'' (ABC
literario, 30-1-1998, pág. 11)
Repito: "esa mania tan occidental de simplificar las cosas para hacernos la
ilusión de que las dominamos". Creo que ya esta todo dicho. Como escribiera el
Nobel Saul Bellow en una de sus novelas: "For God's Sake, open the universe a
little more". La traducción, o en tal caso, la no traducción de los términos
culturales es un paso hacia ello.
Referendas bibliográficas
BORGES, J. L. 1939. "Pierre Menard, autor de El Quijote". Obras completas, 274-281.
Barcelona: Circulo de Lectores.
KUREISHI, H. 1990. The Buddha ofSuburbia. London: Faber and Faber. Traducción espanola: El
buda de los suburbios. Barcelona: Anagrama, 1994 (1996). Trad. de Mónica Martin Berdagué.
LUCAS, R 1992. The Ice Factory. London: Heinemann. Traducción espanola: La fäbrica de
hielo. Barcelona: Anagrama, 1995. Trad. de Maribel de Juan.
RUSHDIE, S. 1981 (1995). Midnight's Children. London: Vintage. Traducción espanola: Los
hijos de la medianoche. Madrid: Alfaguara, 1984 (1989). Trad, de Miguel Saenz.
RUSHDIE, S. 1991. Imaginary Homelands. Essay and Criticism 1981-1991. London: Granta.
RUSHDIE, S. 1994 (1995). East, West. London: Vintage. Traducción espanola: Oriente,
Occidente. Barcelona: Plaza y Janés, 1997. Trad. de Miguel Saenz.
EL MUNDO DE [LA FILO]SOFIA
LA TRADUCCIÓN DE CONCEPTOS CULTURALES Y
LA FUNCIÓN DEL TEXTO - EL CASO DE LA
TRADUCCIÓN DE EL MUNDO DE SOFIA DE JOSTEIN
GAARDER.
Åse Johnsen
Universidad de Bergen, Noruega
Abstract
The paper analyses the English and Spanish translations of Jostein Gaarder's novel
Sophie's World (Norwegian original Sofies verden, 1991), focusing on the work's
double function as indicated by its subtitle: A Novel about the History of Philosophy.
The two translations differ in their strategies for translating cultural aspects. The
English version seems more to emphasize the novel's didactic function as a history of
philosophy, while the Spanish version presents it more as a work of fiction set in
Norway.
Zusammenfassung
Der Artikel besteht aus einer Analyse der englischen und spanischen Übersetzung des
Romans Sofies Welt (norwegisches Original Sofies verden, 1991) von Jostein Gaarder.
Im Fokus der Betrachtung steht die in dem Untertitel indizierte Doppelfunktion des
Romans: Roman über die Geschichte der Philosophie. Die zwei Übersetzungen
unterscheiden sich in bezug auf die gewählten Strategien zur Übersetzung kultureller
Aspekte. Die englische Version scheint in größerem Maß die didaktische Funktion des
Romans als eine Geschichte der Philosophie hervorzuheben, während die spanische
Version den Roman eher als ein in Norwegen situiertes fiktives Werk präsentiert.
Résumé
Notre contribution analyse les traductions anglaise et espagnole du roman de J.
Gaarder: Le monde de Sophie (original norvégien Sofies Verden, 1991), en mettant
l'accent sur la double fonction de l'ouvrage, telle qu'elle est révélée par son sous-titre:
Roman sur l'histoire de la philosophie. Les deux traductions diffèrent dans leurs
stratégies pour rendre les aspects culturels. La version anglaise semble faire ressortir
318 Åse Johnsen
Figure 1. Figure 2.
la cultura meta. Los funcionalistas toman como punto de partida que una
traducción es un acto comunicativo intercultural, y que como todo acto
comunicativo tiene su propósito. El propósito o la función del texto origen
(TO) puede diferir del propósito o la función del texto meta (TM), y lo que
debe guiar al traductor en su elección de método no es la función del TO, sino
la función o el propósito que ha de tener el TM, o sea, el skopos de la
traducción. Segûn Nord, la convención actual de la traducción literaria sigue
siendo la traducción documento y exotizante, con la excepción de muchos libros
infantiles (Nord, 1997:103). Prueba de aquello son también las fuertes criticas
que hubo en Noruega cuando se supo que en la traducción al inglés de El
Mundo de Sofia muchas referencias a la cultura noruega habian sido adaptadas
a la cultura angloparlante/norteamericana. Nord (1997:80) propone un
acercamiento funcionalista a la traducción literaria para mostrar que, teniendo
en cuenta los aspectos funcionales, una traducción instrumento o una
adaptación es muchas veces la que da el resultado más adecuado para la
situación comunicativa meta.
Tomando como punto de partida lo arriba expuesto, decidi acercarme a
la traducción inglesa de El Mundo de Sofia para ver si se trataba de una
tendencia etnocéntrica por parte de los angloparlantes o si las elecciones de la
traductora podian basarse en otro fundamento, por ejemplo un acercamiento
funcionalista a la traducción. Como punto de comparación elegi la versión
española de la novela.
La novela fue publicada por el escritor noruego, Josten Gaarder, en
1991. Gaarder ha sido profesor de filosofia, pero en Noruega es conocido sobre
todo por sus novelas para adolescentes. En El mundo de Sofia, que tiene como
subtitulo: Novela sobre la historia de la filosofia, Gaarder combina bien sus
dos profesiones, ya que el libro es una novela de ficción para adolescentes, y a
la vez una introducción a la historia de la filosofia para el mismo público. Asi
por lo menos se ha de entender la intención original del texto, aunque el libro
llegó a ser un best-seller y ha llegado a fascinar a un publico que va mâs allá de
los adolescentes noruegos. En poco tiempo llegó a alcanzar gran éxito también
entre un publico adulto y en pocos años la editorial vendió los derechos de
traducción a nada menos que 43 lenguas. No es nada raro que fuera un escritor
noruego, y ademâs profesor de filosofia, quien haya tenido la idea de escribir
una introducción a la historia de la filosofia para adolescentes, dado que en
Noruega la filosofia no entra en el curriculum ni de la escuela bâsica ni de los
institutos, impartiéndose solamente como asignatura optativa en algunos
institutos o escuelas populares. Sólo es en la universidad, donde la filosofia es
una asignatura obligatoria.
En El Mundo de Sofia conocemos a Sofia, una chica de 15 años, que un
dia empieza a recibir cartas del filósofo Alberto Knox, quien, a lo largo de la
novela, va introduciendo a Sofia a la historia de la filosofia mediante cartas,
videos, encuentros nocturnos, etc. A mitad del libro se introduce a otra chica,
320 Åse Johnsen
Hilde, y a su padre Albert Knag, que parecen ser personajes de fictión dentro
de la realidad ficticia del libro. A medida que avanza la historia, resulta, sin
embargo, que la "realidad" es el mundo de Hilde y su padre, y la fiction el
mundo de Sofia y Alberto Knox. Resulta, ademâs, ser el padre de Hilde quien,
mediante el mundo de Sofia, quiere iniciar a Hilde en los grandes pensamientos
filosóficos, tal como se puede pensar que ha sido la intention de Gaarder:
introducir a su hijo de 15 años entonces y a los demâs adolescentes noruegos a
la historia de la filosofia.
Como novela, el libro tiene, por tanto, una estructura compleja, ya que
se juega con la "realidad "/fictión dentro de la fictión, pero la novela también
tiene una parte o función didâctica dada por el subtitulo: Novela sobre la
historia de la filosofia. Y es basândome en esa doble función del libro: la
novela como fictión y como texto didâctico, como yo me aproximo al análisis
de las traducciones de la novela.
Segün los functionalistas, la función del texto meta no necesariamente es
la misma que la función del texto original. ¿Cuál es el caso de El Mundo de
Sofial Si bien la función del texto original, es la de ensenar la historia de la
filosofia mediante un código artistico, las traducciones o los textos metas, ¿han
de tener, necesariamente, la misma función? Los jóvenes en el mundo
hispanohablante quizâ no lean el libro como una introduction a la filosofia, dado
que estudian la filosofia en el bachillerato o COU, sino mâs bien como una
novela de fictión sobre una chica noruega que vive en una situatión
sociocultural noruega. Los lectores norteamericanos, que al igual que los
noruegos, generalmente no entran en contacto con la filosofia como disciplina
escolar hasta llegar a la universidad, quizâ no lean el libro como un libro
noruego sino como una introductión a la filosofia, etc. Tal como veremos
seguidamente, el traductor, mediante su manera de traducir puede, hasta cierto
punto, decidir como el lector meta ha de leer el texto traducido, enfocando o
bien la fictión o bien la historia de la filosofia con su función didâctica.
En 1997, la traductora Gülay Kutal, que ha traducido la novela al turco,
publicó un articulo en el que critica a la traductora Paulette Möller por haber
cometido graves faltas en la traductión al inglés (Samtiden num. 2/3, 1997).
Señala, entre otras cosas, que se ha traducido una mention a la novela Victoria
de Knut Hamsun por Of Mice and Men de John Steinbeck, Henrik Wergeland
por Byron y hasta en una ocasión se ha reemplazado Noruega por Francia. El
articulo causó muchas reacciones en Noruega y cambiar a Hamsun por
Steinbeck fue caracterizado como una estupidez (Hansen, Aftenposten,
29.05.97) y casi tan criminal como asesinarlo (Kutal, Samtiden, num. 2/3,
1997). En un principio, el autor mismo se adhirió a la critica, pero en un articulo
posterior en la mencionada revista Samtiden (num. 5/6, 1997), Gaarder subraya
que los cambios que se han hecho en la versión inglesa pueden defenderse, ya
que se ha intentado mantener el mensaje filosófico existente detrâs de los
ejemplos.
El mundo de la [filo]sofia 321
Figure 3.
(l)Men den klangbunn som ligger i navn som Norge, England eller
Tyskland, er nettopp dette bunnløse dyp som kalles middelalderen. Det
svømmer mange feite fisker i dette dypet selv om vi ikke alitid klarer å fa
øye på dem. Men Snorre var et middelader-menneske. Det var Olav den
hellige også. Og Karl den store. For ikke å nevne Romeo og Julie,
Benedikt og Årolilja, Olav Åsteson eller trollene i Heddal-skogen.
(p.173)
Saint Olaf and Charlemagne, to say nothing of Romeo and Juliet, Joan
of Arc, Ivanhoe, the Pied Piper of Hamelin, (...) (p. 143)
(2)Til Norge kom ikke kristendommen før på 1000-tallet, men det ville
være en overdrivelse å si at Norge ble et kristent land etter slaget på
Stiklestad (p 173)
(2a)El cristianismo no llegó a Noruega hasta el ano 1000, pero seria una
exageración decir que toda Noruega se convirtió en pais cristiano
después de la batalla de Stiklestad. (p. 209)
(3) - Déjame añadir otra imagen. Si lees una novela, por ejemplo
Victoria de Knut Hamsun ...
- De hecho la he leido...
- ¿Conoces algo sobre el autor leyendo simplemente la novela que ha
escrito?
- Al menos puedo saber que existe un autor que la ha escrito.
- ¿Puedes saber algo más de él?
- Tiene una visión bastante romántica del amor.
- Cuando lees esta novela, que es creación de Hamsun, obtienes una
impresión de la naturaleza de Hamsun. Pero no puedes contar con
encontrar datos personales sobre el autor. Por ejemplo, ¿puedes saber
mediante la lectura de Victoria la edad que tenia el autor al escribir la
novela, dónde vivia o cuântos hijos tenia?
- Claro que no.
- Ese tipo de datos los podrás encontrar en una biografia sobre Knut
Hamsun. Solamente en una biografia, o autobiografia, sabrás mâs acerca
del autor como "persona".
- Si, asi es.
- Más o menos asi es la relación entre la obra de creación de Dios y
la Biblia. Sólo mediante la observatión de la naturaleza podemos
reconocer que hay un Dios. No resulta dificil ver que ama las flores y los
animales, si no, no los hubiera creado. Pero sólo en la Biblia
encontramos informatión sobre la persona de Dios, es decir, en su
"autobiografia".
- iQué ejemplo más bueno!
Referencias bibliográficas
GAARDER, J. 1991. Sofies verden. Roman om filosofiens historie. Oslo: Aschehoug.
GAARDER, J. 1993. Sofies Welt. Roman über die Geschichte der Philosophie. Traducción al
alemån por Gabriele Haefs. München: Carl Hanser Verlag.
GAARDER, J. 1993. Sofies Värld. Roman om filosofins historia. Traducción al sueco por
Mona Erikson. Stockholm: Rabén & Sjögren.
GAARDER, J. 1995. Sophie's world. A Novel about the History of Philosophy. Traducción al
inglés por Paulette Møller. (Published in arrangement with Farrar, Strauss & Giroux Inc.)
London: Phoenix.
GAARDER, J. 1995. 'alam sofi (1996) Traducción al arabe por Haiat Al Houaiek Aatia.
Stockholm: Dar Al-Muna.
HOUSE, J. 1977. "A Model for Assessing Translation Quality". Meta 22(2), 103-109.
REISS, K. 1981. "Understanding a text from the translator's point of view". The Bible
translator 32 (1), 124-134.
Anne Martin
University of Granada, Spain
Zusammenfassung
In dieser Studie werden die Mechanismen der spanischen Presse, mit denen sie sich auf
politische, wirtschaftliche und soziale Institutionen englischsprachiger Länder bezieht,
analysiert. Die Untersuchung basiert auf einem Corpus verschiedener Textabschnitte, die
unter vier Gesichtspunkten analysiert wurden: von der Textangabe ausgehend, nach Themen
organisiert, unter dem Aspekt der Übersetzunsmethoden und schließlich nach Zeitung und
Autor. Das Ergebnis der Analyse zeigt, daß eine große Tendenz zur wortwörtlichen
Übersetzung besteht, sowie zu Anleihen, obwohl innerhalb dieser allgemeinen Tendenz doch
interessante Unterschiede zwischen den verschiedenen Zeitungen auftreten.
Résumé
Cette étude analyse les mécanismes déployés par la presse écrite espagnole pour se référer
aux institutions politiques, économiques et sociales des pays anglophones. Elle s'appuie sur
un corpus d'extraits de textes qui ont éte analysés en quatre temps: référence originale,
thèmes majeurs, procédés de traduction et, finalement, quotidien et auteur. Les résultats
montrent une forte tendance à la traduction littérale ou à l'emploi d'emprunts, même si, au
sein de cette tendance générale, on constate des différences intéressantes entre les journaux
analysés.
Resumen
Este trabajo analiza los mecanismos desplegados por la prensa escrita española a la hora de
referirse a las instituciones politicas, económicas y sociales de los paises de habla inglesa. Se
basa en un corpus de segmentes de texto que se analizaron en cuatro fases: por referencia
original, por temas afines, por procedimientos de traducción y finalmente, por periódico y
autor. Los resultados demuestran una tendencia mayoritaria hacia la traducción literal asi
como los préstamos, aunque dentro de esta tendencia general, se constatan interesantes
diferencias entre los periódicos analizados.
328 Anne Martin
1. Introduction
This paper reports on a research project which analyses the translation strategies
used by the main Spanish daily newspapers when referring to the institutions of the
English-speaking world. The aim of the project was to describe the strategies used
and establish whether any pattern in that use emerged, and if so, to examine any
possible underlying causes.
Our initial interest in this subject was a result of the impression that the
Spanish reader with no knowledge of the English language and customs of at least
the U.S and U.K. would be at a loss to understand many of the issues covered by
the daily press regarding the English-speaking world. We were surprised by the high
number of translations that did not seem to reflect the source culture adequately or
in which the original English term was most obvious, to the detriment of clarity in
Spanish. This trend seemed especially obvious when dealing with those aspects of
British life which have or are perceived to have a specific cultural or semiotic
weight, and presumed to have no target-culture equivalent. Our hypothesis was
therefore that these references were not usually translated adequately. In theory, we
felt that combinatory solutions such as couplets, triplets and functional formulations
were the stategies most suited to informative journalism (Martinez Albertos
1989:69), but our impression was that they were not widely used in the press and
that this was detrimental to clarity.
We thus sought to verify this hypothesis through a corpus-based study
compiled in April 1992 from the three main Spanish dailies at that time: El Pais,
ABC and Diario 16. The text segments registered in the corpus are mainly from
articles by the foreign correspondents of these three newspapers. The final corpus
has a total of 756 entries, each containing what we have defined as an institutional
reference, that is a reference to an institution related to the social, political,
economic or sporting life of a country (Dictionnario de la Real Academia Española,
DRAE 1992:175).
The analysis was carried out in the framework of the functionalist theory of
translation (Reiss & Vermeer 1984, 1996; Nord 1991), viewing translation as an
offer of information in one language about information that has been offered in
another, and the sociological approach to journalism (Martinez Albertos 1989,
1991) of the Spanish Escuela Complutense.
Once the corpus was complete, it was analysed from four different angles. The first
phase of analysis involved completing all the information concerning each entry in
the corpus, and consequently we felt it was necessary to find the original reference
designated by each of our entries. This was a prerequisite if we were to analyse the
Institutional references 329
treatment each had received in transmission into Spanish. In many cases, the original
reference was of course obvious. However, in others, it was difficult and sometimes
even impossible to determine, as with the case of Comité de Revisión Civil {Diario
16). (Corpus examples are given in bold; underlining indicates italics in the original.)
This stage enabled us to establish which original references had received more than
one translation in Spanish, and this in turn enabled us to compare the different
versions.
The second stage of analysis examined the corpus from the standpoint of
translation strategies or options. This phase responded directly to the objective of
this project, which was to systematise the description of the different methods used
to transmit these references in Spanish. We must point out that in this phase, and
indeed in general, we adopted a product-based, descriptive approach, similar to that
of Franco Aixelá (1995), Mason (1994a, 1994b, 1995) and Mayoral & Muñoz
(1996). Thirdly, the corpus was analysed on the basis of subject area, that is to say
grouping together the terms or expressions whose originals came from the same
subject area, as is common in literature on the subject of translation procedures, for
example all the political parties, all the names of government departments
(Newmark 1981:73, 75-77; 1988:99-100; Santoyo 1987:48,49; Álvarez Calleja
1991:226). Finally, the corpus was analysed according to the origin of each
reference, that is to say by newspaper and by author.
Whilst the first two stages of analysis - by original reference and translation
procedure - aimed at establishing how the references had been dealt with and
whether there was any set pattern in the treatment they had received, the last two
stages of analysis - by subject area and by newspaper and author - were aimed at
revealing possible motivations behind this treatment.
We found evidence of nine general procedures in the corpus. These are not
to be considered as water-tight compartments, and there are many cases of overlap
between the categories. But they do represent general trends.
3. Results
In our corpus we found many loans of all kinds. It was inevitable that we should find
proper nouns and in fact they were most numerous, ranging from the names of
newspapers, TV and radio networks, museums, galleries to football teams:
el "Sunday Times" (Diario 16), el The New York Times (El Pais), la Music
Televisión (El Pais), el San Francisco Museum of Modern Art (ABC), los Chicago
330 Anne Martin
We also found many words and expressions simply left in the original:
el "New York Stock Exchange" (ABC), Oxbridge (ABC), constituency (Diario 16).
In this category, we included all the solutions with two or more component parts,
whether one of the components was in brackets or not. Many different combinations
of the component parts were found, for example, original expression and initials in
brackets, loan plus translation, translation plus functional explanation. The common
denominator in this category was that in all of the combinations, the transparency of
at least one of their components permitted the identification of the original English
reference:
This category includes all those translations sanctioned by use, even though they
may be loans, calques etc. In fact, the majority were literal translations:
Departamento de Estado (El Pais, Diario 16), el Partido Conservador (El Pais,
Diario 16, ABC), la Cámara de los Comunes (El Pais, Diario 16, ABC).
This category refers to neutral versions, that is to say, solutions that explain the
function of the English original, without necessarily mentioning the original name of
the reference through a loan or literal translation, for example:
sindicato de actores del Reino Unido (El País), sindicato de futbolistas (El Pais), el
Institutional references 331
The only combinatory solutions not included in category 3 above. The following are
typical examples of this category:
"el periódico The New York Times" (El Pais), "la cadena CNN" (El Pais).
We included this category because we felt that there were various non culture-
specific institutional references that already had their set translation, such as alcalde
for Lord Mayor, reina for queen, senado for senate, and which needed to be
accounted for in some way. There is a difference between established translations
and this category since the former tends to be used for references which exist
exclusively in the culture of origin but which are widely known in the target culture,
for example Partido Demócrata, Partido Laborista, whereas in this category we
included universal references that are not exclusive to the source culture:
concejal (El Pais), bolsa (El Pais, Diario 16, ABC), senador (El Pais, Diario 16).
We included this category because of the large number of entries, which are
probably very common in the press, made up of two parts: one generic part which is
translated, together with a proper noun, for example:
We wished to distinguish this procedure from those entries in which the entire
expression was left in the original, which occurred in many cases.
la federación inglesa de fútbol (El Pais), parlamento autónomo (ABC, Diario 16),
socialistas británicos (ABC).
4. Discussion of results
We feel that two clear tendencies emerge from the corpus: solutions that are mainly
explanatory, and those that favour identification of the original over explanation of
its meaning or function. Having said that, we do not wish to imply that those
solutions which tend towards identification do not communicate functional
information, or vice versa. In any case, the corpus bears witness to the fact that, on
numerous occasions, combinations of these two purposes appear to be present. As
we have stated above, we felt that those solutions that gave maximum information
in a discreet way would be the most suited to the aims of informative journalism
(Martinez Albertos 1989:69; 1991:1002). Therefore, these would be the
combinatory solutions. We also felt that functional formulations would in principle
be more suitable than literal procedures.
As we had anticipated, this theoretical ideal was not entirely borne out by
the results of our analysis. The two most common options found in our corpus are
those that, at least in principle, give precedence to identification: literal translation
(21.56%) and loans (25.39%). This trend is also reflected in the combination
solutions, where literal translations and loans make up the majority of the
components.
We even noted some cases in which certain literal translations which actually
have a different meaning in Spanish have become so firmly entrenched with use that
they can no longer be expressed in any other way and the two meanings coexist. An
obvious example of this is Secretario de Estado norteamericano, the firmly
established way of referring in Spanish to the U.S. Secretary of State, even though
Secretario de Estado means something quite different in the Spanish government
hierarchy. This example also shows us the importance of the press in disseminating
certain usages. On the other hand, the more communicative solutions were not so
frequent, especially those that did not include a specific reference to the original
term or expression.
As far as the distribution of these different translation solutions is concerned,
there was a tendency to use the least explanatory solutions in the economics sections
of the three dailies, and this tendency was even more marked in the sports pages. It
almost seemed as though this was not a question of non-specialist information but
rather the transmission of specialized information to an expert reader. Curiously
enough, this trend was also apparent in the cultural sections of ABC and Diario 16.
Obviously, the classification as we have described it so far does not take into
account the communicative value in context of the solutions adopted. Solutions that
are included in the same category may be suited to the communicative needs of one
Institutional references 333
contextual situation and not to another. We therefore felt it was necessary to carry
out complementary analyses in order to take the context into account. Such analyses
included grouping together all those references that had received various different
translations within one article. By grouping together these sequences, it became
apparent that often different translations had been used for questions of clarity, or as
a style-varying device. Some translations which, seen in isolation, appeared to be
rather obscure, became clear when seen as part of a sequence. Often a full
explanation had been offered on the first appearance in the same text and the
variation was simply due to stylistic reasons. This phenomenon was most frequent in
El Pais,
We shall now go on to discuss the following phase of analysis, that is, analysis by
newspaper and by author. The results of this stage show that, like the general
corpus, the most common strategies used in each individual newspaper are literal
translation and loans. However, despite this similarity of the results in general terms,
there are certain disparities between the different newspapers, which perhaps
constitute the most interesting aspects in the search for some kind of motivation
behind the use of one option or another. Functional and explanatory solutions in El
Pais total 22.12%) of its total corpus, whereas they only account for 14.56%) of
ABCs total corpus. At the same time, the figures for literal procedures are slightly
lower in El Pais than in the other two publications. Diario 16 seems to follow a
middle course between the other two dailies, except in the case of partial
translations, which is a solution that it hardly used at all, and the non-classifiable
entries, the vast majority of which are to be found in Diario 16.
On closer examination, the values of the different solutions within each
category used by each paper become apparent. Many of the loans in El Pais are
names of sporting teams or initials, the meaning of which had already been explained
in a couplet or triplet earlier in the text. By contrast, in ABC, the loans registered are
of an entirely different nature: names of buildings, clubs or cultural centres, the
generic part of which was left in the original, for example The National Gallery of
Art, San Francisco Museum of Modern Art, el Metropolitan Correctional Center
or institutions like Eton and Oxbridge, transcribed without any form of explanation,
as are institutions whose names could very easily have been translated into Spanish,
such as "Hong Kong Trade Development Council" or "New York Stock
Exchange". In Diario 16 the patterns are not as clear cut. Generally speaking the
majority of the loans are metonymic, although one correspondent consistently
transcribes original words and terms which he perhaps perceives as being culturally
unique; this is not in fact the case, as an adequate translation could have been found
for many of these terms, which include constituency, devolution and hung
334 Anne Martin
parliament.
The use of unusual terms in the target language can create the totally false
impression of a difference between cultures that in fact does not exist. The use of
such source-culture lexis hides the fact that the concept itself does in fact exist in the
target culture, attributing a semiotic status to the source term which it actually does
not have. This seems to be particularly the case with British political institutions with
archaic names but similar functioning to their European counterparts.
This difference of value also exists in the category of literal procedures.
Whilst in El Pais, the literal translations tend to be fairly clear, transparent and
connotation-free, 'in Diario 16 and ABC we observe a mixture of clear and adequate
literal translations with those that were simply incorrect or nonsensical. In ABC for
example, we found the following:
In Diario 16, we found Ministro de Deportes, creating the idea in Spanish that there
is a Secretary of State for Sports in the U.K., and la Oficina Galesa, creating the
idea that there is no ministry in charge of Welsh affairs.
This trend is reinforced when we examine the type of loan and literal
translation that formed part of the combinatory solutions. In El Pais, the majority of
these are composed of original initials together with a functional or literal translation
of the meaning of those initials, e.g. Servicio Nacional de Salud (NHS). By contrast,
in ABC most of the combinations consist of literal translations plus explanation.
Some of the explanations offered are rather curious, for example: Carta de los
Ciudadanos, carta de naturaleza del "majorismo " o farol sin consecuencias -
según vaya todo, offering very little real information. Having said this we must point
out that such solutions cannot be considered as a majority trend. In the couplets and
explanations found in Diario 16, the same kind of loans mentioned above were
apparent {hung parliament, constituency, devolution), although on this occasion
they were accompanied by an explanation.
It is interesting to consider the reasons why these differences should arise. In
the case of El Pais, we feel that the results are due in part to the existence of a style
book (1990) that stipulates clear guidelines with respect to any explanations to be
offered, for example regarding the use of initials and foreign words or expressions.
There is also a clear distinction made between the different genres ofjournalism. It is
curious that failure to comply with the style book regulations is notorious in El Pais,
and is even alluded to and denounced by the paper's own journalists themselves
(Arias in El Pais, 25/6/95). However, our results show a clear difference in so far as
consistency and harmonization of style is concerned if we compare El Pais to the
other two newspapers studied. Furthermore, the number of journalists signing
Institutional references 335
articles from which the entries in our corpus were extracted in El Pais is higher than
in the other two dailies, a situation which could have led to greater fragmentation of
criteria in El Pais. Such fragmentation is present in Diario 16. Diario 16 supposedly
has a twenty-page internal style book (Fernández de Beaumont 1988:391) which we
were unable to obtain and the existence of which was even denied on several
occasions by Diario 16 journalists. We were therefore not at all surprised that this
newspaper proved to have the widest variety of solutions, including many of those
that we could not classify, in addition to a total lack of consistency in the use of
italics, inverted commas and capitals. In short, Diario 16 gives the impression of
having fewer internal style regulations together with the absence of any kind of
common philosophy about the kind of journalism the paper represents. As far as
ABC is concerned, there was no style book at the time our corpus was compiled,
although the widespread custom was to follow the guidelines established in the
Manual de Espanol Urgente published by the Agencia Efe press agency (1992,
1994).
We have mentioned the existence of a common philosophy of journalism
and we consider that part of the motivation behind the use of certain translation
solutions could be related to this aspect. This is very clearly stated in El Pais' style
book (El Pais 1990:15-21) and we feel that the results we have mentioned stem
from this clear statement of intent concerning the type of journalism to be pursued
by El Pais. We are of course referring to the higher number of functional (that is to
say, communicative) solutions, the objective nature of explanations, the greater
consistency of solutions regardless of the author of the article, consistency in the use
of markers, greater number of sources, together with other aspects such as the clear
difference between the different genres (information, opinion and analysis) which is
fundamental if a newspaper is to project any kind of united common philosophy.
In ABC this does not occur and it would seem that this publication pursues a
different kind ofjournalism which is more personal and individualistic in nature, with
more opinion and less information. The kind of journalism we see in ABC seems
directly targeted at a previously identified reader with whom there is a high degree
of complicity, rather than the hypothetical general reader specified by text books on
journalism (Martinez Hernando 1990:47).
Moreover, the ideological function of the journalism prevalent in ABC is
more noticeable than in the other two publications, a fact which has been pointed
out by other authors (Martinez Hernando 1990:170-1). In this sense, there would
seem to be a certain hostility towards the U.K. in ABC in general, which is shown by
drawing attention in a sarcastic tone to minority and possibly anachronistic aspects
of British life and traditions. Some of the translation solutions found in ABC would
seem to aim at ridiculing the source culture rather than offering information about
elements of that culture. This is the case with the biased nature and total lack of
functionality of some of the explanations to be found in ABC, for example:
336 Anne Martin
In these cases it almost seems as though the solutions given in ABC seek to
perpetuate an image of British society as being mildly eccentric and largely
anachronistic, instead of explaining the function of the references in objective terms.
Infrequent expressions are used to designate phenomena that are perfectly familiar
to the Spanish speaking reader, such as entrismo, Canciller del Ducado de
Lancaster. In this way, cultures are falsely depicted as being widely different, when
this is simply not the case. In Spain too, political parties are infiltrated by the left and
there are ministers without a specific portfolio.
6. Conclusions
To summarize our conclusions we could say that the most frequent trends noted in
our corpus were loans and literal procedures, whilst the least frequently used were
functional solutions and cultural equivalents, although the proportions differed
according to the newspaper in question, which was in line with the hypothesis that
we set out to test. Contrary to our hypothesis however, this did not generally affect
the clarity of the Spanish version. In most cases the solutions given clearly
transmitted at least one part of the original meaning, although the few cases of lack
of clarity detected were almost always due to the use of literal procedures. In this
sense, our original impression proved to be false, as the incorrect or nonsensical
versions were in a definite minority. There would seem to be a clear trend towards
the inclusion of the form of the original term through the use of literal options, and
only when these have been ruled out for whatever reason are the other options
considered. The impression is that literal translation is the norm, and when it is not
used this is due to a process of elimination.
This priority given to literal translation no doubt responds to the concept
that society in general has of the nature of translation: an exclusively linguistic
operation. This misconception may be due to the fact that most people's conscious
contact with translation is limited to foreign language classes in which translation is
used as a way of underlining the syntax of the language studied and not as a vehicle
for communication (Nord 1991:156; Newmark 1988:8; Bassnett-McGuire 1980:2).
Be that as it may, it is also clear that the high number of literal procedures and loans
Institutional references 337
reflects the pressure brought to bear on Spanish by the Anglo-American culture and
its language. Moreover, it is indicative of the knowledge of this language and culture
that the average Spanish reader is thought to possess, which would be unthinkable
for other languages and cultures.
References
AGENCIA EFE. 1994. Manual de Espanol Urgente (10th ed.). Madrid: Ediciones Cátedra.
ARIAS, J. 1995. "^Por qué traducimos tan mal los vocablos ingleses?" El Pais, 21.06.95.
FRANCO AIXELÂ, J. 1995. "Specific Cultural Items and their Translation". Translation and the
Manipulation ofDiscourse. Selected papers of the CERA Research Seminars in Translation Studies
1992-93. ed. by P. JANSEN, 109-123. Leuven: CETRA, The Leuven Research Center for
Translation, Communication and Cultures.
MAYORAL, R. & R. MUNOZ. 1996. "La traducción como transmisión de información. Problemas
y estrategias comunicativos en la traducción de un texto asmimstrativo". Lecture given at the V
Curso superior de Traducción Inglés/Espanol University of Valladolid, Spain, 18th April 1996
(photocopy).
Real Academica Espanola. 1992. Dictionnario de la lengua espanola (DRAE) 21th ed. Madrid:
Espasa Calpe.
REISS, K, & H, VERMEER. 1996. Fundamentos para una teoria funcional de la traducción.
Madrid: Akal.
SANTOYO, J.C. 1987. "La 'traducción' de los nombres propios". Problemas de la traducción.
Mesa Redonda - noviembre 1983, 91-97. Madrid: Fundación Alfonso X El Sabio.
BRITISH ACCULTURATION OF ITALIAN
THEATRE
Stefania Taviano
University of Warwick, UK
Zusammenfassung
Im Fokus dieses Artikels stehen die wesentlichen Aspekte der britischen Übersetzung
und Rezeption vom italienischen Drama. Der Transfer vom italienischen Theater in die
anglophone Kultur wird unter Berücksichtigung der britischen Haltung zur
mediterranen Kultur, insbesondere zur italienischen Kultur, analysiert. Es wird gezeigt,
wie und bis zu welchem Ausmass der Respons zum italienischen Drama durch die
Gegensätzlichkeit britischer und italienischer Kultur und durch den Diskurs des Südens
beeinflusst wird. Dieses Herangehen an italienische Kultur kommt in zwei offensichtlich
divergierenden Herangehensweisen mit einer Prädominanz in der Zielkultur zum
Ausdruck. Diese werden hier analysiert.
Résumé
Nous aborderons certains aspects de la traduction et de la réception en Grande-
Bretagne du théatre italien. L'analyse porte sur les attitudes britanniques envers les
cultures méditerranéennes, en particulier la culture italienne. Nous démontrerons
comment et jusqu'où la réponse au théatre italien est marquée par l'opposition entre les
deux cultures en présence et par un certain discours méridional. Deux approches, en
apparence divergentes, prédominent dans la culture d'arrivée.
Resumen
Este articulo se centra en los principales aspectos de la traducción británica y la
recepción de las obras teatrales italianas. Se analiza la transferencia del teatro italiano a
la cultura anglosajona teniendo en cuenta la actitud britânica hacia las culturas
mediterráneas, con especial énfasis en la cultura italiana. Se demuestra cómo y en qué
medida se ve afectado el teatro italiano por los contrastes entre la cultura britânica e
italiana y por el discurso del sur. Este enfoque de la cultura italiana se hace patente en
dos enfoques aparentemente divergentes de la cultura término que se analizarân en este
articulo.
340 Stefania Taviano
1. Introduction
This article analyses some of the difficulties inherent in British translations and
productions of Italian plays. I argue that these obstacles are not so much related
to the exotic character of Italian theatre itself, but that they are, on the contrary,
generated by the British attitude to Mediterranean cultures, in particular Italian
culture. Starting from the notion of faithfulness to the original, I show how the
reception of Italian theatre is influenced by an opposition between British and
Italian culture and by what has been called a discourse of the south. This
attitude to Italian culture reveals itself in the two diverging strategies adopted in
British productions of Italian drama: one adheres to a stereotypical image of
Italian culture, which involves reducing Italian characters to caricatures; and the
other consists of transferring Italian plays to a British milieu to ensure the
domestication of those aspects of Italian drama which are deemed too exotic to
be accepted by target spectators.
2. Faithfulness
Patrice Pavis points out that the notion of faithfulness related to the mise en
scène, predominant in critical discourse, "is pointless". He observes that the
British acculturation 341
culture, his themes and characters so local "that it seems almost impossible to
imagine how such a localised folk theatre could ever be transplanted into
another language, ever be performed by actors of another nationality and
background" (Esslin 1973: 41). Esslin's comments can be specifically applied to
the UK since his concern seems to reflect the British belief in a profound
unlikeness between the two cultures. This view about the nature of the relation
between Italy and the UK belongs to what has been defined as the discourse of
the south, to which I will come back later in this paper.
An anonymous critic in the Times Literary Supplement (1925), writing
on the English translation of Pirandello's plays, identifies the opposition
between British and Italian culture as a more universal condition, whereby the
exoticism of Italian theatre will often need to be erased, no matter what the
target cultural context is. Furthermore, the fact that we read the same kind of
lost-in-translation comments separated by a period of fifty years — the TLS
critic writes in 1925, while Martin Esslin writes in 1973 — and that they have
been applied to two very different playwrights, seems to confirm the existence
of a general British cultural approach affecting the reception of Italian drama in
the UK. While Esslin's concern about the Neapolitan nature of Eduardo De
Filippo's work, about its strong links to Naples society and culture and the
subsequent difficulties in translating it into a foreign language can be up to a
point understood, on the other hand, Luigi Pirandello's theatre and his
reflections about the human condition have a more universal value, even if his
work reflects to a certain extent, as for every writer, his native cultural milieu
— Sicily. In other words, despite the uniqueness of each playwright's work and
the differences in their approach to drama, the overall British attitude towards
Italian dramatists is rendered uniform by the above-mentioned cultural frame
through which Italian cultural products are perceived.
But despite the emphasis given to the foreign identity of Italian theatre
as opposed to British theatre, despite the difficulties of presenting Italian plays
to British audiences, as seen in the reviews mentioned, productions staged in the
UK by Italian companies in the original language have been highly successful. In
1925 Pirandello's company, the Teatro d'Arte, came to London to present four
of his plays: Sei personaggi in cerca d'autore, Enrico IV, Vestire gli ignudi,
Così è se vi pare. The productions were very successful and several reviewers
emphasised the response and involvement of the audience despite the fact that
the performances were in Italian. The enthusiasm initiated by the Teatro
dell'Arte company continued for several years, creating what was defined as
"the cult of Pirandello" {Times 1926). Equally successful were the Italian
production of Carlo Goldoni's The Servant of Two Masters directed by Giorgio
Strehler during the 1967 Peter Daubeny's World Theatre Season at the
Aldwych, and the production of Eduardo De Filippo's Napoli Milionaria by his
company during the 1972 World Theatre Season. Writing on this production,
British acculturation 343
child adoring and highly emotional", Richards adds the following joke: "They
talk as we know with their hands. When a social psychologist went round
orthopaedic wards he found those patients with arm injuries were tongue-tied:
they could not talk if they could not move their hands" (Richards 1994:xiv).
For the 1982 British production of De Filippo's Natale in Casa
Cupiello, under the English title Ducking Out, at the Greenwich Theatre, the
adaptor Mike Scott transposed the action from Naples to Lancashire, and Eric
Shorter in the Daily Telegraph makes an interesting comment: "The evening is
enjoyable in an instructive way, because it shows how different in temperament
and humour, character and culture, religion and social manners the British are
from the Italians" (Shorter 1982). Here this particular play and performance are
interpreted as just another example which proves how different these two
cultures are. The domesticating translation strategy of replacing the original
setting by one familiar to British audiences, although considered by some as a
means of avoiding the use of national stereotypes about Italians, becomes a way
of confirming the opposition between source and target culture and of
diminishing the possibilities of interaction between the two societies. The Punch
critic emphasises the consequences of such an approach:
Just as Chekhov's Three Sisters would look a bit odd in Cheltenham trying to get
to London, so does De Filippo's extended family of Neapolitans; they seem to be
auditioning for some minor Ayckbourn comedy about yet another disastrous
family Christmas and it does definitely diminish them as characters. (Ö) Their
behaviour, instead of being rooted in Italian tradition and plausibility, becomes
oddly arbitrary in Lancashire. (Punch 1982)
The opposition between Catholics and Protestants, which was one of the main
elements of the above-mentioned discourse of the south, and the subsequent
attribution of inferiority to Catholic cultures by the British, is another element
which continues to characterise the reception of Italian theatre. Andy de la
Tour's adaptation of Dario Fo's The Pope and the Witch in 1992 was not very
successful mainly because, according to numerous critics, the author's attack on
the Catholic Church was unappealing to English audiences. Reviewers felt that
British people are too "enlightened" and emancipated to be interested in a play
like The Pope and the Witch, which can only be a success in a primitive country
like Italy. It becomes evident that the discourse based on the notion of British
superiority, as opposed to the backwardness of Italy, is still predominant in
20th-century multicultural Britain. The fact that critics choose to focus primarily
on the attack against the Catholic Church, which is only one aspect of Fo's
critique of Italian society expressed in the play, leaving out more central ones
such as drug addiction, proves once more that the interpretation of Italian plays
continues to be manipulated to fit this patronising cultural frame. On the same
British acculturation 345
line, the fact that Andy de la Tour explains in a programme note that he had to
cut out many of the Italian references from the script because they would not be
understood by a British audience, can be interpreted as a "veiled apology" for
an unsuccessful production (Stanfield 1992). The insurmountable differences
between Italian and British cultures are in fact often exploited by directors and
adapters to justify their domesticating approach.
5. A process of appropriation
The relationship between British and Italian culture, or rather the British, and
also American, approach to Italian cultural values, leads on to the issue of the
relation between minority and majority cultures and the latter's tendendy to
appropriate and transform the first (Pavis 1992). This issue has been emphasised
by several critics and scholars analysing the reception of Italian theatre and of
Italian culture in general. Clive Barker condemns "the insular determination of
the British theatre to solve its own problems in its own way" (Barker 1981:28),
and Malcolm Rutherford puts it in more explicit terms by saying that "some of
the English seem to think the Italians are funny just because they are Italians"
(Rutherford 1992).
It cannot be denied that an act of interpretation is involved in any mise
en scène and equally in any translation from one language into another. As Pavis
argues, "a text does not speak on its own, it has to be made to speak. [...] It is
not possible to neutralise the stage so that the text can speak on its own, or be
heard without mediation or without distortion" (Pavis 1992:45-6). Similarly
Sirkku Aaltonen (1996:74) argues that acculturation is an inherent part of the
process of drama translation.
In the case of the British reception of Italian theatre it is more a process
of "appropriation" which seems to emerge, connected to the uneven relation
between majority and minority cultures. What Pavis observes about the relation
between European and non-European theatre emphasising the risk of
Eurocentrism can be applied to the relation between British and Italian theatre,
where there is equally a risk of Anglocentrism.
The insular tendency of British culture and the re-emergence of a
discourse of the south reveals itself in several fields and in several forms, and
the producing of Italian plays is only one of them. In some cases it is expressed
through the emphasis put on Italian accent and gesticulation. In 1925 a British
production of And That's The Truth by Luigi Pirandello was staged, and the
extensive use of clichés is emphasised in the following review: "the English
actors were evidently determined to be more Italian than Italians. [...] Every
gesture, every movement, was so exaggerated as to convey the idea that Italians
are all born film actors, perpetually posing for "close-ups" {Times 1925). The
346 Stefania Taviano
I found it regrettable that in an attempt to present 'sicilianità' the director had his
company speak in foreign accents, thus maintaining an offensive tradition
developing in the English stage presentation of Italians. This gelato/spaghetti
English, a reflection of the British view of the Italian urban proletariat uneasily
absorbed into its culture, is simply not the best way to present the peasants of
Agrigento. (Lorch 1983:101)
There are numerous productions of Italian plays which have generated this kind
of critique, testifying to a widely adopted stereotypical presentation of Italian
culture to British audiences, and to a common tendency to reduce Italian
characters to caricatures. But throughout the history of English language
productions of Italian plays, the opposite kind of approach has also emerged,
that is to say the tendency to diminish those foreign aspects of Italian culture
which are not easily acceptable by British audiences by transposing Italian plays
to a British context. As we have seen, this is the case of Peter Tinniswood's
production of Napoli Milionaria, but there are several more examples of the
adoption of this strategy.
In 1985 Luigi Pirandello's Six Characters was staged in Bristol, directed
by George Brandt, who explained: "I played down the specifically Italian aspect
of Six Characters without making it aggressively English" (Brandt 1987:5), and
he added that he did so: "to denationalise the play a little". In 1987 another
production of the same play was presented at the National Theatre, this time
based on the adaptation by Nicholas Wright, who does not know Italian. Giulia
British acculturation 347
Ajmone Arsan, commenting on the production, writes that "the play has been
anglicised", and concludes that this interpretation is "patronising towards the
public" (Arsan 1987:35). Katharine Worth (1987:36), in a review of the same
performance at the National Theatre, commented on the "homey context" of the
play, which opened it up for "popular consumption in the English theatre".
The anglicising strategy of reducing a "disturbing phenomenon" to a safe
and familiar event is a different but equally patronising result of the discourse of
the south. Given the cultural superiority and authority of the target culture,
British interpreters and adaptors of Italian texts can either choose to offer a
stereotypical representation of the source culture or minimise, and even cut,
those parts which are simply too unusual and exotic from a British perspective
to be culturally accepted.
In Nicholas Wright's adaptation, the actors rehearse Hamlet instead of
The Rules of the Game, since the latter is not very well-known in Britain. As
was explained in the programme, the choice was made 'from the familiar classic
repertoire', therefore the disorientating effect caused by the use of Pirandello's
own play was lost. The audience was presented instead with a familiar play and
was not made uncomfortable as happens in the source text. The play undergoes
further changes in this adaptation; above all the conflict between reality and
fiction, on which Six Characters is based, is essentially ignored and Nicholas
Wright comments: "We decided not to break down the barrier between the
public and the stage" (Arsan 1987), without justifying his choice. The
conclusion is also different: instead of the daughter who runs away laughing,
which was considered too dramatic an ending, this adaptation concludes with
the Director sitting at his desk and simply asking for more light. All these
alterations make the play more accessible and acceptable to British audiences as
they eliminate typically Italian aspects, considered too disturbing for the British
public, as emerges from Giulia Ajmone Arsan's following comment:
Having failed to grasp the meaning of such a conclusion, Nicholas Wright just
thought it was excessively dramatic, echoing those who opposed the first London
production of the play because it was judged to be too upsetting. Indeed
throughout the play the intention to tone down the exuberance of Latin passion is
evident. Isn't it much more English, discreet, restrained, to conclude the play
with the Director asking for a light while quietly sitting at his desk? (Worth
1987)
6. Conclusion
The two diverging approaches to Italian theatre analysed in this article cannot
be considered simply as the result of specific translation strategies, of choices
made by one particular translator or director, but they need to be put into
context and analysed in relation to a more general British attitude to the
Otherness of Italian culture, based, as we have seen, on the discourse of the
south originating in the late eighteenth century. What is important to understand
is that this approach is not confined to the theatre - though it cannot be denied
that it can become more explicit through mise en scène and the critical discourse
of reviewers - but emerges in other fields such as travel writing and also in the
translation and reception of Italian literature in general. Zigmunt Barânski, in an
article about the diffusion of Italian literature in the UK, reveals the
predominance in British culture of the image of Italy as a "pre-industrial country
and idyllic place" (Barânski 1993:260 translation by S.T.) and identifies it as the
main reason for the limited distribution and success of Italian literature in the
Anglophone world. This aspect of the British representation of Italian culture
cannot be separated from the idealisation of the same culture from a British
point of view, as confirmed also by John Julius Norwich's book The Italian
World (1983).
It becomes clear, then, that the British reception of Italian theatre
represents a fundamental field of investigation in the analysis of the relation
between these two cultures, but it also may be instrumental in the identification
of contemporary aspects of a British discourse of Otherness affecting the
reception of a number of Mediterranean cultures.
References
AALTONEN, Sirkku. 1996. Acculturation of the Other. Irish Milieux in Finnish Drama
Translation. Joensuu: Joensuu University Press.
ARCARA, Stefania. 1998. Constructing the South: Sicily, Southern Italy and the
Mediterranean in British Culture, 1773-1926. Doctoral thesis in British Cultural Studies,
University of Warwick.
British acculturation 349
ARSAN, Giulia Ajmone. 1987. "Six Characters at the National Theatre". The Yearbook of
the British Pirandello Society 7, 32-35.
BARÂNSKI, Zigmunt. 1993. "La diffusione della letteratura italiana contemporanea in Gran
Bretagna". The Italianist 13, 255-265.
BARKER, Clive. 1981. "Right You Are (If You Could Only Think So)". The Yearbook of the
British Pirandello Society 1, 26-33.
BASSNETT, Susan. 1998. "The Translation Turn in Cultural Studies'. Constructing Cultures
ed. by in S. BASSNETT & A. LEFEVERE, 123-139. Clevedon: Multilingual Matters.
BRANDT, George W. 1987. "Six Characters in Bristol". The Yearbook of the British
Pirandello Society 7, 1-10.
COVENEY, Michael. 1992. "The Pope and the Witch". Observer 19 April.
ESSLIN, Martin. December 1973. "Saturday, Sunday, Monday". Plays and Players, 41-42.
LORCH, Jennifer. 1983. "Liolà". The Yearbook of the British Pirandello Society 3, 100-101.
NORWICH, John Julius. (ed). 1983. The Italian World. London: Thames & Hudson.
PAVIS, Patrice, 1992. Theatre at the Crossroads of Culture. London & New York:
Routledge.
RUTHERFORD, Malcolm. 1992. "The Pope and the Witch". Financial Times 15 April
STANFIELD, Keith. 1992. "The Pope and the Witch". City Limits 30 April.
TAVIANO, Stefania. Forthcoming. "The Reception of Dario Fo and Franca Rame in the
UK". In Justice and Morality - Visions of change in European Theatre, ed. by Edward M.
BATLEY & David BRADBY, University of London.
350 Stefania Taviano
TAVIANO, Stefania & LORCH, Jennifer. Forthcoming " Producing Pirandello in England",
VENUTI, Lawrence. 1995. The Translator's Invisibility. London & New York: Routledge.
WARD, William. 1990. Getting it Right in Italy. A Manual for the 1990s. London:
Bloomsbury.
WORTH, Katherine. 1987. "Six Characters at the National Theatre (2)". The Yearbook of the
British Pirandello Society 7, 36-43.
PART VIII
Translation history
TOWARDS A HISTORY OF TRANSLATION IN A
(POST)-COLONIAL CONTEXT: AN AFRICAN
PERSPECTIVE
Paul Bandia
Concordia University, Montreal, Canada
Zusammenfassung
Der Verfasser analysiert das Übersetzen zur Kolonialzeit in Afrika und besonderes die
Tradition der mündlichen Literatur. Im Anschluss daran erörtert der Verfasser die
Ambiguität der Diskurse über die Literatur nach der Kolonialisierung: zum einen
erzeugten diese Diskurse die Hierarchie zwischen dem Zentrum (Kolonisator) und der
Peripherie, zum anderen Hessen sie das aus dem Kontakt der Sprachen und Kulturen
stammende Kreolische zu. Bei dem letzteren Standpunkt, mit einer Sprache, die
zwischen zwei Sprachen oder als dritter Kode fungiert, wird das Übersetzen selbst
gleichzeitig Quelle und Ergebnis der Mischsprache - die Frage über die
Ausgangstexte wird auf eine andere Weise gestellt.
Résumé
L'auteur s'attache à considérer le rôle de la traduction en Afrique, au temps de la
colonisation - notamment envers la tradition littéraire orale. Il s'interroge sur les
ambiguités des discours sur la littérature, après la colonisation: en effet, ces discours
tantôt reproduisaient la hiérarchie entre centre (colonisateur) et périphérie, tantôt ils
assumaient la créolisation issue du contact des langues et des cultures. Dans cette
dernière perspective, avec une langue constamment «entre deux» ou «troisième code»,
la traduction elle-même devient à la fois source et résultat du métissage - posant
autrement, entre autres, la question des textes de départ.
Resumen
Este articulo analiza la historia de la traducción desde una perspectiva postcolonialista,
basada en una evaluación critica de algunas prácticas de traducción. Evalua el impacto
del hibridismo lingüistico sobre la teoria de la traduccion tradicional, seãlando las
limitaciones de ésta y haciendo hincapié en la necesidad de explorar una dimensión
etnogrâfica en la búsqueda de una teoria de la traducción mâs global.
354 Paul Bandia
1. Introduction
2. Colonial translations
efforts were also motivated by a more amateur humanist concern for an African
cultural essence endangered by colonialism itself' (Austen 1990:31). Hence, as
noted by Equilbecq, "These traditions are the supreme vestiges of the primitive
beliefs of the black race and, on this basis, deserve to be saved from oblivion"
(1972:22). There is therefore an unbalanced power relationship which in such
an imperialist enterprise as colonialism is bound to dictate the criteria and
parameters guiding the translation process. As a consequence, the authenticity
of 'translated' or 'recorded' oral texts needs to be clarified.
The authenticity of African oral texts produced by pre-tape-recorder
scholars has come under serious attack as these texts had to be 'recorded'
during performances of African oral works. However, some colonial
administrators hardly made the effort to attend such performances, and even
when they did it is doubtful that they understood. Those informants who were
recognized as performers within their own culture seem to have been the
Africans most conveniently available to administrators, particularly African
subordinates of the colonial regime itself. As pointed out by Austen: "Even
when the immediate narrators spoke in their own languages, the translators
were inevitably government interpreters, whose own imperfect European idiom
was then further edited by the administrator/author of the ultimately published
text" (1990:31). One can only imagine the influence of colonialist inscription
on these collected narratives and their obvious philological shortcomings. The
context of performance of the narratives is hardly known and there is no
indigenous text to control the translation.
The attitude of missionaries towards African oral narratives was
significantly different from that of administrators in many ways. For one thing,
the missionaries had no interest in preserving indigenous culture which, by and
large, embodied the very values they sought to eradicate in their bid to convert
the natives to Christianity. However, the missionaries eventually realized that
the inculcation of Christianity and other related secular western values into
African society required mastery of local languages and a good grasp of those
elements of indigenous belief and practice which could help establish the
foundation for spreading Christianity in Africa.
Therefore, missionaries, unlike administrators, recorded oral literature
texts under more reliable conditions, partly due to their commitment to learn
local languages for purposes of evangelization (particularly Bible translation).
The missionaries produced bilingual texts, translated by themselves and
derived from skilled performers. They even produced texts exclusively in local
languages. Vernacular translations (in local languages) were thus available, and
moreover, mission catechists and school teachers were often Africans who
participated in the translation of these religious texts.
However, missionary colonialism also sought to shape the literature
received by the Africans by constructing texts compatible with Christian
conceptions and omitting facts which conflicted with Christian values (Austen
356 Paul Bandia
3. Post-colonial translations
confidence. The colonial subject was forced to learn the colonizer's language
in which he or she was initially trapped and constrained. Postcolonial writers
found themselves languageless, gagged by the imposition of European
languages on their world (Barber 1996:6).
However, the postcolonial writer made a comeback by radically re-
appropriating the colonial language culture. In the words of Barber (1996), "the
periphery now takes on the culture and language of the center and transforms
it, breaking it, infusing it with local registers and refashioning it so that it
speaks with the voice of the marginalized" (6). Instead of one hegemonic
English, for instance, we get a plurality of local Englishes. Marginality has thus
become an unprecedented source of creative energy. And the literature of the
margins is thus at the frontiers of postmodern aesthetics (Tiffin : x). In
postcolonial criticism of African literature, the colonial subject is thus
portrayed as having finally found a 'voice'; the colonized has mastered and
subverted the colonial codes as a means of asserting his or her identity. If the
African writer has to write in a European language, he or she has to transform
the language to make it bear the burden of African experience.
In The African Palimpsest (1991), Zabus makes an extensive and
sustained attempt to bring postcolonial analysis to bear specifically on African
texts. She carries out a detailed analysis of the expression of indigenous
linguistic repertoires within a range of anglophone and francophone novels.
She assesses the role of linguistic hybridization in African Europhone
literature, not only as a means of providing "a distinctively African cocktail",
but also as an attempt "to resist, recreate and subvert the dominant (European)
language" (Barber 1996:8). The African writer is shown to draw on thematic
elements from oral as well as written repertoires in order to achieve the effect
of orality through specific techniques of writing. African Europhone writing
thus takes on its special Africanity, its ability to express specifically African
experience in an alien tongue. African writers crack the standard forms of
metropolitan European languages and re-form them into unique new patterns.
One has the distinct impression that expressions in indigenous languages are
made to live on through translated borrowings and echoes. According to Barber
(1996), the analysis of oral and indigenous repertoires as a pool of resources
for written Europhone texts precariously straddles the world of orature and that
of literature and bridges the two by translating the one into the other (1996:10).
An example of the impact of indigenous repertoires on African
Europhone texts can be seen in the practice of relexification. According to
Zabus, relexification is an "imaginative world-creating attempt at forging a
new literary aesthetic medium or 'third-code' out of the alien, dominant
(European) lexicon. It is also an attempt at appropriating the Other's language
in order to make it one's own and bend its otherness or fixity to artistic and
ideological exigencies" (1991:103).
Towards a history of translation 359
- Who are your people be? ... If you are coming-in people be, then come in.
(Okara 1964:26-27).
- Et vous les gens, qui pouvez-vous être? ... Si vous devez être des gens à
entrer, alors entrez. (Okara 1985:15).
- You cannot a thing I have done not put on my head; How can you on my
head put a thing that happened not? (Okara 1964:66).
- Vous ne pouvez pas une chose que point n 'ai accomplie me faire endosser;
Comment pouvez-vous me faire endosser une chose qui point ne s'est passée?
(Okara 1985:60).
4. Conclusion
References
AUSTEN, R. 1990. "Africans speak, colonialism writes: the transcription and translation of
oral literature before World War II". Cahiers de Littérature Orale, 28: 29-53.
CHEYFITZ, E. 1991. The Poetics of Imperialism: Translation and Colonization from "The
Tempest" to "Tarzan ". New York: Oxford University Press.
JAHN, J. 1972. Leo Frobenius: the Demonic Child. Austin: University of Texas Press.
LAUE, H. VON. 1976. "Anthropology and Power; R. S. Rattray among the Ashanti". African
Affairs 75: 33-54.
MEHREZ S. 1992. "Translation and the Postcolonial Experience: The Francophone North
African Text". Rethinking Translation ed. by L. Venuti, 120-138. London & New York:
Routledge
NGUGI, W. T. 1986. Decolonising the Mind: the Politics of Language in African Literature.
London: James Curry.
NIRANJANA, T. 1992. Siting Translation: History, Post Structuralism, and the Colonial
Context. Berkeley: University of California Press.
ROBINSON, D. 1996. Translation and Taboo. Dekalk: Northern Illinois University Press.
TIFFIN, H. 1991. "Introduction". Past the Last Post: Theorizing Post-Colonialism and Post-
Modernism ed. by I. Adams & H. Tiffin, vii-xvi. Hemel Hampstead: Harvester Wheatsheaf.
ZABUS, C. 1991. The African Palimpsest: Indigenization of Language in the West African
Europhone Novel. Amsterdam & Atlanta: Rodopi.
LA CRITIQUE DE LA TRADUCTION LITTÉRAIRE
EN TURQUIE (1940-1992)
Abstract
The paper surveys the history of literary translation criticism in Turkey during 1940-
1992. The policies, objectives and contributions of three representative translation
journals are analysed: Tercüme (Translation, 1940-66), "Yazko Çeviri" (Yazko
Translation Magazine, 1981-84) and Metis Çeviri (Metis Translation Magazine, 1987-
92). Early progress was followed by a stagnation phase, despite the attempts to
establish an objective and theoritically based discipline of translation criticism.
Zusammenfassung
Die Untersuchung befasst sich mit der Kritik von übersetzten literarischen Werken in
der Türkei. In diesem Rahmen wurden drei Übersetzungsperiodika, die in den Jahren
1940 bis 1992 erschienen sind, untersucht. Diese sind "Tercüme"
(Übersetzungszeitschrift "Übersetzung", 1940-66), "Yazko Çeviri"
(Übersetzungszeitschrift Yazko, 1981-84) und "Metis Çeviri" (Übersetzungszeitschrift
Metis, 1987-92). Diesem Zeitabschnitt mit der aufsteigenden Tendenz in der
Übersetzungskritik der literarischen Werke folgt jedoch trotz intensiver Bemühungen
hinsichtlich einer objektiv, wissenschaftlich und theoretisch fundierten
Übersetzungskritik eine Ruheperiode.
Resumen
En este trabajo de investigatión sobre la critica de la traducción literaria en Turquia se
han analizado tres revistas de traducción: "Tercüme" (Traducción, 1940-66), "Yazko
Çeviri" (Traducción Yazko, 1981-84), "Metis Çeviri" (Traducción Metis, 1987-92),
representantes de la modalidad de la traducción durante el periodo que va desde 1940
hasta 1992. A través de sus politicas, objetivos y aporte a la critica de la traducción,
estas revistas proporcionan una vista panorámica de la evolución/regresión de la
critica de la traducción durante el mencionado periodo. Después de un periodo de
progreso viene una época de estancamiento, a pesar de todos los esfuerzos realizados
para lograr una critica a la traducción objetiva y teóricamente cientifica.
364 Demirei & Yilmaz
Cette liste des revues parues en Turquie, depuis 1940 jusqu'à nos jours, permet
de voir, d'après les titres, l'évolution du domaine de la traduction. Notre
recherche sur la critique de la traduction, se limitera à l'analyse de trois revues
Tercüme, Yazko Çeviri, Metis Çeviri1, représentatives de l'époque et de la
situation de la traduction uniquement littéraire.
La critique de la traduction littéraire en Turquie a-t-elle évolué ou bien
régressé depuis 1940? Afin de répondre à cette question, nous nous sommes
proposé d'analyser en particulier les politiques de ces revues, leurs objectifs et
leur apport à la critique de la traduction littéraire.
Quelle était la situation de la critique de la traduction avant 1940? Les
premiers exemples de critique apparaissent d'abord sous forme de "reproche" et
de "querelle"2. Entre 1923 et 1940, la critique se limite à présenter les œuvres,
avec des jugements subjectifs. A partir de 1940, la critique devient une mise en
valeur de l'œuvre et de l'écrivain. Cette évolution est due en premier lieu au
manque d'œuvres traduites et aussi au fait que le concept de "critique" est à
l'époque étranger à la société turque.
revue. Nusret Hyzyr, l'un des critiques influents de la revue alors, exprime
souvent son désir de déterminer les critères en critique de la traduction. Il
appelle les autres critiques et traducteurs à établir ces normes. C'est en quelque
sorte un premier pas vers la création d'une théorie de la critique de la
traduction .
Ces dernières années, les maisons d'édition, suite à une demande croissante de
traduction de romans, ont engagé beaucoup de traducteurs. Ainsi de bons et de
mauvais traducteurs ont paru sur le marché. Ma crainte à présent porte sur les
traductions de poèmes à la mode de nos jours, car j'ai bien peur qu'ils engagent
cette fois encore des traducteurs incompétents dans ce domaine. Dans ce cas-
là, tous les domaines littéraires, le lyrique, l'épique et le pastoral auront été
sabotés.6
Critique de la traduction littéraire 367
2. 2. Politique de la revue
Yazko, fondée en 1981 par des écrivains et des traducteurs, a été une revue
bimensuelle de traduction, indépendante de l'Etat; elle a eu pour objectif
essentiel "de préserver l'actualité de la problématique de la traduction en tenant
compte de ses divers aspects"13 . A la période de parution de Yazko, l'activité
de traduction est loin d'être alors systématique. Cette désorganisation dans ce
domaine est nettement exprimée par Ahmet Cernai:
Un grand nombre des traductions faites à cette époque peut être qualifié de
"littérature passe-temps". Cette situation est due à un souci de profit
commercial mais surtout au fait que l'on ne sait pas exactement quelle œuvre
traduire.14
Pour mettre fin à cette désorganisation, la revue s'engage à donner une place
importante aux critiques de traduction car les éditeurs de l'époque pensaient
que les traductions n'étaient pas soumises à une critique objective. De fait, ces
critiques se réduisaient alors à des textes de présentation à caractère
publicitaire. C'est pourquoi la revue décide de consacrer une de ses parties aux
"écrits sur la traduction", exclusivement réservée aux critiques. On décide de la
création d'une commission afin de contrôler d'une part la sélection des textes à
traduire et de l'autre les écrits qui seront publiés dans la revue.
En 1982, dans le numéro 7, le comité de rédaction donne des détails sur
les principes et critères qui vont opérer dans la sélection des critiques de
Critique de la traduction littéraire 369
Les critiques de traduction de nos jours, qui sont seulement une énumération
d'erreurs, ont pour but d'exhiber les connaissances et la culture générale du
critique.16
Pour lui, le critique n'énumère pas seulement les erreurs; il doit être celui qui
sait aussi donner un sens profond à l'œuvre, éclairer le lecteur.
A quoi sert de se baser sur des théories scientifiques si vous continuez à faire
'la chasse à l'erreur', si en pratique vous contredisez votre propre hypothèse
théorique.18
3. 1. Objectifs de la revue
Cette revue trimestrielle est parue pour la première fois en 1987, fondée en
grande partie par des académiciens; son objectif est surtout d'encourager et
d'orienter les jeunes traducteurs, sans négliger les théories de la traduction.
Les conséquences de la traduction sur la langue et indirectement la
culture turques, ont amené cependant les intellectuels et les hommes de lettres
à se pencher aussi sur les problèmes de la structure de la langue turque. En
effet, l'utilisation abusive d'expressions et l'assimilation de règles
grammaticales, issue de langues étrangères, entraînaient, peu à peu, une
déformation inquiétante de la langue turque, en train de devenir "un nouveau
jargon indomptable"19, selon la préface même du premier numéro de la revue.
Un des moyens le plus efficace pour lutter contre cette déformation
linguistique est la critique de la traduction.
4. Conclusion
Notes
1. Comme notre recherche se concentre sur l'analyse de ces trois revues de traduction, nous
nous contenterons de reporter les références dans la partie Notes, sans avoir recours à une
partie Bibliographie. Toutes les traductions sont des auteurs du présent document.
4. Ataç, Nurullah. "Ikinci Yila Girerken" (Au Seuil de la deuxième année). Tercüme 7, 19
mai 1941, p.3.
8. Rado, Şevket. "Yeni bir Tercüme Devrine Girerken" (Au seuil d'une nouvelle période en
traduction). Tercüme 45, mai-juin 1948, p.250.
9. Görk, Vecihi. "Telif Hakki" (Droit de Reproduction). Tercüme 46, juillet- août 1948,
pp.338-344.
11. Tuncel, Bedrettin. "Kisa Notlar" (Brèves notes). Tercüme 60, avril-juin 1955, p. 135.
12. Cemal, Ahmet. "Anayasa Taslagi Üzerine birkaç Söz" (Quelques propos sur la nouvelle
loi constitutionnelle). Yazko Çeviri (Traduction Yazko) 8, septembre-octobre 1982, p.6.
13. Cemal, Ahmet. "Başlarken" (En guise d'avant-propos). Yazko Çeviri (Traduction Yazko)
1, juillet-août 1981, p.9.
14.Ibid.
15. Cemal, Ahmet. "Türkiye'de Çevirmenlik" (La profession de traducteur en Turquie). Yazko
Çeviri (Traduction Yazko) 15, novembre-décembre 1983, p.8.
16. /leri, Selim. "Ferit Edgü'yle Çeviri Üzerine" (Reportage avec Ferit Edgü sur la traduction).
Yazko Çeviri (Traduction Yazko) 11, mars- avril 1983, p.85.
17. Voir l'article de Paker, Saliha. "Çeviride 'yanliş/dogru' Sorunu ve Şiir Çevirisinin
Degerlendirilmesi" (Le Problème du juste et du faux en traduction et l'évaluation de la
traduction de poème). Yazko Çeviri (Traduction Yazko) 13, juillet- août 1983, p. 131-139.
18. Özgüven, Fatih. "Thomas Mann Çevirmeninin Cevabi" (Réponse du traducteur de Thomas
Mann). Yazko Çeviri (Traduction Yazko) 15. Op. cit.. p. 153.
20. Le terme de fonctionnalité apparait pour la première fois chez les théoriciens tels que
Gideon Toury et Hans J. Vermeer. Par ailleurs, dans les revues que nous avons consultés,
d'autres noms sont cités: par exemple Anton Popovic, Raymond van den Broeck, etc.
Ainsi, apparaissent plusieurs interprétations de la "fonctionnalité".
21. Tapan, Nilüfer; Tanyeri Ergand, Çaglar. "Critique de traduction". Metis Çeviri (Traduction
Metis) 19, printemps 1992, p. 116.
22. Maier, Carol 1998. "Reviewing and criticism". Routledge Encyclopedia of Translation
Studies ed. by Mona Parker, p.209. London & New York: Routledge.
THE DOMESTICATED FOREIGN
Zusammenfassung
Im Fokus dieses Artikels stehen generelle ÜbersetzungsStrategien und ihr zeitlicher
Wandel. Das Hauptanliegen ist zu zeigen, wie die Einbürgerung in unterschiedlichen
Übersetzungen in verschiedenen Zeiten gewirkt hat. Die Konzeptionen der
Einbürgerung und Verfremdung werden hier auch generell diskutiert. Paloposki befaßt
sich mit der Einbürgerung von William Shakespeare's Macbeth in Finnland Anfang
des 19. Jahrhunderts. Oittinen konzentriert sich auf die Einbürgerung mit Blick auf
Kinder, insbesondere bei den drei finnischen Übersetzungen des Werkes Alice's
Adventures in Wonderland von Lewis Carroll.
Résumé
Nous nous attacherons ici aux stratégies globales en traduction, à leur évolution. Nous
montrerons en particulier jusqu'où diverses traductions de différentes époques ont été
"naturalisées", pour discuter plus globalement de l'opposition entre naturalisation et
exoticisation. Paloposki traite de la naturalisation du Macbeth de Shakespeare, en
Finlande, au début du 19ème siècle tandis que Oittinen analyse la naturalisation des
livres pour enfants, surtout dans les trois traductions finnoises de Alice de L. Carroll.
Resumen
En este trabajo, nos centraremos en las estrategias globales de la traducción y su
evolución a través del tiempo. Nuestro principal objetivo sera demostrar que la
"naturalización" se ha utilizado en distintas traducciones en diferentes épocas, pero
también abordaremos los conceptos de "naturalización" y "extranjerización" en
general. Paloposki aborda la naturalización del Macbeth de Shakespeare a la Finlandia
de principios del siglo XIX. Oittinen se centra en la naturalización para niños,
especialmente en las tres traducciones al finlandés de Alice's Adventures in
Wonderland de Lewis Carroll.
374 Paloposki & Oittinen
1. Introduction
Time can play havoc with our ideas of translation. At some point in history, the
most highly regarded translations were those that conformed to the ideal of
'belles infidèles', whereas at some other stage, accuracy was the hallmark of
good translations. In addition to time, place plays an important role in what is
considered a good translation. These two variables form an (albeit simplified)
skeleton for studying two strategies translators have often put to use in their
translations: foreignization and domestication.
Foreignization generally refers to a method (or strategy) of translation
whereby some significant trace of the original "foreign" text is retained.
Domestication, on the other hand, assimilates a text to target cultural and
linguistic values (Robinson 1997b: 116-117; see also Chesterman 1997:28).
The most recent - and probably also most fervent - critic of domestication,
Lawrence Venuti (see eg. 1995:18-22) has attacked domestication as a site of
ethnocentric racism and violence, "...an ethnocentric reduction of the foreign
text to target-language cultural values..." (ibid.:20). His preferred method of
translation for literary texts is foreignization; or "...resistancy, not merely
because it avoids fluency, but because it challenges the target-language culture
even as it enacts its own ethnocentric violence on the foreign text" (ibid. :24).
For Venuti, there are a number of reasons why foreignizing is desirable
(and domestication is to be rejected). For him, domesticated translations
"conform to dominant cultural values" (ibid. :291), whereas foreignization
"challenges the dominant aesthetics" (ibid. :309) (like when he himself
translates the Italian poet Milo di Angelis). Secondly, foreignized translations
"signal the linguistic and cultural differences of the foreign text" (ibid. :311).
And thirdly, foreignization "seeks to restrain the ethnocentric violence of
translation" (ibid. :20). From this, it can be inferred that foreignness as such is
something desirable, and that domestic values, linguistic codes, and aesthetics
are undesirable (as they should be challenged, not conformed to). Looking at
these claims, it is interesting to note that Venuti actually denies that he would
be interested in the foreign as such (ibid. :41-42), making a point of using
"foreignness" only as a strategic tool. As Anthony Pym (1996:166-167) notes,
Venuti's own language use certainly defies easy understanding, and unearthing
aims and causes in his text is rather like the work of a detective.
Venuti's approach has been criticized before (see, e.g., Pym 1996;
Lane-Mercier 1997). In this article, we would like to call attention to a number
of empirical examples of domestication in practice and to point out that they do
not conform to the simple mechanistics of "bad" (domesticated) versus "good"
(foreignized) translations. While Venuti's examples offer interesting insights
into the background of several translators, foreignizers as well as
domesticators, and his analysis can be seen as a refreshing challenge to (some)
Anglo-American literary translation practices, his generalisations are likely to
The domesticated foreign 375
be less convincing when checked against different kinds of data. It is not only a
question of how texts are translated (whether they are domesticated or
foreignized), but why these strategies have been used. Anthony Pym (1992:
222; 1998: 5-6) addresses roughly the same problem when he contends that the
"what?" and the "how?" questions (what he calls "translation archaeology" and
"translation criticism") are not enough if a study is to be properly historical.
The "how?" questions logically precede the "why?" questions, but it is the
latter that help us understand the phenomena in question.
Lawrence Venuti has recently elaborated on many of his previous ideas
(see, eg., Venuti 1998); among other things, the use of the term
"foreignization" has given way to a new term, "minoritizing translation"
(ibid. : 12), the aim of which is "... never to erect a new standard or to establish a
new canon, but rather to promote cultural innovation as well as the
understanding of cultural difference by proliferating the variables within
English..." (ibid. :11). However, we wish to challenge a view of foreignization
which advocates this method as the only morally acceptable alternative for a
translator conscious of her/his choices and their consequences in a world of
power politics, racism and ethnocentrism. We wish to show that translation is
(yes, quite rightly) a battlefield of many opposing strategies and views, and that
two seemingly opposing strategies can be aiming at similar effects, while one
and the same strategy can be used for diametrically opposed purposes.
Nearly two hundred years ago, as the result of prolonged wars between Sweden
and Russia, Finland became an autonomous part of the Russian Empire. The
earliest roots of Finnish as a literary language go back to the times of the
Reformation and Bible translation, but it was not until the Russian era that the
first translations of fiction appeared in Finnish, in a situation where cultural
patriotism was slowly changing into national awakening. In the early days of
376 Paloposki & Oittinen
the Russian era, not much changed, though, on the political scene - the
legislation and administrative system remained largely Swedish, according to
the promises made to Finns by their new ruler, Czar Alexander the First. The
Finns could not know for sure, either, whether this was to be their permanent
state, or whether Finland would change hands again and return to Sweden. But
the changed status of the Finnish province, together with a gradual tightening
of the grip of the ruler, gave rise to a more acute awareness about Finland's
identity. The search for identity focused on the Finnish language, the
development, standardization and literary eloquence of which rapidly became
important and even controversial issues. Texts written in Finnish - both
translated and original - started to appear outside the traditional categories of
legal and religious texts (religious texts had hitherto counted for 82% of all
published texts). In 1831, the founding of the Finnish Literature Society
institutionalized the aim of making the Finnish language (which even many
Swedish-speaking Finns had started to call their "mother tongue") one of the
pillars of the future nation. Closely related to this elevated status of the
language was the collection and writing down of songs and poems from the
oral tradition, which started as early as the 17th century.
In addition to this major endeavour of compiling a Finnish epic, all
other kinds of written texts became valuable for the aims of the nationalistic
campaign. Translated fiction was to serve the Finnish cause in creating reading
materials for the public, educating and "improving" people, and for polishing
the language so that original writing could then emerge. Although language is
not the only (and often not even the most important) component of nationalism
(see Hobsbawm 1991:20-22), it was at times a very important element in the
nationalist struggle. Finland is not an isolated example: similar cases of
enhancing the status of national languages have been reported in other parts of
Europe at certain stages of a nationalist awakening, and the Czech and the
Slovak language policies of the time seem rather parallel to that of Finland
(Hansson 1984; Hobsbawm 1991; Schulze: 1993:58-59; Steinberg 1987:203-
204).
afterwards, and even newspaper criticism was largely favorable (even the
British press hailed "Bunulinus" (sic). As for wider audiences, Ruunulinna
obviously did not attract any great readership, and when the first plans for a
Finnish theatre were hatched and Lagervall suggested Ruunulinna as an
obvious choice of a play, he was not taken up on his offer.
The overall strategy for translating Macbeth into Finnish in 1834 thus
consisted of changes on two levels: 1) changing the setting (changing and
explaining, if necessary, all names, both people and places) and all references
to historical events, persons etc. to ones from Finnish history or mythology; 2)
replacing the original form with Finnish runic verse.
Domestication is not necessarily dependent on one reason or factor
only. There may be several reasons behind the need or desire to domesticate a
text. I have here looked at one possible explanation for domesticating: the
historical role given to translations in the creation of not only an indigenous
literary tradition but also a history, a heroic past that would justify patriotic
380 Paloposki & Oittinen
feeling. There may have been other reasons behind domestication, less easy to
detect. Translating poetry has often been considered an act of new literary
creation. Perhaps Lagervall was (unconsciously) applying Percy Bysshe
Shelley's famous idea about transplanting the seed of a poem ("The plant must
spring again from its seed or it will bear no flower"; in A Defense of Poetry
from 1821). Ruunulinna is Macbeth born again in a different world, sprung up
from seed. Underlying Lagervall's creative work, there may have been a
conception of art as universal heritage and of other people's work as a seed that
can be transplanted.
While Lagervall uses Macbeth as material for his own literary creation,
he also offers his work for the good of the Finnish language and for the
creation of a literary language, a theatrical tradition and a canon of history. A
tall order, but such was the time that grand visions and high hopes could give
rise to projects like Ruunulinna.
This domesticated Macbeth can thus be seen as an attempt not only at
the improvement of the Finnish language or the enriching of Finnish literature,
but at the creation of a history worthy of admiration on a national scale.
Ruunulinna is far from the imperialist violence attributed to domestication: on
the contrary, it can be seen as one small attempt by one individual to enhance
the status of a minority language with hardly any literary tradition in fiction.
that apart from the names of flowers and capital cities, children learn many
other important things from books, too: children need to be emotionally
involved so that they learn to understand other people's feelings in different
situations. Stepping into someone else's shoes is easier in a book than in real
life.
Seeing adaptation as negative (like Shavit, Klingberg and Bravo-
Villasante all do) is in line with Venuti's ideas on domestication. Yet
adaptations are products of their contexts, and it would be more interesting to
study the justifications made in favour of either adapting (domesticating) or
foreignizing the text. Thus the point is not whether adapting or domesticating is
a negative or positive phenomenon as such. Rather, what is at issue is the
purpose of the whole translation project, the translation situation, and the
translator's child image.
Klingberg - and to some extent Shavit, too - deals with texts and
languages as closed systems with permanent meanings, paying less attention to
the reader's participation and creative understanding. For Mikhail Bakhtin,
texts can be understood in quite a different way: they are unities, the parts of
which are understandable on the basis of the whole (as well as the whole
reading situation) and the relationship between the whole and its parts.
Klingberg and Bravo-Villasante's fears about children "not learning enough"
are akin to "the authoritarian word" described by Bakhtin: given from above, it
would not be understood in an active way (Bakhtin 1990:342). Thus
foreignizing can be, oddly enough, very authoritarian, and against Venuti's
ideas. As Douglas Robinson (1997a: 79-131) points out, foreignizing is akin to
authoritarian rhetoric in schools and politics, designed to stupefy and passivize.
In the following, I will take a closer look at three full translations of
Lewis Carroll's Alice's Adventures in Wonderland. The original story of Alice
was first published in 1865, and has since been interpreted from innumerable
perspectives. As Martin Gardner (1970:8) has pointed out, the Alice books
"lend themselves readily to any type of symbolic interpretation - political,
metaphysical, or Freudian". Yet most scholars seem to agree on one thing: the
story is a parody, intentionally throwing mud on all our "sacred cows" like
school, religion, old age, babyhood, and family life. In Finland, there are three
full "Alice" translations: in 1906 appeared the first translation by Anni Swan;
in 1972, a translation by Kirsi Kunnas and Eeva-Liisa Manner; and in 1995 the
latest translation by Alice Martin (see Oittinen 1997.)
All three translations were created in different situations and served
different purposes, which is easy to understand: Finland has changed a lot in
ninety years. All three translators have domesticated or foreignized their texts
in different ways. Generally, all three have domesticated the British culture and
brought it closer to Finnish culture, language, and society. What differs in these
translations are their different target groups, the future readers of the text. The
The domesticated foreign 383
first two translations are clearly targeted (domesticated) for child audiences,
while the third, most recent translation is far more adult.
When the first Finnish "Alice" appeared in 1906, Finland was part of the
Russian empire. Strong nationalistic pressures in Russia had led to a series of
severe measures against the autonomy of Finland, and more were to come
before independence would be gained 11 years later. In the early 20th century
the number of books originally written in the Finnish language was not yet
very great. There was a need to develop Finland's literary language, and
influences from other languages and other cultures were sought. New themes
and new genres were needed; we could say Finland needed to be "foreignized".
The translator of the first "Alice", Anni Swan, was a children's author
and translator, who took a great interest in the position of the Finnish language.
Swan used foreign literature to nourish the Finnish language and culture.
Swan's Alice is like little girls used to be in early 20 th -century Finland: she is a
polite little country girl with a Finnish name, Liisa. The 1906 translation
clearly mirrors the child image of early 20th-century Finland. Little girls were
supposed to be nice and gentle. She even curtseys in a situation where, in the
1995 version, Alice only solemnly bows. Of the two Alices written for children
(1906 and 1972), this one was written from an adult perspective.
But what is domesticated here really? As I see it, it is the child image:
the story of Alice comes from Great Britain but the child image of the story is
very Finnish. Swan's "Alice" is domesticated for Finnish readers: the story
seems to take place in Finnish surroundings and the main character seems
Finnish, too. On the other hand, there is also an element of foreignization: by
introducing the story of Alice to Finnish readers, Swan also introduces a new
genre, nonsense, to Finland. In this way Swan's translation is, at the same time,
both domesticated and foreignized. I find this "both and" very interesting: the
"two issues" are actually part of the same whole, that of the change taking
place every time a text is translated.
When Kirsi Kunnas and Eeva-Liisa Manner's translation appeared, Finland had
gone through a period of industrialization and urbanization, which had changed
the country thoroughly in a decade. This was, of course, reflected in literature,
both in originals and in translations: the themes gradually became more
urbanized and the change showed in lexical choices and in the metaphors used.
The 1970s were also years of radicalism and political movements. People were
more matter-of-fact, more serious; fantasy was not considered good for
children. The 1960s and 1970s were clearly marked by educational ideals.
384 Paloposki & Oittinen
Thus the new Finnish version of the anarchic story of Alice, parodizing
rules and regulations, was certainly something else than a reflection of the
times: it was an attack against the seriousness of the time. As Kirsi Kunnas
says, she resisted many of the phenomena of the time, especially the prevailing
child image. In the 1970s children were supposed to act like small adults - a
way of looking at children and childhood Kunnas did not agree with.
Kunnas and Manner's Alice - still with her Finnish name, Liisa - is a
very capable, impertinent, even impudent girl: she seldom thinks twice, she
hardly ever ponders on things or thinks things over. She speaks abruptly and
responds very quickly. The translation is full of carnivalistic laughter. This
Alice laughs shamelessly at the adult phenomena of the Finnish 1970s. This is
a translation domesticated for child readers with their viewpoint in mind.
In 1995, the year when Finland became a member of the European Union, the
latest Finnish "Alice" appeared. In the 1980s and 1990s, the Anglo-Saxon
orientation has been very strong in Finland. Today, watching television, we
know (or we think we know) a lot about countries like the United States and
Great Britain. This certainly adds to our toleration of foreign names, places,
and milieux, even if this knowledge about the foreign may be shallow.
This shows in translations. Translators need not add extra explanations
or domesticate the stories to a great extent. For instance, names need not
necessarily be translated. Unlike the 1906 and 1972 versions, the 1995
translation lets Alice keep her British name. There are also bits and pieces
omitted from the first two translations but included in the newest version: the
references to another culture (such as to Shakespeare) are no longer considered
too strange for Finnish child readers.
In the three Finnish "Alice" translations, the reasons behind the
different solutions and ways of domesticating seem to lie in the translators'
different strategies and different audiences, and different child images. The
first two translations were domesticated to make them more accessible, but the
most recent translation is clearly directed to older readers. Through her
accuracy, Martin, the translator of the 1995 version, has also been able to give
a more thorough picture of the story and even the author's background. But
does this really mean that the translation is foreignized, or that the original has
been foreignized through translation? Is it not always the case that when texts
are translated they always - to a certain extent - become domesticated as well?
Alice Martin has especially underlined the otherness, the foreignness of the
book. While the first two translators have, from a certain point of view,
The domesticated foreign 385
As she said these words her foot slipped, and in another moment, splash! she
was up to her chin in salt-water. Her first idea was that she had somehow
fallen into the sea, "and in that case I can go back by railway," she said to
herself. (Alice had been to the seaside once in her life, and had come to
the general conclusion that wherever you go to on the English coast, you
find a number of bathing-machines in the sea, some children digging in
the sand with wooden spades, then a row of lodging-houses, and behind
them a railway station.) However, she soon made out that the sea was in the
pool of tears which she had wept when she was nine feet high. (Carroll
1981:12)
Here we find the description of British seaside life in the 19th century with
quaint things likes bathing machines and wooden spades, from another time
and another culture. The first two translators have omitted the section in bold
altogether, while the third translator has kept all the details and diligently
depicted the sea, the children, the sand, the lodging-houses, and the railway
station.
If we look at "Alice" translations into other languages, we find similar
solutions. In many Italian, German, French, Spanish and Portuguese (Brazilian)
versions the section has been deleted, especially in the versions meant for small
children. This consistency shows that translators have not been sloppy but that
their strategies and audiences differed from those of the original author's.
Throughout her translation, Martin preserves the otherness of the
language, culture, time, place, and gender, while the earlier translators rewrote
the story for Finnish child readers. The 1995 version is much closer to British
culture and history, and also the two sexes are more distinctly present.
Throughout her translation, Martin has wanted to include everything, the whole
story of Alice. It is, paradoxically, this preciseness that makes Martin's text
very funny and very postmodern. Even if the story as such is clearly situated in
19th-century England, it becomes a postmodern combination of old and new,
strange and familiar, even female and male, when it is rewritten in another
language and in another time.
Gender is the issue in my second example. Martin's translation includes
several details that refer to Carroll's love for little girls and hatred of little
boys. The lullaby sung by the Duchess to a baby boy is a good example of the
differences in dealing with gender. The lullaby based on David Bates's original
386 Paloposki & Oittinen
poem is one of the many poem parodies of the book. Bates's original goes like
this:
Carroll's parody goes as follows, changing the baby into a baby boy, who later
turns into a pig:
The Finnish translators have each been able to give expression to the nasty
tones of Carroll's parody. Yet both of the earlier translators have left out the
baby's sex and speak only of a child, while Alice Martin's translation speaks of
a baby boy. Martin Gardner, the author of The Annotated Alice, points out that
"it was surely not without malice that Carroll turned a male baby into a pig, for
he had a low opinion of little boys" (Gardner 1970: 84). Only the third
translation reveals this detail not only of the story itself but also of Carroll's
life. "De-sexing"can be seen as an instance of domestication, too. "Otherness"
involves not just language and culture, but issues like gender and being a child
(a child is an other to an adult).
4. Conclusion
Going back to Venuti, and on the basis of the data presented here, it seems
evident that foreignizing and domesticating are contextual phenomena and
need to be studied as such. Even if we agree that translating is always an issue
of power and politics, we feel that translation is more than that. The word
"foreignizing" in itself might be misleading, at least in the context of
translating: every time we translate we necessarily domesticate, one way or the
other. The text becomes part of the target-language culture and literature. The
direction of this cultural transfer also matters: translating into English is
different from translating from English.
Maybe foreignizing is an illusion which does not really exist. Perhaps
we should only speak of different levels and dimensions of domestication.
Anyway, when we speak of domesticating and foreignizing, we cannot avoid
The domesticated foreign 387
the problem of situation. As Mikhail Bakhtin points out, in every reading act,
we meet with otherness, with other points of view (1990:279-280). This is
meeting with the foreign, which becomes domesticated through translating. If
we see translating from a dialogic point of view, as communication between
human beings, as an attempt to understand, we cannot accept Venuti's views,
as he - intentionally or unintentionally - forgets the future readers of texts. If
we do not translate for our readers, then why translate at all?
Translators normally address different audiences differently (like child
and adult readers): as translators we domesticate for our audiences, taking into
consideration their assumed views and ways of understanding. The future
readers of our translations are our superaddressees: we have a certain image of
them. Of course, these readers never exist in the flesh. Yet they are necessary,
since in this way the translator shapes his/her text into a credible whole. The
translator's child image is one kind of superaddressee.
Throughout this article, we have been referring to different purposes of
texts, different settings, different audiences, and different times. These are all
dimensions of domestication: what is domesticated, how and why. Names can
be domesticated, the setting localized; genres, historical events, cultural or
religious rites or beliefs can be domesticated. Domestication is not an
automatic product of a certain time or place, either: it can be highly
idiosyncratic. We domesticate for Finns, for children, for minority cultures, for
majority cultures, for political ideals, for religious beliefs. Whether it is cultural
imperialism or emergent nationalism, carried out for propriety reasons or for
educational purposes, depends on the situation. Alice (in Wonderland) has
become Liisa in Finland for quite different reasons than Macbeth became
Ruunulinna. Persian poets may have been "shaped" by Edward Fitzgerald
because of his feeling of cultural (imperialist) superiority (see Bassnett
1991:3); classical texts that have been retranslated into Quebecois may have
been treated that way because of a need for self-assertion on the part of a
minority culture (see Brisset 1990). Feminists may get away with "hijacking"
or "womanhandling" texts because these are seen as liberating practices (see
Godard 1990:91, 94; Simon 1996:14-16, 35), whereas their male counterparts
might not so easily find ways of justifying their domestication. Texts may also
be domesticated because of political pressures, censorship, or differing moral
values.
We are not defending any of these practices here, as we feel it is not a
question of justifying one strategy or the other; but they all have their
underlying logic. The point is, domestication does not necessarily conform to
dominant cultural values: it can also bring about the cultural difference
(advocated by Venuti) of a minor language, as in the Finnish Macbeth. Thus,
domestication cannot be explained away with notions such as "wholesale
domestication of foreign values". There is no inherent, tried and tested ethics of
388 Paloposki & Oittinen
References
BAKHTIN, M. [1981] 1990. The Dialogic Imagination: Four Essays (orig. Voprosy literatury
i estetiki, 1975). Trans. Caryl Emerson and Michael Holquist. Ed. Holquist. Austin: University
of Texas Press.
BASSNETT, S. & LEFEVERE, A. (eds.). 1990. Translation, History and Culture. London &
New York: Pinter Publishers.
BOSLEY, K. 1997: To the Reader. Skating on the Sea. Poetry from Finland. Edited and
translated by Keith Bosley. Helsinki: Finnish Literature Society.
BURKE, P. & PORTER, R. (eds.). 1987. The Social History ofLanguage. Cambridge:
Cambridge University Press.
CARROLL, L. 1972. Liisan seikkailut ihmemaassa (orig. Alice 's Adventures in Wonderland,
1865). Illust. John Tenniel. Finnish trans. Kirsi Kunnas and Eeva-Liisa Manner. Jyväskylä:
Gummerus.
CARROLL, L. 1981. Alice 's Adventures in Wonderland and Through the Looking-Glass (orig.
1865 and 1871). Illust. John Tenniel. New York: Bantam Books.
CARROLL, L. 1983. Liisan seikkailut ihmemaassa (orig. Alice 's Adventures in Wonderland,
1865). Illust. Tove Jansson (1966). Finnish trans. Anni Swan (1906). Porvoo: WSOY.
The domesticated foreign 389
DUBOIS, T. 1995. Finnish Folk Poetry and the Kalevala. New York & London: Garland.
FRIDÉN, A. 1986. Macbeth in the Swedish Theatre in 1836-1986. Malmö: Liber Forlag.
GARDNER, M. [1960] 1970. The Annotated Alice. Alice's Adventures in Wonderland and
Through the Looking Glass by Lewis Carroll. Illust. John Tenniel. Harmondsworth: Penguin
Books.
HANSSON, S. 1984. Svenskans nytta, Sveriges ära. Litteratur och kulturpolitik under 1600-
talet. Kungälv: Skrifter utg. av Litteraturvetenskapliga institutionen vid Göteborgs Universitet,
11.
HEYLEN, R. 1993. Translation, Poetics and the Stage. Six French Hamlets. London & New
York: Routledge.
KLINGBERG, G, 0RVIG, M. & AMOR, S.. (eds.). 1978. Children's Books in Translation.
The Situation and the Problems. Stockholm: Almqvist & Wiksell International.
NIDA, E. A. & TABER, C. 1969. The Theory and Practice of Translation. Leiden: E. J. Brill.
SCHULTZE, B. 1993. "Shakespeare's Way into the West Slavic Literatures and Cultures".
DELABASTITA, D. & D'HULST, L. (eds), 55-74.
SHAVIT, Z. 1986. Poetics of Children's Literature. Athens, GA.: University of Georgia Press.
SIMON, S. 1996. Gender in Translation. Cultural Identity and the Politics of Transmission.
London & New York: Routledge.
STEINBERG, J. 1987. "The Historian and the Questione della Lingua." BURKE, P. &
PORTER, R. (eds), 198-209.
VENUTI, L. 1995. The Translator's Invisibility. A History of Translation. London & New
York: Routledge.
VENUTI, L. 1998. The Scandals of Translation. Towards an Ethics of Difference. London &
New York: Routledge.
Subject Index
K
knowledge, 89-90, 93-94, 96-97, 99, 101, paralinguistic features, 276 (see also
229, 263, 273 orality)
passivisation, 180, 184
pidgin, 76, 297
Latin, 21-22, 67, 81 plays, 340, 342, 376
legal translation, 282-284, 289-291 Portuguese/Brazilian Portuguese, 213,
lexicography, 203, 206 239, 246, 264-266
liaison interpreting, 19 (post-)colonialism, 354, 356-357, 360
literal translation, 124, 159-160, 282, 294, premodified constructions, 183
309, 318, 320, 332, 334, 336, 365 problem solving, 43, 57, 110-112
literary translation, 22-23, 208, 246, 248, prototype, 4, 7, 9, 12, 120, 122-123
319,336 pseudotranslation, 36, 247
loans, 329, 332-333, 359 punctuation, 251
localisation, 219
Index 393