Sie sind auf Seite 1von 220

Juliane House

A Model Revisited
Copley Square

Digitized by the Internet Archive

in 2017 with funding from
Kahle/Austin Foundation

, .• Translation Quality Assessment

- ?;> ■*, M

ttf#0 I M'

I, »

• . *!.
U I »

Tiibinger Beitrage zur Linguistik
herausgegeben von Gunter Narr

Juliane House

Translation Quality Assessment

A Model Revisited

gnw Gunter Narr Verlag Tubingen

Die Deutsche Bibliothek - CIP-Einheitsaufnahme

House, Juliane:
Translation quality assessment: a model revisited /
Juliane House. - Tubingen ; Narr, 1997
(Tubinger Beitrage zur Linguistik ; 410)
ISBN 3-8233-5075-7

Gedruckt mit Unterstutzung der Johanna und Fritz Buch-Gedachtnisstiftung.

© 1997 • Gunter Narr Verlag Tubingen

Dischingerweg 5 • D-72070 Tubingen

Das Werk einschlieOlich aller seiner Teile ist urheberrechtlich geschutzt. Jede Verwertung
auOerhalb der engen Grenzen des Urheberrechtsgesetzes ist ohne Zustimmung des
Verlages unzulassig und strafbar. Das gilt insbesondere fur Vervielfaltigungen, Uberset-
zungen, Mikroverfilmungen und die Einspeicherung und Verarbeitung in elektronischen
Gedruckt auf saurefreiem und alterungsbestandigem Werkdruckpapier.

Satz: CompArt, Mbssingen

Druck; Laupp&Gbbel, Nehren
Verarbeitung; Nadele, Nehren
Printed in Germany

ISSN 0564-7959
ISBN 3-8233-5075-7

Introduction • VII

CHAPTER 1: Review of Approaches to Evaluating the Quality

of a Translation 1

1 Anecdotal, Biographical and Neo-hermeneutlc Approaches

to Judging Translation Quality 1

2 Response-oriented, Behavioural Approaches to

Evaluating Translations 4

3 Text-based Approaches to Evaluating Translations 6

3.1 Literature-oriented Approaches: Descriptive Translation
Studies 6
3.2 Post-Modernist and Deconstructionist Approaches 9
3.3 Functionalistic and Action and Reception-theory
Related Approaches 11
3.4 Linguistically-oriented Approaches 16

4 Translation Quality Assessment and Translation Equivalence 24

CHAPTER 2: The Original Model for Evaluating Translations 29

1 Fundamental Concepts 29
1.1 Equivalence and Meaning in Translation 29
1.2 Functions of Language versus Functions of Texts 32

2 Towards a Model of Translation Quality Assessment 36

3 Operation of the Model 43

3.1 Method of Analysis and Comparison of Texts 43
3.2 Evaluation Scheme 45
3.3 Justification of Method 46
3.4 Implementation of the Model: Test Cases 48

4 Refinement of the Model 65

4.1 Suggestions for a Translation Typology 65
4.1.1 Overt Translation 66
4.1.2 Covert Translation 69
4.2 Distinguishing between Different Types of Translations
and Versions 71

CHAPTER 3: Substantiating the Cultural Filter; Evidence from
Contrastive Pragmatic Discourse Research 79

1 Contrastive Discourse Analyses: German-English 79

1.1 Discourse Phases 80
1.2 Discourse Strategies 81
1.3 Gambits 82
1.4 Speech Acts .s. 82

2 Five Dimensions of Gross-Gultural Difference: English-German 84

2.1 Examples in Support of the Five Dimensions 86

3 Some Contrastive Pragmatic Studies Supporting

the Hypothesis of the Five Cross-Cultural Dimensions 88

4 Examples of Translations Featuring Cultural Filtering Along

the Five Dimensions of Cultural Difference 95

CHAPTER 4: The Model Revisited 101

1 Criticism of the Original Model of Translation

Quality Assessment 101

2 Rethinking the Categories for Analysis 105

3 Rethinking the Overt-covert Dichotomy and Integrating

the Results of Contrastive Pragmatic Research Ill

4 The Meaning of the Cultural Filter 115

5 Rethinking the Notion of “Translation Evaluation” 118

CHAPTER 5: Analysis and Comparison of Source and

Translation Texts 121

CHAPTER 6: Conclusion 159

1 Theoretical Aspects of the Model 159

2 Practical Relevance: Some Pedagogic Implications 167

3 Concluding Remarks 168

Appendix 169

Bibliography 194

Author Index : 203

Subject Index 205


Some twenty years ago I finished work on the model of translation quality
assessment which constituted my Ph.D. dissertation. This was submitted to
the University of Toronto, Canada, in February 1976. In 1977, the text was
published, virtually unchanged, by the Gunter Narr Verlag. Four years later,
in 1981, a second edition was proposed by the publisher. As, at that time, the
state of the art in translation theory and particularly in the field of translation
quality assessment did not seem to me to have radically changed, a second
edition of the model duly appeared with very few changes in it, but carrying
a new preface. The book is still selling, apparently, but is practically out of
print. The publisher therefore has kindly pointed to the possibility of a further
re-print. But this time round it seemed more appropriate and indeed necessary
to undertake a more radical re-working of the old text than might be suggested
by a new edition. In other words, so much time has since elapsed, that re-thin-
king the tenets and presuppositions of the original in the framework of the
current Zeitgeist is necessary.
This book is the outcome of such a revisionary process. Two major sources
of new insight have informed the revisionary process. Firstly, of course, review
comments made on the original model, and new views, theories and develop-
ments inside translation theory are to be put into the balance, as are, secondly,
relevant findings from outside translational studies, particularly results of
cross-cultural studies, which are Important for Issues of divergent cultural
norms, and therefore bear, indirectly at least, on questions concerning whether,
how far and under which circumstances such cultural diversification is to be
taken into account in the process of translation and in the evaluation of the
resulting product. In this secondary revisionary undertaking I have focussed
on the language-pair German/English, and an extensive series of contrastive
studies of my own.
Notwithstanding these revisions, what comes out in the following pages
is still on a deeper level the same book, even if the text is scarcely identical. In
other words, I have found it possible to retain the essential features of the
original model in this book. The text is structured as follows:
Chapter 1 of this book reviews and critically examines approaches to
translation and translation quality assessment, both those preceding and those
following the publication of the original model, which is then sketched in some
detail in Chapter 2. The greatest overlap between the current work and the
original is therefore to be found in this second Chapter.

The model sketched in Chapter 2 distinguishes different kinds of trans-
lation, and proposes that cultural filtering is appropriate in some translation
types. Chapter 3 investigates the notion of a cultural filter in some detail for
the language pair German/English, gathering exemplary evidence from con-
trastive pragmatic research, including a series of studies by the author com-
paring these two linguacultures.’ In Chapter 4 the revised model of translation
quality assessment is presented. The situational dimensions of analysis are
reformulated and restructured, the results of the contrastive-pragmatic re-
search presented in Chapter 3 are taken into account, new concepts such as
genre, discourse worlds, and reference frames are Introduced, thereby giving
conceptual clarity to the crucial theoretical distinction between major different
types of translation, and relativlzing the notion of functional equivalence ac-
Chapter 5 is devoted to the illustration of the model and its application
to a small test corpus: original and translated versions of several texts from
different genres are analysed and compared, leading to a discussion of the
quality of the translation in each case. A final, sixth. Chapter summarizes the
results, attempts to evaluate the model itself, and makes some suggestions for
its use in translation teaching programmes.

October 1996 Juliane House

' The term “llnguaculture”, adapted from Agar 1996, will be used throughout this
book, roughly in the sense of language community. The newer term is preferred, as it
Indicates the close relationship between language and culture.


Review of Approaches to Evaluating

the Quality of a Translation

Evaluating the quality of a translation presupposes a theory of translation.

Thus different views of translation lead to different concepts of translational
quality, and hence different ways of assessing it. In trying to make statements
about the quality of a translation, one thus addresses the heart of any theory
of translation, l.e., the crucial question of the nature of translation, or, more
specifically, the nature of (1) the relationship between a source text and its
translation, (2) the relationship between (features of) the text(s) and how they
are perceived by human agents (author, translator, reclpient(s)), and (3) the
consequences views about these relationships have for determining the borders
between a translation and other textual operations.
In the following discussion of different approaches to assessing the quality
of a translation the relative stance these approaches take vis-a-vis these three
important questions will be highlighted.

1 Anecdotal, Biographical and Neo-hermeneutic Approaches

to Judging Translation Quality

Anecdotal reflections on the merits and weaknesses of translation have been

offered by generations of professional translators, poets and writers, philolo-
gists and philosophers. In these essays on translation, the status and relative
weight of criteria such as the “faithfulness to the original”, “retention of the
originals special flavour”, “preservation of the spirit of the source language”
as opposed to concentrating on “a natural flow of the translated text” and the
“pleasure and delight of the reader of the translation” have been discussed at
great length.
The proceedings of a meeting of professional translators (see Cary and
Jumpelt 1963), in which the problem of translation quality assessment was

discussed, also display a puzzling array of basically vague and unverifiable
statements of opinion suggesting, for instance, a connection between the qua-
lity of a translation and the personalities of the author, the translator and the
audience, or asserting that a good translation is one which does not read like one.
A common trend in the anecdotal treatment of translation quality assess-
ment is to first deny the legitimacy of any effort of trying to derive more
general rules or principles for translation quality and secondly to list and
discuss a series of concrete and random examples of translation problems and
their unexplained or inexplicable optimal solutions. A classic example of the
bewildering profusion of both vague and mutually exclusive guidelines that a
translator should heed when he sets out to produce the “best translation” of
a given text is listed in Savory (1968: 50):

“1. A translation must give the words of the original.

2. A translation must give the ideas of the original.
3. A translation should read like an original work.
4. A translation should read like a translation.
5. A translation should reflect the style of the original.
6. A translation should possess the style of the translator.
7. A translation should read as a contemporary of the original.
8. A translation should read as a contemporary of the translator...etc.”

Instead of striving to set up criteria for evaluating translations that are empi-
rically based, transparent and, at least approximating something like Intersub-
jectlve reliability, propagators of this approach believe that the quality of a
translation can most importantly be linked to the “human factor”, the trans-
lator, whose comprehension and interpretation of the original and her decisi-
ons and moves towards “the optimal translation” are firmly rooted in personal
knowledge, intuitions, interpretative skills and artistic-literary competence.
Even in more recent times, such a largely subjective and Intuitive treatment
of assessing the quality of translations is still being followed in the field of
translation studies. In the “neo-hermeneutlc approach” (cf. e.g., Paepcke 1986;
Stolze 1992; Kupsch-Loserelt 1994) the hermeneutic understanding and in-
terpretation of the original and the fabrication of a translation are individual,
creative acts that on principle defy systematization, generalization and rule
giving. In Stolze’s view, a “good” translation can only come about when the
translator “Identifies” herself fully with the text she is translating. Whether
such “identification” enables or in fact guarantees a translation of quality, and
in either case how it is concretely to be assessed, remains unclear. But maybe
this is not really what the neo-hermeneutlcists are after as the following statement
by Paepcke seems to reveal: “Textverstehen und Ubersetzungskrltlk slnd wle
Schlosser, die den Zugang eroffnen und immer wieder zuschnappen” (1986:131).

Kupsch-Losereit (1994) in an article with the title “Die Ubersetzung als
Produkt hermeneutischer Verstehensprozesse” does exactly what her title im-
plies, she denies the existence of Popper’s World Three (objective knowledge
with a degree of autonomy from authors and Interpreters) claiming that the
translator’s comprehension of the source text is “Verstehen von etwas fiir je-
manden” (1994: 46), which I take to mean that the translator only understands
through the glasses o£his potential target reader. According to Kupsch-Lo-
sereit, translation is then above all a “social practice” mainly dependent on the
“Sinn- und Bedeutungshorizont” of the translation recipient. She maintains
that there is no objectively restitutable meaning of the text, that the intention
of the author of the original text cannot be determined, and concludes that
the basis of translation is “die Applikation eines Verstehens und nicht der
Nachvollzug der Autorintention” (1988: 35). Kupsch-Losereit reveals her po-
sition most clearly in the following statement: “Jede Ubersetzung 1st also eine
Bearbeltung eines ATS (Ausgangstexts, J.H.), da der Sinn eines Textes sich in
jedem geschlchtllchen Moment verandert, und z.B. ein inhaltllch gleicher Satz
in zwel Sprachen auf verschiedene Kontexte und Situationen (z.B. Sprecher-
positlonen) trifft.” (1988: 35).
Such an extreme relativisation of content is in my opinion particularly
Inappropriate as a guideline for evaluating translations: a translation is not a
private affair but normally carries with it a threefold responsibility to the
author, the reader, and the text.
To sum up, most of the anecdotal approaches to the evaluation of trans-
lations emphasize the belief that the quality of a translation depends largely
on the translator’s subjective interpretation and transfer decisions, which are
based on his linguistic and cultural intuitive knowledge and experience. With
respect to our three basic questions (relationship between original and trans-
lation, relationship between (features of) the texts and human agents, and de-
limitation of translation from other text-processing operations), we can state
that the subjective, and neo-hermeneutlc approach to translation evaluation
can only shed light on what happens between the translator and (features of)
the original text. With regard to the other aspects, it is unenlightening, as it
represents a narrow and selective view of translation one-sidedly emphasising
one aspect of translation: the process of comprehension and interpretation on
the part of the translator. In concentrating on the individual translator’s process
of comprehension, the original text, the translation process proper, the relation
between original and translation, the expectations of the target text readers are
not given the attention they deserve, and the problem of distinguishing be-
tween a translation and various types of versions and adaptations is not even
recognized. The aversion of propagators of this approach against any kind of
objectivizatlon, systematization and rule-hypothesizing in translation proce-
dures leads to a distorted view of translation and a reduction of translation


evaluation research to examining each individual translation act as an indivi-

dual creative endeavour.

2 Response-oriented, Behavioural Approaches

to Evaluating Translations ^

A second school of translation quality assessment is behaviour- or response-

oriented. While adherents of this approach all eschew considerations of the
translator s creative actions banning them in a “black box”, in this line of re-
search, too, the pronouncement of general, non-verlfled (or non-verlfiable)
principles seems to be the norm rather than the exception. Thus Nida’s three
criteria for assessing the quality of a translation are programmatic and general:
“1. general efficiency of the communication process, 2. comprehension of in-
tent, 3. equivalence of response” (Nlda, 1964: 182). The third and most im-
portant criterion is, of course, closely related to Nida’s well-known basic prin-
ciple of “Dynamic (or Functional) Equivalence of a Translation” (1964, see
also de Waard and Nida, 1986) i.e., the manner in which receptors of the trans-
lation text respond to the translation text must be equivalent to the manner in
which the receptors of the source text respond to the source text. “Equivalent”
here clearly does not mean identical, as the responses can never be identical
given different cultural, historical and situational settings, let alone different
human beings.
Assuming that it is true that a translation should produce equivalent re-
sponses, the question remains, however, whether the degree to which this
requirement is met, can be empirically tested. If it cannot be tested, it seems
fruitless to postulate the requirement, and the appeal to “equivalence of re-
sponse” is really of no more value than the philologists’ and hermeneuticists’
criterion of “capturing the spirit of the original”.
Three similar criteria are suggested by Nida and Taber (1969: 173), the
correctness with which the receptors understand the message of the original,
the ease of comprehension and the involvement a person experiences as a result
of the adequacy of the form of the translation. Again, these behavioural criteria
need to be further explained and put to the practical test: the crucial question
then remains whether the responses in question can be measured. Nida and
Taber (1969) themselves suggested the following practical tests:

1) The cloze technique, in which the degree of comprehensibility of a text is

related to its “degree of predictability”. The reader is provided with a trans-
lation text in which, for example, every fifth word is deleted, and asked to fill
in whatever words seem to fit the context best. The greater the number of

correct guesses, the easier it is to comprehend the text because its predictability
is greater. However, for any detailed qualitative judgment of a translation’s
strength and weakness, the cloze technique seems to be too rough an instru-
ment: the criteria of inteJHglblllty and ease of comprehension surely cannot
be equated with overall quality of a translation. Further, such a test merely
compares several translations, but fails to undertake the more basic task of
judging a translation against its source text.
2) Elicitation of respondents’ reactions to several translation alternatives. As
with the cloze test, such a test cannot establish true criteria for translation
quality because of the non-inclusion of the original text as a yardstick for
3) Reading aloud of the translation text to some other person who will then
be asked to explain the contents to several other individuals who were not
present at the first reading of the text. This test, which boils down to giving
and comparing precis of different translations, relies entirely on the individual
who reports on the translation rather than on the translation which is to be
4) Reading aloud of a translation by several Individuals before an audience.
Any places in the text at which readers clearly have difficulties in reading the
text are taken as indications of translation problems. Again, this test completely
lacks reference to the source text and suffers from the relativity of any
judgment that lacks a norm (which could be provided by the source text).

Other experimental methods, in which observable, verifiable responses are

taken as ultimate criteria of translation quality have been suggested, e.g., by
Carroll (1966) and MacNamara (1967). They Include (1) asking the opinions
of competent judges; (2) testing translations against so-called “criterion trans-
lations”, l.e., translations of “proven excellence”; (3) having respondents answer
questions about a passage when they had seen either its source text or its trans-
lation text. If the answers are equivalent across the respondents, then original
and translation are to be considered equivalent; (4) ratings of discrete, ran-
domly ordered sentences taken from human and mechanical translations of
scientific articles. Two rating scales were used, one for intelligibility and one
for Informativeness.
The major weakness of all response-based suggestions for translation eva-
luation is the weakness of all behaviour-centered approaches: the “black box”,
the human mind is not taken into account, such that, for Instance, tests invol-
ving expert judges, must take criteria for granted that need to be developed
and made explicit in the first place.
Other limitations of these studies seem to be (1) that equating overall
translation quality with degrees of informativeness and intelligibility is some-
what reductlonlstlc; (2) that the assumption that a “criterion translation” exists

throws up more problems than it solves, as the problem of establishing reliable
“criteria of excellence” for the criterion translation still remains: such criteria
cannot be taken for granted as tests involving expert judges seem to assume,
rather their development constitutes the crux of translation quality assessment,
and (3) that there is no provision made for a norm against which the results
of any response test may be measured, l.e., the basic “double-bind” relation-
ship constitutive for any translation, is not tak^n into account.
With respect to the three questions we asked at the beginning of this Chap-
ter, the response-oriented approach to translation quality assessment all but
Ignores the raison d’etre of any translation which undeniably lies in the exist-
ence of an original text, and the need to present that text “in other words”.
Since they ignore the original text, response-based approaches to evaluating
translations have nothing to say about the relationship between original and
translated text, nor can they shed light on whether a translation is in fact a
translation and not a version, an adaptation or another secondary textual pro-
duct derived from an original text.
Both the subjective and the neo-hermeneutlc approaches to translation
quality assessment can be described as focussing on human beings Involved
in translation: the translator and the reclplent(s) of the translation respectively.
In the seventies, a different conceptualization of translation gained ground,
mainly in Germany. Translation came to be seen as mainly a text-induced
process with the source text being the sine qua non in this reproductive activity.

3 Text-based Approaches to Evaluating Translations

Text-based approaches to translation quality assessment may be Informed by

comparative literature (3.1), by philosophy and sociology (3.2), by theories of
action and reception (3.3) and by linguistics (3.4).

3.1 Literature-oriented Approaches: Descriptive Translation Studies

In this “target-oriented” approach, which draws on comparative literature,

the quality of a translation is assessed according to the function of the trans-
lation in the system of the target culture literature. Within so-called “descrip-
tive translation studies” researchers look upon literary translations as part of
the “polysystem” of the target culture literature (“Polysystem Theory”). In
his programmatic article Toury (1985: 20) proclaimed that “a 'translation’ will
be taken to be any target-language utterance which is presented or regarded
as such within the target culture, on whatever grounds". The source text is

thus of little importance in this approach, and the hypothesis that “translations
are in fact of one system only” (Toury, 1985: 19), namely the literary system
of the target culture, is a clear blue-print for how the issue of translation quality
assessment is to be tackled: first the translated text is criticized without refer-
ence to the source text, then specific solutions of translation problems are
analysed (mlcro-analytlcally) by means of the “mediating functional-relatio-
nal notion of translation equivalence” (Toury 1985: 21).
Researchers working in this paradigm are concerned with literary trans-
lators’ typical behaviour patterns and the types of Innovative Influences on
the target culture literary system which translations can and do exert.
In this approach, the existence of a source text that served as a basis for
the translated text is thus played down to a considerable extent. Whether such
a text in fact existed and what the exact relationship between original and
translation is, is of no major interest to the descriptive translation scholar. A
basic problem with this approach is, it seems to me, how one is to determine
when a text is a translation and what criteria one is to use for evaluating a
translation - but these are questions which a descriptive translation researcher
would probably never ask, since he would typically start from the hypothesis
that a translation belongs exclusively to the literary system of the target lln-
guaculture. However, as Roller (1987: 22) has pointed out, even descriptive
translation scholars have to make an argued decision about when a translation
solution actually counts as one, in other words, regularities and conditioning
factors in the process of translation have to be specified. Even in descriptive
translation studies, then, one has to work with some normative categories
whenever equivalence relations are being examined at a microlevel in the se-
cond phase of the analysis.
Not all scholars belonging to the descriptive translation studies paradigm,
however, play down the Importance of the source text. Thus van den Broeck
(1985; 1986), who manages to combine text-linguistic and discourse analytic
work with literary concerns, offers what he calls a “model of translation cri-
ticism and reviewing” (1985: 55), in which he sketches operations necessary
in translation evaluation, among them the comparative analysis of the source
text and the translation text, taking account of the relations between the source
text and the system of similar and/or other texts originating from the same
llnguaculture, between the target text and its readers, and between the target
text and other translations of the same source text. Unfortunately, van den
Broeck’s insightful and promising model has, as far as I know, not gone beyond
these programmatic statements.
In his latest comprehensive work Descriptive Translation Studies and
Beyond, Toury (1995) makes a renewed case for descriptive and historically-
oriented translation studies and their descriptive-explanatory goal of applying
exhaustive accounts of what has been regarded as “translations” in the recei-

vlng or target culture. Toury reiterates his retrospective focus from the trans-
lation to the original, and his main orientation is still towards “actual transla-
tions” and the textual phenomena that have come to be known in the target
linguaculture as translations. He also confirms his belief in the value of detailed
descriptions of translations and their semiotic value in the target linguaculture.
It is interesting that Toury explicitly states that his theory “entails...a clear wish
to retain the notion of equivalence, which y^rious contemporary approa-
ches...have tried to do without” (1995: 61). His historical-empirical concept
of equivalence is not a single relationship but “any relation which is found to
have characterized translation under a specified set of circumstances” (1995:
61), and it is norms which are responsible for the way this equivalence is rea-
lized. As Toury himself states, his view of translation equivalence is “not one
target-source relationship at all, establishable on the basis of a particular invariant,
rather it is a functional-relational concept, namely that set of relationships
which will have been found to distinguish appropriate from inappropriate mo-
des of translation performance for the culture in question” (1995: 86).
The strength of Toury s approach is its emphasis on solid empirical work,
frequently in the form of detailed case study analyses, and its insistence on
contextualization both at the level of the reception situation and the receiving
culture at large. The fact that descriptive translation scholars’ focus on con-
textualization Includes both a “longitudinal” (temporal) and a systemic per-
spective (paying attention to the relations a translated text has with other texts
in the relevant target system), clearly adds to the explanatory adequacy of this
The emphasis placed by scholars in the descriptive translation studies pa-
radigm on the context of culture and the reactions of the recipients of a trans-
lation and their insistence that “translations are facts of the target culture” put
this approach in the vicinity of the response-oriented studies discussed above
as well as the Skopos theorists (see below). In its emphasis on detailed mlcro-
and macro-analyses of texts, however, this approach also shares characteristics
with the textual-linguistic one.
The major problem with taking this approach as a basis for translation
quality assessment is its lack of delimitation of the object of study, or put more
simply: on which criteria are we to legitimately say that one text is a translation,
another one not, and what exactly are the criteria for judging the merits and
weaknesses of a given “translation”? In terms of the three questions asked at
the beginning of this Chapter, it is most prominently question two concerning
the relationship between (features of) the text and the human agents Involved
that are the concern of descriptive translation studies.

3.2 Post-Modernist and Deconstructionist Approaches

Translation theorists who belong to these approaches (e.g., Graham, 1985; de

Man, 1986; Benjamin, 19S9; Derrida, 1985,1992;Venutl, 1992,1995; Gentzler,
1993) attempt to critically re-think translation from a philosophical and so-
ciological vantage point. They undertake to unmask the unequal power rela-
tions that are reflected in the translation directions from and into English, and
the promotion of further English language hegemony through one-sided
translations from English and an ever decreasing number of foreign texts being
translated into English. They also try to make the translator’s activity “more
visible”, attempting to show the real power translators have in shaping national
literatures and influencing literary canons, revealing the hidden process of
selecting texts for translation, and the reasons for, and effect of, certain stra-
tegies of translation. They also critically examine both translation theories and
individual translation acts pointing to their “cannibalistic” and “imperialistic”
A radical “striking back” view of translation proudly and reversely using
the metaphor of “cannibalism” is the one from the viewpoint of the so-called
Third World propagators of post-modern translational aesthetics (see Gentz-
ler 1993: 192ff). In their view, translating means devouring the original, can-
nibalizing, absorbing and transtextualizing it. Cannibalism is understood as a
a break with monological truth as well as a form of nourishment. The original
is “eaten up”, and the boundaries and hierarchies between original and trans-
lation vanish. Cannibalistic philosophy relativizes the traditional concept of
translation as mimetic representation of the original, and the concept of an
“original” is relativised, as are the notions of hierarchy and power. Translation
is seen as a dialogue not only with the original but with other texts as well:
translation is “transtextualisation” with the translator, vampirelike, taking in
the original text as his nourishment. The translator thus loses his underdog,
self-effacing role acquiring a more important voice as “transtextualizer”. Where
is the borderline to simple plagiarising, one wonders ? And are we here still talking
of translation ? I don’t think so. The boundaries between translations and other
text-transforming activities should be drawn clearly and as objectively as pos-
Post-structuralist thinkers have variously taken up and reconsidered Wal-
ter Benajmin’s famous essay “Die Aufgabe des Ubersetzers” - one of the texts
I chose for my exemplary analyses in Chapter 5 - as a quintessentlally mo-
dernist attempt to formulate a theory of translation. Trying to rethink the
dichotomy “original” versus “translation”, Derrida (1985) for instance argues
that the important point about the fact that a text is “an original” is that it is
found to be worthy and valuable enough to be translated, that it is allowed to
aqulre what Walter Benjamin has called “ein Uberleben” in its translation. It

is from this function of providing an “afterlife” that a translation gains its true
value. In post-structural thinking we can thus discover an attempt at an in-
tegrative view: “The translation will truly be a moment in the growth of the
orglnal, which will complete Itself in enlarging itself... And if the original calls
for a complement, it is because at the origin it was not there without fault, full,
complete, total, identical to Itself.” (Derrida, 1992: 188). Neither original nor
translation form a coherent semantic unity, ;hey are made up of different,
pluralistic meanings always going beyond the original author s intention. And
even the notion “original author” is deconstructed, e.g., by Foucault (1977),
who conceives of the author not as an actual individual, but as a series of
subjective positions, determined not by a single harmony of effects, but by
gaps and discontinuities, the authors creative role being reduced with new
questions being raised as to where the discourse of any particular text comes
from. This is reminiscent of Heidegger’s view (in Sein und Zeit) that language
Itself, rather than one person, is speaking and man is listening.
Venutl (1992, 1995) pleads for making translation visible by seeing it as a
reconstitution of another text mediated by linguistic, dlscoursal and ideolo-
gical differences of the target llnguaculture, and he claims that these differences
- which are crucial for poststructural thinking about translation - can be made
transparent by two kinds of close analysis: “comparisons of the source- and
target-language texts which explore the ratio of loss and gain between them
and reveal the translator’s discursive strategy as well as any unforeseen effects,
and examinations of discontinuities in the translation itself, the heterogenous
textual work of assimilating target-language cultural materials that are inten-
ded to reproduce the source-language text, but that inevitably supplement it.
The analysis of translation can also include its ideological and institutional
determinations, resulting in detailed studies that situate the translated text in
its social and historical circumstances and consider its cultural political role”.
(1992: 10-11).
Similarly, Gentzler (1993) pleads for an explicit comparison of original
and translation such that “shifts” from the original can be revealed in the
analysis. Such an analysis should give access to unconscious manipulations
resulting in mistranslation. The agenda suggested consists of an“elaboratlon
of the theoretical, critical, and textual means by which translations can be
studied and practiced as a locus of difference.” (1993: 93).
Critical theorists of translation, who are mainly concerned with what I
have called texts that call for an overt translation, examine the reasons for the
elevated status of the original text, the “Invisibility” of translation and the fact
that translation ranks lowest on the hierarchy of cultural practices: “The “ori-
ginal” is eternal, the translation dates. The “original” is an unchanging monu-
ment of the human imagination ... transcending the linguistic, cultural, and
social changes of which the translation is a determinate effect.” (Venutl 1992: 3).

This is most clearly diagnosed by a practising translator: “the choices made in
translation are never as secure as those made by the author because we are not
writing our own material” (Rabassa 1989: 7). The originality of the translation
rather lies in its self-effacement, and when a translation reads fluently, when
it gives the appearance that it is not translated, it is rated best. This “fluent
strategy” designed to efface the translator’s “intervention with the foreign
text” has been described by myself as “covert translation strategy” where a
translation is not recognized as one, a strategy singularly inappropriate for
texts calling, in my terms, for an overt translation, i.e., one openly and una-
shamedly recognizable as a translation. In critical theory this very process is
revealed as resulting in the translator’s self-annihilation and marginallty. As
Venuti rightly points out “a fluent strategy effaces the linguistic and cultural
difference of the foreign text: this gets rewritten in the transparent discourse
dominating the target-language culture... In this rewriting, a fluent strategy
performs a labor of acculturation, which domesticates the foreign text, making
it intelligible and even familiar to the target-language reader.” (1992:5). Venuti
juxtaposes a resistant translation strategy with a fluent one, by which he means
an attempt to counteract the illusionistic effect of transparency in the translated
text, which makes the translator’s work visible, “inviting a critical appreciation
of its cultural political function and a re-examination of the inferior status it
is currently assigned in the law, in publishing, in education” (1992:13). Such
strategies also preserve the llnguacultural differences between both texts, i.e.,
the translation is deliberately “strange” and alienating, which prevents domi-
nation of cultural values.
This is, in my terminology, an extolling of the virtues of overt translation
- translation that is recognized as such, as a locus of living difference rather
than as a discursively concealed process of covert assimilation.
With respect to the three questions (relationship between original and
translation, and between (features of) the texts and human agents, and delimi-
tation of translation and other textual operations), the critical, post-modern
approaches are most relevant in their attempts to find answers to the first
question, and also to the second one. However, no answers are sought for the
question of when a text is a translation, and when a text belongs to a different
textual operation.

3.3 Functionalistic and Action and Reception-theory

Related Approaches
In their functionalistic or “Skopos-theory” of translation, Reifi and Vermeer
(1984) claim that it is the “skopos”, i.e., the purpose of a translation, which is
overrldlngly Important. Given the primacy of the purpose of a translation, it

is the way target culture norms are heeded that is the most important yardstick
for assessing the quality of a translation. “Der Zweck (der Translationshand-
lung) heiligt die Mittel” (Reifi and Vermeer, 1984: 96, 101) is Reifi and Ver-
meer’s admitted credo, and a translation counts as a “felicitous interaction”
whenever it is interpreted by a recipient to be sufficiently coherent (“koha-
rent”) with his situation and no fault is found with transfer, language and
Intended meaning” (p. 112). Translation is rather mysteriously seen as “ge-
samtmenschliches Handeln” which, as a special case of transfer, also provides
for the possibility of converting linguistic action into “aktionales Handeln”,
and vice versa (p. 91).
Reifi and Vermeer (1984) distinguish between “equivalence” and “ade-
quacy.” Equivalence refers to the relationship between an original and its
translation, whenever both fulfil the same communicative function; adequacy
is the relationship between source and translation where no functional match
obtains and the “skopos” of the translation has been consistently attended to.
Whether such a terminological distinction is necessary, enlightening and sound
is open to debate. Of more relevance for my discussion here is the failure of
the authors to spell out exactly how one is to determine whether a given trans-
lation is either adequate or equivalent let alone how to linguistically realize
the global “skopos” of a translation text. Further, given the crucial role as-
signed to the purpose of a translation, the source text assumes a minor, secon-
dary importance - it is reduced to a simple “Informationsangebot” i.e., an
“offer of information”, with the word “offer” implying of course that it can
be accepted or rejected, or changed and “improved upon” as the translator
sees fit.
However, by its very nature translation is characterized by a double bind-
ing relationship: it is simultaneously bound to the source text and to the pre-
suppositions and conditions governing its reception in the target linguistic and
cultural system.
Operating in the same functionalist paradigm, Holz-Manttari (1984) en-
tertains an equally cavalierly notion of a translation. She states, for Instance,
that it is of secondary importance what exactly one means by a “translation”
(p. 78). Similarly, Honlg and Kufimaul (1982) deny the original its important
status speaking of those who do not share this opinion as having a misguided
belief in the “holy original”.
Kufimaul (1995) situates himself within “a functional approach” stating
that it has a great affinity with the RelfiWermeer approach, and in particular
with their insistence that the function of a translation depends on the target
readers and their cultural environment. Kufimaul’s concept of “function”
(p. 70ff) is, however, difficult to grasp, i.e., he seems to confuse the notions of
lllocutlon, function of a passage and function of an entire text, and falls to
distinguish between micro- and macro-textual analytical levels and to precisely

define what “function” is. In this muddled notion of “function” Kufimaul is
not alone, other “functionalist” theorists (see below) also fail to define their
concept of function unambiguously (see Kohlmayer 1988 for a similar criticism).
In 1986 Kufimaul had already pronounced the following: “Wenn man,
wie wlr es tun, von der vorranglgen Funktion des ZS-Textes ausgeht, kann
man Im Grunde nicht mehr von textbound equivalence und funktionaler
Aquivalenz reden. Denn funktlonale Aquivalenz ware ja elne Ubereinstlm-
mung der Funktion des ZS-Textes mit der Funktion des AS-Textes. Gerade
die aber kann nicht grundsatzlich postullert werden. Der AS-Text als Funk-
tlonsangebot hat vlelerlei potentielle Funktionen. Der ZS-Text aktuallsiert
elne dieser Funktionen...” (1986: 225). From this I deduce that Kuftmaul, like
Reifi (see my criticism 1977) confuses language functions (in the sense of Biih-
ler and others) with textual functions. Why, however, the translation realizes
only one function and the original offers many functions is not clear to me.
Kufimaul s idea of “function” in the context of translation is shared by
his colleague and collaborator Flonlg, who writes: “Zum andern aber muft sie
(die iibersetzungsrelevante Textanalyse, J.H.) ihr Erkenntnisinteresse am AS-
Text durch die Funktion der Ubersetzung, also durch die kommunlkatlven
Voraussetzungen in der ZS-Sprache definleren” (1986: 233). To equate the
function of the translation with the communicative conditions in the target
language - whatever that means - reveals an idiosyncratic notion of function,
to say the least.
According to Relfi and Vermeer (1984), it is the translator who decides
which function he selects for his translation and his route of translation, he is
given an Important new role of “co-author” (Vermeer, 1994: 13). The notion
of function, critical in their theory, is not clear to me at all. I can only hypo-
thesize that they consider it to be the real-world effect of a text (cf. also Inhoffen
1991 who comes to a similar conclusion).
In Reift/Vermeer s Skopos theory, then, the translator is elevated to a much
more Important position than he is normally credited with - a fact that, as
Wilss (1995) remarked - may Indeed be one of the motivations for setting up
Skopos-theory. I would agree that one of the plausible reasons for legitimizing
manipulations of the source text is an attempt to lift the translator up from his
“invisibleness” (Venutl 1995), and that the sub-text in all the target text/target
culture- and response-centered approaches to translation may well “up-grade
the status of the translator”.
The same concern with the status of the translator (and the concomitant
upgrading of his product, the translation text, which must of course be “con-
structed” (never reconstructed) is evident in Honig’s recent book with the
telling title “Konstruktives Ubersetzen” (1995). Flonig worries about the
image of the translator “man konnte allerdings den Elndruck gewlnnen, daft
der Markt lelchter elne schlechte Ubersetzung akzeptlert als selbstbewuftte

Ubersetzer” (p.77), and suggests a qualification profile for a translator that at
once reveals the deep shift in values that has occurred through the influence
of the functionalistic approach in Germany: “die entscheidende Qualifikatlon
der Ubersetzerpersonen liegt demnach nicht auf dem Gebiet der Fremdspra-
chenkenntnisse und des fachllchen Expertenwissens, sondern darin, daft sle
etwas dariiber wlssen, wofiir Texte benutzt werden und wle ihre Wlrkung
zustande kommt” (1995: 76). One can imaging the relief translation students
will feel at their new “translation” training. One is then no longer surprised
to read a little later in the book the dogmatic verdict: “Es kann nur einen
Mafistab fiir iibersetzerisches Handeln geben: einen Text abzuliefern, der fiir
die (zuvor prazlse definierte) Gruppe von Adressaten funktionsfahlg ist.”
(1995: 137).
As for the task of translation quality assessment, Honig associates himself
closely with the response-oriented approach to translation quality assessment:
“Ubersetzungskrltlk, die diesen Namen auch verdlent, sollte immer klar dla-
gnostizieren, welche Wirkung der iibersetzte Text in seinem Umfeld und fiir
seine Reziplenten hat” (1995: 123). As pointed out above (p. 4), it is an empi-
rically open question whether it is in fact possible to “diagnose” precisely the
effect a text has in any valid and reliable manner, given the fact that many texts
are multiply, ambiguously, and Indeed vaguely addressed. Further, such a state-
ment reminds one strongly of the activities of market research and advertising
managers, a similarity that may flatter some translation theorists but depresses
this author, as the focus of translation evaluation is shifted away from im-
portant Issues of translation Itself. (See here Newmark 1991, who also com-
ments on the emphasis on marketing factors in functionalist theorists’ termi-
nology and Interests.)
Vermeer states apodlctlcally that “in translating, priority has to be given
to one factor and the others have to be subjected to it - because one cannot
serve two masters at the same time” (1994:13), and proceeds from this prescrip-
tion (which is as such incompatible with the guidelines let alone ethics of any
empirical-inductive field of Inquiry) to deny the existence of “the” source text,
which he reduces to simple “source material” (Vermeer 1987: 541). Worse still:
“Was es ... gewifi nicht glbt, ist “der” Ausgangstext. Es gibt nur einen je spe-
zifisch interpretierten Ausgangstext, sozusagen den Ausgangstext-fiir-den-
Rezlplenten-X Im Zeitpunkt-tx. “Der” Ausgangstext kann also auch nicht
Grundlage und Ausgangspunkt fiir “die” Ubersetzung seln (die es ebenso
wenlg glbt). Er 1st entthront, die Translation dleser Flktion enthoben” (1986: 42).
First of all, surely nobody has ever claimed that there is something like “die
Ubersetzung”, so why imply that there is? Further, with this “de-thronlng”
of the original, all respect for an original text seems to have evaporated.
Such a relativistic view as can be gleaned from the writings of the adherents
of functionalistic translatologlsts is anathema to anybody - Including myself

- hypothesizing that a text embodies some autonomous meaning, and that this
meaning cannot exclusively be seen as emanating from the mind of the indi-
vidual reader. Also, many texts embody something we might call their “histo-
rical meaning”, and in the translation of certain texts it is essential to try to
render meaning according to the actual words of the text. As Olson (1996) has
pointed out, this type of Interpretation was both at the basis of “the Jewish
Way” of reading the Old Testament and of the Reformation: it requires close
analysis of the verbal form of the text and its context as well as an analysis of
the author, his Intended audience and his choice of expression.
It may be understandable that such a regard for the original text is alien
to all those who seem to one-sidedly turn their attention to texts of quick
consumption, e.g., instructions, advertisements, manuals, leaflets, market and
sales reports, business correspondence, mail order catalogues, tourist brochu-
res, sales slips and the like. Those ephemeral “one-off” texts may indeed have
so little “core-value” that they can easily be completely re-cast for a new au-
dience. In other words, the selection of texts for translation, and translation
analysis and evaluation may well have twisted conceptualizations and the de-
velopment of theories.
Anyone who is Interested in a greater variety of texts, such as academic,
literary, and other preservable texts, will not easily want to give up the hope
that there is indeed what Popper (1976) has called “World Three”, the world
of what he called “objective knowledge” that is embodied in theories, books,
and texts - i.e., visible (and valuable) artifacts “with a degree of autonomy
from their authors and with special properties for controlling how they will
be interpreted” (Olson, 1996: 9) - and how they will be translated, we might
pertinently add. It is these properties of non-ephemeral written texts (different
from speech) that represent and preserve our intellectual world, which should
not be degraded or “de-throned”.
Further, what Walter Benjamin in his famous essay “Die Aufgabe des
Ubersetzers” (1972) has called “die relne Sprache” designating a language of
pure meaning - unobscured and unobscurable by the mediation of any parti-
cular language - may be invoked here (as it has also been in post-structuralist
thinking), a concept that also Implies a freedom from the fetters of arbitrary
interpretation and random subjectivity. As Gentzler (1993:198) has remarked,
Benjamin s belief that translation should neither be considered as source or
target text-oriented, but as a mode of its own - subject to its own laws - is
clearly alien to the functionalist trend in translation studies with its extreme
The functionalistic approach as proposed by ReiB and Vermeer and others
cannot, in my estimation, be said to belong to linguistics (cf. Kelletat 1987,
who holds the same view), which is regarded as an empirical science. Given
the functionalists’ concern with the target culture, the theory might be clas-

sified as part of cultural studies. Since its propagators believe that the original
is a quantite negligable and emphasize the translation s total dependency on
its purpose and its recipients, it is in fact very similar to the response-oriented
approaches to assessing translation quality discussed above, and the criticism
made above of these approaches hold here too. Since Reift and Vermeer in
particular stress that source and translation text act as two offers of informa-
tion, the translator may legitimately make "ei^enverantwortllche schopferi-
sche Entscheidungen” (1984: 75), I can also detect a similarlity to the subjec-
tivist, neo-hermeneutic approach characterized above.
As concerns the relevance of the functionalistic approach for the evalua-
tion of translations, Reifi and Vermeer themselves state that it is “auEerst zwei-
felhaft, ob eine Rechnung in Defiziten und Uberschiissen, die den Zleltextwert
am Ausgangstextwert mifit, iiberhaupt sinnvoll 1st” (1984: 112). Since this is
what I am trying to do, their approach is of limited use to myself as it would
Indeed be to anyone Interested in some sort of yardstick, some norm of as-
sessment provided by the original. With respect to the three questions, the
functionalistic approach is not concerned about the relationship between ori-
ginal and translation, nor is it concerned with establishing criteria for deli-
miting a translation from other textual operations. As it stands, functionalistic
approaches are solely concerned with the relationship between (features of)
texts and the human agents concerned with them.

3.4 Linguistically-oriented Approaches

In these approaches the source text, its linguistic and textual structure and its
meaning potential at various levels (including the level of context of situation
in a systemic framework), is seen as the most Important, Indeed constitutive
factor in translation. To equate linguistic-textual approaches - as has been done
(implicitly or explicitly) by Honig and Kufimaul (1982), Snell-Hornby (1986),
Kupsch-Losereit (1988) and others - with a narrow concept of traditional or
structural syntax and semantics, is Inappropriate. As Roller (1995) has expli-
cated, one may, of course, find “narrow” linguistic approaches (Roller refers
to Rlein 1992) that focus on only one aspect of translation, e.g., the semantic
one, but there are many others that do not fit this description. Indeed, Catford s
1965 classic “linguistic theory of translation” has already gone beyond the
narrow caricature that some members of the self-styled “new orientation in
translation studies” have purposefully set up as an object of disdain.
Linguistic-textual approaches cover many different schools, the most pro-
mising for the development of models of translation assessment being those
that encompass pragmatic, socio-cultural and dlscoursal meanings. That it is
possible to firmly base one’s approach to translation on a linguistic model

which includes textual, situational and cultural aspects of translation was de-
monstrated twenty years ago in the model the revision of which is the major
content of this book.
An early and highly influential linguistic-textual approach is Reifi (1971).
Reifi suggested that the most Important invariant in translation is the text type
to which the source text belongs, as it determines all subsequent choices a
translator has to make.-She claims that different types of texts can be differen-
tiated on the basis of Biihler s three functions of language: content-oriented
texts, e.g., news, scientific-technical texts, form-oriented texts, such as poems
and literary genres, and conative texts, e.g. advertisements and texts of a per-
suasive bent. To cover translations of texts involving other media than print,
Relfi suggested a fourth additional type: subsidiary or audio-medial texts, e.g.,
operas, radio plays etc., for which different rules of translation apply, if trans-
lation adequacy is to be reached. According to Reifi, it is these textual types
which have to be kept equivalent in an adequate translation: in the case of
content-oriented texts, invariance on the content-plane is the primary consi-
deration; in the case of form-oriented texts, invariance on the content-plane
as well as on the expression-plane is to be established to the greatest possible
extent; and in the case of conative texts, the “effect” of the source text is to be
upheld in the translation text above all other features. An adequate translation
of subsidiary texts must keep the adaptation of the “text” proper to such ex-
tralinguistlc components as musical rhythm etc. invariant.
The determination of the textual types presupposes a careful analysis of
the source text, but Relfi’ suggestions are not explicit enough: her pioneer
work is programmatic only, she gives no clear Indication as to how one should
go about establishing language functions and a source text types. Further, at
what level of delicacy this can and should be done is left unexplained.
Two other influential publications in the seventies also fall to indicate the
exact procedure for source text analysis.
Roller (1972) pointed to the necessity of developing a comprehensive
linguistic model for translation quality assessment. Such a model should con-
sist of three main stages: (1) source text criticism with a view to transferability
into the target language, (2) translation comparison in which the particular
methods used in the production of a given translation are described, (3) eva-
luation of the translation not according to vague, general criteria such as “faith-
ful” or “highly intelligible” but according to “adequate” or “not adequate” in
terms of the text-specific features established in (1) and measured by native
speakers’ meta-linguistlc judgments. Although presenting insightful, original
and stimulating ideas. Roller does not go beyond a very general outline with
no suggestions for operationalization.
Wilss (1974; 1977) also stresses the necessity of building a consistent model
featuring criteria both for the detailed description and interpretation of the

source text and for the evaluation of the “dependent” translation text. He
suggests that the area of the “norm of usage” in a given language community
with reference to a given situational context should be taken as a yardstick. It
is the norm of usage which, according to Wilss, as part of any native speaker s
competence, accounts for a speaker s ability to make metalinguistic judgments.
Therefore a translation may be judged according to whether or not it is ade-
quate vls-a-vls the “normal” standard usage native speakers in a given si-
tuation. However, there will always be several variants which are legitimately
possible within the norm of usage and which depend on the individuals crea-
tive choice. Like any linguistic activity, translation is also (apart from being
mainly recreative) a creative process which always leaves the translator a free-
dom of choice between several approximately equivalent possibilities of rea-
lizing situational meaning. Moreover, the given situation in which the source
text was written is by definition unique, and therefore a notion of norm ex-
isting in the source culture is a somewhat optimistic one. Even more optimistic
is the idea that there should exist a “norm” for this unique text in the target
culture. Further, one should not underestimate the immense difficulties of
empirically establishing what any “norm of usage” is.
Another classic linguistic-textual contribution to translation evaluation is
the work by the Leipzig school (Otto Kade, Gert Jager, Albrecht Neubert)
who expressly considered their work on translation as part of linguistics.
“Translatlonsllnguistik” investigated translational processes as essentially lin-
guistic processes with analyses of translations focussing on linguistic mecha-
nisms of transfer (cf. Jager, 1973). Especially relevant for translation quality
assessment is Neubert s (see e.g., 1985) textual and pragmalingulstlc approach,
and in particular his work on the “directedness” of source texts that determine
potential equivalence frameworks and set up pragmatic translation types and
translation procedures. Early on, Neubert stressed the textual and pragmatic
nature of translation (see Neubert 1968) and the obvious relevance of text
linguistics for translation. As a proponent of the textually and pragmatically
oriented approach to translation evaluation, Neubert claims that “communi-
cative values are the proper objects of translation.” (1994: 19), adding “the
often-heard dictum that we translate meanings blurs the fact that it is only
communicative values that can be equivalent. Meanings as well as language
systems cannot be translated. Equivalence turns out to be a textual phenome-
non. It is a relation between texts, source texts and target texts. Textual equi-
valence, again, is the basis of the equivalence at lower-level units such as partial
texts, sentences, phrases, and words ...” (1994: 20).
Neubert s approach to translation, however enlightened it is with regard
to encompassing pragmatic, situational and pragmatic features of meaning, is
not elaborated as a valid model of translation evaluation remaining at an equal-
ly programmatic level as Relfi, Roller and Wilss.

Trying to go beyond the programmatic nature of the approaches to trans-
lation quality assessment criticized in Reif^’, Roller s and Wilss’, and Neubert’s
work in particular, the present author set up a linguistically oriented model
(1977) that aimed at providing a detailed description and explanation of whether
and how a translation is equivalent to its source. This model will be outlined in
Chapter 2.
Noteworthy in the context of a linguistic textual approach to translation
is also Peter Newmark’s (e.g., 1981; 1988) applications of linguistic models
(e.g., componential analysis, and case grammmar) to the analysis of translati-
ons. For translation evaluation, his distinction between semantic and commu-
nicative translation (made at the same time as my own distinction between
overt and covert translation (see below), has also been important and fruitful.
Newmark has always spoken against the conception of translation as so-
lely a “science” maintaining that the translation process is also a “basic artistic
process ... requiring the translator’s taste, wit and elegance” (1981: 137). Con-
sequently he has refrained from setting up a consistent model of translation
quality assessment. In his writings on “quality in translation” he strongly em-
phasizes the fact that “ultimately standards are relative, however much one
tries to base them on criteria other than norms... the difficulty lies not so much
in knowing or recognizing what a good translation is, as in generalising with
trite definitions that are little short of truism, since there are as many types of
translations as there are texts” (1988: 192). Although there is of course always
a subjective factor in evaluating a translation, and intuition and practice do
certainly play a role in translation evaluation, one must not overemphasize
this “subjective factor” denying the value of scientific inquiry, which enables
one to abstract from the mass of details in individual translations leading to
generally valid procedures and evaluative norms. With respect to the three
guiding questions for this Chapter, Newmark is thus close to the hermeneutic
approach to translation evaluation in that he gives priority to the relationship
between (features of) the texts and human beings.
In the nineties, four linguistically-oriented books on translation appeared
in Britain: Hatlm and Mason (1990), Bell (1991), Gutt (1991) and Baker (1992).
Although none of these books treat the evaluation of translations explicitly,
the attempt made by their authors to look at translation from the perspective
of a broad conception of linguistics Integrating recent research on sociolin-
guistics, speech act theory, discourse analysis and pragmatics, makes them
potentially valuable for translation quality assessment. While necessarily con-
sidering words, strings of words and structures, these broadly conceived ap-
proaches also consider language in use, linguistic actions as communicative
events deeply embedded in situations and cultures.
Baker (1992) and Hatlm and Mason (1990) recognize that any theory of
translation and translation assessment must concern itself with how meaning

is generated within and between different groups of people in different
cultures. They emphasize the fact that translators whose raw material is
language must not only have an expert knowledge of the two languages in
which they are operating, but also about what these languages can do, how
they do it and how they do it for their speakers.
Hatim and Mason (1990) give a broad overview of the field of “translation
and translating” describing the relevance of linguistics in general as well as
sub-fields such as register and discourse analysis, text linguistics as well prag-
matics and semiotics. Hatim and Mason (1990) go beyong register analysis on
the grounds that the insights which register analysis affords into the commu-
nicative dimension of context are Insufficient. They distinguish an additional
“pragmatic dimension” with which to take account of the textmakers’s ability
to “do things with words”, capturing phenomena such as speech act sequences,
inference, Impllcature, presupposition, the cooperative principle. Hatim and
Mason also distinguish another dimension with which to supplement register
analysis, which they refer to as “semiotic” - i.e., treating communicative items
as signs inside a system of signs, including considerations of genre, discourse
and textuallty. The separation of a communicative contextual dimension (”re-
gister“) from a pragmatic dimension seems to me to be unmotivated, (what is
communicative if not pragmatic and vice versa?) considering that any register
analysis i.e., correlating user and use dimensions with linguistic phenomena
found in the text must needs touch upon pragmatic phenomena such as the
ones listed by Hatim and Mason under their pragmatic dimension.
Unlike Hatim and Masons (1990) top-down approach. Baker (1992)
adopts a bottom-up one claiming that the former, while theoretically more
valid, may be more difficult to follow, and an excessive emphasis on “text”
and “context” may make one ignore the fact that although “a text is a semantic
unit, not a grammatical one ... meanings are realized through wordings; and
without a theory of wordings ... there is no way of making explicit one’s in-
terpretation of the meanings of a text” (Halllday, 1985: xvii, quoted in Baker,
Both Baker’s and Hatim and Mason’s approaches to translation (and by
extension to translation evaluation) have gone in a direction in translation
studies that follows closely the path linguistics and applied linguistics have
taken in widening its perspectives to embrace the levels of context and dis-
course. In developing an early Hallldayan, pragmatic and discourse analytic
approach as the basis for translation quality assessment, I took a step in the
same direction fifteen years earlier.
Gutt (1991) presents a “relevance-theoretic approach” to translation. In
line with this approach, he stresses the point that meaning is far from being
determinable in advance of the actual performance of an utterance, but depends
on the addressee’s interaction with various contextual factors by means of his

ability to make inferences. Context in this theory is bound to the addressee s
assumptions about the world, which he uses to interpret the utterance. Inter-
pretation is achieved on the basis of the relevance of a given assumption, which
can be roughly described as the likelihood that adequate contextual effects are
achieved with minimum processing efforts. Gutt s approach is rather one-si-
dedly audience related, and can thus also be placed alongside the response-ba-
sed and functionalistic theories of translation. He also shares with repre-
sentatives of these schools a tendency to make prescriptive statements, witness
the following: “whatever decision the translator reaches is based on his intui-
tions or beliefs about what is relevant to his audience” (1991: 112).
Closely following relevance theorists, Gutt distinguishes two uses of lan-
guage: descriptive and interpretive, suggesting it is exclusively interpretative
use that should be the basis for translation. Translation is thus an instance of
“interlingual interpretive use“, with the principle of relevance “heavily con-
straining the translation with regard to both what it is intended to convey and
how it is expressed. Thus if we ask in what respects the intended interpretation
of the translation should resemble the original, the answer is: in respects that
make it adequately relevant to the audience - that is, that offer adequate con-
textual effects.” (1991: 101-102).
Such a view of translation, in which the principle of relevance is seen as
the only guiding principle, is reductlonlstic and simplistic. The theory of trans-
lation evaluation offered here is an Imposition of a linguistic-pragmatic theory
onto translation, which expressly denies the necessity of rules and principles
of translation itself, because “the principles, rules, guidelines of translation are
applications of the principle of relevance; thus the proposal is that all the aspects
of translation surveyed. Including matters of evaluation, are explicable in terms
of the interaction of context, stimulus and Interpretation through the principle
of relevance” (1991: 188).
Gutt claims that all we need in translation studies is to invoke a notion
“believed to be part of general human psychology - the principle of relevance
and the ability to engage in Interpretive use” (1991: 121). The logical conse-
quence of such an attitude is also delineated by Gutt with a remarkable self-
confidence that is the outcome of his limited vision: “Since the phenomena of
translation can be accounted for by this general theory of ostenslve-lnferentlal
communication, there is no need to develop a separate theory of translation.”
(1991: 189).
In summary, Gutt’s approach is both too general and too narrow. His
attempt to subsume translation theory under a cognitive communication theo-
ry, can usefully be compared with another recent attempt to reintegrate trans-
lation studies into a proper “mother discipline", in this case linguistics. I am
referring to Klein s (1992) pronouncement: “am Prozefi des Ubersetzens glbt
es nlchts, was iiber die Erforschung der Sprache und des Sprachgebrauchs

hlnausfiihren wiirde; ich sehe deshalb keinen inhaltlichen Grund fiir eine ei-
gene Disziplin “Ubersetzungswissenschaft” (1992: 105).
Both Klein s and Gutt’s approaches to translation must be seen for what
they are: reducing the complex multi-dimensional phenomenon of translation to
one dimension, cognitive-communicative and linguistic (semantic) respectively.
In German translation studies, the dissertation by Schreiber (1993) marks
an important stage in that it successfully tries to^ifferentiate a translation from
an interlingual adaptation at a time when the lines beween translation and
other text producing actltlvles have become blurred through assumptions
about the non-relevance of the original, the overriding Importance of the trans-
lation s cultural environment and the upgraded role of the translator as a co-
author (Vermeer 1994: 13). Schreiber s criteria for distinguishing a translation
from an adaptation in terms of their respective demands on invariance and
variance, the difference between hierarchical demands in a translation and
hierarchy-conditioned variance in an adaptation, as well as his distinction of
an “Umfeldiibersetzung” (elnbiirgende Ubersetzung), i.e., translation in the
wider sense and a “Textiibersetzung” (verfremdende Ubersetzung), i.e., trans-
lation proper, are plausibly argued (see here my own covert and overt trans-
lation types and the distinctions between a translation and a version).
In his discussion of the so-called Umfeldiibersetzung, Schreiber claims
that there is a primacy of intention versus a primacy of effect - a laudable and
theoretically plausible distinction, especially in the face of much muddled thin-
king about “intended equivalence” vs “equivalence of effect” and the simplistic
equation of “functional equivalence” with “equivalence of effect.”
Two recent linguistically oriented approaches which constitute serious
attempts to objectify translation quality assessment are Gerzymlsch-Arbogast
(1994) and Steiner (1995).
Steiner s (1995) approach to translation quality assessment is to my know-
ledge the only one in Germany - apart from my own - that operates within a
systemic approach to language and linguistics. Steiner applies register analysis
to the evaluation of translations, and considers the register of a text as the
functional variety the text represents, linking up its main variables of field,
tenor, and mode with the context of situation, and beyond this the context of
culture. Steiner assumes that, in as much as translation is considered to be a
phenomenon different from other forms of multilingual text production, the
register will remain relatively constant across original and translation, and
that, the more certain register values have changed in a translated text, the
more the translation wilbno longer be a translation in the narrower sense.
Steiner presents a methodologically transparent and highly detailed analysis
of an English and a German version of an advertisement which provides a
number of interesting Insights thus demonstrating that this type of approach
is an extremely valuable one for both translation research and teaching.

Gerzymisch-Arbogast (1994) presents what she calls an “iibersetzungs-
wissenschaftliches Propadeutikum” as a basis for a more intersubjectlvely ver-
ifiable method for translation evaluation. In describing the translational process,
Gerzymisch-Arbogast considers the tension in any translation between mi-
cro-structural decisions (such as e.g., theme-rheme sequences, reference rela-
tions) and macro-structural ones (e.g., textual type). Basing her work on Mu-
dersbach s (e.g., 1992)-theoretical and methodological studies and especially
his attempt to specify invariance aspects, she develops a methodology in which
she tries to unite both the holistic textual perspective and the detailed micro
perspective. A list of different “aspects” is used according to which both ori-
ginal and translation are analysed and evaluated. As opposed to the present
author’s model the direction is reversed: i.e., the evaluator proceeds from the
translation recording her impression of the translation with regard to certain
“aspects” that have crystallized out of a list of phenomena that commanded
her attention, i.e., were deemed remarkable or odd. The result is a tentative
matrix of aspects, in which each aspect on a horizontal line is correlated with
a certain part of the text (vertical line). It is only on the basis of this matrix
then that the original is consulted, in order to check whether the “salient
points” discovered in the translation are to be found in the original as well, or
whether they originate from the translation. Following this “spot checking”,
a more holistic analysis is carried out investigating the text’s entire linguistic
and artistic make-up. The end result is then a list of aspects Indicating their
specific realizations or values in the original and the translation.
While Gerzymisch-Arbogast’s approach is certainly a rare and valuable
attempt at making translation evaluation more objective and differentiated,
the approach seems too strongly “bottom-up”, with too little “top down”
provided for systematization and generalization. Further, the assumption that
there can be equivalence on the level of culture, while widely accepted in
translation studies, is highly dubious, and notions like “Kulturem” developed
in a different framework cannot simply be transferred to translation and its
concern for equivalence relations. The theoretically crucial distinction be-
tween overt and covert translation and their relevance for the possibility of
using a “cultural filter” might profitably have been taken into account.
In her attempt to demonstrate the operation of her method, Gerzymisch-
Arbogast refers to Haefs’ controversial translation of Lawrence Norfolk’s
Lempriere’s Dictionary trying to present a more objective view of this trans-
lation than had been done in the polemical and polarized discussions in the
press and the profession. The fact that Gerzymisch-Arbogast has refrained
from giving a global judgment of the translation in terms of good or bad stating,
in all modesty, that a “sachliche, wissenschaftllch fundlerte Kritlk noch aus-
steht” (1994: 152) should not be criticized in such an unfair way as was done
by Honlg (1995:124). This unfairness is also evident in his recent review (1996:

188-193) of Gerzymisch-Arbogast s book, where he misrepresents the goals
the author had set herself for this book, claiming it is an “introduction to the
science of translation” and blaming the author accordingly to not have pro-
vided her readers with a comprehensive review of the relevant literature. Ger-
zymisch-Arbogast states unmistakably that she had meant her book as an
introduction to the methodological problems of translating for translation and
philology students, and that the book “will ab^r keinen Uberblick iiber herr-
schende Lehrmeinungen geben. ”(1994:9). How such an unambiguos (and mo-
dest) statement of Intention can be misread is beyond my comprehension.
Most of the representatives of linguistic-textual approaches aim at going
beyond explicating the niceties of individual translation cases and try to es-
tablish regularities in the relationship between pairs of source and target texts
at various linguistic levels. With regard to the three questions asked at the
beginning of this Chapter, linguistic-textual approaches take the relationship
between original and translation seriously, but differ in their capacity to pro-
vide detailed techniques and procedures of analysis and evaluation (Gerzy-
misch-Arbogast (1994) and Steiner (1995) are exemplary in this respect). The
relationship between (features of) the texts and how they are perceived by
human agents has become a concern of all those approaches that consider
language in use. Few of the linguistic-textual approaches, however, have ex-
amined the question of the consequences of these relationships for determining
the differences between translation and other textual operations, a notable
exception being Schreiber (1993).

4 Translation Quality Assessment and Translation Equivalence

The fundamental characteristic of a translation is that it is a text that is doubly

bound: on the one hand to its source text and on the other hand to the reci-
pient’s communicative conditions. This double-binding nature is the basis of
what has been called in many linguistic-textual approaches the equivalence
relation. In other words, the equivalence relation equals the relation between
a source text and its translation text. It is an Important aim of linguistic-textual
approaches to specify, refine, modify and thus to try to operationalize the
equivalence relation by differentiating between a number of frameworks of
equivalence, e.g., extra-linguistic circumstances, connotatlve and aesthetic va-
lues, audience design and last but not least textual and language norms of usage
that have emerged from empirical investigations of parallel texts, contrastive
rhetoric and contrastive pragmatic and dlcourse analyses.
In a recent attempt to make “a case for linguistics in translation theory”,
Ivir (1996) expresses the Inherent relativity of the equivalence relation very

well:“equivalence is ...relative and not absolute,... it emerges from the context
of situation as defined by the interplay of (many different) factors and has no
existence outside that context, and in particular... it is not stipulated in advance
by an algorithm for the conversion of linguistic units of LI into linguistic units
of L2” (1996:155).
The notion of equivalence is the conceptual basis of translation and, to
quote Catford, “the central problem of translation-practice is that of finding
TL (target language) equivalents. A central task of translation theory is there-
fore that of defining the nature and conditions of translation equivalence”
The concept of equivalence is essential for translation criticism, and it will
therefore be examined and sub-differentiated in what follows. I want to first
clarify the relationship between “equivalence” and “invariance”. I here follow
Albrecht (1987; 1990): Invariance in translation captures that which is the
tertium comparationis in translation. The concept of invariance is not an ab-
solute one, but must be decided in each and every individual case by the goal,
the purpose of the translation. Certain demands of invariance are (externally)
set up for a translation, and when these demands are fulfilled, the translation
is “equivalent”. Equivalence is therefore always and necessarily relative, and
has nothing to do with Identity. “Absolute equivalence” would be a contra-
diction in adiecto.
Koller (1992) attempts to describe the equivalence relation in greater detail
and to classify equivalence types according to different “Bezugsrahmen” (p.
216), which play a role in determining the type of equivalence. He comes up
with the following five “frames of reference”:

(1) the extralinguistic referents to which the text relates. The concept of equi-
valence, which orients itself to the extralinguistic referents is called “de-
notative equivalence”.
(2) the connotations conveyed through the specific means of the verbalisati-
ons present in the text. The equivalence relation constituted here is called
“connotatlve equivalence”.
(3) the linguistic and textual norms of usage (“Gebrauchsnormen”) that char-
acterize a particular text. Koller calls this type of equivalence that relates
to certain text types “text normative equivalence”.
(4) the recipient (the reader) of the translation, for whom the translation is
“specially designed”, such that it can fulfill its communicative function.
This type of equivalence that relates to the addressee is called “pragmatic
(5) certain aesthetic, formal and idiosyncratic characteristics of the source
text. The concept of equivalence that relates to these characteristics of the
source text is called “formal-aesthetic equivalence”.

Given these different types of equivalence in translation, it becomes im-
mediately clear that not all five types of equivalence can be aimed at in trans-
lation, but that - true to the nature of translation as a decision process (Levy
1967) is is necessary for the translator to make a choice, i.e., the translator has
to set up a hierarchy of demands on equivalence that the wants to follow.
Given the relative nature of “equivalence” and the fact that it has clearly
nothing to do with “identity” it is more than Hirprising that a polemic attack
should have been directed against the concept of equivalence, in the course of
which an analysis of the English and German dictionary meaning of the term
“equivalence” was presented (seeSnell-Hornby 1986:12ff). Snell-Hornby sin-
gles out one particular dictionary entry, which supports her claim that equi-
valence basically equals identity and promptly proceeds to dismiss equivalence
as “an illusion” in translation studies. She writes that equivalent means “vir-
tually the same thing”. By contrast, I found the following dictionary entries
for “equivalent” and “equivalence” in my own dictionary searches: “having
the same value, purpose... etc. as a person or thing of a different kind (Longman
Dictionary of Contemporary English 1995), and having the same relative po-
sition or function; corresponding...” (Shorter Oxford English Dictionary 1993),
as well as “equivalence is something that has the same use or function as some-
thing else” (Collins Cobulld 1987). And in German, too, “Aquivalenz” is not
only a term in the “exakte Wissenschaften” as Snell-Hornby claims: in my
Brockhaus I read: “das was in gewlssen Fallen gleiche Wirkung hervorzubrin-
gen vermag”. Such entries were not mentioned by Snell-Hornby as they would
clearly not serve her purpose of discrediting the concept of equivalence in
translation studies.
The attack against the concept of “equivalence” in the field of translation
studies has a slightly dated touch: definitions of equivalence based on formal,
syntactic and lexical similarities alone have actually been criticized for a long
time, and it has long been recognized that such narrow views of equivalence
fall to recognize that two linguistic units in two different languages may be
ambiguous in multiple ways. Formal definitions of equivalence have further
been revealed as deficient in that they cannot explain appropriate use in com-
munication. This is why functional, communicative or pragmatic equivalence
have been accredited concepts in contrastive linguistics for a very long time,
focussing as they do on language use rather than structure. It is these types of
equivalence which have become particularly relevant for translation, and this
is nothing new (cf. Catford 1965).
In this Chapter I have discussed different approaches to translation quality
assessment and their relative stance vls-a-vls three questions concerning the
relationship between original and translation, between (features of) the text(s)
and how they are perceived by human agents, and the consequences views of
these relationships have for determining the borders between a translation and

other textual operations. All three questions implicitly touch upon the crucial
concept of equivalence in translation, which was therefore briefly discussed
in this Chapter as well. This discussion will be resumed in the next Chapter,
in which my original model for translation quality assessment will be presented.

f ♦« ’
• \ *.fr-^.‘ i-J. : i ».,!!»/,; . ' * 1 •

('■'' A' ■ ’ ♦?»»'. ■-' 3> I (• A- Kl 'V', ’

.f-^' •‘■‘Jd'JW’fti’ - •■ ’• < , I

<I " ^“ ' ■ -

# v^:


■ m - ■ $^

.|.w «r-r# ia

:'ir ‘J -t# iji

>• • •' ■>j ij® Smhi^t'-,

' - ^ f1 !*?«’ 1

,♦ . ' ••.‘i ' tc^c

rv. 4ia' r . •' ^

• . • tf f , . lij.V » k»i»^ *“ir

• ■ -.». ■ • J— r-. . *•>4 »46 • ■

' ♦ J‘ ' . •*

> ■' j<r.

.. *' ■ •■ jpct c-

“ - ".W ♦ife^


1 11 X

I ' 4

The Original Model for Evaluating Translations

1 Fundamental Concepts
1.1 Equivalence and Meaning in Translation
The model (House 1977, 2nd ed. 1981) is based on pragmatic theories of lan-
guage use, and it provides for the analysis of the linguistic-discoursal as well
as the situational-cultural particularities of the source and target texts, a com-
parison of the two texts and the resultant assessment of their relative match.
The model draws on pragmatic theory, on Hallidays functional and systemic
theory, on notions developed inside the Prague school of language and lingui-
stics, on register theory and stylistics as well as discourse analysis.
The model is also based on the notion of "equivalence”: translation is
constituted by a “double-binding” relationship both to its source and to the
communicative conditions of the receiving llnguaculture, and it is the concept
of equivalence which captures this relationship. The concept of equivalence
as presented above in Chapter 1 is differentiated in my model according to a
empirically derived distinction into overt and covert translation, concepts
which will be discussed in greater detail later in this Chapter, but which may
be defined for an Intitial purpose as follows: in overt translation, the function
of the translation is to enable its readers access to the function of the original
in its original linguacultural setting through another language. This means,
that there can be no simple functional equivalence, rather a type of “second
level” function must be posited, which allows the translation receptor a view
of the original through a foreign language while clearly operating in a different
discourse world. By contrast, the function of a covert translation is to imitate
the original’s function in a different discourse frame, a different discourse
world (Edmondson 1981). In this case, an equivalence is sought in and via the
vessel of the new language for the function that the original has in its lingua-
cultural setting. One of the means of achieving this functional equivalence is
through the employment of a cultural filter, with which shifts and changes
along various pragmatic parameters (e.g. the marking of the social role relatl-
onship between author and reader) are conducted. Note that this crucial dis-
tinction into overt and covert translation is a dine, not an “either-or” dichot-
omy, and that it goes some way towards getting out of the double-bind in that
a relative leaning towards original or translation is implicit in the distinction.
The notion of equivalence, on which the model is based, is related to the
preservation of “meaning” across two different languages. There are three
aspects of that “meaning” that are particularly Important for translation: a
semantic aspect, a pragmatic aspect, and a textual aspect of meaning. I will deal
with them in turn.

1) The semantic aspect of meaning consists of the relationship of reference or

denotation, l.e., the relationship of linguistic units or symbols to their referents
in some possible world, where possible world means any world that the human
mind is capable of constructing. This definition takes account of the fact that
semantically meaningful utterances occur even though the terms of that ut-
terance have no referent in the real world as is for Instance the case in science
To a large extent, the nature of the universe (l.e., the subjective interpre-
tation of possible worlds) is common to most llnguacultures; the referential
aspect of meaning is the one which is most readily accessible, and for which
equivalence in translation can most easily be seen to be present or absent. This
relative ease of access is one of the reasons why it has been given preference
in many earlier linguistic treatments of translation.

2) While semantics studies the relationships between signs and deslgnata, be-
tween “words” and “things”, with the elements of sentences which are theo-
retical constructs being construed into propositions, pragmatics “is the study
of the purposes for which sentences are used, of the real world conditions
under which a sentence may be appropriately used as an utterance” (Stalnaker
1973: 380). Pragmatics thus relates to the correlation between linguistic units
and the user(s) of these units in a given communicative situation. Pragmatics
is about meaning in speech situations (Leech 1983: 34), as it is manifest in social
acts “outside” sentences, and about the “making of meaning as a dynamic
process. Involving the negotiation of meaning between speaker and hearer, the
context of utterance (physical, social, linguistic) and the meaning potential of
an utterance” (Thomas 1995:22). Pragmatic meaning can also be said to belong
to what Widdowson (1973 etc.) and Edmondson (1981) have referred to as
discourse, l.e., “the communicative use of sentences in the performing of social
actions” (Widdowson 19‘73: 69), It is also possible to view so-called “conno-
tatlve meaning”, l.e. the “communicative value an expression has...over and
above its purely conceptual content” (Leech 1974: 14) as part of pragmatic

The distinction between semantic meaning and pragmatic meaning un-
derlies the theory of speech acts developed initially by Austin (1962) and Searle
(1969). Pragmatic meaning is here referred to as the Illocutionary force that
an utterance is said to have, i.e. the particular use of an expression on a specific
The Illocutionary force of an utterance is to be differentiated from its
propositional content, i.e. the semantic information that an utterance contains.
The illocutionary force of an utterance may often be predicted from gramma-
tical features, e.g. word order, mood of the verb, stress, intonation or the pre-
sence of performative verbs. In actual speech situations, it is, however, the
context which clarifies the illocutionary force of an utterance.
Since translation, which handles language in use, i.e. parole, is clearly con-
cerned with instances of acts of speech, considerations of illocutionary force
or pragmatic meaning are of great Importance for translation. In effect, in
translation we do not operate with sentences at all but with utterances, i.e.,
units of discourse characterized by their use-value in communication. In cer-
tain types of translation then, it is both possible and necessary to aim at equi-
valence of pragmatic meaning at the expense of semantic meaning. Pragmatic
meaning overrides semantic meaning in these cases. And we can then consider
a translation a primarily pragmatic reconstruction of its original.

3) The textual aspect of meaning which is to be kept equivalent in translation

has been stressed already by Catford (1965) and also by Gleason (1968). They
had recognized that translation is also a textual phenomenon. In order to ex-
plain this statement, we have to define what is meant by “text”. A text is any
stretch of language in which the individual components relate to one another
and form a cohesive whole. A text is thus a linkage of sentences into a larger
unit. Various relations of co-textual reference take place in the process of text
constitution, e.g., theme-rheme sequences, occurrences of pro-forms, substi-
tutions, co-references, ellipses, anaphora. It is these different ways of text con-
stitution which account for the textual meaning that should be kept equivalent
in translation.
The importance of the textual aspect of meaning has often been neglected
in evaluations of translations, although the necessity of achieving connectivity
between successive sentences in another language while at the same time re-
taining the semantic meaning conveyed in the original is Important especially
in overt translation.
Given these three aspects of meaning which are considered relevant for
translation, I can now proceed to give a (tentative) definition of translation:
translation is the replacement of a text in the source language by a semantically
and pragmatically equivalent text in the target language. Equivalence I take to
be the fundamental criterion of translation quality. Thus, an adequate trans-

lation text is a pragmatically and semantically equivalent one. As a first requi-
rement for this equivalence, it is posited that a translation text has a function
equivalent to that of its source text. However, as we shall see below (p. 65ff)
this requirement needs to be further differentiated given the dine between
overt and covert translation.
Such a use of the concept of function presupposes that there are elements
in any text which - given appropriate analytlq^l tools - can reveal that text s
The use of the term “function” in this context is open to misinterpretation,
mainly because different language functions can co-exist inside what will here
be described as an Individual text’s function and because language functions
have often been directly, and in my opinion Incorrectly, correlated with textual
types. In the following various investigations of language functions and their
applications will be briefly discussed in order to sharpen the crucial notion of
text function used in the model presented here.

1.2 Functions of Language versus Functions of Texts

“Function” can be seen as a fundamental principle of language, in other words,

it can be Interpreted as something going beyond the notion of mere “use of
language”. It is something that is basic to the evolution of human language,
and the organisation of any natural language can be explained in terms of the
functions human language has (Halliday and Hasan 1989: 17). Many different
classification schemes for the “functions of language” have been proposed.
Only some of the most Influential ones will be reviewed to provide some clarity
about the way the concept of “function of language” has been used in the

1) Based on his work on meaning and the context of situation and culture,
Malinowski (1923) classified the functions of language into two basic ones:
the pragmatic and the magical or ritual function, the latter being associated
with religious and ceremonial activities in the culture. The pragmatic or prac-
tical function was further subclassified as active and narrratlve. It is broad
enough to cover what is called the symbolic or representational function in
other classlficatory systems.

2) Ogden and Richards (1946), in their classic work The Meaning of Meanings
differentiate five functions of language:
(1) symbolization of reference
(2) expression of attitude to listener
(3) expression of attitude to referent
(4) promotion of effects Intended

(5) support of reference (1946: 227).

Having grouped together functions (2), (3), (4) and (5) which form a complex
of “emotive functions”, Ogden and Richards (1946: 229) go on to differentiate
two basic uses: the symbolic use of language and the emotive or evocative use
of language. In the symbolic use of language, the essential considerations are
the correctness of the symbolization and the truth of the reference; In the
emotive-evocative use of language, the character of the attitude aroused In the
addressees Is of prime Importance.

3) Karl Biihler (1965:28ff) made use of a conceptual framework inherited from

Plato’s distinction of first person, second person and third person derived from
his rhetorical grammar (l.e., the organization of the verbal system of many
languages around a category of person, speaker, addressee and everything else).
Biihler distinguished three basic functions in his “organon model of language”,
each linked to one of the three variables of his model: (1) the “Darstellungs-
funktlon” (representational or representative function) is linked to objects and
relations in the real world; this function serves to describe extralinguistic rea-
lity; (2) the “ Ausdrucksfunktlon” (emotive-expressive function) linked to the
speaker/writer of the message and (3) the “Appellfunktion” (conative functi-
on) which is centered on the receiver of the message. According to Biihler, the
representational function is the central or unmarked function which is present
in any message (except in a few interjections); the other two functions are
marked functions.
As with Ogden and Richard’s model, we can thus again recognize a fun-
damental division into the absolutely necessary symbolic function and addi-
tional functions.

4) One of the most intuitively plausible as well as best argued models of the
functions of language is the one developed by Roman Jakobson (1960: 353ff).
Jakobson starts out from Biihler’s model taking over Biihler’s three basic func-
tions, but he adds three more functions and fits all of the resulting six functions
into the following schema of verbal communication: The addresser sends a
message to the addressee; the message requires a context (extralinguistic world)
referred to by the addresser, a code at least partially in common to addresser
and addressee, and a contact, a physical channel or psychological connection
between addresser and addressee. From orientations towards addresser, ad-
dressee, or context, Jakobson derives the three Biihlerian functions. From an
orientation towards contact, Jakobson derives a phatic function - this function
is predominant if a message has the predominant purpose of establishing, pro-
longing or discontinuing communication. When speech is focussed on the
code, it has a metalingual function. The poetic function in Jakobson’s model
consists of a focussing on the message for its own sake.

However, even in this elaborate six-function model, the basic dichotomy
between the primary referential function and all the other “non-referential”
secondary functions still holds. Jakobson states that each of the six factors that
he has Isolated “determines a different function of language. Although we
distinguish six basic aspects of language, we could, however, hardly find verbal
messages that would fulfill only one function. The diversity lies not in the
monopoly of some one of these several functions. The verbal structure of a
message depends primarily on the predominant function” Qakobson 1960:

5) Dell Hymes (1968: 115 ff.) sets up a typology of language functions which
is very similar to Jakobson’s, he adds however a new seventh function, the
contextual (situational) one. As opposed to Jakobson, he does not subscribe
to the view that any given message can be analysed in terms of a hierarchy of
functions such that one function is dominant: “The defining characteristic of
some speech events may be a balance, harmonious or conflicting, between
more than one function. If so, the interpretation of a speech event is far from
a matter of assigning it to one of the seven types of function” (1968: 120).

6) Karl Popper (1972), in an attempt to justify the existence of his three worlds
and especially “World Three” as the world of “objective contents of thought”,
of “knowledge without a knowing subject”, postulates a progression from
lower to higher functions in the evolution of human language. He distinguishes
four functions of language: an expressive function (using language to express
internal states of the individual), a signalling function (using language to com-
municate information about internal states to other individuals), a descriptive
function (using language to describe things in the external world) and an ar-
gumentative function (using language to present and evaluate arguments and
explanations). The expressive and signalling functions of language correspon-
ding to what Halllday refers to as the “Interpersonal function”, are uppermost
in more primitive communicative systems, the descriptive and argumentative
functions are those that were responsible for accelerating the evolution of
human knowledge.

7) Halllday (1973; Halliday and Hasan 1989) distinguishes three language

functions - he calls them metafunctions of systemic theory - that are very
similar to the ones distinguished by Popper: the Ideational, the Interpersonal,
into two subfunctions, the experiential function, to be thought of as the real
world as it is apprehended in our experience, and the logical function, through
which language expresses the fundamental logical relations of the semantic
system) language functions as a means of conveying and interpreting experi-
ence of the world, l.e., it expresses content.

Halllday’s ideational function (with its experiential and logical functional
components) thus corresponds to Popper’s descriptive and argumentative
functions. In its interpersonal function, language acts as an expression of a
speaker’s attitudes and his influence on the attitudes and behaviour of the
hearer. Through the interpersonal function, language also serves as a means
for conveying the speaker’s relationship with his Interlocutor(s), and for ex-
pressing social roles including communication roles such as questioner and
respondent. Halliday thus seems to merge Popper’s signalling and expressive
function in his Interpersonal function, and also Biihler’s “ Ausdruck” and “ Appell”
functions, collapsing the speaker and hearer ends of the communication cycle.
Through the textual function, language makes links with Itself and with
the situation: the construction of texts becomes possible because of this link-
age. It Is a kind of “enabling function”, a “resource for ensuring that what Is
said Is relevant and relates to Its context ’’(Halliday 1989: 45). The textual
function is of a different status from the two other functions in that there Is
no corresponding function in the sense of “use”, and because of this one might
argue, as did Leech (1983: 57ff), that It should not really be called a function
at all.
Halliday’s functional theory thus differs from the other approaches men-
tioned above, in that only the ideational and the Interpersonal functions are
comparable to the notion of function used in the other approaches as a basic
mode of language in use. Halliday’s textual function - and here I agree with
Leech (1983) - really relates to a different intra-language level, to a level of
internal organization of linguistic items. Viewed in this way, Halliday’s model
also seems to confirm the basic split of language use Into a referential or con-
tent-oriented function and a non-referentlal, interpersonal function.
This basic division into a cognitive function and an expressive/emotlve-
connatlve function is, of course, paralleled by the customary division of
meaning Into cognitive (or denotative) meaning Including concepts which
people have with regard to the content of verbal communication, and emotive,
connotative meaning covering the emotional reactions which people have with
regard to various linguistic forms.
Having reviewed some ways of characterizing language functions, we are
now ready to examine how language functions have been related to textual
functions. On the assumption that a text is a stretch of language, the simple
equation of textual function with one of the above-mentioned language func-
tions (the “dominant” one) has frequently been undertaken and, as a further
step, textual function has been taken to be the basis for textual type. In the
context of translation studies, a pioneer work and an influential approach In
Germany is Reift (1971) and more recently Rei£ and Vermeer (1984) and Relb
(1989), who took Biihler’s three language functions as determining three dif-
ferent textual types: the referential, the emotive-expressive and the conatlve-

persuasive textual types. Such an equation of language function and textual
function/type is overly simplistic: given that language has functions a to n,
and that any text is a self-contained instance of language, it should follow that
a text will also exhibit functions a to n, and not - as is presupposed by those
who set up functional text typologies - that any text will exhibit one of the
functions a to n (e.g “der Informative Texttyp”). I believe that if the notion of
a functionally based text typology can have an^ empirical validity, it can only
be a probabilistic one as the ground for placing any text Inside text type A can
only be that this particular text exhibits language function A to a greater extent
than it exhibits other language functions. In other words, while some extremes
may be readily characterized, there is a dine between such extremes. This
simplistic probabilistic text typology based on a predominant language func-
tion exhibited in the text is of no use in terms of determining an individual
text s function, let alone of establishing functional equivalence. However, such
a typology may be a useful basis for selecting and classifying texts for analysis
as well as for providing convenient pre-sclentific labels for the two co-present
components of any text’s function which must, of course, be specified more
precisely for each individual text as will be shown in the model to be charac-
terized below. I therefore use the traditional dichotomy of the two broad
(pre-analytical) functions, which were established to be prevalent in all the
theories of functions of language reviewed above, for choosing and grouping
a sample of texts and for labelling the two components of the textual function
discovered in the individual texts. I adopt Halliday’s terms “ideational” and
“interpersonal” as labels for the referential and the non-referentlal functional
components (see here Lux 1981, who used Hallldayan functions as an ordering
frame for “Textsorten” in a way similar to the one I had suggested).

2 Towards a Model of Translation Quality Assessment

In order to characterize the function of an individual text, “function” must,

as we have seen, be defined differently from “functions of language”. I define
the function of a text very simply as the application or use which the text has
in the particular context of a situation. (Lyons, 1969: 434). Establishing such
a function of an individual text Involves the characterization of its “textual
profile”, which results from a systematic linguistic-pragmatic analysis of the
text in Its context of situation. The phrase “context of situation” is critical here
and needs further elaboration.
Context originally means literally “con-text”, l.e., that which is with the
text, and what is “with the text” goes beyond what is said and written. It
includes the situation in which a text unfolds and which must be taken into
account for the text’s interpretation.

The notion of “context of situation” goes back to the anthropologist Bro-
nislaw Malinowski (1923), who in trying to solve the difficulty of translating
texts from a culture (the Trobriand Islands) very different from Western cul-
ture, operated with the notion of a text “in its living environment”, l.e., the
environment enveloping the text, necessary for understanding and interpre-
ting it. While “context of situation” refers to the immediate environment of a
text, the notion of “context of culture” refers to the larger cultural background
to be taken into account in the interpretation of meaning. These ideas were
taken up by Firth (1959), who Integrated them into his own linguistic theory,
in particular into his view of meaning as a function of context.
Firth (1959) set up a framework for describing the context of situation
that contained the participants in the situation, the action of the participants,
the effects of the action and other relevant features of the situation. This seminal
work Inspired different concepts for describing the context of situation. One
of the most well-known and influential ones that is at the same time very
similar to Firth’s concepts is Hymes’ (1968) conception of the “ethnography
of communlcaton”, in which he considers the following factors for describing
a text’s embeddedness in the context of situation: the form and content of the
message, the setting, the participants, the intent and effect of the communica-
tion, the key, the medium, the genre and the norms of interaction.
The basic idea here is that “context of situation” and text should not be
viewed as separate entities, rather the context of situation, the context in which
the text unfolds, “is encapsulated in the text, not in a kind of piecemeal fashion,
nor at the other extreme in any mechanical way, but through a systematic
relationship between the social environment on the one hand, and the func-
tional organisation of language on the other.” (Halliday 1989: 11).
But how do we get from the context of situation to the text, and vice versa?
How is a text to be characterized in terms of its context of situation? Or,
coming back to our definition of a textual function as being the use a particular
text has in its context of situation, precisely how should we go about deter-
mining this function? If we stress the fact that any text is embedded in a unique
situation, it follows from this that in order to characterize its textual function,
a text must be analysed at the appropriate level of delicacy. For the particular
purpose of establishing functional equivalence between a source text and a
translation text, the source text has to be analysed first such that the equivalence
which may be sought for the translation text can be stated precisely. Since
textual function was defined as the use of a text in a particular situation, each
individual text must be referred to the particular situation enveloping it and
for this a way must then be found for breaking down the broad notion of
“situation” into manageable parts, i.e., features of the context of situation or
“situational dimensions”.

A large number of models (often similar or overlapping) have been sug-
gested featuring such dimensions as abstract components of the context of
situation, e.g., in the British Firthian tradition, Halliday et al (1964), Catford
(1965), Gregory (1967), Crystal and Davy (1969), and Gregory and Carroll
(1978). When I reviewed these models at the time of designing the original
model of translation evaluation. Crystal and Davy’s system of situational di-
mensions seemed the most differentiated and ^aborate one, and I took it as a
starting point for the purposes of correlating situations and texts. Crystal and
Davy’s system of “situational constraints”, in which the notion of situation is
broken down into manageable, l.e. analysable parts, thus served as the basis
for my eclectic model of analysis of the two texts Involved in translation.
Crystal and Davy’s scheme is as follows:

A Individuality
B Discourse
a. (Simple/Complex) Medium (Speech, Writing)
b. (Simple/Complex) Participation (Monologue, Dialogue)
C Province
(Crystal and Davy 1969: 66)

Under A, Individuality refers to the idiosyncratic features of language as used

by an individual in unselfconsclous utterance, i.e., features which Identify
someone as a specific person, e.g. a person’s handwriting, voice quality or
certain pet words or phrases which are over-represented in his speech. Dialect
refers to features which mark an author’s geographical origin (regional dialect),
where the unmarked case is the national standard language, or his position on
a social scale (social dialect), where the unmarked case is the construct of the
educated middle class speaker of the standard language. Time refers to those
features which provide clues to a text’s temporal provenance. The three di-
mensions listed under A constitute relatively permanent and stable features
pertaining to the language user.
The features under B are self-explanatory. It is only the simple/complex
option that needs explanation: the category Medium may function in a “re-
moved” or “explanatory”.way whenever Medium is being used as a means to
an end rather than as an end in Itself, l.e., the category is conceived as a tem-
porary device meant to facilitate a later switch to the alternative category. This
phenomenon is called Complex Medium (e.g. language which is “written to
be spoken”, with possible further subclassification) as opposed to the Simple

Medium where language stays within one category, i.e., “spoken to be heard”
or “written to be read” (in the sense of “not read aloud”).
The category Participation may also be complex, i.e., when a text produ-
ced, by only one person (a “monologue”) nonetheless contains features which
would normally be assumed to characterize a dialogue e.g., imperative forms
or question tags.
Under C, Province reflects occupational or professional activity. Examples
of Province would be “the language of advertising”, “the language of science”,
etc. Crystal and Davy point out that “province features should not be identi-
fied with the subject matter of an utterance, as has sometimes been suggested
in connection with the notion of ”register“. Subject matter insofar as this is a
question of the use of distinctive vocabulary, is but one factor among many
which contribute to a province s definition” (1969: 73).
Status is the term used for the relative social standing of the speaker/writer
and listener/reader in terms of formality, respect, politeness, intimacy, etc.
Modality refers to differences in the form and medium of communication such
as the differences between a report, an essay, or a letter. Modality thus corre-
sponds roughly to the traditional term “Genre”. Singularity is a term for oc-
casional personal Idiosyncrasies which are said to differ from features of In-
dividuality in that they are usually deliberately introduced into a person s
speech in order to achieve a specific linguistic effect.
For my purpose of constructing a model for situational-functional text
analysis and assessment of translation, I eclectically adapted Crystal and
Davy’s model in the following manner:
I collapsed the three sections A, B, and C into two sections: “Dimensions
of Language User” and “Dimensions of language use”, featuring the following

A. Dimensions of Language User

1. Geographical Origin
2. Social Class
3. Time
B. Dimensions of Language Use
1. Medium: simple/ complex
2. Participation: simple/ complex
3. Social Role Relationship
4. Social Attitude
5. Province

In the section “Dimensions of Language User” (A), I considered Crystal and

Davy’s two factors under Dialect separately as Geographical Origin and Social
Class respectively. These two dimensions as well as the dimension Time are
defined in the sense of Crystal and Davy. I omitted the dimension of Indivi-

duality on the grounds that the text producer s idiosyncratic linguistic features
would be captured in other dimensions.
In the section “Dimensions of Language Use” (B), the following modifi-
cations of Crystal and Davy’s model were made:

1. Medium : simple/ complex

I refined the category of Complex Medium by drawing on the delicate distinc-
tions suggested by Gregory (1967) with respect to writing only due to the
nature of the task of translation quality assessment, which Involves written
texts only:

to be spoken as not necessarily

if not written to be spoken to be spoken

to be read as if

(adapted from Gregory 1967: 189)

These distinctions between different combinations of spoken and written mo-

des are Important and necessary because, even if a text is meant to be spoken
and is, in fact, at some stage spoken, there is still a difference between genuine
spoken language (as in e.g. a conversation) and the above mentioned “spoken”
subcategories of the written mode. However, my analysis did in fact reveal
that even Gregory’s classification is a relatively unsophisticated analytical tool
for the purposes of a delicate stylistic analyses of source and translation texts.
Therefore I introduced appropriate refinements in the course of the detailed
textual analyses conducted in the original work.
In determining features of the spoken mode in the various manifestations
of a Complex Medium^ I considered phenomena such as structural simplicity.
Incompleteness of sentences, specific manner of text constitution, particular
theme-rheme sequencing, subjectivity (marked, for Instance, through the use
of modal particles and gambits), high redundancy, etc.

2. Participation : simple/ complex

A text may be either a “simple” monologue or dialogue, or a more “complex”
mixture involving, in an overt “monologue”, various means of indirect parti-
cipation elicitation and indirect addressee Involvement manifest linguistically,
for Instance, in a characteristic use of pronouns, switches between declarative,
imperative and interrogative sentence patterns or the presence of contact pa-
rentheses, and exclamations.

3. Social Role Relationship
I subdivided Crystal and Davy’s dimension Status into two categories: Social
Role Relationship and Social Attitude. Under Social Role Relationship I ana-
lysed the role relationship between addresser and addressees, which may be
either symmetrical (marked by the existence of solidarity or equality) or asym-
metrical (marked by the presence of some kind of authority). In considering
the addresser’s social role vis-a-vis the addressee(s), account is further taken
of the relatively permanent position role (teacher, priest) and the more transient
situational role (visitor in a prison, speaker at a given occasion).

4. Social Attitude
Under this dimension I described the degrees of social distance or proximity
resulting in relative formality or Informality. I adopted the distinctions of dif-
ferent styles suggested by Joos (1961), which consists of five different styles
or levels of formality: frozen, formal, consultative, casual and intimate. In the
actual analyses, I provided for the possibility of transitional styles such as e.g.
consultative-casual. In Joos’ schema the most neutral style is consultative. It
is the norm for conversations or letters between strangers and it is mostly
marked negatively, i.e., through the absence of both formal and Informal style
markers. In using consultative style, the addresser does not assume that he can
leave out certain parts of his message - which he might be able to do in a
socially closer relationship where much of the message is “understood”. In
consultative style, the author has to be fairly elaborate in supplying back-
ground information. A further characteristic of consultative style is the parti-
cipation of the addressee(s) - hence the term “consultative” - either directly
or implicitly.
Casual style is especially marked by various degrees of implicitness, in
which the addresser may indulge because of the level of intimacy between
himself and the addressee(s). Background information is not necessary: casual
style is used with friends or “Insiders” of all kinds with whom the addresser
has something to share or desires or imagines that there is something to share.
Ellipses, contractions, and the use of lexical items and collocations marked
[- formal] are characteristic linguistic markers of casual style.
The consultative and the casual style levels which can also be called col-
loquial styles are used to deal with public information. By contrast, intimate
style excludes such public Information, it is the language used between people
who are personally very close to each other with a maximum of shared back-
ground information being available. Its major feature is referred to as “extrac-
tion” by Joos, i.e. an extreme type of ellipsis.
Formal style deviates from consultative style in that addressee participa-
tion is to a large degree omitted. Formal texts are well-structured, elaborate.

logically sequenced, and strongly cohesive. They clearly demonstrate advance
Frozen style, like intimate style an extreme style, is the most formal, pre-
meditated, often “literary” style. Frozen texts may be consummate products
of art meant for the education and edification of the readers, but it may also
be used in business letters, in which the social distance between writer and the
reader is thus given expression. ^

5. Province
The dimension of Province is more comprehensive in my scheme than in Cry-
stal and Davy’s model, as I subsumed both their categories Province and Mo-
dality under Province. Province is thus very broadly defined referring not only
to the text producer’s occupational and professional activity but also to the
field or topic of the text in its widest sense of “area of operation” of the language
activity as well as details of the text production as these can be deduced from
the text Itself (the notion of “register” is relevant here).
I omitted Crystal and Davy’s dimension Singularity just as I omitted In-
dividuality and for the same reasons.
Returning now to the earlier discussion of a textual function which is to
be kept equivalent in translation, it is now posited that the function of a text
can be determined by opening up the linguistic material (the text) in terms of
the set of situational constraints discussed above. The evidence in the text
which characterizes it on any one particular dimension is, of course. Itself
linguistic evidence.
The situational dimensions and their linguistic correlates are then consid-
ered to be the means by which the text’s function is realized, l.e., the function
of a text is established as a result of an analysis of the text along the eight
situational dimensions as outlined above. The basic criterion of functional
match for translation equivalence can now be refined: a translation text should
not only match its source text in function, but employ equivalent situational-
dimensional means to achieve that function, l.e., for a translation of optimal
quality it is desirable to have a match between source and translation text along
these dimensions which are found - in the course of the analysis - to contribute
in a particular way to each of the two functional components, ideational and
interpersonal, of the text’s function.
By using situational dimensions for opening up the source text, a parti-
cular textual profile is. obtained for the source text. This profile which char-
acterizes the function of the text is then the norm against which the quality of
the translation text is to be measured, l.e., a given translation text is analysed
using the same dimensional scheme and at the same level of delicacy, and the
degree to which its textual profile and function match or do not match the

source text’s is the degree to which the translation text is more or less adequate
in quality.
This was an outline of the provisional theoretical model, which had, in
the original model, the status of a hypothesis to be tested with a corpus of texts.

3 Operation of th*e Model

I will now briefly describe the method of operation of the model, outlining
the method of analysing and comparing texts by indicating how the various
situational dimensions of the model are realized syntactically, lexically, and
textually, drawing eclectically on a number of concepts deemed useful for the
establishment of linguistic correlates to the situational dimensions. Further, I
will briefly present the evaluation scheme for the measurement of mismatches
between source and translation texts. Finally, the method of operation of the
model will be justified.

3.1 Method of Analysis and Comparison of Texts

Starting from the assumption that in order to make qualitative statements

about a translation text (TT), TT must be compared with the source text’s (ST)
textual profile which determines the norm against which the appropriateness
of TT is judged, the first task in my model is a detailed analysis of ST. Using
the set of situational dimensions as outlined above, it is necessary to establish
text-specific linguistic correlates of the situational dimensions.
The grammatical model used for the analysis is a Neo-Firthlan one. In
seeking to extend the descriptive power of the model I also made use of the
convention of expressing the components of meaning by means of feature
symbols such as [+/- human], [+/- abstract]. I also made use of rhetorical-stylistic
concepts such as alliteration and anacoluthon, concepts from speech act and prag-
matic theory, discourse analysis as well as the concepts of “foregrounding”
and “automatization” developed by Prague school linguists (see Flavranek
1964, who coined the terms). Foregrounding is a linguistic device for making
the reader conscious of a particular linguistic form such that the linguistic form
itself attracts attention, and is felt to be unusual or “de-automatlzed”, as is the
case, for instance, in alliteration, assonance, omatopoeia, puns, and wordplays.
Automatization is the opposite of foregrounding referring to the conventional,
“normal” uses of the devices of language such that the linguistic forms them-
selves do not attract special attention.
On each of the situational dimensions, I differentiated syntactic, lexical
and textual means, although it might not always be the case that all three

categories are found to be operative on a particular dimension. As all the texts
considered in the analyses were written texts, phonology did not play a role.
Important in the conception in this model was the inclusion of textual means;
which were not considered in Crystal and Davy’s approach, nor were they
generally given much attention at the time I was working on the original model.
In fact, one of the more serious objections (e.g. by Widdowson 1973) to the
Crystal and Davy approach is that they proc^ded too atomlstlcally as they
were only concerned with breaking down stretches of language into their con-
stituent linguistic elements without seeking to establish the meaning construed
via different ways of sentence connections, thematic movements etc. This ob-
jection is not valid for my own approach, as I did take account of textual
I based my treatment of textual means of realizing a particular situational
feature eclectically on Enkvist’s work on linguistic stylistics (1973), on work
done in the Prague school on theme-rheme distribution and on the Insightful
work on texts in spoken and written language by Soil (1974), as well as on
Edmondson’s early work on discourse analysis later (1981) to be published in
his model for discourse analysis.
In eclectically adapting strands of the above research, I distinguished three
main textual aspects:

1) Theme-dynamics
Theme dynamics charts the various patterns of semantic relationships by
which “themes” recur in texts (e.g. repetition, anaphoric and cataphoric ref-
erence, pro-forms, ellipsis, synonymy, and near-synonymy) and takes account
of “functional sentence perspective”, a concept first used by Matheslus (1971).
For my purposes, the notion of functional sentence perspective was rather
slmplistlcally interpreted as follows: any utterance consists of two basic parts
which differ in the function they have in carrying Information: (a) the theme
which refers to facts either taken for granted, universally known, or given
from the context, and which therefore do not, or only marginally, contribute
to the new information conveyed by the total utterance; (b) the rheme con-
taining the main “new” information conveyed by the utterance. Word order
is the primary formal means of realizing the theme-rheme distribution: in
“normal”, unmarked speech, the theme precedes the rheme (Matheslus’ “ob-
jective position”), in emotive speech, however, the rheme precedes the theme
(“subjective position”).

2) Clausal Linkage
Clausal Linkage is described by a system of basically logical relations between
clauses and sentences in a text, e.g. additive, adversative, alternative, causal,
explantory, or illative relations.

3) Iconic Linkage
Iconic Linkage or structural parallelism occurs when two or more sentences
in a text cohere because they are, at the surface level, isomorphic.
Following Soil (1974: 51), I distinguished between two basic types of text
constitution which in analogy to a distinction Introduced by Pike (1967) are
referred to as “emlc” and “etlc” texts. An emlc text is one which is solely
determined by text-lrrfmanent criteria, and an etic text is one which is deter-
mined through text-transcending means, i.e., temporal, personal, or local delc-
tics pointing to various features of the situation enveloping the text, the ad-
dresser, and the addressee(s).
I also considered textual features such as the overall logical structure, the
presence of narrative or other routine formulae, and the presence or absence
of redundancy.
Following the analysis of ST, TT was analysed in the same manner, and
the two resulting textual profiles were compared for their relative matching.
In the presentation of the results of the analysis of TT, I restricted myself to
listing the mismatches along the various dimensions.

3.2 Evaluation Scheme

If a translation text, in order to be adequate, is to fulfill the requirement of a

dimensional, and as a result of this, a functional match, then any mismatch
along the dimensions is an error. Such dimensional errors were referred to as
covertly erroneous errors. These were differentiated from those overtly erro-
neous errors which resulted either from a mismatch of the denotative meanings
of source and translation text elements or from a breach of the target language
system. Cases where the denotative meaning of elements of ST were changed
by the translator were further subdivided into omissions, additions, and sub-
stitutions consisting of either wrong selections or wrong combinations of ele-
ments. Cases of breaches of the target language system were subdivided into
cases of ungrammaticality, i.e., clear breaches of the language system, and cases
of dubious acceptability, i.e., breaches of the norm of usage which I defined
as a bundle of linguistic rules underlying the actual use of language as opposed
to the language system, which is concerned with the potentialities of a language.
Both groups of overtly erroneous errors have traditionally been given
more attention whereas covertly erroneous errors, which demand a much
more qualitative-descriptive, in-depth analysis, have often been neglected. The
relative weighting of individual errors both within the two categories and
across them is a problem which varies from individual text to individual text.
The final qualitative judgment of a translation text consists, then, of a
listing of both covertly and overtly erroneous errors and of a statement of the

relative match of the Ideational and the interpersonal functional components
of the textual function. The notion that a mismatch on a particular situational
dimension constitutes a covert error presupposes:

1) that the socio-cultural norms, or more specifically the norm-conditioned

expectations generated by the texts, are essentially comparable. Obvious
differences in the unique cultural heritage must, of course, be stated ex-
plicitly and discussed in each particular te^t;
2) that the differences between the two languages are such that they can
largely be overcome in translation, i.e., basic inter-translatability between
the two languages is assumed. Again, exceptional cases such as the non-
availability of the German Du-Sie distinction in English must be stated
explicitly and treated as exceptions;
3) that no special secondary function is added to the translation text, i.e.,
works translated for special audiences (e.g., classical works “translated”
for children") or special purposes (e.g., “interlinear translations” which
arc designed for a clarification of the structual differences between the
two languages involved) arc explicitly excluded. Such translations arc no
longer translations but are defined as overt versions of an original text.

Given these three presuppositions, it is thus assumed that the addressees of a

translation text form a comparable sub-group in the target community to the
sub-group formed by the addressees of the source text in the source language
community both being defined as speakers of the contemporary standard lan-
guage, i.e., that supra-regional variety which is (commonly) used by the educa-
ted middle class speaker and which is at the same time accepted by the majority
of the whole language community.

3.3 Justification of Method

Apart from using the objectively fixed set of situational dimensions as a sort
of tertium comparationis, this method of determining the appropriateness
of a TT depends of course on the analyst s intuition and on the intuitive judg-
ments of further judges asked to help substantiate certain points. This approach
of relying on the analyst’s (as a native speaker, a near-native speaker, and at the
same time an expert in translation) seemed to be the only feasible method of
putting this type of model into practice.
Th is reliance on the analyst-cum-cxpcrts’ evaluation of a translation docs
not lead us into the vicinity of the type of nco-hcrmcneutlc approach criticized
above, because all the Intuitive judgements involved in this method arc argued,
i.e., taken as hypotheses which are being validated as objectively as possible
by the reasons given for them. The use of the fixed set of situational dimensions

and the use of authentic texts (rather than pre-fabrlcated examples), with which
the model was tested, makes my investigation more objective.
However, it is of course undeniably true that the decisions about the ap-
propriateness of linguistic elements in any TT must necessarily always contain
a subjective and hermeneutic element. Further, it is Important to stress again
that equivalence relationships between items belonging to two languages be
considered non-absolme ones falling on a dine of more or less equivalent with
a range of equivalents in both directions running from more or less probable.
(Halllday, McIntosh and Strevens 1964: 124.)
To give priority to native speaker and translation expert intuitions and
evidence gained partially through Introspection is a legitimate undertaking, if
it yields useful Insights. To be concerned with “objectivity” as a goal in itself,
l.e., to aim at strictly objective and exhaustively empirical procedures at the
cost of gaining useful insights into a phenomenon seems a futile undertaking.
In view of the strictly experimental suggestions and studies of translation eva-
luation inside so-called “response-oriented approaches”, which fail to take
account of the complexity of the phenomenon of translation, I can only repeat
that a reduction of the problems of evaluating translation to features which
can be objectively measured seems less desirable than an intensive analyls of
the kind suggested in my model.
In the last analysis, translation is a complex hermeneutic process. Trans-
lation evaluation - despite the attempt in my model to objectify the process
by providing a set of categories - must consequently also be characterized by
a necessarily subjective element, due to the fact of course that human beings
are Important variables. It seems unlikely therefore that translation quality
assessment can ever be completely objectified in the manner of the results of
natural science subjects. Within the social sciences, the method I developed
may be placed among one of the major modes of social scientific Inquiry, the
case study approach, in which an intensive in-depth examination of the many
characteristics of one unit is conducted. Case studies have been used with
benefit to supplement traditional experiments involving extensive observation
of large samples. The case study method which rests on the recognition of the
crucial Importance of specifying the complex contextual embeddedness of the
phenomenon under study, has two general purposes: (1) to arrive at a com-
prehensive understanding of the phenomenon on hand, and (2) to develop
more general theoretical statements. My work had the added purpose of ve-
rifying a scheme of concepts with which I proposed to analyse and evaluate a
set of texts and their translations.

3.4 Implimentation of the Model: Test Cases

In House (1977), the above model of translation quality assessment was tried
out with a corpus of eight English and German textual pairs, four belonging
to the ideational functional category, and four to the interpersonal functional
category. The texts covered a wide range of different “provinces”: a scientific
text, an economic text, a journalistic article and^ tourist information brochure
make up the ideational set of texts; an excerpt from a sermon, a political speech,
a moral anecdote and a dialogue taken from a comedy belong to the interper-
sonal set of texts.
In my original work, I had felt it necessary to exclude from the interper-
sonal category all those texts which may be considered to be predominantly
poetic-aesthetic or “form-oriented”, l.e., in which the form of their lingui-
stic units has taken on a special autonomous value , e.g., poems. In a poetic-
aesthetic work of art, the usual distinctions between form and content (or
meaning) no longer holds. In poetry, the form of a linguistic unit cannot
be changed without a corresponding change in (semantic, pragmatic and tex-
tual) meaning. And since the form cannot be detached from its meaning, this
meaning cannot be expressed in any other way, l.e., through paraphrase, ex-
planation or commentary, borrowing of new words etc. In poetry the slgnlfiers
have an autonomous value and can therefore not be exchanged for the slgnlfiers
of another language, although they may in fact express the same signified con-
cept or referent. Since the physical nature of slgnlfiers in one language can
never be duplicated in another language, the relations of slgnlfiers to signified,
which are no longer arbitrary in a poetic-aesthetic work, cannot be expressed
in another language. Jakobson (1966) has made the same point, which culmi-
nates in his statement that poetry is by definition untranslatable. Similarly
Nlda and Taber (1969: 4) state that “anything that can be said in one language
can be said in another, unless the form is an essential element of the message.”
Poetic-aesthetic texts are characterized by a maximum of foregrounding:
in fact, foregrounding is here used for its own sake such that it is the case that
language is then not being used “in the service of communication but in order
to place in the foreground the act of expression, the act of speech Itself” (Mu-
karovsky, 1964: 19). True, in many other texts. Indeed in some of the texts
Included in the corpus of my original work, e.g., the religious sermon, the
political speech, foregrounding also occurs, l.e., there are cases of alliteration,
wordplays etc., which, are difficult or impossible to translate. However, in
these cases, foregrounding is always subordinate to communication. The basic
purpose of these non-poetlc texts which use foregrounding occasionally is
always to draw the addressee’s attention more Intensely to the subject matter
expressed by the foregrounded linguistic item but not to the expression Itself.
This is the reason for the possibility of translation in cases of non-poetlc texts.

In a predominantly poetic-aesthetic text, however, the limits of translatability
are reached: a TT is then no longer a translation but a kind of creative trans-
To exemplify the operation of the model for the benefit of readers not
familiar with the 1977 book, here are two of the original analyses (see Appendix
for the actual texts).

II. COMMERCIAL TEXT (ST English; TT German), see Appendix.

1. Analysis of ST and Statement of Function

Dimensions of language user:

(1) Geographical Origin: non-marked, Standard American English

(2) Social Class: non-marked. Educated Middle Class
(3) Time: non-marked, contemporary American English

Dimensions of language use:

(1) Medium: simple: written to he read, as realized by the following linguistic


syntactic means:
a. absence of elliptical clauses, contractions, contact parentheses and com-
ment parentheses, and any kind of spoken language signals such as well,
you see, you know, etc.,
b. placing of expanded subordinate clauses of purpose before the main clause:
this is a focussing device typical of the written mode as its use in spoken
language is restricted by performance constraints, e.g., V3,
c. presence of expanded postnominal modification resulting in the separa-
tion of the head of the subject noun phrase and the corresponding finite
verb. This construction is typical of the written mode as there are perform-
ance constraints in spoken language: IXl.

lexical means:
a. absence of qualifying modal adverblals. Interjections, and other subjecti-
vity markers typical of the spoken mode.

textual means:
a. the text is predominantly emlc. There are a few pronominal references to
the addresser and the addressees; however, the Immediate circumstances
of the production and reception of the text are clearly irrelevant for the
organization of the message. As a result of this, the text is largely deter-
mined through text-immanent criteria and is marked by an explicitness
and elaborateness typical of the written mode,

b. lack of repetitions resulting in a lack of redundancy,
c. frequent use of passivization as a typically “written” means of complex
syntactic linkage for text-constitutive purposes, i.e., especially for the pre-
servation of the theme-rheme sequence, e.g., II2, IV2, V2,4, Xi, XI2.

(2) Participation: complex: monologue with addressees being directly addres-

sed and given instructions. However, the addj;essees’ potential reactions are
not being taken into account by the addresser. The addressee-oriented nature
of the text is thus limited to the direct address and the request for action. This
characterization is manifest in the following linguistic means:
syntactic means:
a. presence of second person personal and possessive pronouns for direct
address: Ii, Vi,3,4, VIi,2, VIIi,
b. presence of requests put to the addressees through the use of the verb
require in the passive, modal auxiliaries of obligation, and the mandative
subjunctive in a that-c\iuse.\ V3,4, VIi,2,
c. absence of interrogative sentences. This is indicative of the predominantly
monologous character of the text which - with the exception of the par-
ticipatory devices listed above in a and b - allows for no direct (even
Imaginary) participation of the addressees.

(3) Social Role Relationship:

Asymmetrical role relationship: addresser has de facto economic authority
over the addressees.
Position role of addresser: president of an international financing com-
pany, of which the addressees are shareholders.
Situational role of addresser: representative of the Interests of the company
Informing the shareholders about recent developments in the company.
The role relationship as manifest in the text may be detailed in the follo-
wing way: in the interests of his company (I.O.S.), the president is diploma-
tically indirect, non-committal, and evasive, avoiding any statement of an as-
sumption of direct responsibility for the new VCL-company on the part of
I.O.S. The relationship is an impersonal, distant one: the shareholder is not
being approached as an individual but as a type, as a member of the class of
shareholders. The text’s role relationship is further characterized by a delibe-
rate attempt on the part of the addresser to downplay his own and the com-
pany’s power status and give the addressees the illusion of possessing more
Influence than they really have. The addresser flatters the addressees and tries
to create a feeling of security, loyalty, and trust in the well-being of the com-
pany. This characterization of the text’s role relationship has been derived
through an examination of the following linguistic means:

syntactic means:
a. use of second person singular personal pronoun you and possesive pro-
noun your in a specific way, i.e., for addressing corporate members not
“persons” as such (witness the substitution of each shareholder {or you,
e.g., in II3): Vi,3,4, VIi,2, VIIi. Notable is the use of these personal and
possessive pronouns in “flattering contexts” only, i.e., in connection with
possible rights, actions, etc., on the part of the addressees,
b. use of the first person plural personal pronoun we to refer to the addresser
or the company (I.O.S.) or the Board of Directors, i.e., the addresser avoids
referring to himself as an individual (although the letter is personally
signed): V4, VIi,
c. frequency of impersonal constructions using impersonal it and existential
there as well as passives: the use of these devices is Indicative of a desire
on the part of the addresser to be cautious and “hedgy” and to avoid
specifying a causer or agent. In using these devices, the addresser also
intends to give the addressees the impression that it is not the company
that requires them to do something (e.g., fill out a form) but that they are
agents of their own free will merely obeying some ulterior abstract neces-
sity: V3,4, VII, VIIl,4, VIIIl, Xi, xili,
d. preponderance of [-human] subject noun phrases adding to the imperso-
nal character of the text: IIi,2, IIIi,2, IVi,2, Vi,2,3, VI2, VII2, etc.,
e. use of subjunctive in a r/7<^f-clause: VII - asked that you designate as op-
posed to the alternative: asked you to designate. This is a marked choice
in English. The effect of the that+V subjunctive construction is such that
the addressee is not the direct recipient of a request or command by the
addresser but is left his own free agent. In other words, this structure has
the illocutionary force of a suggestion whereas the structure asked you to
designate would have the illocutionary force of a request in the context
of this text.

textual means:
a. deliberate attempt to underplay the role of I.O.S. through putting I.O.S. in
non-focussed position in prepositional phrases IV2, VIIIi,2, XI2,
b. deliberate overall organization of the text such that the addressees are first
being presented with the change as a fait accompli and its many positive
sides, and that they are only later (paragraph IX) being given the reasons
(negative ones) for the change.

(4) Social Attitude

Consistent with the Impersonal, distant relationship as outlined above, the
social attitude of the addresser towards his addressees as reflected on the level
of style, is a formal one:

syntactic means:
a. frequency of complex noun phrases showing both multiple premodifica-
tion, postmodification, and discontinuous modification which add to the
text’s abstractness and impersonality. Examples may be found in nearly
every sentence, therefore a specific listing is unnecessary,
b. deletion of conjunction if plus subject-auxiliary inversion: XIIi ... than
would have been possible, had those operations ...,
c. completeness of clauses (no elliptical clauses); absence of contractions (cf.
d. 'frequency of impersonal constructions using it, there, and passives; pre-
ponderance of [-human] subject noun phrases: use of subjunctive in a
that-clause (for all of which see above Social Role Relationship).

lexical means:
a. presence of words and phrases marked [+formal] due to their restricted
use in Impersonal - in this case, business - situations: e.g.. Ill - declared,
payable on and after, shareholders of record as of the close of;V}- expedite
the distribution', IX2 - precluded the maintenance of, XIIIi - Very truly
yours, a [+formal] letter closing formula, etc.
b. absence of interjections, qualifying modal adverbials and other subjecti-
vity markers (cf. above, Medium).

textual means:
a. frequent use of passlvlzation as a means of complex syntactic linkage spe-
cifically for preserving theme-rheme sequence (cf. above Medium).

(5) Province
Commercio-financial circular letter issued by the president of an international
financing company to the company’s shareholders. In this letter, the sharehol-
ders are being Informed about changes in the set-up of the company. The
preliminary label “language of commerce” with which one usually associates
a preciseness in giving data of all kinds, textual cohesion, and explicitness,
especially explicit allowance for possible alternative interpretations to avoid
potential (costly) misunderstandings, can be further explained and justified
by examining the use of the following linguistic features in this text:

lexical means:
a. use of precise technical terminology, l.e., special commercio-financial le-
xical items and collocations, e.g., IIi - pro-rata, dividend, holding com-
pany, IVi - stockholder's equity, IV2 - historical earnings, performance,
b. presence of phrases which precisely define the information given or ex-
plicitly state conceivable alternatives: \\\ - on and after December 20,1971;

to all shareholders of record as of the close of business on December 17,
1971; VII - a bank (or broker); VI2 - stamp (or seal),
c. absence of foregrounded words and expressions, and of any kind of figu-
rative language. '

textual means:
presence of strong textual cohesion due to the employment of several mecha-
nisms of theme-dynamics and clausal linkage:
a. repetition of lexical items, e.g., IIi,2 - dividend; Hi,2,3, - share; IIIi,2 - con-
tribute, contribution; Vi,2,3, and VII - certificate(s) etc.,
b. frequency of anaphoric referencing by means of pro-forms for noun
phrases, adverbials, predicates, clauses or sentences, e.g., III2 - in return
for its contribution; in turn; that company; all of these shares; V2 - That is;
V4 - this, etc.
c. organization of thematic movement in sequences of theme-rheme to in-
sure given-new ordering, e.g., II1-2; Vi-2-3, V4; VI1-2, VII1-2.

clausal linkage
achieved through logical connectors: II3 - of course; IV2 - since; V2 - That is,
VIIIi - as a result of; VIII2 - therefore, etc.

Statement of Function
The function of the text consisting of the two components - ideational and
interpersonal - may be summed up in the following way: the addresser s in-
tention is (a) to Inform the addressees of a collection of facts as precisely and
efficiently as possible and to request action; (b) to establish a positive rapport
with the addressees, to convince and reassure them of the appropriateness and
advantages of certain moves by the company, to give the addressees a feeling
of importance and power, and at the same time to always attempt to be indirect
and non-committal as to the moves announced and their potential consequences.
This summary statement of the text’s function has been derived by an
examination of the ways in which the dimensions are marked in this text, and
the manner in which they contribute to the two functional components:
On the dimension Medium, the written to be read mode supports the
ideational component of the text’s function by facilitating a condensed, unin-
terrupted and premeditated information flow unimpeded by any direct pre-
sence of the addressees in the act of communication. Similarly, on the dimen-
sion Participation, the lack of addressee participation, l.e., the Infrequency of
addressee-involving structures, also acts in support of the ideational compo-
nent by making for a linear, non-alternating and premeditated organization
of the message. However, on the same parameter, the few attempts at involving

the addressees by addressing them directly, and by putting requests to them,
do support the interpersonal component of the textual function.
On the dimension Social Role Relationship, the impersonality of the
relationship reinforces the ideational component by promoting an economical
transmission of facts disregarding the social circumstances of addresser and
addressee. However, equally strongly supported on this dimension is the in-
terpersonal component: the same linguistic dcA^ces which create the imperso-
nality are also used to “manipulate” the addressees; e.g., the avoidance of a
specification of a responsible causer or agent is used to give the addressees the
illusion of their obeying an abstract necessity and not I.O.S.’ interests. Further,
the attempt at flattering the addressees, which we discovered on this dimension
also obviously filters into the Interpersonal component of the textual function.
The dimension Social Attitude, which we defined as formal, operates in
support of the ideational component of the textual function in that the fre-
quency of complex, abstract noun phrases and Impersonal structures, and the
exclusive presence of complete, well-planned and well-structured sentences
provide for an efficiently condensed and objective Information flow.
The dimension Province, marked by the use of clearly defined, automat-
ized technical terminology, an explicit consideration of alternative interpreta-
tions of certain terms, and strong textual cohesion, clearly supports the idea-
tional component of the text’s function as well.

2. ST and TT Comparison and Statement of Quality

Mismatches on the following dimensions have been discovered as a result of
the analysis of TT and the comparison of ST and TT:

TT lacks the excpllclt Involvement of the addressees in a few Instances:
Vi(2) - as you know ^ bekanntlich; VI i - your dividend certificate die
Aktienzertifikate VI2 - Your hank ^ die Bank\ VIIi your new company ^ die
Value Capital Limited.

Social Role Relationship

TT IS in certain specified Instances less reassuring and flattering and less non-
committal and diplomatically indirect vis-a-vis I.O.S.’ role and responsibility:
IV2 - TT uses active voice: this has the effect of stressing I.O.S. as theme,
which is undesirable in this context, because it is thus more strongly suggested
that I.O.S. is important with respect to VCL’s future earnings. In view of I.O.S.’
fate, this is certainly not reassuring to the addressees. Vi(2) - as you know ^
bekanntlich: TT is less implicitly flattering to the addressees. V3,4(3) - your
assistance is required... for your completion ^ bitten wirSie ... auszufiillen. TT
is more direct and forceful. ST expresses the action to be done by the addressees

more abstractly and indirectly (nominally); the utterance in ST has the illocu-
tionary force of a subtle suggestion, while the utterance in TT has one of a
request. ST tries to suggest that it is not the company that wants something
done, but that some extqrnal necessity suggests a course of action to the ad-
dressees. VIi - your dividend certificate ^ die Aktienzertifikate: TT does not
make an attempt to create in the addressees an idea of their own possessions
and is thus less implicitly flattering. VIi - asked that you designate ^ hahen
wirSie gebeten: In ST, the addressees are not direct recipients of a request, but
are left agents of their own initiative. The utterance has the Illocutionary force
of a subtle suggestion; TT lacks this nuance and is thus less careful and indirect;
the illocutionary force in TT is one of a request. VIi - will he sent ^ geschickt
werden sollen: In ST, the relative clause is a non-restrlctive one, l.e., the sending
of the certificates follows automatically from the naming of the bank, and the
sending is the company’s responsibility. In TT, the relative clause has to be
understood as a restrictive one, such that the Instruction that the certificates
should be sent to the designated bank, is the shareholders’ responsibility.
Hence ST is more reassuring, while TT undiplomatically throws the onus onto
the shareholders. VI2 - your hank (or broker) should indicate Sie miissen die
Bank (oder einen Makler) bitten: The lack of the possessive pronoun renders
the expression in TT less Implicitly flattering; also, the illocutionary force of
the utterance in TT is, mainly through the use of the modal rniissen, one of an
order. The addressees thus appear to be dependent on the addresser. Such an
illocutionary force is directly opposed to the cautious and diplomatic tenor
in ST. VIIi - your new company ^ die Value Capital Limited: TT is less impli-
citly flattering, i.e., it fails to suggest that the addressees are “owners of the
company”. VIIi - ST’s impersonal it-clause, which reinforces the non-com-
mittal and detached tenor of the text, is not matched in TT, which features
Value Capital Limited as agent. TT gives an impression of greater certainty,
which is unwarranted given the evasive, impersonal structure it is anticipated
in ST. VII4 - present intention ^ z.Z. (zur Zeit): TT’s expression z.Z. has the
negative connotation of temporariness and fickleness, which is undesirable
given the addresser’s Intention of reassuring his addressees and building up
their good-will.
VIIIi - ST’s Impersonal there-clause is rendered in TT by a “personalized”
construction featuring I.O.S. as subject-agent. Xi - new facilities being estab-
lished ^ von neuen Einrichtungen: TT suggests that these facilities are, at the
time of utterance, already established. TT loses the be+V-ing connotation of
’being set up right now’, a subtle difference, but in TT the addresser again
appears to be less non-committal, and carefully evasive. XIi - ST focusses on
Value Capital Limited which is in theme-position; the role of I.O.S., from whose
failures the addressees’ attention is to be detracted, is thus underplayed. In TT,

Value Capital Limited appears in non-focussed position after 1.0.S. is men-
Social Attitude
TT is in very few instances less formal, i.e. - consistent with the findings on
the dimension Social Role Relationship - TT appears to be less distant, and
more personal and direct:
II4 - er hleiht natiirlich. In this position, n^tiirlich gives the sentence an
almost colloquial tone. Initial position of natiirlich or the use of the [+formal]
selhstverstdndlich would have been more adequate.
V3 - your assistance is required ^ bitten wirSie: TT is more personal, i.e.,
less socially distant and formal. VI2 - Sie miissen die Bank bitten - a persona-
lized, Informal expression. VIIi - ST’s impersonal it-clause: it is anticipated
that is not matched in formality by TT s more direct, non-lmpersonal struc-
TT is, in a few Instances, less clear, precise, and less textually cohesive than ST:
VI2 - Ihre Unterschrift auf dem Dividenden-Zustellungsformular zu be-
stdtigen - the prepositional phrase auf dem ... is ambiguous; it may either be
an adjectival or an adverbial phrase of location, i.e., it may either qualify be-
stdtigen or Unterschrift. Thus TT appears to be less unambiguously clear. IXi
- TT does not preserve the theme-rheme sequence as it starts the clause with
the rheme thus losing the textual linkage to the preceding paragraphs. IX2 -
TT lacks ST’s anaphoric noun phrase these operations (a consequence of the
different thematic organization of IXi in TT). XIi beabsichtigt diese - unde-
sirable ambiguity of the anaphoric pronoun diese's referent.
Overtly erroneous errors:
There are two mismatches of the referential meanings of ST and TT items:
II2 - wrong selection: newly established holding company ^ eine nachdem
Recht der Bahamas neu gegriindete Gesellschaft Bahamische Holding Gesell-
schaft would have been more adequate). VIIi - wrong selection: it is anticipated
^ wird die Value Capital Limited. The choice of the future tense in TT does
not express the uncertainty of an anticipation (the adverbial voraussichtlich
should have been Included).
Further, we discovered one breach of the target language system, to be
subcategorized as a case of dubious acceptability:
III2 - erhielt die I OS 6.2 Mill. Aktien ... die alle von der I OS ... This is a
confusing and illogical structure because I.O.S. is the subject of the main clause
and it appears in a prepositional phrase in a passivized relative clause. Hence
passlvlzation serves no real purpose as it does not omit the agent. We claim
that this structure is counter-intuitive, and of dubious acceptability. The fol-
lowing similar example seems to confirm our assumption: Each of us received

$20 which was spent by each of us on the spot. This example - and III2 in the
present text - is only acceptable if the agents in the main clause and the passi-
vized relative clause are non-identical.

Statement of Quality
The comparison of ST and TT along the eight parameters shows that there are
mismatches on all dimensions of language use but Medium; however, by far
the greatest number of mismatches occur on the Social Role Relationship
parameter rendering TT in the specified Instances less flattering to the addres-
sees, less diplomatically polite and deliberately non-committal, i.e., blunter
and more direct. Clearly, the interpersonal component has been altered
through these mismatches. The few mismatches on the dimension Participa-
tion which result in TT’s involving the addressees less directly and explicitly
in a few (for the addressees positive) Instances also detract from the interper-
sonal functional component. The few mismatches on Social Attitude which
render TT less formal also alter the interpersonal component of the textual
function by making TT less socially distant, and carefully polite. The mismat-
ches on Province, which result in TT being less unambiguously clear and
textually cohesive, as well as the three overtly erroneous errors affect the idea-
tional component of ST s function by detracting in these few Instances from
a clear and efficient passing on of information.
From this configuration of mismatches, it becomes clear that, while the
ideational component of ST s function is violated to a minor degree only, ST s
Interpersonal functional component is violated to a considerable extent as evi-
denced by the pattern of mismatches along the dimension of Social Role
Relationship. Thus, we may say that with regard to the addresser’s Implicit
attempt at giving the addressees a feeling of Importance and his desire to be
non-committal, indirect, and diplomatic about the consequences of the chan-
ges in his company, TT has serious shortcomings which we have specified in
detail above.

III. JOURNALISTIC ARTICLE (ST English; TT German), see Appendix.

1. Analysis of ST and Statement of Function

Dimensions of language user:

(1) Geographical Origin: non-marked. Standard American English
(2) Social Class: non-marked. Educated Middle Class
(3) Time: non-marked, contemporary American English

Dimensions of language use:

(1) Medium: complex: written to he read as if heard. There are at least two
ways in which this text may have been produced: a) the text is an adaptation

of lecture notes or even an (edited) transcript of a lecture given by the addresser,
or b) the text has been specially prepared to appear as though it originated in
the manner of a) above because of the addresser s assessment of the addressees’

The above considerations are substantiated by the following linguistic


syntactic means: ^
a. presence of anacolutha: IIi, IVi
b. presence of elliptical structures: II 3,4
c. presence of loosely-structured clauses featuring parenthetical and appo-
sitional structures either inside the main clause or extraposed, thus creating
an impression of lack of premeditation, typical of the spoken mode: IVi,
VI2, VIIl2,3.4.

textual means:
a. the text is predominantly etlc, l.e., showing links to the particular parti-
cipants in the situation in which it is embedded through the frequent use
of deictic personal pronouns, l.e., we, involving addresser and addresses
together, and you in direct address. See especially paragraph II, in which
the addressees’ physiognomy is referred to for Illustration purposes,
b. looseness in the logical structuring of the text, which is indicative of the
lack of premeditation typical of the spoken mode (and which may have
been introduced deliberately into this text to simulate this Medium).

(2) Participation: complex: monologue with direct address and personalized

Instructions indicative of addressee participation.

syntactic means:
a. use of first person plural personal and possessive pronouns as inclusive
terms; involving the addressees directly in order to heighten their Interest
in the subject matter: Ii, IIIi,2, IVi, VII2, VIIl2,3,
b. use of second person singular personal and possessive pronouns as direct
address forms: IIl2,3,4,
c. presence of rhetorical, addressee-directing utterances such as: VIIIi -
What brought the split about}: A rhetorical question through which the
addressees are invited to participate in the argument put to them. VIII3 -
we cannot simply say: which constitutes an admonition of the addressees.

(3) Social Role Relationship:

Asymmetrical role relationship: addresser has professional authority over the
addressees. Position role of addresser: professor of anthropology.

Situational role of addresser: writer of an article for the general lay public
about his field of specialization.
The role relationship in this text may be characterized as follows: the
addresser is fully aware-of the “authority gap” between himself and his ad-
dressees in terms of knowledge of his special field, and he does not make any
attempt at concealing this fact. This results in a sometimes slightly conde-
scending tone. The following linguistic means have suggested this charac-

syntactic means:
a. presence of structures in which instructions given to the addressees acquire
a connotation of mild condescension: VII2 - or between animals properly
called'. The addresser informs the lay public of the true way of referring
to a certain type of animal in a schoolmasterly manner. VIIIi - What
brought the split about}'. Rhetorical question exemplifying the distinction
between the informed and the uninformed. VIII3 - we cannot simply say:
The addresser points out to the unenlightened that things in his field are
not as simple as their common sense might suggest they are.

textual means:
a. insertion of a whole paragraph (II), clearly with illustrative, educational

(4) Social Attitude

Consultative-casual, an informal style which might be glossed as ’conversa-
tional, friendly, chatty’. This style level seems to be consistent with the role of
an Instructor who adopts a mildly condescending tone vis-a-vis his addressees,
i.e., the addresser “steps down” to reach the addressees at what he assumes is
their level:

syntactic means:
a. presence of anacolutha, elliptical structures as well as parenthetical and
appositional structures (cf. above Medium),
b. simplicity of noun phrases, i.e., lack of multiple premodification or post-
c. use of ’s-genitive with a [-human] object. This results in personification
of the object and achieves an informal, personalized style level: I2 -India’s
Siwalik Hills.

lexical means:
a. use of abbreviations for educational institutions. I2 - of Yale., V\-at Yale,
b. presence of words and phrases marked [-formal] due to their common
occurrence in more intimate social situations, e.g.. Ill - these things', III2

- belong in (Instead of belong to); IVi - put away in a drawer; Vi- was (instead of investigating or examining); VIIi - on one hand
(omission of article the); VII2 - anything on the human side; VII3 -
looking for (instead of investigating., etc.); VIIIi - What brought the split

(5) Province:
Journalistic science article written by a special!^ for the general, non-specialist
public. The article is published in a general, “Readers Digest” type magazine
which is characterized by its instructional Informative nature. Descriptions of
scientific discoveries are not so much factual reports but attractive stories.
There is an all-pervading tendency of personalization, dramatization, and con-
cretization of scientific facts. Typical of this “popularization of science” is a
concern with linking general life experiences to abstract scientific notions.
The following linguistic means are evidence for this description:

syntactic means:
a. use of be+Ving forms suggesting an interest in the process of discovery
involving a human being (as opposed to a concern with the bare facts of
the discovery): Vi, VI2, VII3,
b. use of 's-genitive for personalization of [-human] noun phrase (cf. above
Social Attitude): I2 - India’s Siwalik Hills,
c. use of the Instantaneous present to achieve a dramatic, theatrical quality:
VII2 - thus we also see (the logical connector thus also adds to the dramatic
forces of the clause),
d. use of a rhetorical question for dramatic force (cf. above Social Role Re-
lationship): VIIIi,
e. omission of the article preceding a noun in apposition. This results in
giving the appositive the status of a title and achieving a bombastic, dra-
matic effect: VII1 - Dryopithecus, ancestor of the apes;
f. use of personal and possessive pronouns (cf. above Participation) when-
ever the addressees are being Invited to relate scientific facts to their own
range of experience: I1/III2 - our ancestor/our ancestry; and entire para-
graph II.

lexical means:
a. use of figurative language, e.g., IIIi - as though he had just set his foot on
a path (personification of species); IVi - the tide of scientific opinion...was
against Ramapithecus (cliche metaphor plus personification); VIi-Simons
rescued...Ramapithecus from burial (personification: only animate beings
can be burled), etc.
b. use of scientifically imprecise words and phrases which are typical of the
Province of popularized science text, e.g., Ii - and pointed to some man-

like features; II] - these things; IV2 - after almost 30 years; VI1 - other
pieces of Ramapithecus; VI2 - in various places from the US. A. to India...
a few more fragments; VII2 - anything on the human side, etc. All these
items provoke a logical follow-up question: Which/what exactly...}
c. popularized way of giving bibliographic references, i.e., lack of precision
in the data given: I2 - G. E. Lewis of Yale; V1 - Elwyn Simons at Yale; IV2
— after almost 30 years, L. S. B. Leakey found a similar fossil at Fort Ternan.
Reference is being made to the place where the researcher worked but not
to the publication in which he describes his findings.

textual means:
a. use of an introductory formula (as in a fairy tale) to achieve dramatic force:
V\- it happened that at the same time;
b. use of cleaving as a device for giving thematic prominence to the new
element in a clause and adding a dramatic note to the utterance: \\-It was
out of..that...and, in fact it was;
c. presence of textual cohesion through repetitions: IIIi - foot; IVi - tide(s),
facts; and iconic linkage: It - it was; II3/4 - an ape's are longer ...
yours is straighter. These features add to the attractiveness of the text and
thus help to catch the addressees’ Interest.

Statement of Function:
The function of the text consisting of an interpersonal and an Ideational com-
ponent may be summed up in the following way: the addresser’s intention is
to Inform and Instruct the addressees about scientific facts in such a way that
the material presented to the addressees is made “non-technical”, attractive,
Interesting, and easily digestible in order to suit the addressees’ level of know-
ledge and understanding of the subject matter.
The Interpersonal component of the textual function is well marked on
each dimension, whereas the ideational component is never visibly marked,
but is, of course, implicitly present because the text obviously aims at passing
on factual Information to the addressees. The individual dimensions operate
in the following way: On the Medium dimension, we found that the text has
a complex mode: written to be read as if heard. This mode acts in support of
the interpersonal component of the text’s function by promoting a direct in-
volvement of the addressees, suggesting that the addresser is speaking to the
addressees (i.e.. Implying his immediate presence). The interpersonal functio-
nal component is also supported by the dimension Participation because of
the anticipated or implied addressee participation manifest in the frequent use
of personal and possessive pronouns and the presence of addressee-directing
utterances in the text. Similarly, on the dimension Social Role Relationship,
the presence of structures showing a connotation of condescension and the

insertion of a whole paragraph of addressee-related illustrations of facts also
clearly support the interpersonal functional component. The Social Attitude
dimension is marked for consultative-casual style, a style level suitable for
informally passing on information to a lay public, l.e., specially geared to the
addressees’ assumed need of a colloquial presentation. This dimension there-
fore also filters into the interpersonal functional component.
On the dimension Province, the attempt aNdramatizlng and personalizing
scientific facts also promotes the Interpersonal component by making the ma-
terial attractive and palatable to the addressees.

2. ST and TT Comparison and Statement of Quality

Mismatches on the following dimensions have been discovered as a result of
the analysis of TT and the comparison of ST and TT:

TT has fewer features characterizing the spoken component of the complex
medium written to be read as if heard:
a. TT lacks anacolutha: IIi, IVi,
b. In two Instances, TT lacks parenthetical and appositional structures
which, in ST, express the type of afterthought or “in-between commen-
tary”, typical of the lack of premeditation in the spoken mode: VIIl2(3),3(6).

TT appears to be less concerned with deliberately involving the addressee:
a. lack of personal pronoun we, l.e., lack of addressee - inclusion: IVi, VII2,
b. lack of second person singular personal and possessive pronouns: IIi,2,3,4.
The use of wir {{or you) in IIi invites addressee participation less markedly.
The use of Mensch {{or you) in II3 puts the utterance onto a more gene-
ralized, l.e., less personalized level, which is not desirable here,
c. absence of rhetorical, addressee-directing utterance: VIII3 - the utterance
in TT lacks the subtle overtones of the utterance in ST where the adressees
are being warned against having a too facile approach to the subject matter.

Social Role Relationship:

TTd oes not exhibit ST’s slightly condescending tone. Also, the relationship
projected in TT is a more symmetrical one, witness the following linguistic
a. lack of connotation of condescension which we established to be present
in ST: VII2 - the phrase properly called is omitted in TT; VII3 - whole
utterance is omitted in TT. This is perhaps the most striking illustration
of the fact that the social role relationship as portrayed in TT is markedly

different: the jovial, rather trivial remark in VII3, which is clearly designed
to make the addresser’s special field palatable to the Ignorant lay public, is
omitted in TT. TT thus appears to indicate a more symmetrical relation-
ship. VIIl3(5) - omission in TT of the personal pronoun, i.e., use of an
impersonal es-structure; TT is therefore more neutral and less conde-
scending. Further, the naive value judgement expressed in better is omitted
and the meanings of better and more successful are collapsed into vorteil-

Social Attitude:

TT’s style level is, in a considerable number of instances, more formal:

syntactic mismatches:
a. lack of anacolutha and parenthetical structures (cf. above. Medium);
II2/II4 - Derjemge des Affen, jene der Affen: demonstrative anaphoric
pronoun marked [+formal] in German; III2 - use of subjunctive in TT
and of conjunction obschon, both of which are marked [+formal]; use of
past tense instead of present perfect, the former differing from the latter
in terms of formality; presence of extensive prenominal modification in
TT: in eine vom Dryopithecus abweichende Richtung, which renders TT
more abstract and formal; IV3 - das Fossil, als eine weitere Art von Dryo-
pithecus abgetam -ed participle clause in which the antecedent head is
Identical with the deleted subject of the -ed post-modifying clause. The
participle is firmly linked with the passive voice, which makes the expres-
sion impersonal and [+formal]; VIII5 - es ware...gleichsetzen zu wollen:
impersonal es-structure plus infinitive in TT, which is typical of a formal
style level.

lexical mismatches:
Il4(5) - this length makes an ape's face projecting die Lange der Molaren
bedingt, dafl\ the structure is marked as f+formal] in German; IIl2(3) - So Lewis
thought ^ dies bewog bewiegen is strongly marked as [+formal] in German);
W\i- anything on the human side ^Arten mit Ansdtzen von menschenartigen
Merkmalen: TT’s phrase is much more precise and formal; VIII2 - evolution
has reasons ^ Die Evolution folgt einem Plan; VIII7 - lichte Walder, gewisse,
teils: all of these items are marked [+formal] in German.

TT shows in many instances fewer traces of personalization, dramatization,
and concretlzation through illustrations and imagery. TT is therefore less true
to the Province of “popularized science text”; it is also less journalistically
attractive, and less vague and imprecise in its descriptions of scientific facts.
Consider the following mismatches:

syntactic mismatches:
a. lack of personal and possessive pronouns rendering the tcxtless addres-
see-oriented (cf. above Medium),
b. TT lacks equivalent for he+Ving form, e.g., V+a modal adverbial such as
gerade, nun, etc., thus falling to emphasize the process of discovery by a
human being: Vi, VI2, VII3 (altogether omitted in TT),
c. 4
dramatic quality achieved In ST through> elctlon of the definite article
preceding a noun phrase In apposition Is lost in TT: VIIi (an equivalent
structure In German is available),

lexical mismatches:
a. lack of scientifically Imprecise words and phrases: IIi - omission of an
equivalent structure for these things in TT; II3- molar teeth ^ Molaren, a
scientific term in German (as opposed to the general term Backenz'dhne),
VII2 - anything on the hitman side ^ Arten mit Ansatzen von menschen-
artigen Merkmalen (cf. above. Social Attitude); apes ^ Affen ttnd Men-
b. lack of, or use of less evocative, figurative language: path diverging
- abweichende Richtung; IVi - tide of scientific opinion ^ Meinitngen der
Wissenschaft; VIi - rescued ...from burial - verhinderte ... daf: VIIJ2 -
Evolution has ''reasons” - Die Evolution folgt einem Plan', i.e., TT is far
less personalized. Plan, in this case, being something pre-existent and im-
personally fixed.

textual mismatches:
a. lack of double cleaving in TT which adds dramatic quality to the ST ut-
terance, and lack of iconic linkage: Ii - TT has foregrounded word order
only in the second part of the sentence, therefore the emphatic effect of
iconic linkage is lost,
b. lack of equivalent dramatizing Introductory formula in TT for \\ - It
happened that,
c. omission of German equivalent for the logical connector thus in VII2,
which adds to the utterance’s dramatic effect (although it really lacks a
logical foundation).

Overtly Erroneous Errors:

There is one mismatch of denotative meaning to be subclassified as a wrong
selection: 11-4(5)- makes an ape's face projecting, yours is straighter ^ bedingt,
daf das Gesicht der Affen - im Vergleich zum menschlichen - an der s ist.

Statement of Quality:
I he comparison of ST and TT along the situational parameters shows that
there arc mismatches on all five parameters of language use. Our analysis of

ST has shown that, while the Interpersonal component of the textual function
is marked on all dimensions, the ideational component is-although implicitly
present - not specifically marked. This relative markedness of the two com-
ponents is different in TT.
On the dimension Medium, we found that ST s mode written to be read
as if heard is not always matched in TT, which appears to lack several “spoken
mode elements”. On this dimension, the interpersonal component is less marked,
whereas the ideational component is strengthened because the informational con-
tent is transmitted more straightforwardly. Similarly, on Participation, TT
appears to be less geared to elicit addressee participation; therefore, the inter-
personal component is weakened while the ideational one is strengthened
through a greater concentration on the information to be transmitted. On the
dimension Social Role Relationship, the addresser’s attitude towards the ad-
dressees appears to be less condescending, but more neutral and symmetrical.
Again, the ideational component is stronger in that TT is more obviously
designed to pass on factual information than to adapt the facts to the particular
needs of the addressees.
On Social Attitude, the ideational functional component is again strength-
ened at the expense of the interpersonal one, since TT’s style level is, in certain
specified instances, less personalized and more formal.
On Province, TT appears to be less personalized, dramatized, and jour-
nalistically attractive than ST, but rather more precise and sober especially
with respect to the use of scientific terms. Again, the interpersonal component
of the textual function is violated while the ideational one is developed more
There is one overtly erroneous error, which detracts from the ideational
component of passing on factual information.
Taken together, Td’ suffers from a distinct mismatch of both the interper-
sonal and ideational components of the textual function as revealed on all five
dimensions of language use. The intention of the addresser to make his material
attractive, interesting, and easily digestible for his addressees has not been
realized in TT to the same degree as it has been in ST; the addresser’s concern
with passing on scientific facts is, on the other hand, clearly - and unwarrant-
edly - more visible in TT than in ST.

4 Refinement of the Model

4.1 Suggestions for a Translation Typology

I had suggested above that attempts at setting up a text typology as a means

of gaining insight into, and accounting for, different types of invariance de-

mands and (as a result of these) different types of translation equivalence re-
lationships are theoretically not tenable and not fruitful. Underlying such an
approach is the presupposition that translation quality is somehow determined
by the nature of the source text while the process of translation is itself a
constant, i.e., it is presupposed that if one can classify texts successfully, then
one will have successfully accounted for differences in translations and the
theoretical problems surrounding translation evaluation. This presupposition
seems to have been the basis of Relft’ early influential work (who later switched
from a source text focus to the other extreme: target text centeredness). In the
approach suggested here a different procedure seemed more likely to be in-
sightful, namely to set up a translation typology on the basis of the eight case
studies undertaken in the original study, whereby different texts were found
to have been treated in different ways in translation. There is however clearly
some relation between source text type and appropriate translation type.
I suggested a basic division into two major translation types: overt trans-
lation and covert translation.

4.1.1 Translation
An overt translation is one in which the addressees of the translation text are
quite “overtly” not being directly addressed: thus an overt translation is one
which must overtly be a translation not, as it were, a “second original”. In an
overt translation the source text is tied in a specific manner to the source lan-
guage community and its culture. The source text is specifically directed at
source culture addressees but at the same time points beyond the source lan-
guage community because it is, independent of its source language origin, also
of potential general human interest. Source texts that call for an overt trans-
lation have an established worth or status in the source language community
and potentially in other communities. Such source texts may be divided into
two groups:

1) overt historically-linked source texts, i.e., those tied to a specific occasion

in which a precisely specified source language audience is/was being addressed.
Examples from the test sample I analysed in the original corpus are a sermon
(by Karl Barth given at a Basel prison) and a political speech (given by Winston
Churchill on the steps of the townhall in Bradford in 1942);

2) overt timeless source texts, i.e., those transcending as works of art and
aesthetic creations a distinct historical meaning while, of course, always
necessarily displaying period and culture-specificity because of the status of
the addresser, who is a product of his time and culture. Examples in the test
corpus are a moral anecdote (a nineteenth century “Kalendergeschlchte” by
the well-known German author Johann Peter Hebei) and a comedy dialogue
(an excerpt from Sean O’Casey’s one-act play: “The End of the Beginning”).

Both these texts - although timeless and transmitting a general human message
- are quite clearly source-culture-specific because they are marked on the
language user dimensions (presence of a particular etat de langue and a geo-
graphical dialect respectively) and because they have independent status in the
language community through belonging to the community’s cultural pro-
ducts. Both texts are “literary” texts having the feature [+ fictional], i.e., the
texts are “sltuationally abstract” in that they do not Immediately refer to a
unique historic situation, in which both author and readers find themselves.
Fictional texts describe a kind of “fictlve reality”, which is, in every reception
by an individual reader, newly related to the specific historic reality in the
concrete situation in which the reader finds himself. The message in a fictional
text is text-contained, which gives the text its independent value: the message
“presupposes a wider context so that everything necessary for its interpreta-
tion is to be found within the message Itself” (Widdowson 1974: 203).
Both groups of source texts, then, historically-linked and timeless ones,
conventionally lead to an overt translation. The requirements for this type of
translation led to an important modification of the model of translation quality
assessment as outlined earlier: a direct match of the original function of the
source text is not possible in overt translation, either because the source text
is tied to a specific non-repeatable historic event in the source culture (for
example, Karl Barth’s sermon or Winston Churchill’s speech, both given at a
particular time and place to a particular audience) or because of the unique
status (as a literary text) that the source text has in the source culture.
In the case of texts that are bound to a specific historic occasion, it is quite
obvious that a translator cannot set out to match the original function that ST
had for the original addressees but that he must try to match what I called a
“second level function”, one that recognizes the “displaced situatlonality” of
the two texts and one not only holding for the contemporary, educated middle
class native speaker of the target linguaculture but also for their potential coun-
terparts in the source culture who are also not the original addressees.
Similarly, in timeless overt translations a second level function must be
met. For Instance, in the case of a Kalendergeschichte by Johann Peter Hebei,
we have a basically comparable case in that this text, although as a piece of
fiction timeless and of general human interest, is tied on a language user di-
mension to a specific bygone period of time in the source culture. The trans-
lator has to operate again with a level of function which this work has for the
contemporary addressee in the source culture, for whom the text would clearly
be marked as [+archaic], when it was clearly not marked in this manner for
the original addressees, who lived in the particular period of time in which the
author produced the text.
In the case of Sean O’Casey’s particularly Irish brand of comedy, it is the
fact that the text is marked for a culture-unique geographical dialect on a

language user dimension which necessitates a “topicalization” of the function.
The text must be transposed from one cultural area to another. Put more ge-
nerally, the function that the source text has for a contemporary educated
middle class standard English speaker can never be equivalent in the translation
text because of the uniqueness of the Intralingulstic variations in any particular
In cases of overt translation a similar secoq^d-level function, l.e., a kind of
“topicalization” of the original function may have to be posited as a criterion
for adequate translation. This second-level function is then the function hol-
ding for the contemporary standard language speaker of the target culture and
frequently also for their potential counterparts in the source culture, who may
also not be the original addressees.
In overt translation, the source text as a piece of work with a certain status
in the source language community must remain as Intact as possible given the
necessary transfer and recoding in another language. On the other hand, cases
of overt translation present difficulties precisely because their status in the
socio-cultural context of the source language community, which must be to-
picalized in the target culture, necessitates major changes. It is this dialectical
relationship between preservation and alteration which makes the finding of
translation equivalents difficult in cases of overt translation. The two examples
mentioned above illustrate this difficulty: in the case of the comedy dialogue
it is extremely difficult to achieve (even a) second-level functional equivalence
as it is necessary to select an “equivalent” target language geographical dialect,
l.e., a dialect equivalent in “human or social geography” (cf. Catford 1965: 88).
If we consider standard English speakers to be the majority of the potential
addressees of O’Casey s comedy, and if we regard Hiberno-English as a rather
heavily marked dialect (especially on the grammatical level), whose speakers
share a strong folk tradition and are characterized by a marked striving for
provincial, or national Independence and distinctiveness, anybody trying to
translate O’Casey’s play into German would have to search for an approxi-
mately comparable German dialect, e.g.. Bavarian. Many speakers of this dia-
lect share the separatist intentions and the “rootedness” in folk tradition with
the Hiberno-English speakers. However, it should be stressed that such marked-
ness on the dimension Geographical Origin clearly presents often Insoluble equi-
valence problems, and always entails a second level function.
In the case of the nineteenth century moral anecdote text, the presence of
a marked temporal dialect also presents considerable equivalence problems.
Again, the principle of second level or toplcalized function necessitates a se-
lection of approximately comparable [+ archaic] items in the translation text
to ensure an overt translation of the source text. An employment of markers
of a “comparably archaic” temporal dialect will create “to some extent a trans-
lation equivalent of the source language etat de langue” (Catford 1965: 89).

Catford s phrase “to some extent” clearly points to the Impossibility of achieving
a perfect match because of the uniqueness of the cultural-historical context,
and its non-transferability from the source language to the target language.
Texts that are linked io a specific historic occasion in the source language
community (such as Churchills speech and Karl Barth s sermon in the original
work s test corpus) can also present such topicalization problems (although
they did not present such problems in the orignal corpus I examined). For
instance, it might well have been the case that Churchill or Karl Barth had
spoken a regional dialect of their respective mother tongue. However, in view
of the fact that the o'yer^-historically-llnked texts have the status of a document
of a historical event in the source culture, where the culture specificity and
uniqueness is more strongly marked than in the timeless [+ fictional] texts, it
seems to be more appropriate in these cases to abstain from finding approxi-
mate equivalents for culture-specific geographical, temporal, or social class
markedness on the language user dimensions and to provide explanatory notes
to the members of the target culture who are exposed to a translation text.

4.1.2 Translation
A covert translation is a translation which enjoys the status of an original
source text in the target culture. The translation is covert because it is not
marked pragmatically as a translation text of a source text but may, conceivably,
have been created in its own right. A covert translation is thus a translation
whose source text is not specifically addressed to a particular source culture
audience, i.e., it is not particularly tied to the source language and culture. A
source text and its covert translation text are pragmatically of equal concern
for source and target language addressees. Both are, as it were, equally directly
addressed. A source text and its covert translation have equivalent purposes,
they are based on contemporary, equivalent needs of a comparable audience
in the source and target language communities. In the case of covert translation
texts, it is thus both possible and desirable to keep the function of the source
text equivalent in the translation text.
In the sample texts analysed in my original work, a scientific text (an
excerpt from a coursebook in mathematics), a tourist information booklet
(advertising brochure on Niirnberg), an economic text (a letter written by the
President of an international investment company to the shareholders), and a
journalistic text (an article on anthropology, which appeared in a popular mag-
azine the (English Language) UNESCO Courier and the (German Language)
UNESCO Kurier (the latter two texts are included in this book) exemplify
the category of source texts that led to a covert translation. All these translation
texts have direct target language addressees, for whom they are as immediately
and “originally” relevant as is the source text for the source language addres-

In the case of the economic text In the test sample, for instance, both source
and target language addressees are shareholders of the same (Internationally
operating) Investment company, i.e., they differ only accidentally in their re-
spective mother tongues. Similarly, a text taken from a mathematics course-
book is potentially of equal concern for German and English science students;
and a journalistic article on an anthropological topic Is of potential equal in-
terest to both German readers of the journal IONESCO Kurier who are Inte-
rested in anthropology and English speaking readers of the UNESCO Courier
who share this Interest in anthropology. The tourist information booklet is
(generally) as much directed at German speaking visitors of Niirnberg as It is
directed at English speaking ones.
While it Is thus clear that such texts are not source-culture specific. It is
the covert type of translation that such texts require which presents more
difficulties, and many more subtle, cultural translation problems than those
encountered In the case of overt translation, where the particular source culture
specificity had to be either left intact and presented as a culturally and histo-
rically linked monument, or overtly matched in the target culture setting. If
the source text and its translation text are to have equivalent functions, how-
ever, which is necessary in a covert translation, the translator has to take dif-
ferent cultural presuppositions in the two language communities Into account
in order to meet the needs of the target language addressees in their cultural
setting, and In order to keep the textual function equivalent in source and
target cultures. When talking about “culture” and “cultural” In this context,
I mean the anthropological concept of all those traditional, explicit and Implicit
designs for living which act as potential guides for the behaviour of members
of the culture. Culture Is seen here as a group s dominant and learned sets of
habits, as the totality of Its non-biological Inheritance involving presupposi-
tions, values, and preferences.
In a covert translation, the translator has to make allowances for under-
lying cultural differences by placing what I call a cultural filter between the
source text and the translation text. The translator has, as it were, to view the
source text through the glasses of a target culture member.
A glance at the rich anecdotal literature on translation may lead one to
believe that there are Indeed many crucial inter-cultural differences complica-
ting any translation process. However, upon closer examination, most of the
Impressive examples of differences are drawn from comparisons of a European
language and languages of South East Asia or American Indian languages, etc.,
where the socio-cultural differences are obviously remarkable. Thus Catford
(1965: 90-91) illustrates the importance of considering cultural “undercur-
rents” in translation by describing how an oriental youth may use honorific
forms when talking (at a specific occasion and on a specific matter) to his father.

and how an English speaking youth would, in the “same situation”, use only
casual style.

4.2 Distinguishing between Different Types of Translations

and Versions

I hypothesized that, in the case of European cultures such as e.g., the German
and British English ones, the differences in the soclo-cultural norms between
the two cultures were not substantial and basically knowable. Existing and
verified differences of the soclo-cultural norms and presuppositions of cultural
knowledge were to be taken care of in covert translation through the applica-
tion of a cultural filter. It then seemed to me reasonable to assume that the
contemporary Western European and North American middle class speakers
of the respective standard language, closely related through socio-political and
economic ties, did not differ in relevant ways concerning, for instance, their
reception of a scientific text, a journalistic article, or a commercial circular
letter. Unless presuppositions concerning cultural differences were substan-
tiated by ethnographic, socio-cultural and discourse research, it seemed more
reasonable in translation to follow an assumption of basic comparability for
such closely related cultures as the Western European and North American
ones than to take the liberty in translation of changing the source text on the
assumption of existing cultural differences. This did not amount to a claim
that differences in values and habits, in understanding, emphasizing or disre-
garding certain emotions, attitudes etc. did not exist between any two, however
closely related, cultures. However, given the goal of achieving functional equi-
valence in a covert translation, assumptions of cultural difference should be
carefully examined before any change in the source text is undertaken. In cases
of unproven assumptions of cultural difference, the translator might be led to
apply a cultural filter whose application resulting in possibly deliberate mis-
matches between the source text and the translation text along several situa-
tional parameters, was seen as not justified. In other words, I then advised a
“non-risk taking” strategy in covert translation when applying a cultural filter,
l.e., “when in doubt, leave it out”, or more respectably put, the unmarked
assumption is one of cultural compatibility, unless there is evidence to the
contrary. As will be demonstrated in the next Chapter, in the case of the Ger-
man and Anglophone llnguacultures, such evidence seems now to be available,
which has important consequences for cultural filtering.
In the original study, such evidence was not yet available and consequently
I had considered the translation of the commercial letter (see above) as a clear
example of unjustified filtering. In this text, the President of Investors Overseas
Services (a fraudulent company as it was revealed much later) Informs the

shareholders about changes in the set-up of the company that will not exactly
be to their advantage. Dimensional changes on the parameter Social Role Re-
lationship are responsible for the fact that the English original s carefully eva-
sive, hedging, and distantly polite tone is changed into a much more direct,
blunt, and undiplomatic tone which clearly acts against the textual function,
which is characterized in the following way: the addresser’s intention is a) to
Inform the addressees of a collection of facts as precisely and efficiently as
possible and to request action (ideational functional component), and b) to
establish a positive rapport with the addressees, to convince and reassure them
of the appropriateness of certain moves undertaken by the company, to give
them a feeling of Importance and power and at the same time always to stay
indirect and non-committal as to the moves announced and their potential
consequences (interpersonal functional component).
In the analysis of this Commercial Text and its TT, it was found that TT,
along the dimension Social Role Relationship, does not contribute to the
interpersonal functional component in the same manner: for example, as you
know is translated as hekanntlich, which is potentially less flattering to the
addressees as it does not address them personally; In order to avoid the possi-
bility of accidental misdirection of your certificate... your assistance is required. We
have enclosed a 'Dividend Instruction Form' for your completion; this should
be returned in thepre-addressed envelope... is translated as: Umzu vermeiden,
daf Ihre Zertifikate versehentlich fehlgeleitet werden ..., bitten wir Sie, das
beigefugte Dividendenzustellungsformular auszufiillen und in dem ebenfalls
beigefugten adressierten Umschlag zuriickzuschicken ...
In the German translation, the writer appears to be much more forceful,
active and direct, while the source text expresses the action to be done by the
addressees more abstractly and indirectly (nominally). The utterance in the
English source text has the Illocutionary force of a subtle suggestion, while in
the translation text it appears to be a request. The original tries to intimate that
it is not the company that wants something done, but that some external ne-
cessity proposes a course of action to the shareholders. Similarly the German
rendering of Your bank (or broker) should indicate as Sie miissen die Bank
(oder einen Makler) bitten ... clearly flouts the Interpersonal function of the
source text as characterized above: the lack of the possessive pronoun renders
the expression in the translation text less implicitly flattering; also, the utteran-
ce in the translation is an order mainly through the use of the modal ‘miissen’,
which makes the shareholders appear dependent on the President. Such an
illocutionary force is directly opposed to the cautious and diplomatic tenor
of the original.
Many similar dimensional differences could be listed, all of which change
the social role relationship between addresser and addressees. The point I made
in the original study was that an assumption of the German shareholders’

different expectations with regard to such a letter (which may have caused the
translator’s changes) was unwarranted since it was not substantiated by facts
and only acted to perpetuate the cliche assumption of German addressees
preferring such manifestations of a social role relationship. Given my own and
others’ contrastive pragmatic research, the model will now have to be “re-vi-
sited” in the light of this research.
Another example of what I had called “unmotivated mismatches” in a TT
is the case of the Journalistic Article (see above), where I observed changes on
the dimensions Social Role Relationship and Province, and consequently all
other language use dimensions. I had hypothesized - because of a whole pat-
tern of mismatches that my analysis had revealed - that the translator had
entertained presuppositions about different social role relationships between
writer and reader in the target linguaculture, in this case a professor of anthro-
pology and the interested lay public. The TT which results is markedly diffe-
rent in that a slightly condescending, lightly entertaining and very informal
tone in ST is converted into a more serious, scientifically more accurate and
certainly less entertaining TT. The presuppositions that German readers prefer
this kind of textual profile in a TT of this Province was judged to be unjustified.
Again, in the light of the results of contrastive pragmatic research, the appli-
cation of the cultural filter must be re-assessed.
From these examples of culturally conditioned differences the following
conclusions were drawn: if a covert translation accommodates, unwarrantedly
and in a patterned way, for the target culture group’s different presuppositions
about the Social Role Relationship and Social Attitude of addressees vis-a-vis
the addresser in a particular Province, then such a translation is no longer a
translation but will be defined as a covert version. A covert version is by de-
finition an inadequate translation in that the application of the cultural filter
is unjustified.
Thus, in the original analyses, the translation texts of the commercial cir-
cular and the journalistic article discussed above were judged as covert versions
because the translator - in order to preserve the function of the source texts -
had applied the cultural filter non-objectively and consequently undertook
changes on the situational dimensions. Since these changes were at the time
not substantiated by research, the translation was judged to be a covert version
of the source text.
Covert versions must be clearly differentiated from overt versions., which
are produced whenever a special function is overtly added to a TT, l.e., (1) in
particular when a translation is to reach a particular audience. Examples are
special editions for a youthful audience with the resultant omissions, additions,
simplifications, or different accentuations of certain features of ST etc., or
popularizations of specialist works designed for the lay public, and (2) when
TT is given a special added purpose. Examples are Interlingual versions or

“linguistic translations”, or resumes and abstracts, where it is the express spe-
cial purpose of the version producer to pass on only the most essential facts.
An altogether different type of systematic mismatching along the dimen-
sions of analyses proposed in the model seems to have occurred in another
example taken from the original corpus, namely the case of the Tourist Infor-
mation Booklet praising the attractions of the city of Niirnberg originally
written in German for tourists from all over Qermany and later translated for
an English speaking public. Here it was judged that the translator rightly as-
sumed that the addressees of the translation text did not necessarily share the
knowledge of the source culture’s inventory of cultural phenomena such as
nationally well-known artists or works of art. In this case, then, the supply of
added Information was deemed justified, as it seemed to be clearly distinct
from the type of assumptions of different expectations on the part of the target
culture addressees in the case of the commercial circular. The changes on the
parameter Social Role Relationship through the addition of information for
the benefit of members of the target culture, were not regarded as covert errors
but were explained through the necessity of applying a cultural filter to the
source text, which accounted for the changes.
In this case, then, the cultural filter operates differently from the cases of
the commercial letter and the journalistic article, the two texts included in this
volume: it operates primarily on the content, the semantic level, in that it adds
information in a more obvious explanatory “surface” way. In the case of the
other two texts, the changes which operate on the Interpersonal functional
component, go “deeper” and affect values, mentalities, communicative prefe-
rences and expectation norms. Differences at this level are more difficult to
handle in translation and translation evaluation, and it is only through the
type of empirical research decribed in Chapter 3 that relevant hypotheses can
be suggested.
In the discussion of different types of translation and the distinction be-
tween a translation and a version, it was implicitly assumed that a particular
text may be adequately translated in only one particular way. The assumption
that a particular text necessitates either an overt or a covert translation may,
however, not hold in every case. Thus, any text may, for specific purposes,
require an overt translation, l.e., it may be viewed as a document which “has
independent status” and exists in its own right: for Instance, the commercial
circular discussed above may be cited as evidence in a court of law, or its author
may, in the course of time, have become a distinguished political or literary
figure. In these two Instances, the texts would clearly not have an equivalent
function in translation, i.e., in both cases an overt translation would be appro-
priate, and it should be evaluated as such.
Further, there may well be source texts for which the choice overt-or-
covert translation is a subjective one, e.g., fairy tales may be viewed as folk

products of a particular culture, which would predispose a translator to opt
for an overt translation, or as non-culture specific texts, anonymously produced,
with the general function of entertaining and educating the young, which
would suggest a covert translation; or consider the case of the Bible, which
may be treated as either a collection of historical literary documents, in which
case an overt translation would seem to be called for, or as a collection of
human truths directly relevant to Everyman, in which case a covert translation
might seem appropriate.
Moreover, it is obvious that the specific purpose for which a “translation”
is required, i.e., the spelcific brief a translator is given, will determine whether
a translation or an overt version should be aimed at. In other words, just as
the decision as to whether an overt or a covert translation is appropriate for a
particular text may be conditioned by factors such as the changeable status of
the text producer, so clearly the initial choice between translating a given source
text and producing a version of it, cannot be made on the basis of features of
the text, but is conditioned by the arbitrarily determined purpose for which
the translation/version is required.
The original assumption in the model that a TT in order to be adequate
should have a function equivalent to the function of its ST had to be refined
in the light of the crucial distinction between overt and covert translations: it
is thus only in cases of covert translation that it is in fact possible to achieve
functional equivalence. This functional equivalence is, however, extremely dif-
ficult to achieve because differences in the socio-cultural norms of the two
linguacultures have to be taken into account, and a cultural filter must be
applied. As became clear from the analyses of the test corpus in the original
work, such filtering is crucial whenever the original has a well-marked inter-
personal component of the textual function. It is the interpersonal component
which presents the most difficult (and interesting) problems of translation
equivalence. In the case of the Scientific Text (taken from a mathematics text-
book) the problems were relatively reduced - precisely because the interper-
sonal component of this text s function was not strongly marked. In the case
of other texts’ calling for a covert translation (the commercial text, the tourist
brochure and the journalistic article) however, the matching of the interper-
sonal component of the textual function (especially on the Social Role Rela-
tionship dimension) clearly presented the most subtle problems for the trans-
lator who had to apply a cultural filter in his translation. From the viewpoint
of the translation evaluator, the lack of objective knowledge about differences
in the socio-cultural norms makes it difficult to assess the legitimacy of any
changes made as a result of the application of the cultural filter. Empirical
cross-cultural pragmatic research is of critical importance here in that it can
add to our knowledge of cross-cultural differences.

In the case of overt translation a second level function must be aimed at
in translation. Since in an overt translation the ST is, in a way, “sacrosanct”
due to its status (as a work of art or a historical document) the translator cannot
strive for simple functional equivalence in the target culture, which would
involve the undertaking of adjustments of cultural presuppositions, he has to
restrict himself to “simply” transposing the ST from the source to the target
culture, giving target culture members the opportunity to have access to the
original via the medium of the foreign language. Overt translations are more
“straightforward” since their STs are taken over unaltered, l.e., are merely
transplanted into a new environment with no provisions being made for the
TT addressee s (potentially different) norms of expectation. This can be de-
monstrated by the fact that in one of the texts of the original corpus, the
Religious Sermon, the two references to local source culture phenomena are
best translated by the exact source language terms “Fastnacht” and “Muster-
messe” to be explained, outside the body of the text in a footnote. The source-
culture orientation is clearly brought out through this procedure.
The difficulty of evaluation in the case of overt translation is also reduced
since considerations of “cultural filtering” can be omitted. The major difficulty
in translating overtly is, of course, the finding of linguistic-cultural “equiva-
lents” on the language user dimensions. However, here we deal with overt
manifestations of cultural phenomena which must be transferred only because
they happen to be manifest linguistically in ST. A judgement whether a “trans-
lation” of culture-specific user characteristics is adequate in an overt transla-
tion cannot be objectively given, i.e., the degree of correspondence in terms
of social status between dialects in two different cultures cannot be measured
at the present time since no completed cultural-comparative studies exist. Such
an evaluation must therefore necessarily remain to a certain degree a subjective
matter. However, as opposed to the difficulty of dealing with differences of
cultural presuppositions with respect to Social Role Relationship, Social At-
titude etc. in a particular text, which characterizes the evaluation of covert
translation, the explicit overt transference necessary in an overt translation is
still easier to pin down and diagnose.
As regards the evaluation of different translation texts of the same source
text, my model facilitates an evaluative statement only to the extent that the
relative importance of the individual situational dimensions has been demon-
strated in the analysis of the source text. A relative weighting of covertly er-
roneous and overtly erroneous errors can only be achieved through a consi-
deration of each individual textual pair. However, the subgroup of overtly
erroneous errors referred to as “mismatches of the denotative meanings of
elements of the source and translation texts”, will detract more seriously from
the quality of a translation text, whenever the source text has a strongly marked
ideational functional component, e.g., mismatches of the denotative meaning

of items in a science text are likely to be rated higher than a mismatch on Social
Attitude. A detailed hierarchy of errors for any individual case can, however,
only be given for a particular comparison of two or more texts depending in
any particular case on the objectives of the evaluation, or in the case of trans-
lation in the context of foreign language teaching, on the objectives set by the
In the original study I chose my corpus of sample texts in such a way that
they represented a wide range of different “genres”. Given the range of genres
and subject matters, I believe that I was able to demonstrate successfully the
practicality of the suggested model, and that I was able to make some valid
generalizations concerning the establishment and the nature of translation equi-
valence being closely tied to the two basic translation types I hypothesized. I
believe that these results are valid for the particular corpus I analysed but also
for other STs and TTs that can, on account of the criteria suggested in my
model, be placed along the covert-overt continuum, with clear-cut endpoints
marking texts that unambiguously call for an overt or a covert translation.
A particular ST does not necessarily require once and for all either a covert
or an overt translation, given the different, dynamic ways of viewing a text
and different purposes for which a translation may, in the course of time, be
required. However, in clarifying the distinction overt-covert, and in detailing
the consequences in translation practice and evaluation which follow the choice
of a particular predominant translation type, the original model has succeeded
in shedding some light on a theoretically problematical area of translation.
In evaluating translations, a version must be distinguished from a trans-
lation, and an overt translation from a covert translation. These categories are
designed to clarify the nature of the equivalence required for a translation of
optimal quality. In translation, this equivalence is to be one of function, func-
tion being determined by a pragmatic analysis, i.e., a detailed opening up of a
particular source text and its translation text.
Returning to the three questions (relationship between original and trans-
lation, between texts and human agents, distinction between translation and
other textual operations) asked initially (p. 1) to judge the differences in theo-
retical and empirical potential between different approaches to translation and
translation evaluation, the model presented here is firmly based on the as-
sumption that translation is a double-bind operation. As opposed to views
that show a one-sided concern with the translation, its receptors and the con-
ditions holding for the translation s reception in the target culture, the model
takes account of both original and translation by positing a dine along which
it can be shown which tie of the double-bind has priority in any particular
case, the two endpoints of the dine being marked by the concepts overt trans-
lation (source text focussed) and covert translation (target text focussed). The
relationship between (features of) the text(s) and the human agents Involved

(as author, translator, reader), is explicitly taken care of in the model through
the provision of an elaborate system of pragmatic-functional analysis of ori-
ginal and translation, with the covert-overt dine on which a translation is to
be placed determining the type of reception sought and likely to be achieved.
Finally, the model explicitly provides for the means of distinguishing a trans-
lation from other types of textual operations by specifying the conditions
holding for a translation to turn into a version^The notion of a cultural filter
which is to be applied in covert translation cases needs to be substantiated
through language-pair specific contrastive pragmatic research. In the next
Chapter I will present examples of such research relevant for German-English
and English-German cultural filtering.


Substantiating the Cultural Filter:

Evidence from Contrastive Pragmatic Discourse Research

One of the most intriguing “open questions” that emerged from the original
model, is the nature of the cultural filter to be applied in covert translation,
where the differences in communicative preferences, mentalities, and values
are taken into account in translation. In the original model, it was hypothesized
in the two exemplary analyses of covert translations (which are reprinted in
Chapter 2) that the cultural filter was not legitimately applied in these cases.
This hypothesis needed closer examination, the assumption being that research
into differences in the socio-culturally determined communicative preferences
in the two linguacultures (English and German) involved might give more
substance to the notion of cultural filter^ and consequently a better basis for
judgments of the legitimacy of the application of a cultural filter in covert
translation. In this Chapter I will therefore outline some contrastive pragmatic
and discourse studies I have conducted over the past twenty years and indicate
the results and hypotheses I have formulated on the basis of these studies.

1 Contrastive Discourse Analyses: German-English

In the seventies and eighties I conducted a number of contrastive pragmatic

analyses comparing the discourse of German and English native speakers, the
subjects being students at German and English universities. (Cf. House 1979;
1984; 1993; 1996a,b; Blum-Kulka, House and Kasper 1989; and see the refer-
ences below.) The data collection methods were open, self-directed dyadic
role plays, often followed by retrospective interviews and discourse comple-
tion tests combined with meta-pragmatic assessment tests, as well as natura-
listic data drawn from interactions between English and German native speak-
ers. In the following I want to briefly describe the discourse phenomena
Investigated and to summarize the results indicating their relevance for the

cultural filter (English-German) in the model of translation quality assess-
The analyses undertaken were based on the discourse analysis model pro-
vided in Edmondson (1981) and the interactional grammar of Edmondson and
House (1981) as well as on the categories of analysis developed within the
Cross-Cultural Speech Act Realization Project (CCSARP) (Blum-Kulka,
House and Kasper 1989). ^
The data was elicited in the framework of two projects to be described in
greater detail in what follows. The first project supported by the Deutsche
Forschungsgemeinschaft dealt with the acquisition of communicative compe-
tence by German learners of English (cf. Edmondson et al 1984). It Involved
dialogic speech, which was elicited in open role-play from German native
speakers, English native speakers and German learners of English. The dis-
course type was dyadic face-to-face conversations in everyday situations of
some interactional complexity. The role plays were based on a number of
abstract interactional bases and the role relationship between the two partici-
pants was such that the two parameters [+/-authority] and [+/-famlliarlty] led
subjects to produce a wide variety of different speech acts in different social
constellations. In this framework I conducted various contrastive discourse
analyses using the data produced by the German and the English native spe-
akers. These contrastive pragmatic analyses were conducted mainly in order
to establish the presence or absence of pragmatic differences in the verbal
behaviour of English and German speakers.
The analyses were conducted on one of the three following levels:

(1) On the most “superficial” level, those tokens were compared which cor-
respond pragmatically in the two llnguacultures, given the different sys-
tems of selection holding for the various types in the two llnguacultures.
(2) On a “deeper” level, norms of expectation with regard to certain illocu-
tionary acts and their sequencing were taken into consideration.
(3) On an even “deeper” level, the social functions of the analytic categories
were investigated, l.e., their respective positions alongside parameters such
as [+/-authority] and [+/-famlliarlty], as well as the resulting degrees of
“directness”, types of politeness, and formality.
A comparison was made in the following areas:

1.1 Discourse phases

Discourse phases, l.e., discourse opening and closing phases were analysed in
terms of their various structural elements, their sequencing, the interactional
functions of these structures, and their linguistic realizations (House 1979;
1982a; 1984).

In the analysis of opening and closing phases in'English and German
interactions I found that certain opening acts such as the Territory Invasion
Signal {Excuse me) or closing acts such as the Extractor (/ really must go now)
tend to be realized by routine formulae by the English native speakers, whereas
the German speakers tend to prefer lengthy ad-hoc content-oriented expres-
sions, which result in a noticeably greater verbosity in the German interactions.
In opening and closing-phatic talk a whole series of routine formulae have no
readily available equivalent formulae in English. Thus: Nice to see you^ Nice
to have met you, Lovely talking to you, Let’s keep in touch etc. tend to be
formulated in an ad-hoc manner, in which greater emphasis on the respective
content or subject matter of the interaction is placed, as in e.g., Wie schon, dajl
ich Sie mal wieder treffe, wir haben uns ja schon iiher einen Monat nicht mehr
gesehen etc.
Also, the use of moves with which to explicitly introduce a topic was
found to be more frequent in German.

1.2 Discourse Strategies

The analysis of discourse strategies (House 1979; 1982b, 1984) i.e., supportive
moves used in an anticipatory and prophylactic manner (such as e.g., Checks-
on-availability {Are you free tonight), Getting-a-pre-commitment {Can you
do me a favour). Disarming the Interlocutor (/ don’t want to bore you with
unnecessary details), or muddling your way through an issue by expanding it
verbosely) revealed the following tendencies:

(1) German speakers tend to prefer the use of content-oriented and self-ref-
erenced strategies with which they explicitly introduce a topic {Ich habe
eine Frage) expand on it {Also mein Hauptpunkt hier ist folgender, und
ich willversuchen, die wesentlichsten Punkte in allerKiirze darzustellen...)
or provide reasons and justifications of different kinds, whereas English
native speakers tend to select strategies which are more interpersonally-
oriented such as e.g., a Dlsarmer (/ hate bothering you but...). The prefer-
red self-referencing in German as opposed to the “other-orlentedness” in
English can also be demonstrated with the following two examples: Are
you busy at the moment vs Store ichf Kann ich Sie einen Moment storenf
and Is this seat taken/Is anyone sitting here^ vs Ist dieser Platz noch frei
(i.e., “for me”).
(2) There is more ad-hoc formulation in German and a greater reliance on
pre-fabrlcated expressions or conversational routines in English. For ex-
ample: Can you do me a favour vs ich wollte mal fragen, ob du mir da mal
aushelfen kannst.

1.3 Gambits

The analysis of gambits (House 1979; 1982c; 1984) i.e., discourse elements
such as well, you know, you see, question tags, etc. that function to “lubricate”
an ongoing discourse in different ways, and also serve the speaker as time-gai-
ning routines during speech production revealed that Germans tend to prefer
content-oriented and self-referenced types of gambits such as e.g., starters {ja
also) used in prefacing ones message or undersborers {und das ist mein Haupt-
punkt hier) used to emphasize the importance of the content of a message,
whereas English speakers prefer gambit types with which they explicitly ad-
dress their conversational partner such as e.g., cajolers {you see) used to coax
the Interlocutor into heightened attention or sympathy.

1.4 Speech Acts

The analysis of the realization of speech acts, especially requests and com-
plaints (House 1979; 1984; 1985; House and Kasper 1981) as potentially face-
threatening acts was one of the most important parts of my work. Phenomena
such as directness and politeness in the use of speech act were investigated,
and different “levels of directness” were suggested ranging e.g., in the case of
requests from the most direct level, the raw imperative to the most Indirect
“hints” with which speakers skirt around a subject. German subjects tend to
prefer more direct expressions, a tendency which must however not automat-
ically be equated with a preference for utterances reduced in politeness. The
Pikean distinction between emic and etic perspectives of culture-conditioned
behaviour is relevant here. For example: (Situation: A wants B to to P): Don't
you think it might he a good idea if you did P vs Also ich finde Du solltest P
In the second, larger and internationally operating project, the “Cross-
Cultural Speech Act Realization Project” (Blum-Kulka, House, Kasper 1989),
which involved cross-cultural pragmatic contrast between seven different lan-
guages and language varieties, I again conducted a number of German-Engllsh
contrastive pragmatic analyses (House 1989a,b; Blum-Kulka and House 1989;
House and Kasper 1987). The data was elicited in this project through the
method of using a (written) discourse completion test, i.e., everyday dialogic
situations (16 altogether) with a blank left in each situation for the insertion
of a certain speech act were presented to subjects (university students) on a
questionnaire. This data was triangulated with retrospective interviews and
so-called metapragmatic assessment tests, i.e., tests in which subjects were
asked to assess e.g., the appropriateness of certain utterances in a given situation
or the level of politeness or directness in a set of utterances, the social role

relationship, or the rights and obligations of participants holding in a given
The result of my analyses with 200 German and 100 English subjects
basically confirmed the results of my work in the previous project: German
subjects tend to opt for more direct realizations of requests, and they prefer
to surround their requests with more content-oriented discourse strategies
than the English nativ.e speakers (who tend to prefer interpersonally active
and routinized strategies). In their realization of apologies, Germans tend to
choose self-referenced moves wheras the English native speakers more fre-
quently select moves conventionally expressing concern for alter, e.g., YouWe
not upset, are you vs Ich wollte Dich nicht krdnken, are the most frequently
realized English and German expressions of apology in a work-place situation
where one collague had Insulted the other with a careless remark.
Germans were again found to use discourse strategies with which to ex-
plicitly Introduce topics, and to justify and give reasons for e.g., a request more
frequently than English native speakers.
English native speakers also tend to use more routinized expressions in
realizing apologies than German speakers, i.e., in many of the apology situa-
tions presented to subjects in the project, English native speakers use the il-
locutionary force indicating device: sorry in places where the German subjects
realize a surprising variety of different tokens: (Oh) Entschuldigung, Entschul-
digen Sie hitte, Verzeihung, Tut mir leid, Pardon, Sorry, as well as a number
of different combinations of these tokens.
More recently, I have tried to look into the manifestations and causes of
cross-cultural misunderstandings between German and English speakers
(House 1993; 1996a,b). Data was collected in authentic conversations, inter-
views, self-reports and field-notes. Results so far suggest that a substantial part
of German-English cross-cultural misunderstandings result from differences
in the pragmatic areas outlined above.
Further, a recent investigation into attempts to develop pragmatic fluency
in German learners of English in an Instructional setting has clearly demon-
strated that many of the pragmatically inappropriate utterances on the part of
these learners stem from the difference in orientation in the German and Anglo-
phone linguacultures: towards a message to be transmitted as opposed to an in-
teractional, addressee-oriented and conventionally routinized orientation (see
House 1996c).

2 Five Dimensions of Cross-Cultural Difference:

From all these individual results of a whole series of cross-cultural pragmatic

analyses based on different subjects, data and methodologies, a consistent pat-
tern emerges: in a variety of everyday situations, German subjects tended to
interact in ways that were more direct, more e:{plicit, more self-referenced and
more content-oriented. German speakers were also found to be less prone to
resort to using verbal routines than English speakers.
The pattern of cross-cultural differences that has emerged from my Ger-
man-Engllsh contrastive pragmatic analyses can be displayed along five di-
mensions as in Tab. 1:

Dimensions of Cross-Cultural Difference


Directness Indirectness

Orientation <=> Orientation

towards Self towards Other

Orientation <=> Orientation

towards Content towards Persons

Explicitness <=> Implicitness

Ad-hoc Formulation <=> Use of Verbal Routines

Table 1: Dimensions of Cross-Cultural Difference (German-English)

Along these dimensions, Germans were found to give preference to positions

on the left hand side. It must be stressed again, that these are continua or dines
rather than clear-cut dichotomies, that we are dealing here with tendencies
rather than categorical distinctions.
At this stage of the discussion one might also pose the question whether
results such as the ones described above and especially the hypothesized five
dimensions as generalizations from these results increase existing prejudices
and stereotypes. Admittedly, this is a real danger. However there are at least
three ways how this danger can be minimized:

(1) Conduct further empirical cross-cultural pragmatic research, which

draws on different and larger samples and is designed to falsify the results
so far achieved;

(2) Involve blcultural experts in the interpretation and evaluation of the data;
(3) Extend bicultural contrastive studies towards multicultural ones, which
might relativise the results of blcultural contrastive studies.

Returning to the results of the cross-cultural German-Engllsh studies described

above, they Imply that in German discourse, “transactional language use” (to
use Brown and Yule’s 1983 terms), which Is primarily message-oriented, tends
to be valued differently from interactional, addressee-oriented language use.
And in terms of the two Hallidayan language functions, the ideational and the
Interpersonal, It Is the ideational one, which seems to be given a different focus In
German interactions - often at the expense of the interpersonal one.
Further, one might conclude on the basis of the hypothesized five dimen-
sions that some of the Gricean Maxims (Grice 1975) seem to operate differently
in the German linguaculture, i.e., the Maxim of Quantity (“Make your con-
tribution as informative as is required” and “Do not make your contribution
more Informative than is required”), and the Maxim of Manner (“Be brief”)
do not seem to have the validity they have for Anglophone cultures, for which
they were conceived.
In looking for a “deeper” explanation for these results, it is necessary to
look for the cultural-historical roots of the differences found along the hypo-
thesized dimensions, and try to link up differences in interactional style and
cultural ethos with a richer ethnographic context considering the history, phi-
losophy, religion, educational system, political and social developments, con-
ventionalized cultural practices of the llnguacultures involved. Preliminary
hypotheses explaining the roots of some of the German communicative pre-
ferences discussed above point to phenomena in German society such as the

(1) a break-up of a so-called national identity and a loss of a sense of com-

munity (“Gemelnschaft”) (with an ensuing heightened Individualism),
(2) a break-up of etiquette, of a behavioral canon of rules of conventional
behaviour in post-war Germany,
(3) an educational system that has traditionally tended to place greater em-
phasis on the transmission of knowledge, of content and subject matter,
and on the development of “inner virtues”, feeling and character, rather
than on conventional social values, let alone the transmission of the par-
ticular Anglo-Saxon brand of an “etiquette of simulation”, where rules of
communicative behaviour are implicitly handed down from generation
to generation “to sound as if you mean it” when, e.g., expressing thanks,
apologies, compliments and other “face-lifting” speech acts. The results
of my cross-cultural (and misunderstanding) research has pointed to the
generalization that such “impression management” is less strongly valued
in the German linguaculture.

(4) as in other European countries, the legal system features pre-established
statutes, laws, and regulations. This is very different from the Anglo-Saxon
pragmatic and negotiable case law system (cf. Legrand 1996).

Influences on communicative patterns such as the ones briefly sketched above

(for a more extensive discussion see House forthcoming), need further syste-
matic Investigation, and an equally systematic attempt to link them to empi-
rically established cross-cultural communicative differences.
An Important research task for the future is, it seems to me, to try to
investigate the underlying reasons for the historical development of such deep-
seated differences in communicative orientations. What we would need here
are longitudinal studies to supplement existing cross-sectional studies.

2.1 Examples in Support of the Five Dimensions

To elucidate the dimensions I have distilled from a variety of pragmallnguistlc

studies, I will now list a number of examples taken from various different
cultural practices and discourse types. The first set of examples is taken from
a corpus of German signs, which appear in many different walks of life. In
many cases these signs were also provided with an official translation that
nicely reveals the German-English dimensional differences with respect to a
focus on content vs an Interpersonal focus in the demonstration of German
explicitness, and the apparent need felt by those responsible for the signs to
explain and justify them:

(1) Sign at Frankfurt airport (displayed at a building site; original German)

Damit die Zukunft schneller kommt! vs We apologize for any inconveni-

ence work on our building extension is causing you!

The difference in perspective, l.e., focus on content in German, Interpersonal

focus in the English translation is clearly noticeable here.

(2) Sign displayed in many different places in the University of Hamburg:

Zur Vermeidung von Gesundheitsschaden und unzumutharen Beldstigungen

ist in den Hallen, Fluren, Treppenhdusern und
Veranstaltungsrdumen dieses Gehaudes
mit Ausnahme der Cafeteria
und der Eingangshalle
Bitte nehmen Sie Rucksicht auf die Gesundheit Ihrer Mitmenschen!

The English translation of this sign which not only explicitly names the areas
where it is operative but also justifies its own existence as it were (“Zur Ver-

meidung von Gesundheitsschaden”), would most likely be a simple “No smo-
king” (this was suggested to me by Anglophone observers puzzled by the sign).

(3) Traffic sign in Hambjjrg s Ohlsdorf cemetery:

30 km/h
Anf dem Friedhof
This sign seems to explicitize the obvious: when the addressees are driving
through this cemetery, one would assume that they know where they are, so
the addition “Auf dem Friedhof” seems overly explicit.

(4) Sign in a hotel bathroom

Lieher Gast! Weniger Wdsche und weniger Waschmittel schUtzen unsere
Umwelt. Bitte entscheiden Sie selhst, oh Ihre Handtiicher gewaschen wer-
den sollen. Nochmals henutzen: Fiandtucher bitte hdngen lassen. Neue
Handtiicher: Handtiicher auf den Boden legen.
Dear guest, will you please decide for yourself whether your towels shall
he washed. Use again: please leave your towels on the towel rack. Clean
towels: please put your towels on the floor.
In the German original, but not in the translation, an explicit justification for
the request is offered in the first sentence. Further, the German original seems
slightly less polite than the translation, l.e., mentioning “bitte” twice may have
seemed too much for the German writer, whereas the English translation in-
serts a “please” in each of the requests.

(5) Advice given by airport personnel

Ladies and Gentlemen, please do not leave your luggage unattended at
any time!
Meine Damen und Herren, bitte lassen Sie Ihr Gepdck aus Sicherheits-
griinden me unbeaufsichtigt.
Here again the German addressees are provided with an expllct reason or
justification for the request made.
The following examples are taken from the TV comedy series Mr. Bean,
which was dubbed in German TV and also shows the difference in focus on
(explicit) content and on the level of directness in utterances:

(6) Everybody out now please! vs Die Badezeit ist zu Ende

The translation shows a focus on a rule or rather the content of the rule,
whereas the English original focusses on the human beings and their actions.

(7) Will you get out of there please! vs Weg da!

The level of directness in the English original is much lower than the one in
the German translation.

(8) Watch your step! vs Vorsicht Stufe!

Even in this conventionalized form, the German translation reveals a content-

focus, the English original an Interpersonal orte.

(9) Mind the Baby, Mr. Bean! vs Babysitter wider Widen!

The last example, (9) is an example of what I would call the German predilec-
tion for “letting the cat out of the bag” in the interest of being explicit. Many-
more examples can be found in the translations of TV film titles, of which I
will only list a few here:

(10) Where are the Children^ vs Grenzenloses Leid einer Mutter

{\\)Jack the Bear vs My Dad - ein ganz unglaublicher Vater

(12) A Gunfight vs Duell in Mexico

(13) Trapped and Deceived vs Wenn Eltem ihre Tochter verraten

(14) Mommie Dearest vs Meine Hebe Rabenmutter

In the German translations of the film titles in (10) to (14) the content of the
films is “given away”, i.e., the titles explicitly tell the reader what to expect in
the film.
Having provided a few examples of the operation of these dimensions in
“real life” in a number of different textual types, parallel texts and covert trans-
lations, I will now briefly refer to some studies relevant for, and supportive
of, the findings of my own cross-cultural work.

3 Some Contrastive Pragmatic Studies Supporting

the Hypothesis of the Five Cross-Cultural Dimensions

While there is direct support for these findings for instance, from Byrnes
(1986), Kotthoff (1989), Blellck (1991) and Fetzer (1996) who also compared
English and German oral discourse, it is interesting to see that my results,
which were achieved with oral data, are comparable to the results of a number
of contrastive pragmatic analyses in the area of written text production as well.
Cross-cultural comparisons of texts and investigations of attitudes towards
specific text types are fields of research which are becoming increasingly
popular. They are of particular relevance for the teaching of writing and for

translation theory and practice. (Cf. House 1994 for an overview, and see
Kaplan 1987; 1996; Purves 1988; Connor and Kaplan 1987.)
Research methodologies include comparisons of texts written by mem-
bers of different cultures^ in terms of cohesive devices, textual organization,
presence and nature of reader or writer perspective, and judgments of a text’s
effectiveness by expert raters in the two cultures.
With regard to the two linguacultures, with which I am here primarily
concerned, Michael Clyne (1981; 1987; 1994) has revealed the impact of cul-
tural values on discourse pointing to the culture-specificity of discourse struc-
turing in written academic texts. Clyne compared a large corpus of academic
discourse (linguistics and sociology texts) written by Germans in both German
and English and by English native speakers. All texts compared were chosen
and matched according to the discipline and sub-discipline, type of text (article,
book chapter, published conference paper), topic, intention of the text, and
the author’s gender. Clyne’s results show the following;
1) Texts composed by Germans tend to be less “linear” and show more
digressions, which were found to fulfill particular functions in German
academic texts: “they enable writers to add a theoretical component to an
empirical text, a historical overview, ideological dimension, or simply
more content, or engage in a continuing polemic with members of a com-
peting school ... these are all crucial aspects of German Intellectual style
and German culture” (Clyne 1994: 163).
2) Germans focus much more on the content they are presenting than on
how this content is to be presented and/or received by readers.
3) German texts are often “writer-oriented” rather than “reader-oriented”,
or, put differently, it is the reader’s responsibility to make sense of a text,
with the writer not being burdened with the task of making his text pala-
table to his readers (see Hinds 1987 on reader- vs writer- responsibility).

Clyne’s finding can be Informally confirmed through the obvservation that in

German translations of reader-friendly, “simple” English terms or titles one
notices a trend towards “complexlflcation”, see e.g., John Austin’s famous
work entitled How to do things with words and its German rendering as Zur
Theorie derSprechakte. Similarly, consider the German adaptation of an Ame-
rican course of studies called Public Health as Gesundheitswissenschaften, or
the way Social Management is rendered as Betriebswirtschaft im Sozialwesen.
Clyne also pointed out that there seems to be a much clearer division
between academic writing and non-academic writing in the German lingua
culture. By contrast, native English speakers’ academic articles tend to be more
similar to non-academic texts and thus more reader-oriented.
Clyne (1994) set out to interpret the contrast he found in terms of different
essay-writing norms that are transmitted by the educational systems in Engllsh-

and German speaking countries. According to Clyne, then, more weight is
attached to form in the English school tradition, which has an effect on style
preference, i.e., whereas the German scholar wishes to present his/her creden-
tials as a “Wissenschaftler”, the English-speaking scholar alms to show that
his/her argument is sound and well thought out.
Other researchers, who compared so-called “intellectual styles” or “cul-
tural thought patterns” (i.e., ways of turning thoughts into language) have
suggested deep-seated differences between makers of different llnguacultu-
res that determine different discourse styles in speech and writing. Thus Ka-
plan (1966) in his classic “doodles article” entitled “Cultural Thought Pat-
terns” suggested the existence of five discourse types based on genetic language
types: the linear and logical progression in English, the parallel constructions
with the first idea completed in the second part of a composition in Semitic
languages, the circularity with which a topic is looked at from different angles
in Oriental languages, the licence to digress and Introduce “extraneous” ma-
terial in Romance languages, and lastly and similarly to Romance discourse,
the freedom to digress lenghtily using parenthetical amplifications and sub-
ordinate elements in Russian. German would be placed somewhere in between
Romance languages and Russian, such that the difference between the typical
German discourse type and the typical English one hypothesized by Kaplan
is marked. Indeed. Kaplan s hypotheses also confirm my own German-Engllsh
contrastive studies.
As opposed to the more linguistically oriented Investigations of text pro-
ductions, a more speculative, non-emplrlcal approach to looking at cross-cul-
tural differences in writing is the one exemplified by Galtung (1985). On the
basis of his familiarity with scholars from different cultural backgrounds, Gal-
tung hypothesized four different “Intellectual styles”, which he called “Sach-
sonisch”, “Galllsch”, “Nlpponisch”, and “Teutonisch”, designating their re-
spective backgrounds. Galtung correlates his styles with their relative
strengths and weaknesses in terms of the analysis of paradigms, the production
of theses, the formation of theories, and the ability to provide commentaries
on other Intellectuals. For Instance, the Saxonic Intellectual style is very strong
on the production of (hypo)theses, where the Teutonic one is weak. The Sa-
xonic style is weak on theory formation, where the Teutonic style is very
strong, and the Teutonic style is also superior to the Saxonic one in terms of
paradigm analyis. Teutonic Intellectual style, according to Galtung, is more
elitist, individualistic, and monologue-oriented as well as polarized, involving
an attack on others’ weak points, whereas the Saxonic style is more “democra-
tic”, non-polarlzed, and alms more at dialogue and a harmonization of diffe-
rent viewpoints. Commenting on Galtung’s speculations, Clyne states that
“while (Galtung’s) categories help account for some aspects of German aca-
demic discourse, the ’’democratic'* characteristics of the ’’Saxonic** are not

necessarily accompanied by tolerance for variation, cultural and otherwise”
Relevant is here also a study by Kusch and Schroder (1989), who compared
German and English texts by Habermas and Davidson confirming the hypo-
thesized differences between Teutonic and Saxonlc “Intellectual styles” as well
as Clyne s differences between German and English academic discourse. They
also admitted, however, that it is extremely difficult to operationalize concepts
such as “Intellectual style” or “linearity”, let alone generalize from results
based on a very small corpus of texts.
Another attempt to provide empirical support for Clyne s and Galtung’s
theses is Gnutzmann and Oldenburg (1991), who compared Introductions and
conclusions in academic articles in the journals Language and Linguistische
Berichte. They found that introductions in both journals are highly conven-
tionalized but conclusions display culture-conditioned differences: in Lan-
guage, authors routinely highlight the importance of their own work, in the
German journal, however, a (more modest) summary is provided.
Other influential approaches investigating cultural differences in value
orientations are the psychosoclally oriented cross-cultural studies by Hofstede s
(e.g. 1984; 1991) on comparative “work-related” values, and Halls classification
of cultures according to certain dimensions (cf. Hall 1976; Hall and Hall 1983).
Hofstede Investigated differences in cultural values on the basis of a
questionnaire study with over a 100.000 employees in a large multi-nation-
al company in 40 countries, triangulated by other empirical studies. He sug-
gested four cross-cultural dimensions along which cultures differ: Individual-
ism vs collectivism, power distance, masculinity vs feminity, and uncertainty
avoidance. The dimension “uncertainty avoidance” captures the way culture
members handle uncertainty and ambiguity. Wherever there is a low tolerance
for uncertain situations, there will also be a need for explicit rules and struc-
tures, with which events can be anticipated and controlled. According to Hof-
stede, members of the German culture tend to have a high value on the uncer-
tainty avoidance parameter.
The parameter “uncertainty avoidance” can be related to my dimensions
“directness vs Indirectness” and “explicitness vs implicitness”, in that the need
to express oneself directly and explicitly may be said to stem from a drive to
reduce uncertainty. I can also detect a similarity with the “self vs other” ori-
entation in my dimensional scheme and Hofstede’s dimension “collectlvistic
vs Individualistic” cultures, respectively displaying an “us vs an I-feeling” or
a “responsibility to others/society vs responsibility to self”. According to my
findings, Germans appear to be more on the individualistic end of the scale
than Anglo-Americans.
From an anthropological vantage point, Edward T. Hall (1976) attempted
to classify cultures and Indeed mentalities according to “hidden differences”

In terms of preferences of “high or low context” and “monochronic or poly-
chronic time” in culture-conditioned communication. In high-context com-
munication “most of the information is either in the physical context or in-
ternalized in the person, while very little is explicitly coded...” (1976: 79)
whereas in low-context communication “the mass of the information is vested
in the explicit code” (1976: 79). Hall (1983) further distinguishes polychrone
and monochrone cultures. Individuals in monochrone (“M-time”) cultures
tend to do only “one thing at a time” taking'schedules, plans and temporal
commitments very seriously, whereas Individuals in polychrone (“P-tlme”)
cultures do several things at the same time, taking plans, schedules, dates etc.,
with a grain of salt. German communication patterns and preferences would
then be more on the low-context and the monochrome end of Hall’s continua
than anglophone communication patterns and preferences. Hall and Hall
(1983: 65) go as far as deducing from these findings a culturally conditioned
difference between American “friendliness” as opposed to German “serious-
ness”. One of the most important themes in a corpus of interviews I conducted
with American exchange students (House 1996a) was, in fact, “friendliness”,
and how and why it is felt to be absent in the context of the German culture.
The link to the Anglophone focus on persons versus the German focus on
content is obvious. Further, there is certainly a similarity between Hall’s two
dimensions of context and time and my hypothesized dimensions of implicit-
ness vs explicitness and orientation towards content vs orientation towards
Cross-cultural differences in the realization of spoken and written dis-
course and the communicative preferences of members of different llnguacul-
tures (with particular reference to German vs English discourse) have also
been the topic of much recent “textual-pragmatic” research, involving a variety
of different text types. I will very briefly sketch some of the results of studies
which I see as particularly relevant for my attempt here to substantiate the
cultural filter in German-Engllsh/ English-German translations. These studies
can thus also be taken to provide converging evidence for some or all of my
hypothesized dimensions.
Relevant support for my hypotheses of German-Engllsh communicative
differences is given in Agar (1992) who in a review of Werner Holly’s “Poll-
tikersprache ...” - a review of a book from one intellectual tradition in the
context of another - spelled out Anglosaxon and German differences very
similar to the ones I have hypothesized: Content-focus versus reader-focus,
explicitness vs implicitness, bottom-up data-driven approaches vs top-down
priority of fundamental, theoretical Issues. Leaving out other nationalities.
Agar quotes a joke that makes this points nicely: An American and a German
go off to study elephants. The American comes back with a brief essay entitled

“How to Use an Elephant”. The German comes back with 11 volumes entitled
“Introduction to Elephant Science” (1992: 161).
In a short insightful article, Luchtenberg (1994) contrasts American and
German software manuals, coming up with results that basically confirm my
hypotheses about German and English content vs Interpersonal orientation,
as well as directness vs indirectness in communicative orientation. Her title
says it all: “A friendly voice to help you vs. working thoroughly through your
manual: Pragmatic differences between American and German software ma-
Luchtenberg gives a number of examples taken from German software
manuals, that reveal how “the friendly helping voice” in the English original
is changed for the German user such that he is made to feel guilty for not
having properly read the Instructions running the risk of needlessly bothering
the Customer Support Department. The German software user is given orders
as though he were a schoolboy who failed to do his homework, whereas the
American user’s cry for help is seen as perfectly legitimate, even though he is
expected “to have exhausted all other help avenues”. Compare:

WordPerfect is hacked by a customer support system designed to offer you

fast, courteous service. If you've exhausted all other Help avenues and
need a friendly voice to help you with your problem, follow these steps ...

The German reader has to make do with the following:

WordPerfect hat ein Support-Zentrum eingerichtet, dessen Mitarheiter Ih-

nen hei Prohlemen kompetente Unterstutzung anhieten. Wenn Sie trotz
der m WordPerfect zur Verfiigung stehenden Hilfsquellen ein Problem
nicht losen konnten, wenden Sie sich an unser Support-Zentrum (1994:

Luchtenberg’s study touches on an important difference in the notion of “ser-

vice” in the Anglo-Saxon and German linguacultures, which might be ex-
plained with reference to the hypothesized dimensional differences of self-
orientation and content-orientation versus Interpersonal orientation in
communicative styles.
Another study supporting my results is Ventola (1995a). She contrasts the
use of English as a lingua franca in written communication by Finnish and
German native speakers confirming the German predilection for content-rich,
author-focussed communicative styles.
Another contrastive pragmatic study relevant for my discussion here is
Schmitt’s (1995) comparison of German and English safety precautions, war-
ning labels etc., in which he finds that the usage in the English manuals is

practically standardized, whereas usage in German manuals seems to vary
widely. I take this finding to support my hypothesis of the difference between
German and English speakers’ reliance on routinized linguistic tokens in many
discourse areas. Similarly, Busch-Lauer (1995), who compared English and
German articles in medical journals, comes to the conclusion that the English
texts she examined display a certain normative pattern, whereas the German
texts are more random and variable, and that German authors often Introduce
additional propositions into the text in the form of “Exkurse”.
Gopferich (1995) analysed Instructions and product descriptions and found
the German exemplars more reticent and objective, whereas the English in-
structions and descriptions were found to be much more interpersonally ori-
ented frequently resembling advertisements.
Finally, Fiedler (1994), who compared linguistic articles in Esperanto,
English and German, found that the German articles were “less personal”.
However, she also observed that German publications tend to become more
interpersonally focussed due to the dominance of English in the scientific
All the above studies provide converging evidence for the assumption that
culture-conditioned preferences of focus, structuring and the use of routinized
or ad-hoc tokens strongly influence writing styles. Fiedler’s (1994) finding of
the changeable nature of language-specific styles should however sound a note
of caution, l.e., itshould not be forgotten that, as Mauranen (1993) has pointed
out, “in contrasting texts and cultures, it is necessary to abstract from a con-
siderable amount of variation... categorical, that is absolute, differences would
be a surprising result... not only owing to the Inherent variability of national
cultures but also because of culture contact and Intercultural Influences.
Writers and cultures have different kinds and amount of contact with other
cultures. The differences... are therefore broad tendencies rather than exact
dichotomies” (1993: 2).
In other words, in cross-cultural studies focussing on differences in com-
municative styles, one should always remind oneself that these styles are
historically conditioned and dynamic, and that one is therefore most pres-
singly “faced with the question about the possible directions and effects of the
transmission of intellectual values across languages and cultures” (Duszak,
1994: 295). This question is, of course, also highly relevant for translation
studies and the evaluation of translations.
In setting up my five dimensions of cross-cultural difference, I started
with a series of detailed contrastive pragmallngulstlc analyses, suggesting dif-
ferences in English and German interactional norms, from which I hypothe-
sized differences in discourse orientations. In doing so, I implicitly suggest,
of course, that language use is linked to culture and mentality, and that lingui-
stic differences in the realization of discourse phenomena may be taken to

reflect “deeper” differences, at a conceptual-cognitive and emotive level, in
cultural preference and expectation patterns.
This type of “deep difference” can have serious consequences for the process
of translation in that they are likely to influence a translator s decisions about
changes in the original text. The translator may consciously or unconsciously
apply a cultural filter in covert translation to account for cross-cultural diffe-
rences in the expectation norms holding in the two cultures concerned.
In the following I want to give a few examples of translations and/or
parallel texts in order to show how the empirically derived dimensions can help
understand and explain the occurrence of cultural shifts in covert translation.

4 Examples of Translations Featuring Cultural Filtering

Along the Five Dimensions of Cultural Difference

The examples are taken from different genres. The first two textual pairs (15) and
(16), are instructions for use: one for using oven ware, the other one for setting
up a perpetual calendar. In (15) the original is German, in (16) it is English. In
both cases the German preference for greater explicitness is noticeable.

{\5)KERAFOUR ist in unahhdngigen Priifungsinstituten auf Ofenfestigkeit

und Mikrowellenhestdndigkeit getestet warden. Damit Sie lange Freude
an ihm hahen, gehen wir Ihnen einige kurze Gebrauchshinweise:
(1) Stellen Sie nie ein leeres, kaltes Gefdjl in den erhitzten Ofen (als leer
gilt auch ein nur innen mit Fett hestrichenes Gefdjl).
(2) Lassen Sie tiefgefrorene Kost vor dem Uberbacken erst auftauen, da
sich so die Lebensdauer ihres KFRAFOUR-Teiles wesentlich erhoht.
(J) Spiilen Sie das Gefdjl nach Gebrauch nicht mit kaltem Wasser. Bei
Gebrauch sollte die Ofentemperatur 300 Grad nicht iiberschreiten.
(4) Stellen Sie das KFRAFO UR-Teil nicht auf den Flektroherd oder gar
in die offene Flamme.
Viel Spajl und guten Appetit.
KFRAFOUR oven-to-table pieces have been tested by independent re-
search institutes and are considered ovenproof and micro-wave resistant.
Here are a few simple rules for using KFRAFOUR.

(1) Never put a cold and empty piece into the heated oven.
(2) Do not put deep frozen food immediately in the KFRAFOU R-piece
into the oven.
(3) Clean the piece with warm water after use. The working temperature
in the oven should never exceed 300 oc.

(4) NeveruseaKERAFOUR-pieceon the open flame or on the hot oven.
Enjoy cooking with KERAFOUR.

In the second sentence, the German original gives a reason for this instruction
“Damit sie lange Freude an ihm haben”, which is left out in the translation.
And under (1), the original - unlike the translation - defines precisely the
conditions under which the Kerafour-plece is to be considered “empty”. One
is reminded of (the then fire-inspector) Wh^f s famous example of a fire
breaking out because of an erroneous conception of a gas-filled vessel being
“empty” - whether the German text producer thought of the potentially costly
consequences of a misinterpretation of “empty” or whatever else the reason
for his greater explicitness may have been, the interesting fact remains that the
entire bracket is left out in the English translation (but is retained in the French
Example (16) also shows that the German translator is more explicit and
provides more details than is the case in the English original.

{\6)Perpetual Calendar
You have in your possession a very sensitive calendary do not damage it by
riskily pressing the buttons. Follow our instructions to allow it to live ...
Push colored tabs to uppermost position once they have come to an end.
vs Dauerkalender
Sie sind der Besitzer eines leicht zu beschadigenden Kalenders; drucken
Sie nicht auf die Fasten bevor der Kalender an der Wand hdngt, sonst
zerbrechen Sie ihn. Lesen Sie die Gebrauchsanweisung ... Schieben Sie
jedes farbige Pldttchen mittels eines Schiebeknopfes nach oben, sobald das
Ende der Skala erreicht ist.

In specifying the consequences of inappropriate handling (“sonst zerbrechen

sie ihn”) and the time during which such handling should be avoided (“bevor
der Kalender an der Wand hangt”), as well as the Instrument to be used (“mittels
eines Schiebeknopfes”) in the instruction, the German translation is much
more explicit as regards the Information conveyed than the English original.
The next two examples, (17) and (18), are advertisements. (17) is an extract
from an advertisement for coffee from Colombia. I suspect (but was not able
to verify this) the English text is the original and the German text is the trans-
lation, but even if they were both translations from Spanish, the points I want
to make still hold:

(17) (The advertisement features a picture showing a young man tasting coffee)
Look at the concentrated face of this coffee taster. This is how conscientious
Colombians are about their mild and noble Highland coffee.

Our friend’s profession is an important one in Colombia; his tongue is
incorruptible. He himself is severe, but today he is just as mild as his much
praised Highland coffee. This year’s crop can only be rated excellent. Find
out for yourself over'there in far-off Europe.
Wie ernst man in Kolumbien seinen edlen, milden Hochlandkaffee nimmt,
konnen Sie an der konzentrierten Miene dieses Herrn erkennen.
Er ubt einen der wichtigsten Berufe hierzulande aus. Er ist Kaffeetester.
Seme "Zunge ist unbestechlich. Er selbst ist sehr streng. Doch heute ist er so
milde wie der viel geriihmte Kaffee. “Dieser Ernte kann man nur das
Pradikat excellent geben”. Wohl bekomm’s im weitentfemten Europa.
The English text is interpersonally more potent than the German one: in the
first paragraph, the reader is directly asked to do something, which immedi-
ately involves him personally. In the second sentence the reader is helped along
with an interpretation of the picture he has looked at. The first paragraph in
the German text consists of one long sentence which is more difficult to grasp
than the two short English ones. The reader is only indirectly addressed in
German, the imparting of information is given greater weight, and a greater
social distance is erected between the coffee taster and the reader, he is “dieser
Herr” in German, whereas he is simply “our friend” in the second paragraph
in English. In German he is “sehr streng”, in English only “severe”. And the
ending, too, is more personal in the English text, as the reader is again personally
and directly addressed, the German “Wohl bekomm s” is more conventionally
Example (18) is taken from Smith and Kleln-Braley (1997). Both the Eng-
lish and the German translations are based on a French original.

(18) (Picture of a vampish lady)

The chances of her being seated next to you are so slim that you won’t
regret the extra space between our seats.
L’Espace Europe
We know how hard it is for business travellers to have to concentrate on
their work while waging the eternal battle of the armrest, so we have
re-arranged the space between our L’Espace Europe seats. Where there
used to be rows of three seats, there are now two seats separated by a little
table. Your seat is much wider, more comfortable and the total space more
conducive to a little privacy. Now, when you take a seat in one of our
planes, you take your seat in space. (Picture of seats with table).
The Right to Privacy. Air France introduces passengers’ rights.

Ihre Chancen stehen schlecht, dajl sie nehen Ihnen sitzt. Ihrem Komfort
zuliehe hahen wir den Abstand zwischen den Sitzen spiirhar vergrojlert.
UEspace Europe
Geschaftsreisende wollen im FlugzeugAkten studieren, Zeitung lesen oder
sich in Ruhe auf eine Sitzung vorhereiten. Am liehsten ohne Tuchfuhlung
zum Nachbarn. Oder zur Nachbarin. Darum haben wir unsere UEspace
Europe von Grund auf neu gestaltet. Grdjfer, schoner, bequemer und vor
allem mit viel willkommener Ablagefldche zwischen den Sitzen. Fur viel
Ellbogenfreiheit beim Lesen, Essen und Entspannen genau die richtige
Distanz. Und auch fur ein anregendes Gesprdch. (Picture of seats with table.)
Ihr Recht auf Distanz. Bei Air France sind Sie mit Recht Fluggast.

Smith and Klein-Braley (1997) comment that the German advertisement pro-
vides more precise specification of activities than the English one. While the
passengers in the English advertisement genetically “concentrate on their
work”, their German counterparts are involved more explicitly in “Akten
studieren, Zeitung lesen oder sich in Ruhe auf eine Sitzung vorhereiten”, and
whereas the English readers are not informed about the reasons why passen-
gers need space, the German reader learns that space is needed to enable the
passenger to “lesen, essen und entspannen”. Further, Smith and Klein-Braley
also point out in their perceptive analysis that the English advertisement at-
tempts to directly address the reader as potential consumer through frequent
repetition of “you” and “your”, whereas such an Interpersonal appeal is only
present in the heading and slogan in the German advertisement.
We have here a good example of the provision of more Information for
the German reader coupled with an interpersonally less effective addressee
Involvement. These findings support two of my hypothesized dimensions of
cross-cultural differences in communicative preference - focus on content and
explicitness (see also Smith and Klein-Braley 1997 for further examples of the
strategy of giving German consumers more information in advertisements
than the English one).
In this Chapter, I have presented my own and others’ research on cross-
cultural differences in communicative preferences, norms and values. I have
presented five dimensions of cross-cultural differences English-German as
well as a number of studies that provide converging evidence for these dimen-
sions. I have also discussed various examples of English-German, German-
English covert translations and parallel texts, in which a cultural filter seems
to have been applied. The notion of a cultural filter was Introduced in the
original model of translation quality assessment in order to capture cultural
shifts in translation. The filter has now been given substance through the re-
search results and the cross-cultural dimensions presented in this Chapter.

How the cultural filter In its substantiated form can be integrated into the
model for translation quality assessment will be shown in the next Chapter
which presents the model in its revised form.

v* '^ni . ‘T .'W r.'ifx. -'iM' . <•{ »i» '■« .w* 5.‘it woH
\ I * - ni-'^ ••?-.■ '»;■ v -‘r .. ,-;#p
■ • ■ ' ■• /iA<juS« r> ♦((

fe »

» . ' , • . '. < » ji /• ’ j.

»'./'. - *• ♦ ■ - *■ - .

* ii.— ,J J. 1*»'

. V-. . ■ -^



. •

: r » . _

- t I ' !•(:• ■■ - .

. ♦ X .1 ' ' 'lij

• . V

« ' * » « .►.. • 'V*'

• K, ‘ I , , _ . ,

' > . c

. •!■ I I I . ■

I .-

> . >i

) 4' . . *■ * I

: 1 ^ !».* ' '< •«<>

11 '2

The Model Revisited

In the 20 years since the model of translation quality assessment was first
published, it has been used by practising translators, by students of translation
and by translation scholars. The model was favourably reviewed as a pioneer
work and as an empirically based attempt to make the evaluation of translations
more objective. However, it has, of course, also been criticized. These criti-
cisms will be briefly discussed and, if found appropriate, taken into account
in the revision of the model. The revision will concern the categories for ana-
lysis, in particular those originally used for register analysis, the dis-tinctlon
between overt and covert translation including the cultural filter now substan-
tiated by empirical research, as well as a re-consideration of the whole notion
of “translation evaluation”.

1 Criticism of the Original Model of Translation Quality


Four major aspects of the model have been criticized:

1) The nature of the analytical categories and the terminology used.

2) Lack of Intersubjective verifiability of the analyses.
3) The “limits of translatability”.
4) The distinction between overt and covert translation.

I will deal with these points in turn.

1) The nature of the analytical categories and the terminology used

Newmark critically remarks that the “categorizations are too rigid” (1981:
182), as well as “jargon-ridden” (1979: 61), and that in fact the whole critical
apparatus is too complicated, but he has not elaborated on this point. In a
number of “personal communications”, however, he has made it clear that he
objects mostly to needlessly jargonized terms such as “overtly erroneous er-
rors .
Slote (1978) and Crisp (1981) also mention the complexity, unwleldlness
and cumbersomeness of the entire categorlal apparatus that would make it too
complicated to be used in translation classes or by practicing translators working
under time pressure: “To be truly useful, the eight categories should be tightened
up, reduced to six at the most. The very bulk o^the analytic tools would prove
to be a stumbling block to pragmatic use of the model” (Slote 1978:175). A
certain complexity and differentiation of the categorlal apparatus is, however,
necessary if the analysis is to facilitate insightful results, and any reduction of
analytic delicacy through less finely-grained analytical categories may yield
less reliable results. Thus, while certainly attempting to dejargonlze the ana-
lytic categories, I do not think it is possible to substantially reduce their com-
plexity. What can be achieved though is a re-grouping into fewer more general
categories (such as Field or Tenor), which, despite a necessary subcategorlsa-
tion, make the overall organization of the categorlal system more transparent
and simpler.
A more serious criticism is made by Brotherton (1981). Brotherton claims
that “the situational dimensions do not seem to provide any alignment for an
interpretation of the basic content of the source text, particularly as no
allowance is made for implicit elements” (1981: 19). He also finds fault with
the categorization of features as textual, syntactic, and lexical claiming that, in
many examples, the features allocated to one of these categories could with
equal justification be ascribed to either or both of the other two categories.
As to the point about the non-consideration of the basic content of the
source text: apart from being explicitly accounted for through the parameter
Province, content is implicitly considered in the way the other dimensions
feed into the ideational functional component. Therefore the type of textual
analysis provided in the model clearly does take account of content.
The claim that the general fit of linguistic, l.e., textual, syntactic and lexical
features with the situational categories is certainly correct, however this is due
to the nature of language and does not point to a basic shortcoming of the

2) Lack of intersubjective verifiability of the analyses

Konigs (1981: 205) criticises the lack of intersubjective verifiability of the ana-
lyses provided in the model. What is to be “verified”? The linguistic facts are
laid out, and can be argued in detail, should differences of interpretation arise.
So I do not think the analyses as such lack “intersubjective verifiability” - on
the contrary, the whole analytical apparatus is designed to explicate the basis

of any global judgement. It is conceded, of course tha't the Interpretation of
mismatches in the formation of an assessment of quality “contains” an inter-
pretative hermeneutic element. Indeed, I expressly stated that, in the last ana-
lysis, “translation is a copiplex hermeneutic process” (House 1981: 64). This
is an important point, and I will take it up again, when I discuss the nature of
translation evaluation in general (see 5 below).
It is this “relativisation” of the model which surprises Wilss (1980), who
states: “Ferner fiillt auf, daft House ihren modelltheoretischen Ansatz im Ver-
lauf ihrer Untersuchung in erheblichem Umfang relativiert. Gerade weil sie
dezidiert und mit Recht die Notwendigkeit eines rationalen Umgangs mit
Ubersetzungstexten betont und - Gott sei Dank - den theoretisch und me-
thodisch vagen Begriff der ”Wirkungsidentitat“ oder ”Wirkungsanalogie“
zwischen ausgangs- und zielsprachlichem Text nicht thematisiert, ist ihre ...
Parteinahme fiir die Subjektivitat des libersetzungskritischen Urteils eine
Uberraschung” (Wilss 1980: 5). However, as emphasized above, the whole
analytical apparatus enables the evaluator to make the analyses and interpreta-
tions transparent, explicit and non-subjective- but only to a certain point, i.e.,
the ultimate judgment of quality resulting from the analyses contains neces-
sarily a hermeneutic, subjective component.

3) The “limits of translatability”

The fact that I deliberately excluded poetic-asthetic texts from my analyses

(basing my views on Jakobson, Politzer and Wellek) disturbs a number of
critics, among them Slote (1978) and Brotherton (1981). The exclusion of pure
poetic-aesthetic texts from the scope of the analytic procedure provided inside
the model was then justified with reference to the fact that in the case of
poetic-aesthetic texts, the usual distinction between form and content (or
meaning) no longer holds, i.e., the physical nature of the slgnlflers in one
language can never be duplicated in another language.
While I still believe in the truth of this viewpoint, the model is and has to
be applicable to literary texts, indeed the tensions that arose in the process of
revising the model between covert and overt translations and the issue of cul-
tural filtering and cultural transferability are distinctly relevant with regard to
the translation of works of aesthetic and literary value. One of the reasons
why such texts were (and are) excluded from further consideration is that they
are outside my professional focus of Interest. This does not mean, of course,
that poetic-aesthetic text cannot and should not be considered by anybody
using the model in translation quality assessment.

4) The distinction between covert and overt translations
Newmark (1981) states that the two “translation methods” are not analysed
properly in terms of their real differences, and he clearly prefers his own distinction
between semantic and communicative translation claiming that “House confus-
ingly distinguishes ‘overt" (l.e., semantic) from ‘covert’ (i.e., communicative)
translation... unfortunately she does not distinguish stylistically between the
two types of translation” (1981: 52). Newmarl^ emphasizes, however, the ne-
cessity of making the distinction, as it clearly sets up an opposition against
assuming that all translating is (nothing but) communicating, with the less
effort expected from the reader, the better. Newmark felt that this crucial di-
stinction represents “a challenge to the prevailing view that everything must
always be done for the reader of a translation, that he must have everything
served up to him, that he is therefore the unifying and generalizing factor for
every text type and translation procedure. Which I cannot accept” (1981: 68).
I cannot either, I may add, and I cannot agree more with Newmark today, 15
years later, when what he called “the prevailing view” seems to be more “pre-
vailing” than ever.
Relfi and Vermeer (1984: 48ff) claim that three definitional criteria for a
translation, l.e., form, meaning and function are not clearly distinguished one
from the other In the model. To substantiate this claim, they point to the {overt)
translation of Hebei’s moral anecdote, the function of which was originally to
give a moral lesson. In translation, however, the functions might be many
different ones, e.g., to allow linguistic comparisons, to demonstrate the aesthe-
tics of the composition etc. This statement shows that Relft and Vermeer mi-
sunderstood the notion of an overt translation, which has a second level func-
tion that Is not different but is in fact closely related to the function of the
original text in that it allows target culture readers access to the original func-
tion. If, however, different secondary functions were added to the translation,
an overt version would result in the terms of my model, i.e., a bilingual textual
operation which Is no longer considered to be a translation.
Fal£ (1981) makes the point that what was called refinement of the model
(following the analyses of the test corpus) was In fact “a mere relativlsation of
the model, and the distinction made between covert and overt translation is
valuable only in so far as it has to be conceded that two such types of translation
exist. Moreover, the usefulness of this differentiation to the translator himself
seems practically null, since a particular source text ’’does not necessarily re-
quire either a covert or,an overt translation, given the different ways of viewing
a text and different special purposes for which a translation may be requlred“.
House, p. 210” (Faifi 1981: 80). Surely, this criticism is Invalid: if “elther-or”
holds, then this makes the distinction particularly relevant rather than the

2 Rethinking the Categories for Analysis

In the following theoretical argument I wish to review the framework for the
analytic categories, and, specifically clarify the relationship between textual
function, linguistic characteristics and social use of a text by introducing the
category genre.
The framework in the original model for textual analysis and the estab-
lishment of textual function was register analysis. The notion of register pro-
poses an intimate relationship of text to context. Register, l.e., functional lan-
guage variation, refers to what the context-of-sltuation requires as appropriate
linguistic realizations in a text. Register is thus basically a “contextual category
correlating groupings of linguistic features with recurrent situational features”
(Gregory and Carroll 1978: 4).
In the literature reviewed in Chapter 2, registers are also described as sets
of particular foregrounded lexicogrammatlcal choices. Bhatia (1993: 5) speaks
of registers enabling “surface-level linguistic descriptions” of texts. However,
one may need another category to explain “deeper” similarities or differences
between texts. Assigning entire texts to different registers on the basis of the
realized choices revealed in the analysis may be forbiddingly complex. Further,
two texts may have similar linguistic features but may still be intuitively felt
to be texts of a different type. Thus, it may be advisable to look for a text’s
overall patterning in order to be able to decide which texts might belong to
the same register. If texts are found to belong to the same register, they can be said
to belong to the same “deeper”, underlying generic structure (Halllday 1978),
which defines them as belonging to the same genre. In other words, the concept
of genre refers to discourse types, it is a category “superordinate” to register.
While Hasan (1978) seems to have looked upon register and genre as syn-
onyms, Halllday originally saw genre as “language as projection of higher-
level semiotic structure” (1978: 134) suggesting that “the concept of generic
stucture can be brought within the general framework of the concept of regi-
ster” (1978: 145). Following Halllday and Hasan’s seminal work, text-based
register analysis has thus more recently (Halllday 1989; Halllday and Hasan
1989; Martin 1989; 1993) been extended to take account of the semiotic di-
mension and of different types of discourse or genres - although the notion
of “genre” was already foreshadowed in Crystal and Davy’s (1969) situational
dimension “Modality” - a dimension I had not considered in the original
model. In the revised version of the model I will, however, integrate genre into
the categorlal system. Before demonstrating how this is done, I want to give
a brief review of some recent relevant literature on genre.
Genres such as e.g., the academic paper or the market report, are conceived
as cultural discourse types featuring different configurations of lexical and

grammatical units characterized as registers, with different register choices
realizing different genres. Or, put differently, genres result from social iden-
tifications in terms of use, source or function, and registers are the result of
decisions inside a genre choice concerning field, mode and tenor. For example:
suppose a publisher tries to sell a children s book that no child can understand,
then this is an unsuccessful Instance of the genre, because the register is wrong.
Martin (1989) argues that genres constrain the ways in which the indivi-
dual register variables operate in combination with each other in a particular
culture and in a complete text (cf. also Couture 1986, who stresses the point
that genres are completable structured texts, while registers represent more
generalizable stylistic choices). Different topics may thus be conventionally
more or less suitable for an academic lecture, and such conventions may vary
across cultures. Genre links texts to culture in that they “refer to the staged
purposeful social processes through which a culture is realized in a language”
(Martin and Rotherby 1986: 243).
For the comparison and evaluation of texts in translation evaluation,
questions of genre are potentially Important in that each of the texts will be
“related to ‘certain shared knowledge’ about the nature of the other texts of
the same kind, that is, to the concept of ‘genre’.” (Ventola 1995b: 121).
According to Martin (1989; 1993) register and genre are both semiotic
systems realized by language, a special kind of semiotic system. The relation-
ship between genre-register-language is seen in terms of semiotic planes which
relate to one another in a Fljelmslevian “content-expression” type, i.e., the
genre is the content-plane of register, and register is the expression-plane of
genre. Register in turn is the content-plane of language, with language being
the expression-plane of register. This relationship can be represented as in
figure 1 (adapted from Ventola 1984: 277). The fact that a plane functions as
an expression-plane of higher semiotics is indicated by setting the semiotic
boxes placed on top of each other and by showing realization on the lower
planes through the dotted lines.


Reg ister


Fig. 1: Genre-Register-Language/Text

The relationship between genre and register is then such that generic choices
are realized by register choices, which in turn are realized by linguistic choices

that make up linguistic structures in the instantiation of a text. We are able to
categorize texts into particular genres on the basis of what we know about
texts and their intertextuality and the social uses of texts, l.e., genre is taken as
the basis on which we arguable to classify texts as of the same kind.
Genre has been approached from many different angles: literary, sociolo-
gical, folklorlstic, psychological, anthropological. I will here restrict myself to
considering the linguistic ones.
Inside linguistics, as we have seen above, it was mostly systemic (Halll-
dayan) linguists who explored the concept of genre. However, other linguistic
approaches have also concerned themselves with genre: in the framework of
ethnographically oriented linguistics, Hymes juxtaposes genre and speech event:
“Genres often coincide with speech events, but must be treated as analytically
Independent of them. They may occur in (or as) different events. The sermon
as a genre is typically identified with a certain place in a church service, but its
properties may be invoked for serious or humorous effects in other situations ”
(Hymes 1974: 61).
Hymes insight that genres are to be treated as analytically independent of
speech events is particularly relevant for my purpose of establishing categories
for the analysis and comparison of original and translation: a translation em-
beds the text and its genre in a new speech event in the case of an overt trans-
lation, and it recreates an “equivalent” speech event in the case of covert trans-
lation. Theoretically, the genre of a text is something to be kept equivalent in
both overt and covert translation.
From a sociolinguistic vantage point, Ferguson (1994) sees genre as a con-
ventionalized message form that “recurs regularly in a community (in terms
of semantic content, participants, occasions of use, and so on) and will tend
over time to develop an identifying internal structure, differentiated from other
message types in the community.” (1994: 21).
To summarize this brief discussion of genre which is to serve as a new
category in the revised model, I will define genre for my purposes here as
follows: genre is a socially established category characterized in terms of oc-
currence of use, source and a communicative purpose or any combination of
these. Inside my model, genre might serve as a category linking register (which
realizes genre) and the individual textual function (which exemplifies genre).
The resultant revised model consists then of four levels: function of the indi-
vidual text, genre, register and language/text.
Whereas the original model adapted Crystal and Davy’s (1969) system of
categories for register analysis, the categorial system will now be revised in
such a way that, while retaining a number of both Crystal and Davy’s and my
own “old” categories, I now subsume them under the simplifying Hallldayan
“trinity” Field, Tenor, Mode (see Halllday and Martin 1993). I also make use

of Biber s (1994) “Analytical Framework for Register Studies” for some con-
ceptual differentiations.
On the basis of the above considerations and in adaptation of Martin
(1993: 120) the following scheme results:

Fig. 2: A Scheme for Analysing and Comparing Original and Translation Texts

The register categories FIELD, TENOR and MODE are explained and/or
subdifferentiated in the following manner:

refers to the nature of the social action that is taking place, it captures “what
is going on”, i.e., the field of activity, the topic, the content of the text or its
subject matter. Here I will differentiate degrees of generality, specificity or
“granularity” in lexical items according to rubrics of specialized, general, po-
pular. This dimension is similar to Crystal and Davy’s Province, which how-
ever specifies some features that will now be subsumed under Genre (e.g.,
professional character of a field as in a religious sermon).

refers to who is taking part, to the nature of the participants, the addresser and
the addressees, and the relationship between them in terms of social power

and social distance, as well as the “degree of emotional charge” in the relation-
ship between addresser and addressee(s) (Halliday 1978: 33). Included here
are the addressers temporal, geographical, social provenance as well as his
intellectual, emotional or affective stance (his “personal viewpoint”) vis-a-vis
the content he is portraying and the communicative task he is engaged in.
Stance is reminiscent of Crystal and Davy’s category Individuality - which I
once rejected, but now feel is an Important, if often neglected, variable (see
Ochs 1989 and Taavitsainen 1994 for recent convincing pleas to take the cate-
gory of “subjectivity”, “personal affect”, “stance” and the role of affect in
meaning making more seriously).
The subdivisions of the dimension Social Attitude in the original model
in terms of Joos’ five styles will be simplified such that a tripartite division
into formal - consultative - informal is now adopted for a more economic

This category captures both parts of Crystal and Davy’s (1969) parameter
“Discourse”, i.e.. Medium and Participation in their simple and complex
manifestations. In my scheme, therefore, MODE refers to both the channel
- spoken or written (which can be “simple”, e.g., “written to be read” or
“complex”, e.g., “written to be spoken as if not written”), and the degree to
which potential or real participation is allowed for between the Interlocutors.
Participation can also be “simple”, i.e., a monologue with no addressee parti-
cipation “built into the text” or “complex” with various addressee-involving
mechanisms characterizing the text (see also Chapter 2, p. 40 for a more de-
tailed description of these categories).
In taking account of the differences in texts between the spoken and the
written medium, I will additionally, when appropriate, make use of the empi-
rically established (corpus-based) oral-literate dimensions hypothesized by
Biber (1988). Biber proposes correlates of medium by suggesting dimensions
along which linguistic choices may reflect medium. These parameters are as

(1) Involved vs Informational Text Production

(2) Explicit vs Situation-Dependent Reference
(3) Abstract vs Non-Abstract Presentation of Information.

Along the dimension of Involved vs Informational text production (cf. also

Chafe 1982), spoken genres tend to veer towards Involved production, written
genres towards informational production. Thus, with respect to each of these
three dimensions, the poles characterize academic exposition and conversation
respectively. However, as regards dimension (1), among written genres, personal
letters are clearly marked by “Involvedness”, and among the spoken genres.

prepared speeches and broadcasts are strongly marked for informational pro-
duction. Along dimension (2), written genres are strongly marked for explicit
reference, spoken ones for situation-dependent reference. But public speeches
and interviews behave like written genres along this dimension, and fiction
genres resemble spoken ones. Along dimension (3), written genres tend to be
full of abstract information, spoken genres tend to lack it. But fiction genres
and personal letters resemble spoken genres a^ong this dimension.
Thus, none of the dimensions defines an absolute spoken/written distinc-
tion, l.e., Biber (1988) confirms (on the basis of large-scale corpus analyses)
the pioneer insights and register analyses by Gregory (1967), Crystal and Davy
(1969) and others who set up categories such as “complex medium” and “com-
plex participation”. Biber (1988) shows that it is possible within each mode to
override the salient situational characteristics of the mode and to overcome
the situational constraints operative in each medium, thus creating discourse
that is atypical for that medium.
Such findings and insights are, of course, directly relevant to the type of
cross-linguistic textual comparison that I propose as an Integral part of trans-
lation quality assessment. However, cross-linguistic and cross-cultural com-
parisons of texts require first of all considerable research into the range of oral
and written speech situations and the distribution and functions of linguistic
features Inside each of the linguacultures involved (cf. e.g., Besnier 1988, for
a careful ethnographic analysis preceding contrast).
The type of linguistic-textual analysis in which linguistic features disco-
vered in the text are correlated with the register categories FIELD, TENOR
and MODE is not changed from the original model as outlined above, neither
is the way a “statement” of the INDIVIDUAL TEXTUAL FUNCTION -
consisting of an Interpersonal and an ideational functional component, is de-
rived from the register analysis. Additions and modifications Include the in-
troduction of the category GENRE, “in between” as it were the register char-
acterisation and the textual function, as well as the subdivisions of the register
categories TENOR {Author's Provenance and Stance^ Social Role Relationship
and Social Attitude) and MODE {Medium as reflected by linguistic choices
along the three parameters “involved vs Informational”, “explicit vs situati-
on-dependent”, and “abstract vs non-abstract Information”). The exact ana-
lytical procedure will be exemplified in Chapter 5 with a small corpus of texts.
Before I can proceed to present an application of the new categorlal system
to a corpus of texts, however, the Important distinction between overt and
covert translation needs to be re-considered.

3 Rethinking the Overt-covert Dichotomy and Integrating
the Results of Contrastive Pragmatic Research

One of the more important theoretical distinctions resulting from the trans-
lation analysis and evaluation of the original corpus was the dichotomy overt
versus covert translation. Related but not Identical distinctions have cropped
up in the literature on translation providing different variations on the century
old theme of literal vs free translation, in particular Dlller and Kornelius’ (1978)
prlmare vs sekundare Ubersetzung, Newmark’s (1981) semantic vs commu-
nicative translation, Nord s (1988) dokumentarlsche vs instrumentelle Uber-
setzung; Gutts (1991) direct vs indirect translation; Pyms (1992) observational
vs participative posltlonings; and Schreiber s (1993) Textiibersetzung versus
While certainly valuable in their own rights, few of these distinctions are
an integrative part of a model or a theory of translation, in which the relation
of the two hypothesized types of translation to certain translation procedures,
types of equivalences and possibilities for change along extrallngulstic dimen-
sions are systematically accounted for. However, it is exactly such a link be-
tween theory and practice which is attempted through the empirically derived
distinction between covert and overt translation. The idea behind the oppo-
sition covert vs overt translation is thus not only to present the translator with
two possible lines of action, it also involves important conceptual distinctions.
While the basic original distinction between overt and covert translation
(see Chapter 2) will be retained, I will try to develop the distinction and achieve
greater explanatory adequacy by relating it to the revised analytic model and
to the concepts of “frame” and “frame shifting” as conceived by Bateson (1972)
and Goffman (1974), as well as to Edmondson’s (1981) concepts of “discourse
worlds” and “world shifts”. These concepts are of explanatory value for the
notions overt and covert.
Translation involves text transfer l.e., “the material moving of texts across
space and time” (Pym 1992). Whenever texts move, they also shift frames and
discourse worlds. “Frame”, for Bateson (1972) a psychological concept and
in a sense the psychological pendant to the often more “socially” conceived
concept of context, delimits a class or set of messages or meaningful actions.
It often operates unconsciously as an explanatory principle. “Like a picture
frame, the (psychological) frame tells the Interlocutor that he must not use the
same line of thinking in interpreting the picture that he might use in interpre-
ting the wallpaper outside the frame.” (Bateson, 1972: 187-188). A frame is
metacommunlcative: any message that defines a frame gives the receiver in-
structions in his interpretation of the message Included in the frame. An explicit
frame-setting message would, for instance, be “This is ironic”.

Similarly, the notion of a “discourse world” refers to a superordinate frame-
work for Interpreting meaning in a certain way: “A discourse world is to be un-
derstood as an application of the notion of a possible world derived from
logical semantics to the pragmatic interpretation of conversational behaviour”
(Edmondson, 1981; 201). In Edmondsons model of discourse analysis the
locutionary act acquires an Illocutionary value by reference to an operant dis-
course world. World switching or the handllngspf two discourse worlds occurs
for Instance when the teacher moves from an “unreal” instructionary world
to the “real” world of the classroom.
Returning now to the twin concepts of overt and covert translation: in
overt translation only a “second level function” can be reached because the
translation embeds the text in a new speech event, which gives it also a new
frame. An overt translation is a case of “language mention” (as opposed to
“language use” in covert translation). Overt translation is thus similar to a
citation or quotation.
The notions overt translation and second-level function in overt transla-
tion can be clarified through reference to the four levels in the revised analytic
ginal text and its overt translation are to be equivalent at the level of LAN-
GUAGE/TEXT and REGISTER (with its various dimensions) as well as
tional equivalence is still possible but it is of a different nature: it can be described
as enabling access to the function the original text has (had) in its discourse
world or frame. As this access is realized in the target llnguaculture via the
translation text, a switch in the discourse world and the frame becomes neces-
sary, i.e., the translation is differently framed, it operates in its own frame and
discourse world, and can thus reach at best what I have called “second-level
functional equivalence”. This type of functional equivalence is achieved
through the required equivalence at the LANGUAGEA’EXT and REGISTER
levels, which facilitates the co-activation of the original’s frame and discourse
world. In this way members of the target llnguaculture may eavesdrop, as it
were, i.e., be enabled to appreciate the original textual function, albeit at a
The work of the translator in overt translation is Important: the results of
her work are clearly visible. Since it is the translator’s task to allow persons in
the target culture to gain access to the source text and its cultural impact on
source culture persons, the translator puts target culture members in a position
to observe, be worked upon and evaluate the original text’s function as mem-
bers of the target culture.
In presenting this analysis of the relationship between overt translation
and psychologically and socially conceived notions of context (frame, discourse
world and register), I basically agree with Gutt’s (1991: 165) statement that

direct translation (the term he uses as a rough equivalent for overt translation)
should be processed with respect to the original context. However, I believe
that my analysis is more differentiated and has thus greater explanatory value.
The case of a literary*work, which I had originally classified as “timeless,
of general time/space-transcending human and aesthetic interest” seems to be
a special one. While it is of course true that the literary work carries/creates
its own cultural frame of reference, textual interpretation is, in part, both time-
bound, and culture-bound. The suggestion is then that for an established work
of literature, two discourse worlds co-exist for the contemporary reader
situated in the llnguaculture of the writer, and both are co-actlvated by the
readership of the translation, such that three discourse worlds can be said in
this instance to co-exist. This situation is represented in Figure 3, which, how-
ever, takes no account of the location of a translation (and Indeed of different
translations) along the temporal axis. In other words. Inside the framework
developed inside Fig. 3, the contemporary reader of bowdlerised Shakespeare
translated into German can access the text via four discourse-worlds, possibly
all at once.
For example, Joseph Conrad s place in English literary tradition gives his
work a particular “function” for the (Informed) English reader. This “function”
presumably does not hold for the German reader. Nor should it! The trans-
lation is, on the whole, to be overt^ in the sense that no cultural filtering is in
principle licensed. The German reader can, however, in theory, get the relevant
intertexts by simply reading a number of English writers in translation.

Then Temporal Axis Now

The Text
(functioning in linguaculture 1)

Original readership Contemporary readership

(discourse world 1) (discourse world 2)


(functioning in linguaculture 2)
Contemporary readership
(discourse world 3)

Fig. 3: Reading Historically-established Literary Texts

In overt translation the reader can/will operate in different worlds (as will
the quality assessor). Clearly, in the overt case, a reader can be moved, upset,
horrified as though he were in fact a member of the source linguaculture - this
is what Wordsworth called the “willing suspension of disbelief”. But more
often than not (perhaps via footnotes), the reader will be made aware of entering
into the “as though” game.
As distinct from overt translation, in cove^ translation, the translator at-
tempts to re-create an equivalent speech event. Consequently, the function of
a covert translation is to recreate, reproduce or represent in the translation text
the function the original has in its llnguacultural framework and discourse
world. A covert translation operates quite “overtly” in the different frame and
discourse world provided in the target linguaculture without wishing to co-
actlvate the discourse world in which the original had unfolded. Covert trans-
lation is thus at the same time psycholinguistically less complex than overt
translation and more deceptive. The translator s task is, in a sense, to cheat,
and to be hidden behind his feat, the transmutation of the original. The cultural
filter he employs is so skillfully integrated into the fabric of the text that the
seams do not show. Since functional equivalence is aimed at, changes at the
levels of LANGUAGE/TEXT and REGISTER may if necessary be underta-
ken. The result may be a very real distance from the original text, which is
why covert translations are received as if they were in fact originals. This char-
acteristic belying of their origins in covert translation texts creates ethical pro-
blems because of the deception of the origins of the text. Pym (1992) points
to the danger Inherent in such a procedure, and he points out that what he
terms “radical misrepresentations of transfer” should also be critically ex-
amined as “an eclipse of intercultural distance and thus as potentially perni-
cious imposition of cultural homogeneity.” (1992: 185). One is reminded here
of Venuti’s (1992 and see above p. 11) preference for a translation to openly
show points of difference.
In the terms provided in the revised analytic model, the following situation
seems to hold in the case of covert translation: at the levels of LANGUAGE/
TEXT and REGISTER (with its dimensions of FIELD, TENOR and MODE)
the original and a covert translation need not be equivalent, i.e., the original
can be manipulated using a cultural filter to be based on cross-cultural empi-
rical research.
At the level of GENRE, and at the level of the INDIVIDUAL TEXTUAL
FUNCTION, equivalence is necessary, i.e., the distinction between a trans-
lation and an overt version (where a special secondary function is added overtly
to the translation) as well as between a translation and a covert version (resul-
ting from an unjustified application of a cultural filter) made in the original
model, still holds in the revised model. Consider also the following case: if, in
a covert translation situation, the GENRE established for the original text

does not exist in the target linguaculture, then translation {covert or otherwise)
IS impossible, i.e., a version results. In an overt translation situation, it is irre-
levant whether the GENRE holds in the target linguaculture. In the embedded
or “secondary” world of course, GENRE is noted “ + i.e., kept constant.
Overt and covert translations differ on the dimension of REGISTER and
the demands of equivalence of the communicative values of the linguistic units
in the two texts. They also clearly differ in terms of the possibility of reaching
equivalence of the INDIVIDUAL TEXTUAL FUNCTION. These dif-
ferences can be displayed in the following diagram:

Is strict equivalence the translational goal?

Overt Translation Covert Translation

Primary level function NO YES

Secondary level function YES (does not apply)


Register YES NO

Language/Text YES NO

Fig. 4: The Dimension Overt-Covert Translation

4 The Meaning of the Cultural Filter

The cultural filter to be applied in a covert translation has been given some
substance in my own work on Anglophone and German communicative dif-
ferences and priorities along a set of hypothesized dimensions. Converging
evidence from a number of cross-cultural (German-Engllsh) studies suggest
that there are German communicative preferences which differ from Anglo-
phone ones along five dimensions: directness, self-reference, content-focus,
explicitness, and routine-reliance.
To see what this means for the analysis of covert translations, let us look
again at the two texts from the original corpus, the commercial text and the
journalistic article, which were analysed and judged as covert versions (cf.
Chapter 2 for the original analyses of these texts and the Appendix for the
texts themselves). In both cases, the analysis and the resultant evaluation may
now be revised in the light of results of the contrastive pragmatic German-

English research described above. It must be pointed out at this stage (and see
also below 5) - and this is another “revision” of the original model - that it is
necessary to distinguish much more between the analysis proper and the
judgment, the evaluation. Unlike the scientifically (linguistically) based ana-
lysis, the evaluative judgment is ultimately not a scientific one, but rather a
reflection of a social, political, ethical, moral or personal stance.
With this qualification in mind, we can ^e-analyse and re-evaluate the
commercial and the journalistic text. In the commercial text, the finding that
the Interpersonal functional component had been drastically altered due to
changes along the parameters Social Role Relationship and Participation may
now be considered to have a basis in the different communicative preferences
of readers in the two llnguacultures. In other words, although the German
translation text was found to be more direct and blunt, l.e., to exhibit a much
less notlcable attempt to explicitly involve the addressees rendering it less flat-
tering to the addressees, less diplomatically polite and less deliberately non-
committal, these changes may well have been achieved in accordance with the
type of directness and content-focus (vs an interpersonal focus) discovered in
German communication norms. The changes, which were called “mismat-
ches” in the original analysis leading to “serious shortcomings” of the trans-
lation text, can now be seen in a different light, l.e., as the natural consequence
of the differences in communicative norms in the two llnguacultures.
Similarly, what was diagnosed as a mismatch in the original analysis of the
translation of the journalistic article (see Chapter 3 and Appendix) must be
re-analysed in view of the results of their German-Engllsh contrastive find-
ings. As in the commercial text, the Interpersonal functional component had
been changed or rather substantially weakened in the German translation of
the journalistic text through differences along the parameters Medium, Parti-
cipation, Social Role Relationship, Social Attitude and Province. Taking ac-
count of all the changes along these parameters, the analysis had revealed that
the Ideational functional component had been “upgraded” to a remarkable
degree in the German text making it much more “content-focussed”, more
concerned with the transmission of Information than with a consideration of
the readers’ reception of this information as had been the case in the original.
The transmission of information being of prime Importance, the German
text has become less personalized, less dramatized, less journalistically attrac-
tive but rather more precise and sober, which was reflected for Instance in the
use of scientific terms. In the final “statement of quality” (see above p. 65) it
was pointed out that the American author’s intention of making his material
easily digestible and interesting for his audience, had failed to come across in
the German translation, mainly because the author’s concern with passing on
scientific facts had become much more important in the translation - “unwar-
rantedly” so as was then found. Given the results of the contrastive pragmatic

analyses, however, this analysis must now be revised, T.e., the translator may
well have been aware of the differences in communicative norms and expec-
tations in the German and English-speaking linguacultures, and his translation
consequently did no more than reflect these differences. In other words, what
was listed as a cluster of mismatches may no longer count as such, and the
translation originally diagnosed as a covert version can be seen as a covert
The Integration of an empirically verified cultural filter into the revised
model would at first sight mean that there is generally greater certainty as to
when a translation is no longer a translation but rather a covert version. Given
the fact that for the language pair English-German some relevant work exists,
and that many studies have been conducted over the past ten years or so in
the area of contrastive pragmatics involving other languages as well, there
surely exists a better basis for making and evaluating covert translations in a
non-arbltrary way. However, given the dynamic nature of communicative and
societal norms and the evolving process of research, both translators and trans-
lation critics will still have to be maximally aware of research results and hy-
potheses in the area of cross-cultural pragmatics to help them judge the ap-
propriateness of changes through the application of a cultural filter in any
given language pair.
Further, as mentioned above, it is Important to be aware of the difference
between (scientifically based) analysis and (social) judgment in evaluating
translations. To differentiate between a covert version and a covert translation
presupposes a judgment of social norms, but such norms are not stable and
static but dynamic and subject to manifold influences. In the final analysis,
however, such judgments of norms are of limited value for an Individual text,
and the type of case study judgments I am making here on the basis of an
individual textual profile or norm. In other words, despite the cross-cultural
pragmatic evidence provided for the cultural filter, I am strongly cautioning
against making quick decisions about what is a covert version and what is a
translation. While theoretically upholding the distinction between a covert
translation and a covert version because it is necessary for conceptual clarity,
it may be difficult in practice to make an unambiguous judgment. While such
a pronouncement may be taken to be “relativistic”, I believe it is advisable to
maintain such position in the face of our never-completed knowledge of com-
municative norms in any two linguacultures, rather than make precnptlve
statements preclosing our minds to the complexity of translation.
This touches of course on the more general point of the nature of evalua-
ting the quality of a translation as opposed to comparing textual profiles, and
describing and explaining differences established in the analysis.

5 Rethinking the Notion of “Translation Evaluation”

My approach to translation quality assessment is thus not “absolutely evalua-

tive”. Indeed, the relativism commented on by Wilss (see above, p. 103) merits
elaboration and reinforcement. The model was never and is not now intended
to provide for” absolute evaluation” and the following statement still holds:
“a detailed hierarchy of errors for any lndivl4ual case can only be given for a
particular comparison of two or more texts depending in any particular case
on the objective of the evaluation” (1981: 209). Gutt’s (1991: 13) criticism that
the model provided only the basis for systematic comparison, but not for value
judgments, as those would have to follow an “assumed detailed hierarchy of
errors which is specific for the set of texts to be compared and the objective
of the evaluation” is, however, not valid as the model clearly lays open the
many factors that might theoretically have Influenced the translator in making
certain decisions and rejecting others, thus providing the basis for evaluation
in a particular case, which is much more constructive than feebly claiming, as
Gutt does, that “whatever decision the translator reaches is based on his in-
tuitions or beliefs about what is relevant to his audience.” (1991: 112).
Instead of taking the psychological category of readers’ “intuitions and
beliefs” as the cornerstone of translation evaluation, the model presented here
is a text-based, linguistic one in the broad Hallldayan conception of linguistics,
which looks at language in social life and focusses on texts, the products of
human decision processes that are the most tangible and least ambiguously
analysable entitles.
In the type of detailed comparison and evaluation of source and translation
texts provided for in the model, the evaluator is not put in a position to give
easy judgments of “good” or “bad” in translation. Rather, the model prepares
the ground for the analysis of a large number of evaluation cases that would,
in any Individual case, not be totally predictable, however. This is so because,
in the last analysis, any evaluation, depends on a large variety of factors which
condition social evaluative judgments. These judgments are dependent on, or
rather follow from, the analytic, comparative process in translation evaluation.
The model, however, is based on the assumption that translation is a linguistic
phenomenon (in the Hallldayan sense of linguistics), and the linguistic analyses
provide a basis for judgement and grounds for arguing an evaluative judgment
- which, in fact, means that there is less an opposition between analysis and
judgement, rather the latter follows from the former.
The choice of an overt or a covert translation depends not just on the
translator himself, or on the text or the translator’s personal interpretation of
the text, but also, and to a considerable extent, on the reasons for the transla-
tion, on the implied readers, on publishing and marketing policies. In other

words, in translation there are many factors that cannot be controlled by the
translator and have nothing to do with translation as a linguistic procedure or
with the translator’s linguacultural competence. Such factors are social factors,
they concern human agepts and socio-political or even ideological constraints
that normally have far greater power and Influence than the translator. Still, a
translation is also a linguistic-textual phenomenon and can be legitimately
described, analysed and assessed as such.
I consider the concept of “quality” in translation - after all, the point of
the whole model - problematical if it is meant to refer to value judgments
alone. It is problematical especially if one does not know anything about, or
does not take into account (for various reasons), the Ideals and ideas about
translation quality the translator, reviewer, or researcher entertains. It is dif-
ficult to pass a “final judgment” of the quality of a translation that fulfil the
demands of objectivity. Can one therefore refer to the model as “twenty years
of interesting errors”? Surely not! As a field of Inquiry, translation criticism
will always have to move from a macro-analytical focus to a micro-analytical
one, from considerations of Ideology, function, genre, register to the commu-
nicative value of individual linguistic units in order to enable the reconstruction
of the translator’s choices and his decision processes in as objective a manner
as possible. This is a highly complex and, in the last analysis, probabilistic
undertaking. If one refrains from giving prescriptive, dogmatic and global
judgments rather than reveal exactly where and with what consequences and
possibly why a translation in an individual case is what It is in relation to its
original, one proves that one has some respect for both the subject of transla-
tion and the translator.

iV I '.M -1 f. Ws;. ■*lj.jV»

V.* ’’fj )*^tx'- Tt^t' »■ *-* I’ lf' jsi tas II "'jtw rr ■ ' ^ ’;:2

‘I. ■ns^W'w'»■* > ‘ . ■

«rii/ ‘ "'' *'

#Oi6ti ■’'■ '-.4.'**!'''* ; /-. • **y t - ^ i.

yi ..- '.I I.'* ri . v: ", */ •' 't •'

■•. ■ 1. ' ii .■* C* "■-V

<4, fn«o.\ *:■- 1M. i-i^nr /Kacvklua^'i r-* ''•ft’fj ,• vi> ^/v^'. ••> i .i>-
tJc *<:. rr, ^If ;ltJU'^ <yl *.4tr»i ; 7 ?■• ‘ ^Jlf( Jt t?-”* • ...

’. f SO'Jtt ’t V»> •

<'t» 4 *1 ♦!' ' '*‘j . .*1/1 fi ■■''4’■ * ^«(|| *. ■»* ®

wi »..*■' ^ ' ‘ ■' ^

goaatj * . . li,* ► i* ’ ' ". H*'' n • i ‘ f--'Oi .4|j,.>

. i. i. 'i; '‘ii£>|>jji , i l <i» ‘ ^ - r. V * ' *»'“> •vft.fsyiv* ' ^

I. t*»<J. l\ !L^<<>. ^ ' '» - ' »■ . ■ .C 0 v! > ' ' !#l/i'S

•».*nfi'".'* ^<^' T'' ■>■' ■ . ■« J . ^.■•" •• . *. ^/i.'-

*• <!.» .’ iU/ .»'7iii; * j " ‘ fvu*.;*.*’ .. y fH’£l£V«fl

I*. .•r_ t _ , • . ..* «wl -v.? 7 » • ' . • t*»►''‘■(M ^

!i'^ :..i VO ' .i>rf« f I* '- ! ' • 1^

! .il* r-j |l t If,"- , >1 .) U

brtit ai ••■- 'ipTH. wj, ii ■■ aiAfc •i'^v . •'v-:- * ’V T* -. * ■* (<ift^0VjrU^

Ui ».. 1. - .! ' d^ »' i' . . U ' «■' *. ». 1'* I 11 ft , yiw

• »J ,«( .. , , i rt5i>'i 1/ I • '! "> ■ ■ . ^ ti ^t/O !*.!i* » >/O’‘0 : "5 . ' ‘4 7<*

, . Ite.^ . '.***

, «*«■

• K-t

't. • « : a

* ■ M (y > -


■• • ■ * ,
; 1 • ' r n i
. I .W . ■ "V

"'t* K ..

t ► »


Analysis and Comparison of Source and Translation Texts

In this Chapter I want to demonstrate the viability of the revised model of

translation quality assessment by presenting a number of “model analyses” of
original texts and their covert or overt translations. The texts are to be found
in the Appendix. For easy reference in the presentation of the analyses para-
graphs are numbered sequentially in Roman numerals.

The texts are:

1. An English children’s book translated into German (taken from a larger
corpus of children’s books and their translation, cf. House (forthcoming).
2. An excerpt from an autobiography by a Nobel prize winning scientist
written in English and translated into German.
3. An excerpt from Walter Benjamin’s famous essay “Die Aufgabe des Uber-
setzers” and its English translation.
4. A passage from Daniel Jonah Goldhagen’s book Hitler’s Willing Execu-
tioners and its German translation.

I have deliberately refrained from including in my sample analyses texts for

immediate practical use and fast consumption, such as Instructions for use,
advertisements, signs, day-to-day journalism etc., l.e., texts in whose transla-
tion the step from a covert translation to different forms of multilingual text
production is dangerously small. I have exluded such texts because 1)1 have
already given a number of examples of translations of such textual types in
connection with the discussion of cultural filters in covert translations (Chap-
ter 3), 2) I have Included two such texts and their translations from the corpus
of the original model for translation quality assessment presenting in full the
original analyses and re-discussing the employment of the filter and their status
as covert translations or versions (Chapter 2), and 3) I want to make a case for
texts that go beyond ephemeral textual exemplars.

TEXT l: Childrens Book
Jill Murphy Five Minutes^ Peace (original English, translation German) see

Analysis of the Original

This is a simple little picture book story about an elephant family: Mrs Large
and the children Lester, Laura and the baby. Mrs Large wants “five minutes
peace” and retreats to the bathroom but is soon disturbed by all three children.
This is however not taken as a dlsquletenlng occurrence but rather seen as part
of ordinary family life with young children. The story designed to be read to
young children is told with warmth, good-natured common-sense, a dry sort
of humour and realism.

Lexical means:
Preponderance of lexical items that are likely to be part of the competence of
young children from their interactions in the Immediate home and neighbour-
hood environment.

Syntactic means:
Short clauses with simple structures throughout the text.

Textual means:
Strong textual cohesion to make the text easily comprehensible and digestible
for young children. Textual cohesion is achieved through iconic linkage as well
as theme dynamics.

Iconic linkage between a number of clauses highlighting (for the children’s

benefit) a similarity of actions and behaviors, and heightening the dramatic
effects, which is also often further intensified through foregrounded, rhematlc

IX: Can I see ... Can I have ... Can I get...

VI; VII: So Lester played. So Laura read.
VII; VIII: In came Laura. In came the little one.
Ill; XI: Where are you going ... To the ... Why ... Because I want five minutes
peace from you lot. That's why. Structural parallelism between II and XL

Theme Dynamics: Thematic movement frequently arranged in sequences of

theme-rheme movements to ensure given-new ordering, e.g., I.

Foregrounded rhematlc fronting

XII: And off she went... VII: In came Laura

Author's Temporal, Geographical and Social Provenance:
Unmarked, contemporary, standard middle-class British English

Author's Personal {Emotional and Intellectual) Stance:

The author clearly views the family life she portrays with humour, involve-
ment, empathy and a sense of the absurdly funny characterizing the domestic
life a mother with young children routinely experiences. However, she never
sentimentalizes her characters: although elephants, the characters keep their
dignity and their names: Mrs Large, Lester, Laura.

Syntactic means:
Presence of structures describing typically stressful situations with laconic
humour and empathy.
I: The children were having breakfast. This was not a pleasant sight. (The
picture nicely attests to this statement!)

Social Role Relationship

Author - reader(s): author puts herself on a par with her double-tiered ad-
dressees, i.e., both the parents reading this story and the children as the “end-
receivers” or secondary addressees. The relationship if one of empathy and
Author - characters in the story: respect for the individuality of the children
shown through the names they are given, and the fact that they are allowed to
act Individually not only collectively as “the children”.
Mother-children in the story: however stressed and thwarted in her intentions,
Mrs Large is still humourously indulgent with her children, very human in her
inconsistency and reluctant tolerance.

Lexical means:
Humourous effect through precise specification of time allowed for Mrs Lar-
ge s moments of peace: XII: three minutes and forty-five seconds, and precise
specification of the children’s performance in the bathroom: VI;VII: three and
a half times, four and a half pages.

Syntactic means:
Presence of short-clipped, “matter-of-fact” clauses producing an effect of re-
signed dry humour: III: That's why, VI: So Lester played, VII: So Laura read.
IX: Mrs Large groaned', XII:... before they all came to join her.

Textual means:
Humourous effect of “setting the scene” through detailed and deliberate enu-
meration of all the things Mrs Large put on her tray to Indicate the level of
preparation for her five minutes’ peace, when the following, second paragraph

presents a stark contrast in its dialogic make-up describing the thwarting of
her intent.

Social Attitude
Informal style level: conversational, the type of talk commonly occurring in
a family.

Syntactic means:
Presence of elliptical structures: IV: downstairs, by yourselves.
Simplicity of noun phrases: lack of pre- and post-modification.
Use of s genitive:/2:^^e minutes' peace

Lexical means:
Use of lexical items marked as informal through their use in settings marked
by a lack of social distance:
II: stuffed, sneaked', \W\from you lot, IV: trailed up the stairs', V: plonked on ...
got in ... got out.

complex: written to be read aloud as if not written, creating for the child hearer
the illusion that the person doing the reading aloud is Inventing it simultane-
ously with the reading. This medium, then, is designed to simulate real-life
spontaneous language. Along Blber’s three dimensions. Involved vs informa-
tional, explicit vs situation-dependent, and abstract vs non-abstract, this text
is clearly on the involved, situation-dependent and non-abstract end of the

Syntactic means:
Frequency of elliptical structures typical of oral conversational encounters
where the situation automatically disambiguates the Incomplete utterances;
frequencies of contractions; frequency of short coordinated clauses linked
with and.

Phonological means:
Presence of emphatic stress frequent in oral encounters, and marked in writing
through italics.

Textual means:
Ample use of repetition for redundancy throughout the text designed to make
comprehension easier.

Complex: monologue with built-in (fictional) dialogic parts.

Syntactic means:
Predominance of first and second person personal and possessive pronouns
to indicate direct conversational Interaction inside the story, frequent switches
in the text between declarative, interrogative, exclamatory and Imperative ut-
terances, which is typical of the creation of an on-going interaction between
the participants in the (fictional) family scenario.

Textual means:
Interactants alternate very frequently in the encounters portrayed in the sce-
narios; in fact, the text as a whole is characterized by heavy use of direct speech.

Children’s picture book designed to be read aloud to children by adults, often
as a bed-time story. The primary goals of such books are, then, to entertain
children, to calm them down and to give them reassurance so they can settle
down. In the English tradition, children’s books often use humour to gently
socialize the young into a not always humourous world in the family and
beyond it. In the text presented here and in Its analysis I have omitted the
pictures as they would not have added anything that the text Itself did not
make explicit. In fact, the pictures are the same In the German translation and
the English original, except for the front cover where the German edition adds
more pictorial detail.

Statement of Function

The function of this text consisting of an Ideational and an interpersonal func-

tional component may be summed up as follows: although the ideational func-
tional component is not marked on any of the dimensions, it is nevertheless
Implicitly present In the text, in that the text informs its readers about a certain
event involving the protagonists depicted In the text. In other words: it tells a
story! But the ideational component is clearly less Important than the inter-
personal one, which is marked on all the dimensions used for the analysis of
the text.
The particular GENRE, a children’s picture book of which this text is an
example, determines that the interpersonal function Is primary. Its purpose
being to provide a feeling of “ Geborgenheit”, a sense of belonging, reassurance,
warmth and good-feeling to settle a child down for the night or otherwise.
On the dimension of FIELD, too, the interpersonal component is strongly
marked: the description of a typical piece of family life where a mother struggles
with her three young children Is presented In a light-hearted, good-natured and
humourous way, at the same time making the story easily digestible and com-
prehensible. On TENOR, we have seen that the Author’s Personal Stance as

well as the social role relationship and social attitude strongly mark the inter-
personal functional component: the relationship between both author and
readers and between the (fictional) mother and her children are characterized
by common sense and good humour. The informal style level also clearly
operates interpersonally by enhancing the text’s intimate humorously human
quality. On MODE, the medium characterized as “written to be read as if not
written” being marked as Involved, situatlon-;dependent and non-abstract, as
well as the many dialogic stretches in the text very clearly feed into the inter-
personal functional component because of the emotive effect of (simulated)
spontaneous directness, with the readers’/listeners’ Interest being focussed on
the dialogue participants as living “human” beings.

Comparison of Original and Translation and Statement of Quality

Mismatches along the following dimensions were discovered in the analysis

of the translation and the comparison of translation and original:

Lexical mismatches:
Greater explicitness through the addition of more finely granulated, descrip-
tive verbs and through greater detail and explanation, as well as the introduc-
tion of generic terms:

I: Wenn die Elefantenkinderfruhstiickten, ging es meistens laut und unordent-

lich zu vs The children were having breakfast. This was not a pleasant sight.
In this first paragraph the difference between original and translation is par-
ticularly marked: the German sentence explicitly describes what the English
sentence implies.
II: Mutter Elefant holte ein Tahlett und stellte ihr Friihstuck drauf: Teekanne,
Milchkrug ... vs Mrs Large took a tray from the cupboard. She set it with a
large teapot, a milk jug. The German text Introduces an explicitly explanatory
generic term Friihstuck ...
IV: “Nein ” sagte Mutter Elefant bestimmt vs “A/o ” said Mrs Large’, IV: murrte
vs muttered’, V: gemiitliches heifles Bad vs deep, hot bath’,
VI: Du sagst immer ich soli fleifig iiben vs Tve been practising told me
to’, VIII: Baby Elefant schleppte sovielSpielsachen an, wie er nur konnte vs...
a trunkful of toys',
X: Scblieflich sprangeh ... Baby E. war so aufgeregt, dafl
vs they all got in. The little one was in such a hurry that...

Syntactic mismatches:
Loss of structural simplicity in the translation: paragraph I

Textual mismatches:
Loss of Cohesion: III: Wohin gehst du Mama vs where are you going with
that tray (when the tray and what was on it was described in detail in the
preceding paragraph); III: Because I want five minutes peace from you lot vs
weilich 5 Minuten Ruhe hahen mochte; IV: “Diirfen wir mitkommenf” fragte
Elefantchen vs “Can we corned” asked Lester as they trailed up the stairs behind
her. The German translation fails to link up with the clause where Mrs Large
asked them to play downstairs; VIA^II: Elefantchen hegann zu floten vs So
Lester played] Elefantinchen hegann zu lesen vs So Laura read (missing clausal
Loss of iconic linkage resulting in both loss of cohesion and aesthetic
pleasure: VII/VIII: Darf ich dir eine Geschichte vorlesen ... - Baby Elefant
schleppte vs In came Laura - In came the little one. The dramatic effect achieved
through iconic linkage and the rhematic fronting is lost in the translation.
Explicit addition of a closing “punchline” not present in the original: XII:
bevor die Kinder kamen, damit ihre Mutter nicht so alleine ware vs before they
all came to join her.

Author's Personal (Emotional and Intellectual) Stance
Loss of humour, sentlmentallzation and Infantillzatlon of characters in the

Lexical mismatches:
All the names of the characters have been sentimentalized and de-individua-
lized: Mutter Elefant vs Mrs Large] Elefantchen vs Lester] Elefantinchen vs
Laura] the little one/baby vs Baby Elefant.

Textual mismatches:
Loss of humour in the initial paragraph setting the tone of the book: Wenn
die Elefantenkinderfriihstuckten ging eslautund unordentlich zu vs The child-
ren were having breakfast. This was not a pleasant sight.

Social Role Relationship

between author-readers, between author and the characters in the story, and
between the story’s protagonists: these three role relationships, which are all
changed in the translation, are clearly interdependent such that the relationship
between the story protagonists is a reflection of the author’s assessment of her
readers and the author’s view of her characters.

Lexical mismatches:
The relationship between the mother and her children reveals a lack of respect
for the individuality of the children, and with it a lack of politeness, and both

a sentimentalization and a certain “negativisation” of the children and their

This is evident from the names chosen in the German translation (see above)
and the fact that individual children s answers are changed in German to col-
lective answers: XI: ‘"Wamm” fragten die Kinder wie aus einem Mund vs
'‘Why” asked Lester.
The characters are to represent children, 6f course - they are elephants
only in the picture - the original text does not tell us this, the German trans-
lation, however, does! The whole point in the original is the usual duality
between distancing and identity that is at the core of children s stories. In the
translation, this duality and complexity is lost to a one-sided explicitness.
Further, whereas the children throughout the original are seen as indivi-
duals building up a strategy together in order to outwit their mother, the trans-
lation frequently lumps them together as "die Kinder”.

In the following examples, the children are not addressed by their names in
VI.- ...fragte ein Stimmchen vs asked Lester (which also sentimentalizes the
IX: "Darf ich mir die Bilder in der Zeitung ansehen ...” vs "Can I see the
cartoons in the paper” asked Laura.
VII: "Nein”, sagte Mutter Elefant vs "No, Laura”.

We can thus discover a consistent pattern in the translation of leaving out the
children s names. The lack of politeness, which this omission clearly Implies,
is reinforced, if we look at another example also showing how the children are
not taken seriously as individuals the way they are in the English original: not
only is the individual child not addressed by her name, the mother also falls
to answer her: VII: Du magst ihn lieher als mich. Das ist nicht fair - Also fang
schon an vs You like him better than me. It's not fair - Oh, don't be silly, Laura
...Go on then.

Ill: from you lot vs 0 realization. Although the children are regarded as one
collective entity by the authorial voice in the German translation, the transla-
tion does miss out on the mother s fake-indignant, goodnaturedly humourous
lumping the children together asyow lot. This phrase is omitted in the German
translation - the humorous tone in the original is lost. The point here is also
that for the mother the children are both Individuals and you lot - in other
words, of course she loves them. This complexity of the mother’s feelings is
not realized in the translation.
The role assignment is made explicit in different ways in German and

IV: “Ich will nicht mehr das Baby sein” murrte Baby Elefant vs “Tm not a
baby” muttered the little one. In the German translation, this remonstration
is possible, because the translation says the little one is a Baby, the original
does not make it explicit that he is a baby. Also, the German resembles an adult
reasoning, the English original is a little one*s utterance.
VI: Darf ichf Nur eine Minute vs Qan If Please just for one minute.
VII: Darf ich Dir eine Geschichte... vorlesen vs can I read you a page and VI:
was vorspielen vs my tune.
Loss of subtlety of the children s individual and collective strategy in the

Syntactic mismatches:
The relationship between the mother and the children is sometimes marked
by a greater directness in the translation:
IV: Geht spielen und zwar ins Kinderzimmer vs you can play. Downstairs. The
difference in the directness levels is one between the illocutionary force of an
order and a suggestion.
VI: Du sagst immer ich soli fleijlig iiben vs Fve been practising, you told me
to. In German, the complaint about not being allowed to play is much more
direct than the English one.

Social Attitude
Loss of informal style in some Instances, which reduces the humourous effect:

Lexical mismatches:
III: No German equivalent {or you lot
III: in das Badezimmer: oddly formal in its avoidance of the more common
contracted form ins
V: setzte die Badehaube auf vs plonked on her bath-hat

In as much as the translation is still a children’s picture book to be read to a
child, there has been no change in the GENRE. However, the “framing” is
totally different in German: both the initial and the final paragraph set a very
different tone. The children were having breakfast. This was not a pleasant
sight vs Wenn die Elefantenkinder fruhstuckten, ging es meistens laut und un-
ordentlich zu.
The first paragraph sets the frame “this is how this story will be told”. In
the original, the immediacy of the events taking place in the story is reflected
in a simple detached statement, which, at the same time, sets the story’s tone
as humourous and affectionate. In the translation, the first paragraph is a
description not an occurrence, with reflection substituting for affection. Com-

pare also the ending, where there is again a commentary in the German trans-
lation, i.e., the story ends as it began, the frame is complete:
And off she went downstairs, where she had three minutes and forty-five
seconds of peace before they all came to join her vs Und sie hatte drei Minuten
undfiinfundvierzigSekunden Ruhe, hevor die Kinder kamen, damitihre Mut-
ter nicht so allein ware.
In the translation, the story is thus framec^ifferently, and in keeping with
this frame, the child addressees are led through the story, the characters are
explicitly elephantlzed for them and they are provided with commentaries,
whereas the original insinuates, implies through laconic statements.
As the analysis of a larger corpus of German-Engllsh and English-German
children s books translations has revealed (House, forthcoming) there may
well be differences as regards children s books between the German and Eng-
lish linguacultures: there is a German tendency towards depicting a different
role relationship between children and adults, there is a much more sentlmen-
talizatlon and infantilization in German children s books, as well as less hu-
mour, a far greater explicitness and a greater need to impose edifying ideas and
ideology on the stories told in children’s books. Some of these tendencies have
been revealed in the above register analysis of original and translation.

Overt Errors
I: wenn die Elefantenkinder friihstuckten, ging es meistens unordentlich und
laut zu vs the children were having breakfast. This was not a pleasant sight.
II: stellte thr Fruhstiick drauf vs 0
III:0 vs with that tray
IV: ich will nicht mehr das Baby sein vs Tm not a baby
V: gemutlich vs deep; gofl Tee in die Tasse vs poured herself a cup of tea
VI: was vs my tune
VII: eine Geschichte vs a page
VII: ins Kinderzimmer vs downstairs
VIII soviel Spielsachen wie er nur tragen konnte vs a trunkful of toys
IX: omission of asked Laura, asked Lester, asked the little one
X: war so aufgeregt, daft vs was in such a hurry that
XI: warum, fragten die Kinder wie aus einem Munde vs Why, asked Lester
XII: addition: damit ihre Mutter nicht so allein ware.

Statement of Quality

The analysis of original and translation has revealed a number of mismatches

along the dimensions of FIELD and TENOR, and a consequent change of
the interpersonal functional component, but also various Overt Errors which

detract from the ideational component and change the transmission of infor-
On FIELD a greater explicitness in the translation was established in a
number of cases and a loss of cohesion through the omission of referential
Identiy, repetitions and iconic linkage. The Interpersonal functional compo-
nent is changed in that explicitness of content guides and directs a reader’s/
listener’s imagination and interpretation much more closely. The loss of co-
hesion reduces the aesthetic pleasure a well-made story will elicit.
On TENOR the Author’s Stance is changed such that the dry, subtle hu-
mour is often lost and a new sentimentalized, infantilized key is Introduced.
The role relationship portrayed (as well as the one implicit in the author-
reader/listener relationship) is different in that the characters’ individuality
is lost, their image is more negative and there is a greater directness in requests
and complaints. The style level is in certain Instances less informal and less
designed to communicate “closeness”.
These register differences reflect a culturally conditioned difference in the
realization of GENRE between English and German children’s books (as es-
tablished by analyses of a larger corpus, see House forthcoming). This diffe-
rence is most clearly visible in the different framing evident in the German
translation: adult commentaries and explicit “interpretation guides” are pro-
vided, where the English original trusts the reader’s creative imagination.
The greater explicitness in the German translation is clearly in line with
the differences in German and English communicative preferences established
in cross-cultural pragmatic research, so is the markedly stronger interpersonal
focus in the English original. The translation can then be described as a covert
one with a cultural filter having been applied. One wonders, however, if the
translator’s choice might not have been different, l.e., why did she not opt for
an overt translation? 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? (In my corpus of 52 children’s books I found that all the translations
examined are covert translations.) 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 is there not a greater respect for the
original children’s book - especially if the original is a little literary masterpie-
ce? One answer to these questions may be that the current one-sided recep-
tion-oriented climate may have played a role, another is, of course, market-
oriented strategies pursued by publishers who squeeze the originals via their
translations into the GENRE realization dominant in the target culture.

TEXT 2 Autobiography
Extract from Richard P. Feynman Surely you’re joking, Mr. Feynman (original
English; translation German), see Appendix.

The Dignified Professor

Analysis of the Original ^

Anecdotal autobiographical account about events of the private life of a famous
Nobel-prize winning physics professor. In this excerpt, the author describes
in a readable and attractive way his knowledge of and experiences with diffe-
rent types of university jobs for people with “great minds”.

Lexical means:
Preponderance of everyday “human Interest” lexical items, no technical terms.

Syntactic means:
Short simple clauses throughout this excerpt (and the text as a whole).

Textual means:
Strong cohesion achieved through a) repetition and iconic linkage, b) the or-
ganization in sequences of theme-rheme to ensure given-new ordering and c)
anaphoric referencing by means of pro-forms for noun phrases, adverblals,
clauses or sentences.
a) II: Nothing happens ... still no ideas come ... nothing happens ... Nothing
I, II, IV: When I don’t have any ideas ... so they don’t get any ideas ... and
they’re not getting any ideas ... you’ve got wonderful ideas ... you’re not
getting any ideas...
V: you can think about elementary things... These things... The elementary
things are easy to think about...
V: The elementary things are easy to think about... if you can’t think of a
new thought... If you do think of something new
V: Is there a better wayf Are there any new problems^ Are there any new
IV/V: Tm teaching my class... If you’re teaching a class
b) V: If you ’re teaching a class, you can think about elementary things... These
c) Those great minds at the Institute for Advanced Study ... These poor

Author’s Temporal, Social, and Geographical Provenance:
Unmarked, contemporary middle class standard American English

Author’s Personal (Emotional and Intellectual) Stance:

Humourous, light, detached, and down-to-earth, with a keen eye for the ab-
surd in human nature.

Lexical means:
Humourous lexical items such as II: these poor bastards for Nobel-prize win-
ning scientists in one of the most prestigious institutions of higher learning;
III: the experimental guys', IV: teaching described as the greatest pain in the
neck; it drives you nuts.

Syntactic means:
Lively description with frequent changes between declarative, interrogative,
imperative and exclamatory structures.

Social Role Relationship

Despite the fact that the author is a famous “dignified professor” of consld-
crabl e Intellectual standing and world-wide importance in the world of science,
he comes across as a humble and unpretentious person. However, at times the
authority of the famous professor also makes Itself felt in his pronouncements
about teaching and life in general.

Syntactic means:
Author makes himself more approachable through frequent insertion of si-
mulated direct speech, e.g., I: At least I’m living.
Presence of multiple rhetorical questions to drive home the author’s own
ideas of the function of teaching for a university professor: V: Is there a better
way to present them^ Are there any new problemsf Are there any new
Use of utterances expressing a firm conviction stated by the author as
expert, V: If you do think of something new, you’re rather pleased ... VI: It’s
not so easy to remind yourself of these things.

Textual means:
Humourous effect of opposition in VII and VIII, l.e., presenting himself as
very “human” in that first a certain belief or standpoint is described only to
show immediately afterwards how it was broken: / would never accept any
position ... But once I was offered such a position.

Social Attitude
consultative, conversational, friendly, revealing no great social distance be-
tween author and audience.

Lexical means:
Use of colloquial lexical items throughout, sometimes Informal ones those
poor bastards etc. (see above under Author^s Stance)^ and use of modal particles,
gambits and lexical items indicating vagueness, e.g., kind of, just etc.

Syntactic means:
Frequent use of contractions and simple noun phrases lacking pre- and post-
modification. ^
Presence elliptical structure: II, VII.

Textual means:
Ample use of repetition for redundancy throughout the text; clausal linkage
with conjunction so marked for informality.

Written to be read as if spoken: in fact, as indicated in the preface, the entire
text was originally spoken i.e., “told to Ralph Leighton”, transcribed and edited.
This medium can be described along Blber s oral-written continuum as invol-
ved, situation-dependent and non-abstract.

Lexical means:
Use of colloquial lexical items as well as vague and loose formulations frequent
in oral styles: I: something ..., Vm making some contribution', II: this lovely
house by the woods there; V: these things ...

Syntactic means:
Presence of anacolutha: I: Vm making some contribution - it's just psycholo-
gical. V: if you can't think of anew thought, no harm done. Presence of elliptical
structures (see Social Attitude). Frequency of coordinated and apposltional
structures strongly creating an Impression of impromptu formulation, i.e., lack
of premeditation. Emphatic-emotive use of language: V: if you do think ...

complex: monologue with many Instances of readers being directly addressed.

Syntactic means:
Predominant use of first and second person possessive and personal pronouns
indicating a situated interactional context.
Frequent switch of clausal mood.
Use of rhetorical addressee-directing utterances: V.

Textual means:
This is a predominantly etlc text drawing its readers into the text through
ample use of deictic pronouns.

Autobiography of a famous person. The goals of such works are to share with
a larger public the person s private life, his habits, convictions and beliefs in
order to satisfy a certain voyeurism on the part of the public, and to give the
public the opportunity to learn something from an exemplary life.

Statement of Function *

The function of this text consisting of an ideational and an interpersonal func-

tional component may be summed up as follows: the author s intention is to
tell a good story about his personal experiences. He also wants to convey to
his readers his philosophy of life and work in such a way that his thoughts are
presented in a totally non-technlcal. Interesting, entertaining and highly rea-
dable fashion.
The interpersonal functional component is strongly marked through the
GENRE of the text and on each of the dimensions, whereas the ideational
component, although of course present, recedes into the background. On
FIELD, the abundance of colloquial lexical items, the absence of technical
academic terms, the presence of simple clause structures and the massive oc-
currence of redundancy through repetition and iconic linkage - all acting to
make the text pleasant to read - clearly feed into the interpersonal functional
On TENOR, the author s humourous and unpretentious “presentation
of self” but also his occasional emphasis of his role as experienced professor
is evidenced linguistically by the use of gambits, vagueness particles, rhetorical
questions, funny colloquialisms and humourous utterances giving expert ad-
vice re the connection between teaching and research clearly support the in-
terpersonal component. The consultative style level marked through Informal
lexical items, contractions, gambits, ellipses and repetitons also helps to make
this text Interpersonally successful.
On MODE, both the fact that the Medium of this text is “written to be
read as if spoken” (involved, situation-dependent and non-abstract) and that
Participation is marked by frequent dialogic parts Interspersed in the mono-
logic framework supports the Interpersonal functional component. Linguisti-
cally this is achieved through many instances of directly involving the readers
via the easily digestible form of a personal narrative (elliptical and coordinate
structures, emphatic and emotive lexical items, eticness of text, rhetorical ques-
tions and frequent switch between declarative, interrogative and Imperative

Comparison of Original and Translation and Statement of Quality

Greater abstractness of nouns in German: IV, V: Ich hahe ja meinen Unterricht.
Wenn man unterrichtet... vs Vm teaching my class. If your teaching a class ...
However, this type of mismatch cannot be seen as a covert error, since there
is no conventionally established German equivalent for “classroom teaching”
other than “Unterricht”. Still, the effect is surely one of reduced Interpersonal
force, as “class” suggests a group of human beings, whereas “Unterricht” refers
to the abstract “process of Instruction”.

Author's Personal Stance

Lexical mismatches:
Reduced humour through neutralized, “flattened” lexical items:
II: arme Kerle vs poor bastards; III: die Leute, die Experimente machen vs the
experimental guys; IV: macht einen wahnsinnig vs it's driving you nuts; grofte
Geduldsprobevs greatest pain in the neck; II: heschleicht einen ein Schuldgefuhl
oder eine Depression vs a kind of guilt or depression worms inside of you.

Social Role Relationship

Syntactic mismatches:
Lack of personal pronouns throughout the text. Impersonal pronouns used
in German Instead, e.g., man, einem vs you produces a less personal “touch”.
This is not to be counted as a mismatch though, as this usage is normative in
German. Still, the effect is one of reduced Interpersonal effectiveness.

Social Attitude
The style level is less personal {see Social Role Relationship and Personal stance),
also certain lexical items are markedly more formal in German: VI: zu Zeiten
vs at times; V: ...dajl man eine neue Methode hat, die Dinge zu hetrachten vs
a new way of looking at it.

The German translation resembles the “spoken” part in “written to be spoken”
less than the English original does:

Syntactic mismatches:
Lack of anacolutha V: Wenn einem nichts Neues dazu einfdllt, so schadet das
nichts vs If you can't think of a new thought, no harm done.

Textual mismatches:
The etic nature of the text is reduced through lack of deictic pronouns in German,
i.e., lack of direct address of reader(s), which is a hallmark of spokenness.
We can thus say that the text is less involved and less situation-dependent, to
use Biber s criteria.
The translation is sometimes less dialogic through a lack of direct address of
the readers (see TENOR and Medium). The German Impersonal pronouns
(however normative in German in this type of text) and the English “you”
clearly differ in their potential of Involving readers.

Statement of Quality

This text’s Ideational and Interpersonal functional component are kept up to

a large extent in the translation resulting in a basically compatible second level
function. In other words, this is clearly an overt translation - an autobiography
is a classic candidate for an overt translation in that the author’s original voice,
the whole point of an autobiography, is to be respected as much as possible,
and the readers should be given the opportunity to have unimpeded access to
this voice. The ideational functional component is left unaltered. The inter-
personal one was changed slightly in the translation along the following di-
On FIELD, a more abstract noun had to be chosen in the German trans-
lation {Unterricht vs class)., a noun typically used in German whenever class-
room learning is thematized. Along TENOR, the humourous, easy-going tone
and the consultative style is rendered in places slightly less humourous and
more formal, thus less interpersonally oriented. On MODE, the translation
is less efficient in its simulation of “spokenness” (it is less involved and less
situation-dependent) through the lack of second person pronouns, again a
choice made in accordance with German norms of usage.
In general, given the differences in German and Anglophone communi-
cative preferences, a cultural filter may have been applied in the translation. If
we were dealing with a case of covert translation, the application of this filter
would clearly be justified considering the empirical findings of a generally less
strong Interpersonal focus and a less well developed reader-involvement in
German texts. But if one believes that this is a case for translating overtly,
where the readers of the German translation should be given “direct”, unfiltered
access to the original, then the cultural filter, however empirically verified,
should not have been applied. On such a view, then, the (admittedly very few)
mismatches, i.e., those wherever German norms of usage did not allow maximal
dimensional matches, would have to be seen as distracting from the interper-
sonal force of the translation and distracting from its second level equivalence.

TEXT 3: Philosophical Essay
Excerpt from Walter Benjamin “Die Aufgabe des Ubersetzers” (original Ger-
man, translation English) see Appendix.

Analysis of the Original and Statement of Function

This is one of the most Important 20th century statements about translation.
Walter Benjamin, a German literary essayist, art critic and philosopher wrote
“Die Aufgabe des Ubersetzers” in 1923 as a preface to his own translation of
Baudelaire’s Tableauxparisiens. It is a rich, multi-layered text with a potential
for various different interpretations. Most recently, the essay has been
discussed in postmodernist circles. It is both about a theory of language and
about the nature of the complex task of translation, which Benjamin himself
had engaged in abundantly. Because the essay prefaces Benjamin’s own Bau-
delaire translation and is supposed to explain this word-for-word translation,
in which the syntactic structures of the original are kept up as far as possible,
the essay must be seen in conjunction with this translation. Indeed, it can be
seen as a justification for it: Benjamin wrote to Rang and Hofmannsthal that
he was well aware of the problems of this translation. Others such as Werner
Kraft critically commented on the translation saying that it does not have its
“own language” (Kraft quoted in Wiesenthal, 1973: 109) - a judgment in line
with the “Wortlichkeit” which Benjamin praises, and with his rejection of the
“zuchtlose Freihelt schlechter Ubersetzer”.
The essay Itself is an academic text, a very “well-made” text, l.e., one which
is extremely carefully formulated and thought out, with the language - due to
the careful planning that has gone into the construction of the text - acquiring
a great importance: it is part of the message.

Lexical means:
Use of precise, explicit and unambiguous lexical items and phrases revealing
care, planning and a concern for clarity, e.g., I: ... den Aufnehmenden\ indicating
precisely the process in which the recipient is finding himself at the moment
of perception; ein hestimmtes Puhlikum oder dessen Reprasentanten; ample
use of quotation marks to indicate how exactly a term is to be (or not to be)
Interpreted; II: das Unfajlhare, Geheimnisvolle, “Dichterische”; III: oder, und
eigentlicher ...; IV: ... aus seinem “Uherlehen”: the frequent use of Inverted
commas in the text can be taken as evidence for the author’s search for preciseness,
and his unwillingness to be satisfied with terms that are unable to capture the
exact meaning he had Intended to convey.

Syntactic means:
Complex syntactic structures featuring long sentences, subordinated clauses
and multiple embedding.

Textual means: •»
Presence of strong textual cohesion due to the employment of several mecha-
nisms of theme dynamics, clausal linkage and Iconic linkage.
a) Repetition of lexical items and phrases throughout the text and use of
anaphoric and cataphoric pronominal reference.
b) Ample use of opposites and synechdoches.
c) Frequency of clausal linkage through a multitude of different logical con-
nectors such as denn, daher, wenn dann, dem gegeniiher, demnach etc.
resulting in logical structuring of entire text.
d) Chains of theme-rheme sequences to secure given-new ordering.

I cannot list here all the examples elucidating the cohesive phenomena as there
are far too many of them. One stretch, the beginning of paragraph III may
suffice: Uhersetzung ist eine Form. Sie als solche zu erfassen, gilt es zuriickzu-
gehen auf das Original Denn in ihm liegt deren Gesetz als in dessen Uber-
setzharkeit heschlossen. Die Frage nach der Ubersetzbarkeit eines Werkes ist
doppelsinnig. Sie kann bedeuten: ob es unter der Gesamtheit seiner Leser je
seinen zuldnglichen Ubersetzer finden werde? oder, und eigentlicher, ob es
seinem Wesen nach Ubersetzung zulasse und demnach - der Bedeutung dieser
Form gemdjl - auch verlange. This stretch of text features repetition {Uber-
setzbarkeit, Ubersetzung), anaphoric reference (e.g., denn in ihm liegt deren
Gesetz als in dessen Ubersetzbarkeit, sie kann bedeuten) and cataphoric ref-
erence {doppelsinnig) and clausal linkage {denn, demnach).

e) Frequent iconic linkage.

f) Explicit numerical structuring ...die erste, die zweite etc.

Author's Temporal, Social and Geographical Provenance
Unmarked geographically and socially, but marked temporally, l.e., through
the frequent use of the subjunctive, which is much more Infrequently used in
contemporary German.

Author's Personal Stance

The author is emphatically convinced of the truth of his theory of language
and translation. With almost messianic zeal he defends this theory and the
method of translation that goes with it, a theory and method that was and is
not uncontested. The author s style (as well as the content of his essay) is often
provocative as if to deliberately pre-empt protest and contradiction. A good
example is the very first sentence: Nirgends erweist sich einem Kunstwerk oder

einer Kunstform gegenuher die Riicksicht auf den Aufnehmenden fur deren
Erkenntnis fruchthar.

Syntactic means:
Frequent use of rhetorical questions for emotive emphasis. These questions
are not so much directed at the text s readers but seem to be directed at the
author himself. For Instance: II: Gilt eine Ubersetzung den Lesem, die das
Original nicht verstehenf... Was sagt denn eine Dichtungf

Textual means:
Frequency of foregrounded rhematic structures for emotive effects, e.g. IV:
1st dock die Ubersetzung spdter als das Original und bezeichnet sie dock...;
Dajl eine Ubersetzung niemals, so gut sie auch sei, etwasfiir das Original zu be-
deuten vermag, leuchtet ein; In vollig unmetaphorischer Sachlichkeit ist der Ge-
danke vom Leben undFortleben der Kunstwerke zu erfassen... II: Was aberaufer
der Mitteilung in einer Dichtung steht... II: denn in ihm liegt deren Gesetz ...

Social Role Relationship

The author Is an expert who explicates his theories with the authority of someone
who Is convinced of their correctness. However, the relationship between author
and readers recedes into the background: It Is considered of no great importan-
ce and is definitely subservient to the cognitive content of the text. The rela-
tionship can thus be described as highly impersonal. This impression of im-
personality is based on the following linguistic features:

Lexical and syntactic means:

Great care has been taken to construct this text: every word and phrase seems
to be deliberately selected, as if the “message” of the text had to be proven In
the text Itself, I.e., Its lexical and syntactic realization.
a) Absence of first and second person pronouns.
b) Frequency and complexity of [-human] noun phrases.
c) Use of verbs which merely signal logical connection and thus mainly point
to other semantically more significant textual content, i.e., they shift at-
tention away from themselves to the noun phrases and therefore heighten
the impression of abstractness and impersonality. I:... ist sogar der Begriff
eines '‘idealen” Aufnehmenden in alien kunsttheoretischen Erorterungen
vom Ubel, wed diese lediglich gehalten sind, Dasein und Wesen des Men-
schen iiberhaupt vorauszusetzen. So setztauch die Kunst selbst dessen leib-
liches und geistiges Wesen voraus - seine Aufmerksamkeit aber in keinem
ihrer Werke. IV.- In vollig unmetaphorischer Sachlichkeit ist der Gedanke
vom Leben und Fortleben der Kunstwerke zu erfassen...
d) Use of rhetorical questions to make the material presented more salient
(see Personal Stance).

Social Attitude
Formal, i.e., marked by social distance and (resulting from this) impersonality.
Syntactic means:
Complex sentence structures featuring long clauses, ample subordination, ap-
positional and parenthetical constructions, and especially the placing of sub-
ordinate clauses before the main clause, a focussing device typical of formal
style: e.g., I: Nicht genug, dajl... IV: Dajl eine Uhersetzung niemals, so gut sie
auch seiy ... V: Dajl man nicht der organischen Leiblichkeit allein Lehen zu-
sprechen diirfe..., etc.
Lexical means:
Frequency of words and phrases marked [+abstract] due to their restricted
use in formal academic treatises. Absence of interjections, qualifying modal
adverblals and other subjectivity markers.

Simple: written to be read
This Medium can be characterized on Blber s continua as being strongly in-
formational, explicit, and abstract.

Syntactic means:
Absence of any kind of spoken language phenomena such as anacolutha, struc-
tural redundancy, short and simple coordinate structures.

Lexical means:
Absence of gambits, interjections and other subjectivity markers typical of the
spoken mode.

Textual means:
This text is completely emlc, i.e., there are no explicit references to the texts s
author and its readers. Indeed, the Immediate circumstances of production
and reception are clearly Irrelevant for both the organization and the reception
of the message. As a result of this, the text is constituted solely through text-
immanent criteria, and is marked by an explicitness, elaborateness, and calcu-
lated unamblguity typical of the written mode. The text is also strongly cohe-
sive (see FIELD) due to a plethora of cohesion-creating devices.

Simple: Monologue
No participation of addressees is evident in this text. Readers are not taken
into account at all. Even the frequent rhetorical questions are not directed at
the readers, but are designed to make the message more expressive, more
salient, more aesthetically pleasing.

Philosophical essay in which the author attempts to develop his own theory
of translation, and with this, a theory of langauge.

Statement of Function

The function of this text consisting of the two components, the ideational and
the interpersonal, may be summed up as follows: the author s intention is to
explain and justify (via the presentation of a theory of language and translation)
the manner in which he had translated Baudelaire s Tableaux parisiens. In fact
this text is a preface to these translations. The text is totally writer-oriented,
with no consideration of the reader being manifest in the text. The “message”
of the text is extremely carefully formulated, every word seems to be well
thought out, the structure enables therefore a well-organized flow of thought.
Given the above characterisation and the text’s GENRE, the ideational
functional component is overall strongly marked, the Interpersonal one, while
co-present, is not marked, except on the dimension of TENOR. On FIELD,
the careful choice of precisely apposite lexical items, the use of intricately
embedded structures and the frequency with which various cohesive devices
are used, clearly feeds into the ideational functional component.
On TENOR, the Author's Stance is characterized by an intensity, an al-
most messianic zeal and a conviction that the theory he is presenting justifies
his own translation. While this authorial stance supports the Interpersonal
component in that the author’s own involvement potentially affects his
readers, the readers do not seem to be “present” in the text at all, and the
Role Relationship is Impersonal to the point of being absent. On Social Attitude
the style level is formal. Taken together, the dimension TENOR supports the
Ideational functional component more strongly because the information flow
is given priority and it is unimpeded by social considerations Involving the
On MODE, the written to be read Medium with its highly informational,
explicit and abstract nature supports the ideational functional component by
facilitating a condensed, complete and premeditated Information flow. On
Participation^ the totally monologous nature of the text, where even the fre-
quent rhetorical questions are non-dialoglc, solely serving to emphasize the
message, also supports the ideational functional component. In this text, therefore,
author and message are more Important than the conditions holding for the text’s
receivers, and this is of course consistent with the author’s message, l.e., “con-
siderations of the receiver” do not seem to exist.

Comparison of Original and Translation and Statement of Quality

The translation I examine here is the “standard” one by Harry Zohn, which
has been heavily criticized in the literature, e.g., by de Man (1986), who prefers
Jacobs’ (1975) translation, and Johnston (1992). Given the content of this essay.
It would seem to be of prime importance that the translation be consistent
with that content. In the particular case of the analysis of this text, treating the
topic of translation in a particular way, this criterion should be taken into

Lexixal mismatches:
Less precise terms are chosen: I: first sentence: receiver vs der Aufnehmende\
the processual aspect is not captured in the translation (’’Percelver” as suggested
by Jacobs might be more adequate). Although the English translation manages
to express the notion of “process” through the phrase In the appreciation of
a work of art..., it deviates from Benjamin’s original expression, in which the
Kunstwerk is personalized. This is important in the context of Benjamin’s
theory, as the work of art is thus given more prominence. The translation,
where the work of art is hidden as an object in a prepositonal phrase, clearly
loses this nuance. The next sentence is also not precise enough in that the lexical
item gehalten sind clearly expresses some obligation which gets lost in the
translation : "... since all it posits is the existence and nature of man as such vs
weil diese lediglich gehalten sind, Dasein und Wesen des Menschen Uherhaupt
vorauszusetzen ...
There are many similar lexical infelicities. I cannot list them all, a few
examples may suffice: I: vom Wege abfiihrt vs misleading-, vom Uhel vs. de-
trimental, III: In solcher Loslosung vs in this sense. The translation is not as
precise as the original in that it lacks an equivalent expression for the original’s
Loslosung, which through being more anaphorically specific also creates co-
hesion in the original. IV: Flowering vs Entfaltung: the English term adds a
nuance not present in the original’s more neutral Entfaltung. (“Unfolding”
would be more suitable); IV:... ein Zusammenhang des Lehens vs a vital con-
nection. In all these cases, the use of equivalent English metaphors would have
been possible.

Textual mismatches:
Lack of Cohesion through
1) frequent lack of clausal linkage, e.g., I: Denn kein Gedicht... vs No poem-,
II: Das ist denn auch das Erkennungszeichen vs This is the hallmark ...;
Ill: Ihm gegenuher ist darauf hinzuweisen, dafl... vs it should he pointed out
...-,W-.Denn von derGeschichte, nichtvon derNaturausvs In the final analysis,
the range...; V: Daher entsteht dem Philosophen die Aufgahe ... vs The philo-

sopher’s task consists... There are many more examples in which the translation
simply omits the conjunctions.
2) Loss of repetition e.g., in II, where the English equivalent Mitteilung, mit-
teilen vary (unnecessarily): communicate, imparting and transmitting infor-
mation, statement i.e., Was teilt sie mitf Ihr Wesentliches ist nicht Mitteilung...
die Mitteilung vs What does it communicate... its essential quality is not state-
ment or the imparting of information.

Author's Personal Stance
Textual mismatches:
Frequent loss of foregrounded, dramatic (rhematic) structures in the transla-
tion. In the case of the essay s first provocative sentence, this loss is critical:
Nirgends erweist sich einem Kunstwerk oder einer Kunstform gegeniiber die
Rucksicht auf den Aufnehmenden fur deren Erkenntnis fruchtbar vs In the
appreciation of a work of art or an art form, consideration of the receiver never
proves fruitful. The emotive Impact of the German sentence could have easily
been retained in English: Nowhere does consideration for the perceiver with
respect to a work of art or an art form prove fruitful for their understanding.

IV: Dajl eine Ubersetzung niemals etwasfur das Originalzu bedeuten vermag,
leuchtet ein vs It is plausible that no translation ...; Ist doch die Ubersetzung
spdter als das Original... vs For a translation comes later than the original,
Vielmehr nur wenn allem demjenigen, wovon es Geschichte gibt... Leben
zuerkannt wird, kommt dessen Begriff zu seinem Recht vs The concept of
life is given its due only if everything that has a history of its own ... is
credited with life.
W: In ihnen erreicht das Leben des Originals seine... Entfaltung vs The life of
the originals attains in them its ... flowering.

Loss of simulated emphatic elliptical answer to rhetorical questions: II: Was

tedt sie mitf Sehr wenig dem, der sie versteht vs What does it communicated It
“tells" very little to those who understand it.

Social Role Relationship

Syntactic mismatches:
The translation is often rendered less Impersonal through the use of personal
pronouns: IV: (der Zusammenhang) darf ein naturlicher genannt werden vs
we may call this connection ...; II:... welche man demnach ... definieren darf
vs which consequently 'we may define ...; Was aber aufler der Mitteilung in
emer Ubersetzung steht... vs But we do not generally regard ...; V: Die Ge-
schichte der grojlen Kunstwerke kennt... vs The history of the great works of
art tells us about...

Social Attitude
The style level is slightly less formal due to the use of personal pronouns (see
Social Role Relationship), and a lack of focussing devices typical of formal style
such as placing subordinate clauses before main clauses.

Textual mismatches:
The translation text is not completely emic through the use of first person
personal pronouns, i.e., more situation-dependent.

The use of inclusive we Involves readers minimally, but still more than the
original does.

Overt Errors
The following lexical error is a serious blunder (which is also critically com-
mented upon by de Man), because it changes the meaning of the original com-
pletely and should have been avoided by anyone having really understood the
original: II:... ist darauf hinzuweisen, daft gewisse Relationshegriffe ihren gu-
ten ... Sinn behalten, wenn sie nicht von vorneherein ausschlieftlich auf den
Menschen bezogen werden vs ... that certain correlative concepts retain their
... significance, if they are referred exclusively to man.
Interestingly enough, the same error was made by the French translator
in his translation of Benjamin s essay into French, which causes de Man to
furiously claim that both translators didn’t seem “to have the slightest idea of
what Benjamin is saying; so much so that when Benjamin says certain things
rather simply in one way - for example he says something is not - the trans-
lators, who at least know German well enough to know the difference between
something is and something is not - don’t see it! and put absolutely and literally
the opposite of what Benjamin has said. This is remarkable because the two
translators ... are very good translators ...” (de Man 1986: 79). Is it possible
that the translators own ideas of translation and especially a translation’s “re-
cipient design” has played a role here? One can only speculate, but the fact
remains that this is a serious overt error.

Statement of Quality

The comparison of original and translation along the situational parameters

shows that there are a number of mismatches on FIELD in the translation,
where Benjamin’s precise, thoughtful formulation is frequently not rendered
with the same care and attention to detailed shades of meaning, and where the
original’s intricate web of textual cohesion is often destroyed due to a lack of

repetition, explicit clausal linkage, and theme-rheme sequencing. These mis-
matches detract from the text s ideational functional component. On TENOR,
Benjamin’s Personal Stance which comes across forcefully e.g., in the original’s
foregrounded emphatic clause structures is not always kept up in the transla-
tion, and the impersonality characterizing the original’s Social Role Relation-
ship is thwarted through the introduction of inclusive we-pronouns in the
translation. Both the Interpersonal and the Ideational functional components
are affected. The original’s great (and deliberate) social distance between author
and readers and the resulting formal style leA^el characterizing the dimension
Social Attitude is sometimes, due to the presence of de-complexified clause
structures, less formal in the translation. On MODE, the translation’s occa-
sional use of personal pronouns deviates slightly from the entirely non-par-
tlclpatory written nature of the original, also affecting both functional com-
ponents, but in a minor way. One serious overt error was found in the passage
Taken together, the ideational functional component and, to a much lesser
degree, the Interpersonal one (which is, with the exception of the Author's
Personal Stance, not marked at all in the original) are affected through changes
in the translation. The changes of the ideational functional component are
critical in a text whose Ideational functional component is marked strongly in
the first place. The analysis of the entire text, of which this passage is but a
small part, confirms this finding.
This text would seem to be a classic case for an overt translation because
of the status of the author and the status of the text, because of the particularly
close connection in this text of content and the language used to express it,
and last but not least because of the nature of the content; a particular theory
of translation. The translation examined here might however have been trans-
lated covertly, in that a cultural filter may have been consciously or uncon-
sciously applied. The original was found to be more explicit and (slightly) less
Interpersonally focussed than the translation which appeared to be less precise
and less explicit, l.e., weaker on the Ideational functional component. This is
in line with the English-German communicative preferences. One might, of
course, argue that a covert translation and an application of a cultural filter is
justified for this text as it makes it easier for Anglophone readers to understand
this difficult text. However, given the importance of this text, the status of its
author and the nature of its content, an overt translation would do more justice
to it. This is the stance I am taking here. However, this is a social not a scientific
judgement, l.e., it is not based on the type of close linguistic-textual analysis
of the source text the model provides. The plausibility of allocating a text to
be translated overty or covertly depends on social factors l.e., the status of the
author and the text.

TEXT 4: History Text

Excerpt from Daniel Jonah Goldhagen/Zii/er 5 Willing Executioners. Ordina-.

ry Germans and the Holocaust. (Original English, translation German) see
Appendix. ^

Analysis of the Original

This is an academic text from the field of history based on the author’s PhD
dissertation. The author, a young Harvard historian, puts forward a new, pro-
vocative hypothesis designed to better explain the holocaust. According to
this hypothesis, the holocaust did not originate with a small group of Nazi
maniacs, rather it was made possible because “ordinary Germans” willingly
cooperated and executed the crimes. Paragraph III in the selected excerpt de-
monstrates this hypothesis well in that Goldhagen here explicitly equates Na-
zis with Germans: he introduces a quote about how ^a poison of diseased
hatred permeates the blood of the Nazis’^ but interprets and prefaces it in the
following way:... to Jewish victims it appeared as if its hold on Germans could
he captured and conveyed only in organic terms ...”.
The book is now a bestseller. It has had a controversial reception not only
in Germany but worldwide. While being clearly an academic historical text,
it also shows the author’s strong personal, emotional and Intellectual involve-
ment (see TENOR).

Lexical means:
Given the central thesis of this work, the author makes sure that the word
German (often in the collocation ordinary Germans or in opposition to Jew-
ish and Jews) is mentioned frequently in order to drive the message home. An
important piece of evidence (see paragraph IV) for this deliberate, regular and
insistent use of the lexical item German is the express Insertion, in brackets, of
the word “German” in a quote where it was not present originally and where
this non-presence was unlikely to cause a misinterpretation given the disam-
biguating co-text. Especially marked is the frequent (negativising) collocation
these Germans as in VI: After successful kills, these Germans were in the habit
of rewarding themselves...

Syntactic means:
Frequency of long, complex clauses featuring subordination, coordination and

Textual means:
Strong cohesion achieved through repetition of (emotive) key terms such as
German, ordinary Germans, Jews, Jewish (see above), antisemitic, antisemi-

tism, genocide, eliminationist, exterminationist, annihilation, hatred, slaughter,
kill, cruelty, metastized etc.

Author’s Temporal, Social and Geographical Provenance:
Unmarked contemporary middle-class standard American English

Author’s Personal (Emotional and Intellectual) Stance:

The author has both the professional Interest of a historian In his subject and
a personal one - his father, to whom he has dedicated this book, Is a holocaust
surv'lvor - and his deep personal Involvement Is strongly noticeable In the
book. The following linguistic phenomena provide evidence of this Involve-
ment, which Is unusual In an academic text, and Is also likely to have an even
stronger emotive effect on the readers of the book than would have been the
case anyway given the enormity of the topic.
Lexical means:
a) extremely frequent use of Intenslflers, superlatives and lexical Items marked
either [+concrete] or [+emotlve] or both. Due to this frequency, I can only
list a selection here: h strongly tended, virulent eliminationist antisemitism,
fully two thirds, enthusiastic support, enormous number of ordinary Ger-
mans, genocial assault, significant dissatisfaction and (principled) dissent,
... trenchantly put it, exceedingly close; II: virulent exterminationist form,
act barbarously against Jews; III: So profound and near universal..., pro-
found hatred; IV: gigantic catastrophe, depth of hatred, hatred of emotion;
V: systematic killing; VIII: near universal, terrorize, orgy of killing; IX:
concentrated slaughter, mass annihilation, slaughter of the Jews; X: vast
majority of the German people, inescapable truth, enormous number of
ordinary, most of the rest of their fellow Germans.
b) Ample use of metaphoric language to express emotive Involvement: e.g.,
I: metastize, the putative Jewish illness, the German body social, the social
pathology ofJewry; II: metastized, given free rein; III: poison of diseased
hatred, lain dormant; VI: slaughter ofJews; IX: slaughter of the Jews.
Textual means:
a) Frequency of Iconic linkage for strong emotional effect: II: Genocide was
immanent in ... It was immanent in ... And it was immanent in... VI, VII,
VIII, IX, X (and more frequently In textual passages not examined here):
These were the beliefs that... The beliefs that... These beliefs that were ...
b) Use of foregrounded, rhematlc structures for rhetorical effect, e.g.. Ill: So
profound and near universal was ...; V: It is the masses, the ordinary Ger-
mans ... whom Kaplan exposes ...

Social Role Relationship
The text IS both designed to address historians and tKe general public. As a
professional historian in his field, the author clearly demonstrates that he is
an expert in the subject he is treating. He puts forward his main thesis which
IS provocative and new with an air of full authority, quoting extensively from
primary sources he has had access to.

Lexical means:
Presence of [+abstract] noun phrases, often Latin based, e.g., eliminationist
antisemitic ideology. Markedly frequent use of intensifiers, superlatives and
[+emotive] lexical items and metaphoric language (see Personal Stance)

Syntactic means:
Absence of personal pronouns as behooves an academic text. Presence of
lengthy, multiply embedded clauses.

Textual means:
Frequency of effective rhetorical opposition through adversative structures -
often featuring iconic linkage for heightened rhetorical effect: e.g., VIII: they
did not choose ... hut instead... that prepared men ... not to hate, hut to esteem
...; X:... that prepared not just the Germans... hut also the vast majority of the
German people...
Use of iconic linkage and foregrounded rhematlc structures for emotive
effect on readers (see Personal Stance).

Social Attitude
A carefully pre-medltated style with well planned, logically constructed clauses
and carefully selected and combined lexical items. However, the label “con-
sultative” in conjunction with “formal” is justified as the author - despite
engaging in scientific discourse with its structural density and close lexical
packaging of information - also deliberately Involves his addressees through
an Intensely emotive language, (see TENOR) which reduces social distance.
The texts style may thus also be characterized as “rhetorical”: the material is
presented for a certain effect, l.e., to convince readers of the truth of the author’s
thesis. This characterization is based on the following linguistic evidence:

Lexical means:
Frequency of [+abstract], often Latin-based noun phrases which arc often
heavily pre- or post-modified, e.g., I: eliminationist antisemitic ideology, the
Germans* twentieth century conception of the Jews, putative social pathology,
the latter part of the pre-genocidal nineteenth century, VI: the men of Police
battalion 61 *s First company, IX: the German perpetrators rejoicing proudly in
their mass annihilation of the Jews.

Syntactical means;
a) Absence of contractions.
b) Frequency of long sentences featuring multiple embedding and revealing
careful planning for rhythmic and aesthetic effect.

Simple: Written to be read
While the medium in this text can be characterized along Biber s dimen-
sions as explicit and abstract but also - which is marked for this medium - as
highly Involved.

Lexical means:
Absence of interjections, anacolutha and other characteristics of the spoken
mode. Presence of [+ abstract] noun phrases.
Due to the high frequency of [+emotive] lexical items the text is marked
for involvedness.

Syntactic means:
Absence of elliptical clauses, contractions, contact and comment parentheses
and other kinds of spoken language signals such as gambits, modal particles etc.
Frequency of long and complex clauses featuring subordination, apposition
and multiple coordination.

Textual means:
The text is emlc, l.e., there is no pronominal reference to author or readers,
because the immediate circumstances of the text’s production and reception
are Irrelevant for the organization of the message. As a result of this, the text
is determined through text-immanent criteria and is marked through the ex-
plicitness and elaborateness of the monologously written mode.
Still, through the presence of a number of rhetorical devices such as pat-
terned repetition, iconic linkage, foregrounded rhematic structures etc., this
text is also strongly “Involved”.

Simple: Monologue
The reactions of the readers are never directly elicited, l.e., there are no overt
participation devices such as e.g., pronouns, questions directly addressing readers,
speech acts with the illocutionary force of e.g., a request, etc. In this text, it is
clearly the content and the way this content is presented which is of utmost
Importance. Reader participation in this text is thus implicit and is expressed
through the values on the other situational dimensions.

Academic text featuring a provocative hypothesis and re-lnterpretation of the
causes of historical facts.

Statement of Function r

The function of the text consisting of an ideational and an Interpersonal com-

ponent may be summed up as follows: the author’s intention is to (1) inform
his readers of his research into a period of German history presenting a new
hypothesis about the causes of certain historical facts, (2) to convince his readers
of the correctness of this hypothesis and to make an emotional plea for its
plausibility. The interpersonal functional component is strongly marked, with
the Ideational one being of course also present, as the GENRE of the text is a
history text. On FIELD the hammering repetition of key lexical items (Ger-
man, Jews, antisemltlc, genocide) strongly support the interpersonal functio-
nal component. The clustering of abstract lexical terms and the frequency of
complex clause structures mark the text Ideationally as an academic history
On TENOR, the Author’s Personal and Emotive Stance is very strongly
marked in this text. It is the reason for the author’s refraining from presenting
the results of his research in the conventional Impersonal academic manner,
and for the fact that he presents in detail vivid and concrete Images of atrocities
and horror. The Interpersonal component of this text’s function is therefore
heavily marked. On Social Role Relationship, the author’s assumption of pro-
fessional and moral authority supports the interpersonal component. The fact
that there is a complete lack of direct Involvement of the readers marks the
text Ideationally as an academic text. On Social Attitude, the formal-consulta-
tive style level marked by premeditated complex structures and carefully selected
lexis as well as the text’s emotionally involving and rhetorically effective lexical
and syntactic choices also support both functional components.
On MODE, the written to be read Medium, which is here marked as
explicit and abstract but also highly Involved, feeds into both the Ideational
and the Interpersonal functional component. The totally monologous nature
of the text along Participation feeds into the Ideational functional component,
as it is the efficient organization of content which such a value supports.
Taken together, the strong marking of the Interpersonal component is
remarkable for an academic text.

Comparison of Original and Translation and Statement of Quality

Mismatches along the following dimensions were discovered following a de-

tailed analysis of the two texts:

Lexical mismatches:
The haunting repetition of the word German and the various collocations with
German (such as ordinary Germans) are not kept up in the German translation.
Given the central claim of this work, this is a serious change, and if one con-
siders that, given the text s and the author s prominent status, it might be trans-
lated overtly, it is a crucial mismatch, indeed.
This critical change (or mismatch) can be found not only throughout this
passage but also in the entire book. I will glv6a few examples of the numerous
substitutions for, and omissions of, equivalents of the word German in the
translation. I: derfUr das zwanzigste Jahrhundert typischen Vorstellungen von
den Juden vs the Germans' twentieth-century conception of the Jews;... losten
jedenfalls weder Unruhe noch Dissens in der Bevolkerung aus vs failed to
produce significant dissatisfaction and (principled) dissent within the general
German populace;... die erschreckenden Diagnosen und Prognosen vs The dire
diagnosis and prognosis for Germany; III: Germans' profound hatred ofJews
vs der tiefe... Haf; IV: Die Massen ... vs the (German) masses (here the author
explicitly Inserts ""German" in brackets in a quote, although the cotext disam-
biguates perfectly; VI: Sie brachten die Offiziere...dazu, sich wie so viele Be-
teiligte ... vs ... that prepared officers ... to boast like so many other Germans
engaged in the slaughter; VI: ... diese Polizeireservisten vs ... these German
reservists ...; Nach erfolgreichen Mordeinsatzen belohnten sie sich gern ... vs
after successful kills, these Germans were in the habit of...; VIII: Die immer
gleichen Vorstellungen und Bilderbewogen auch die ganz gewohnlichen Man-
ner des Polizeibatallions 307, den Hauptmann nicht ... vs ... these were the
beliefs that prepared the men of Police Battalion 309, ordinary Germans ...
Especially Important in Goldhagen’s work is the constant juxtaposition
of the words German and Jews (and their various derivations). This opposition
is of course destroyed if the word German is deleted as e.g., in I: den jiidischen
Biirgem zundchst den gesellschaftlichen Einfluf zu nehmen und sie dann ganz
aus der Gesellschaft auszuschlieflen vs to eliminate German Jewish citizens first
from the influence of German society and then from society itself. There are
many similar examples.
In paragraph IX, the omission of (the equivalent of) German is most no-
ticeable: in IX, the word German is mentioned 8 times in the English original
with Deutsch figuring only 3 times in the translation.
Apart from the word German, other key terms such as antisemitism, ge-
nocide, eliminationist, exterminationist, are also not repeated with the same
frequency as is the case in the original: Paragraph I exemplifies this tendency.
A comparison of the first two sentences in the original and the translation
makes it already clear that the translation often avoids direct mention of the
horrors - which is the exact opposite of one of the main purposes of the orl-

ginal: Even though the eliminationist antisemitic ideology was multipotential
in action, it strongly tended, given the Germans^ twentieth century conception
of the Jews, to metastize into its most extreme, exterminationist variant, pro-
mising a commensurate political “solution''^ to the putative ‘‘problem”. The
elective affinity between a person subscribing to a racially based, virulent eli-
minationist antisemitism and a person concluding that an exterminationist “so-
lution ” was desirable could already be seen in the latter part of the pre-genocidal
nineteenth century vs Diese Ausprdgung des Judenhasses liejl also viele Hand-
lungsmdglichkeiten offen. Abersie tendierte, gerade vordem Hintergrund der
fiir das zwanzigste Jahrhundert typischen Vorstellungen von den Juden, zur
extremsten Variante, zurdefimtiven “Losung”. Daf zwischenrassistischen An-
tisemiten, die fiir eine gewaltsame Aussonderung der Juden pladierten, und
solchen, die die “Losung” in der Vernichtung sahen, eine Wahlverwandtschaft
bestand, wurde bereits Ende des neunzehnten Jahrhunderts deutlich.

Author's Personal Stance
Lexical mismatches;
The translation fails to provide equivalents of intensifiers, upgrading adjectives
and superlatives as well as [+emotlve], and starkly concrete lexical items de-
signed to both express the Author's Personal Stance and emotionally involve
the readers. In the original these lexical items are used to drive home the extent
of the horror and to stress the fact that it was the vast majority of ordinary
Germans which were willing executioners of the Nazi crimes. In falling to find
equivalents of these words, the German translation tones the entire text down,
a serious difference as it thwarts the author s intention to make a provocatively
strong case.

I: zwei Drittelws fully two thirds; wurde von der deutschen Bevolkerung geteilt
vs produced enthusiastic support among the German population; jeder Schritt
des Programms vs every major feature of the evolving eliminationist program;
eine grofe Zahl von ganz gewohnlichen Deutschen vs an enormous number
of ordinary Germans; ...die Juden zu ermorden vs ... their willing slaughter of
the Jews; II; das spdtere Programm vs the society's protogenocidalprogram;
III;... so allgemein verbreitet vs So profound and near universal...; VI; Vor-
stellungen und Bildervs belief. Belief \s much stronger than Vorstellungen und
Bilder. Belief, unlike Vorstellungen und Bilder, refers to something one is con-
vinced of and that guides or conditions ones behaviour. The choice of such a
“strong” word as belief is therefore important in that it helps cement Gold-
hagen s thesis. The fact that the word belief‘m the structure these are the beliefs
is repeated very frequently (with some minor variations), makes its toned-
down versions in German all the more critical.

V: Kaplan heobachtet ganz gewohnliche Deutsche vs It is the masses, the ordi-
nary Germans... whom Kaplan exposes. The connotation of “expose” is totally
lost in the neutral German “heobachten”\ VI: wie so viele Beteiligte vs like so
many other Germans engaged in the slaughter, VII:... die Deutsche dazu be-
wogen ... und spdter auch wie die Manner des Polizeibataillons 101 am Ver-
nichtungsprogramm teilzuhaben vs... thatprepared the men of Police Battalion
101 and so many other Germans to be eager killers. The abstract German
version flattens and avoids the impact of “eager killers”’, VIII: eine kleine Min-
derheit, die Zuriickhaltung iibte vs the tiny minority...; X: die grofie Mehrheit
der Deutschen vs the vast majority of the German people.
There are also two omissions which can be interpreted as being in line
with the general pattern of flattening and neutralizing the author’s main thesis,
the omitted stretches being particularly strong versions of it: II: 0 in German
vs Under the proper circumstances, eliminationist antisemitism metastized into
its most virulent exterminationist form and ordinary Germans became willing
genocidal killers', VIII: 0 in German vs the cruelty in the camps having been
near universal.

Textual mismatches:
a) Iconic linkage: II: no equivalent structural parallelism for: Genocide was was immanent..And it was immanent.
b) Loss of foregrounded rhematlc structures in III: Der Antisemitismus war
wahrend der Nazizeit... vs So profound and near universal was antisemi-
tism-, V: Kaplan beobachtet ganz gewohnliche Deutsche ... vs It is the ma-
sses, the ordinary Germans ... whom Kaplan exposes.

Social Attitude
The style level in the German translation is more formal, more neutral, and
socially distant due to the lack of rhetorical intensity and emotionally gripping,
addressee-involving linguistic devices as characterized under AUTHOR’S
STANCE and FIELD above.

Due to the lack of rhetorical devices such as repetition, iconic linkage, fore-
grounded structures and [+emotlve] lexical items including Intenslflers and
superlatives the translation is less Involved and more informative than the

The translation is even more monologous than the original due to the absence
of (+emotive) linguistic devices (cf. the values along the other situational di-

The GENRE of the original - an academic history text with a provocative new
thesis and a strong emotional appeal is realized differently in German in that
the Author*s Stance^ as well as his rhetorical strategies are toned down consid-
erably in the translation - at least in the passages I selected. Having conducted
a more cursory analysis of the complete text, I would hypothesize, however,
that this analysis holds for the whole book.

Statement of Quality

There is a pattern of differences between translation and original along the

dimensions of FIELD and TENOR, which strongly affect the interpersonal
functional component of the translation. The differences on the dimension
FIELD, where the repetition of the key word German is consistently ignored,
and other key concepts and terms are also either missed out or toned down,
detract from the interpersonal component making the text less provocatively
effective and emotionally intense. On TENOR, there is a pattern of changes
in the author’s intellectual, emotional and moral stance due to the omission
and/or downtonlng of Intensifiers, superlatives and other upgrading devices
as well as [+emotive] lexical items, and the lack of Iconic linkage and fore-
grounded rhematic structures in the translation. These changes also Influence
the translation’s style level, making it more formal, and the original’s mono-
logous character is reinforced in the translation due to a lack of rhetorical and
emotive devices. All these changes have an effect on the realization of the
GENRE: the translation is more of a serious academic document rather than
a provocative text expertly trying to convince its readers minds and hearts.
While this analysis, which is based on an analysis of only a few pages taken
out of a 600 page volume is of course not representative as a sample, I still
believe that I am in a position to hypothesize - and a cursory analysis of the
entire text supports this hypothesis - that the German translation of Goldha-
gen’s book reduces the Impact of the original in the areas I mentioned. I thus
essentially confirm the claim made in an article in the magazine Der Spiegel
(August 12, 1996) that “die Ubersetzung glattet Goldhagens Thesen”. It is
Interesting to consider the author’s rejection of this claim in a letter to the
editor in the following issue of Der Spiegel (19.8.1996), where he calls the claim
“absurd” and denies the Spiegel’s brief exposition any validity stating that the
cited examples were trivial “Spitzfindigkeiten” taken out of context. Seeing
that my analysis has exposed a pattern of mismatches consistent with the Spie-
gel’s claims, one wonders why this denial was made. An Interview with the
translator revealed the appropriateness of the Spiegel's and my own analysis,
and a claim was made that the changes had been imposed by the publisher for
political and marketing reasons.

In his letter to the Spiegel, Goldhagen also comments on the book’s title
Hitler's Willing Executioners and its translation into German as Hitlers willige
Vollstrecker, once again a toned-down rendering: Vollstrecker in German con-
ventionally collocates with Testament as e.g., in “Testamentsvollstrecker”,
whereas the English word executioner in the context of this book is more
equivalent to a “Henker” or “Scharfrichter” in German. Despite Goldhagen’s
claims that these lexical items are too narrow in meaning given the wider
meaning potential of executioner, the fact remains that there is a whole
pattern of differences, i.e., there is convergingevidence for the hypothesis that
the German translation is systematically toned down in many specifiable in-
stances and on different linguistic levels.
The last sentence of the book is also revealing in this connection. In the
English original we read; The camp reveals the essence of the Germany that
gave itself to Nazism, no less than the perpetrators reveal the slaughter and
barbarism that ordinary Germans were willing to perpetrate in order to save
Germany and the German people from the ultimate danger “DER JUDE”.
The German translation renders this sentence as follows: Das Lager offenhart
das Wesen Deutschlands, das sich dem Nationalsozialismus ausgeliefert hatte
- so wie die Morde und die Barharei der Tdter die Bereitschaft ganz gewohn-
licher Deutscher offenharen, Deutschland und das deutsche Volk vor seinem
vermeintlich gefdhrlichsten Feind zu retten: “DEM JUDEN”.
The German translation of this last sentence again removes the Germans
out of the unbearably strong accusatory light that Goldhagen throws on them,
toning the original down and, in this Instance, even commenting upon it. “To
give oneself” implies a free will on the part of the one doing the giving, and a
belief that this giving is something positive, whereas the German “ausllefern”
implies something negative, it is frequently used in the passive voice implying
that one can’t help it as in e.g., “ich war Ihm ausgeliefert”. Even in the active
voice, this connotation is retained, i.e., “Deutschland hatte sich dem Natio-
nalsozialismus ausgeliefert” implies that it is the Nationalsozialismus that is
negative and bad, not so much the Germans themselves.
A more serious change of the meaning of the original’s last sentence, how-
ever, is the following: “the perpetrators reveal the slaughter and barbarism
that ordinary Germans were willing to perpetrate” surely means: “die Tdter
offenharen das Abschlachten und die Barharei, die ganz gewohnliche Deutsche
willentlich durchfiihrten”, and not, as in the translation "... so wie die Morde
und die Barharei der Tdter die Bereitschaft ganz gewohnlicher Deutscher of-
fenharen, Deutschland... zu retten”. As it stands, the translation clears those
ordinary Germans of the blame and guilt, which is of course in stark contrast
to the preceding 600 pages (at least in the original).
With regard to my assessment of the translation and the problems sur-
rounding the decision for an overt or a covert translation to be made and the

placing of a cultural filter between original and translation in the process of
covertly translating, one might argue that the German translation of the Gold-
hagen text Is covert, and that a cultural filter has been used changing a strongly
Interpersonally active text Into a more content-based sober and scientific one
that Is more In keeping with the German conventions for academic texts (see
Chapter 3). On the other hand, given the status the author and his book has
attained through the Intense discussion of his theses especially In Germany,
one might argue that an overt translation would have been more appropriate,
whose function It is to give German readers access to the original (in an un-
adulterated fashion). In order to enable them to judge the book for themselves.
This Is the position I am taking here - knowing full well, however, that
political and marketing considerations on the part of publishers and all those
exerting an Influence on them are often more powerful and can overrule any
translator’s professional and ethical decision to allow the readers of the trans-
lation undlstorted access to the author’s original voice.

:. /H 2.^’ V *' *f^ * * '".*’* ■ ■*'

’: irT3->i

.,' ^rr>'r'*^ ,Shil|^»f| fQ

'♦*/ ,.> -ir- , .i • ■- :i .

■.,:!;.'I'x- ..>>jis«...;.. ., ^ • 'i

f,iu. i;«- V-;

- “!• •, I I tM • I* • ■ .•:

.: v" •►« .* ' ^ :

^*•4 i»» fr>4*

■#. iS' ^ » »v«^J' '■*

1 i • '■

I ' I <»*

• 4. 'i V
. . 'I ‘ f»/ I

Ts " ’ ' I . > '.' ^ .•* - *

.,' i I" *

t , •%. ’ .•!«<*%•“» < • • ^ iir'^t|«*

' «. ‘^/•»-' ■• «. ,| .V ». lk?>'

I» j ' . f '«•. ^. »• — j fI»' * K <v* ■ <

, ‘ . I , •

• ' ■ ■" * * - ■ . ■ »-«J ♦ 1 ■

••* f .•■■*• T"’'!’’ ' ^ ^ J -*0 i

' : ■ 4 .•* • ^ • ♦ A > • ;

. . •. i ,• » V ». { ( ,i . *./»<' If ,.v ..'!^

,- .4, *.• -•■■^ - *■ / Vs, •. ! -

• .■lit.*.!-*

.!:, i • ^ #* ► . S. AiV

* ^*1 * "* ■bf «r.f

.'' •-i dk^Xt' t • ■ ■».



In this final Chapter I shall both re-present the central theses that have been
developed in the preceding Chapters, and develop selected theoretical and
practical aspects of the model presented in some detail.

1 Theoretical Aspects of the Model

In this book I have presented a revision of a model of translation quality

assessment. Central concepts of the original model have been retained, but
made more accessible, and more relevant to actual concerns in translational
theory and practice. The book has therefore both Introduced new concepts -
specifically in terms of the analytic apparatus suggested for categorial linguistic
analysis, and at the same time retained the central notion of source and target
text comparison as the basis for translation quality assessment, even when this
text-based approach has for many specialists in the translational field been
overtaken by a more target-audience-oriented notion of translational appro-
priateness. I believe this recent shift of focus in translational studies to be
fundamentally misguided.
The analytical categories are revised in such a way that the three levels of
analysis: LANGUAGE/TEXT, REGISTER and GENRE are related expon-
entially, and at the same time each level contributes to the characterisation of
a functional profile for the individual source or target text.
The notion of GENRE is of theoretical Interest, though this interest has
not been developed in detail in these pages so far. I have proposed to use the
term in its everyday sense, while applying some restrictions. What counts as
a GENRE cannot therefore be manipulated by academic whim, but is to be
discovered in the everyday practice of the llnguaculture in question. The ca-
tegory remains therefore a socially-determined, pre-scientlflc category in the
sense that its parameters cannot be set by scientific decree. Consequently, of

course, it Is conceded that the concept remains fuzzy-edged. Two factors con-
tribute to this lack of clarity. Firstly, we have the issue of degree of delicacy in
distinguishing one GENRE from another. If “speech” constitutes a GENRE
does “speech given at a dinner-party held for fund-raising purposes” also con-
stitute a GENRE? The category is for my purposes to be conceived very
broadly, but precisely how broadly is a priori difficult to establish. Secondly,
we have the issue of criteria for distinction - as GENRES can be characterised
and distinguished on many dimensions simultaneously, the use of different
distinguishing criteria will lead to different categorial results. Take for example
the instance of a verbal delivery in a formal setting which is designed to intro-
duce another person, who will later address the assembled company. Infor-
mally we could assign such an event to the category “speech” and/or the cate-
gory “Introduction”. It is both. Note that the categorlal assignment “Introductory
speech” does not resolve the theoretical issue at stake, firstly because it simply
transposes the categorlal issue to one of degree of delicacy, and secondly be-
cause it leaves open the hierarchical issue, l.e., the question as to whether we
are handling a speech used as an introduction, or an introduction realised via
a speech. I have adopted the position that GENRE is to be conceived so broad-
ly that an inventory of generic categories subsumes all texts across all cultures.
However, I have not attempted the rather daunting task of spelling out what
this Inventory might contain. Further, while any text in any culture can, in
principle, be assigned to a GENRE, it is not the case that every GENRE has
textual realisations in every culture. The category GENRE is therefore uni-
versal in the sense of universal grammar - it can accommodate all Instances,
but not every instantiation would realise all its exponents: in different cultures,
certain GENRES might have “null realisations”.
It follows that inside the theory developed in this book, a translation of
a particular text will belong to the same GENRE as the original, be that trans-
lation overt or covert. In the case that the GENRE to which the source text is
to be assigned has a null realisation in the target llnguaculture, and a covertly
functional equivalence is desired, then clearly translation as such is not possi-
ble: a different version of the source text will have to be substituted, or Indeed
some non-verbal realisation. Crudely, if a greeting ritual in one culture is per-
fomed by extensive, conventionalised verbal means, in another by the ex-
change of gifts, in another by the rubbing of noses, bowing or the shaking of
hands, it is inappropriate to describe these different behavioural modes as
“covert translations” one of another.
Leaving such theoretical cases aside, I wish to propose that in the trans-
lational process, GENRE stays Intact, although register-shifting and cultural
filtering may accompany the translation process in the covert case, and al-
though an additional GENRE may be superimposed in the case of the overt
translation (the translation serves, for example, to exemplify cultural, textual.

rhetorical or other features of the source culture). By definition, then, if the
GENRE is changed with no apparent necessity, one may no longer be able to
speak of a translation, but should refer to a version.
The category of GENRE relates an individual pair of texts to this broader
category, and enables thereby intra-generic comparison, and, possibly, gene-
ralisation. In my own research (House forthcoming), it has for example been
possible to examine a larger corpus of texts belonging to the specific GENRE
of children’s books. The analysis and comparison of 52 German and English
texts inside this GENRE, together with their translations, gives explanatory
weight to the analysis of any one single textual pair, in that the results of the
analysis of one original and its translation can be related to a system of trans-
lational norms and options operating within this particular GENRE, across
this intercultural gap. The model developed in this book thus opens up the
possibility of research into culturally-conditioned translational norms, which
are likely to differ for different GENRES. More research with large corpora
of texts belonging to different GENRES would thus give greater explanatory
value to any individual translation analysis.
The second major theoretical thrust of the model developed in this book
concerns cultural filtering, and the overt-covert distinction.
As indicated at several points in the book, and particularly in Chapter 3,
an empirical rationale for cultural filtering has been proposed for the transla-
tional pair German/English, in that extensive cross-cultural pragmatic re-
search into differences of communicative preferences in the German and Ang-
lophone llnguacultures has established a number of differences. The
establishment of such differences does not though in itself resolve questions
of cultural filtering in the evaluation of translations between German and Eng-
lish. It allows us rather to pin them down more concretely. Let us accept for
example that at a particular point in time, children’s literature in German ex-
hibits systematic differences from texts of the same GENRE in another lin-
guaculture, for example English. Suppose we then establish empirically - as
in fact I claim to have done - that these differences can be given a plausible
explanation - grounds are proposed for “skewing” in the translation of English
children’s books into German. Do such empirical facts “justify” skewing, or
cultural filtering? The answer is of course affirmative - I have already revised
my earlier negative judgement concerning similar cases in Chapter 4 of this
book. But does this mean that a case of non-filtered (but covertly intentloned)
translation thus becomes inappropriate? There is, it seems to me, no simple
answer to this question. In the case of children’s books, it would be a rash
decision indeed to suggest that non-filtering leads to inadequate or non-ap-
propriate translation. Presumably two factors would Influence the degree of
freedom to be tolerated in cultural filtering: the rigidity of the norms holding
for books of the relevant GENRE written in the target language, and the

general rigidity with which social norms are observed and upheld in the lin-
guaculture in question. We need, of course, also to bear in mind that social
norms may either be imposed top-down, or simply be established over time
bottom-up. In either case they are and should be open to change: there is no
ground whatsoever, then, for assuming or decreeing that a translation may not
by definition be a potential agent for such change.
A further theoretical issue here is the question of reciprocity of skew. In
other words, if it is deemed appropriate and normative to filter when transla-
ting (in a certain way, namely covertly) frorrlslinguaculture A to linguaculture
B, then we might logically suggest that it is equally warranted and indeed
necessary to filter in the opposite way when translating from B into A. Is this
logical conclusion justified? Firstly, how far such bi-directional skewing takes
place is of course an empirical issue, but I know of no studies that explicitly
address it. My own work - especially with children’s literature - suggests for
the translation pair English and German that skewing is in this case often but
not necessarily always reciprocal, in that German children’s literature trans-
lated into English is not consistently filtered towards the norm for children’s
literature written in English to the extent that children’s books translated into
German are so filtered. I hypothesise that this is not an isolated case, and that
there is such a thing as preferred direction of skew. If this is so, and if in fact
bi-directional skewing is found not to obtain with other language pairs, it
would be of interest to Investigate why. One might then develop a dlrectlon-
of-skew hypothesis, predicting, possibly, skewing on a probability scale, given
genre, language pair, and direction of fit. These proposals are not in the least
fanciful. Inside research into language teaching and learning, for example, a
so-called contrastive hypothesis was developed in the sixties, which suggested
that the degree of difference between two languages correlated with the degree
of difficulty with which a speaker of one of that pair of languages could be
expected to learn the other. It did not take long to establish, however, that
direction of fit, and type of difference were, amongst other things, important
factors Inside any contrastive hypothesis (see Edmondson and House 1993:
208ff for a brief overview).
The issue of cultural cross-over in translation is also related to the issue
of behavioural norms inside a foreign language culture. In other words, if
research (and experience) establish that speakers of language X have rather
different interpersonal behavioural norms in everyday interaction than speak-
ers of language Y, this clearly does not entail that it is the task of foreign
language teachers to train speakers of X learning Y to adopt the norms of that
target linguaculture, and vice versa (see for example Edmondson and House 1982
on the issue of politeness, and see House 1997 on politeness and translation).
Questions of cultural filtering are of course directly relevant to the ques-
tion as to which type of translation is being attempted. The central distinction
IS between overt and covert translation. These translational types are seen,
however, as endpoints along a continuum, such that unclear cases will in prac-
tice arise. An overt translation aims at what I call second-level-functional equi-
valence. At this secondary level the target text should attempt to match the
This equivalence occurs at a second level., however, precisely because, given
the envisaged readership, the translated text must have a different reception
from that which obtains for the original, source-culture readership. An overt
translation allows members of the target culture access to the function of the
original. The translation lets them eavesdrop, so to speak, and invites them to
perceive the text, as though they were direct addressees enabling them to ap-
preciate the original s function, albeit at a distance.
The metaphor of “levels” has further been given a cognitive, informati-
on-processing interpretation in this book inside frame theory, or the postulate
of co-existing discourse worlds. Thus in an overt translation, the discourse
world of the original is activated together with, and at the same time as, the
cultural and contextual discourse world actually operant for the reader in the
target linguaculture. She may well be moved, enthralled, engrossed by the
translated text, only to be reminded by the occasional reference (possibly com-
mented on in a footnote), or indeed by a “mistranslation” that the reading
experience occurs across a cultural gap.
In overt work, then, the translator is explicitly a mediator. Her role is
Important: the resultant text is clearly her work. This transparent role derives,
however, not from the changes to the original text caused by the translator’s
hand, but by virtue of the fact that the reader is thereby given access to a text
in a new language (the most obvious case is the translation of literary works).
The paradox is then that in the overt case, the translator has the least leeway
to alter the fabric and content of the text, but has a clearly recognisable role
and function for the reader. In covert translation, on the other hand, it is the
task of the translator to be invisible, but at the same time to transmute the
original such that the function it has in its original situational and cultural
environment is re-created in the target linguaculture. To this end, various “fil-
tering” devices, and translational compromises may well be necessary. At the
levels of LANGUAGE/TEXT and REGISTER, the translation does not
need, then, to be equivalent with its original. An equivalent function is to be
aimed at, although it is differently framed, and operates in the target culture
and discourse world. In order to achieve this, changes at the levels of REGIS-
TER, and the LINGUISTIC/TEXTUAL realisation thereof, may be warran-
ted. In covert translation, the translator’s work is hidden. It is, essentially, the
translator’s task to cheat, i.e. to achieve in translation a second original, hiding
its source. This can be done relatively easily with authorless texts or texts that
have dispensable authors, as is the case with light-weight journalistic texts.

Instructions, tourist brochures or advertisements. However, it is important to
clearly demarcate the lines between what is still a translation and what is “a
new text for a new event in a new culture”. In the terms provided in my model,
a covert translation despite changes (brought about for example via cultural
filtering) on the LINGUISTIC and REGISTER levels must be functionally equi-
valent to the original.
The paradox can now be reformulated as follows: in an overt case, the
translator has to make, as it were, as few changes as possible, and will be held
openly accountable for the degree of succes^with which this is achieved; in a
covert translation, the translator is implicitly licensed to make as many sub-
stantive changes as necessary, and in fact is only likely to be “caught out” and
held accountable in the case that not enough change was wrought, such that
the resultant text is in fact perceived to be a translation.
The concept of “functional equivalence” is thus differentiated inside the
covert-overt distinction. The above discussion of categories of GENRE, types
of cultural filtering, and the overt-covert distinction feeds naturally into a
discussion of evaluation. What makes one translation more “successful” than
another, how can the analytic apparatus developed in this book, together with
the model for which it is to operate, embracing the concepts and distinctions
discussed above, in fact give us an assessment of translation quality?
A first point is that there is no scientific basis for determining that a given
source text requires a translation of the one kind or the other by virtue of its
status, function, content or anything else. In other words, it is not warranted
to claim that in translating any text, only an overt (or only a covert) translation
is possible. This is not to say that “anything goes” along the covert!overt
continuum. There are for example Issues of consistency and transparency that
can be invoked in assessing translation value. The point concerns not arbitrary
fusion of overtly and covertly translated elements in one piece of translational
work, but the decision to follow the one translational strategy or the other.
The reader is further reminded that I have been at pains to draw the line be-
tween a translation in the technical sense of a “functionally equivalent” text
in one of the senses adumbrated, and new texts or versions based in different
ways on a source text. It seems to me however that a model of the kind pre-
sented in this book, and Indeed a model of any other kind that aspires to be
open to professional discussion and evaluation, cannot be prescriptive on this
issue. The reason is that just as the provenance of a source text is Itself em-
bedded in a soclo-cultural milieu, which of course will be taken into account
in analysing that source text, so too is the translational Imperative culturally-
embedded. It is the outcome of a cultural/business negotiation in a particular
social context, and not the result of the application of a theory of translational
quality assessment to a given set of social parameters. The task of a model for
translation quality assessment is therefore not to ask whether a particular trans-

lational blueprint, task, or attempt is or was licensed, but to assess as objectively
as possible what has emerged from the set of circumstances Inside which the
translation task or purpose was conceived.
Different texts and other cultural products are put to many different uses
in society, be it the playing of Beethoven in a radio clip to sell chocolate, the
use of some advertising jingle based on Shakespeare, pictures of Einstein on
posters for anything, or Indeed the use or misuse of literary quotation by
politicians. Such facets of our current society are doubtless of professional
interest to sociologists, marketing experts and students of rhetoric amongst
others in different scientific communities, but not necessarily of interest to
musicologists, students of Shakespeare, theoretical physicists and literary cri-
tics. In other words, while one may decry the fact that certain texts are trans-
lated and used in certain ways, this has as far as I can see nothing to do with
their evaluation inside translation quality theory.
If, then, it is the case that the translation theorist cannot qua expert decree
that a specific text requires an overt as opposed to a covert translation, or vice
versa, it follows that as regards mismatches which result from some form of
cultural filtering, an evaluative judgement can only be relative - despite the
evidence provided in empirical validation of the cultural filter^ as is the case
for the language pair English-German. The case of the Goldhagen text,
discussed at length in Ghapter 5 may be seen as a crucial example. I have shown
that systematic changes to the original English text have been perpetrated in
the translation of this book. It is plausible to suggest that the grounds for such
filtering have little if anything to do with German conventions concerning
historical texts or indeed with German communicative preferences. It seems
more likely that the “skewing” in this text is in part at least occasioned by the
nature of the content of the text. Thus we have in this sense a “covert” trans-
lation of what must be classified as a work of historical scholarship in terms
of GENRE, although a priori an overt translation might be deemed to be
normative. The point here is that a theory of translation quality assessment is
not into the business of postulating marketing motives, publishers’ power,
best-seller aspirations, or indeed simple decent-mlndedness, a desire to avoid
unnecessary offence, a sense of cultural tact or any other fancied motives to
the translational strategy pursued in this or any other translational instance.
Nor is it the business of a translation quality theory to say, regardless of motive,
that filtering in this or any other particular case, was unwarranted, resulting
in a qualitatively inferior translation. We may note here en passant that in our
everyday use of language the idea of tailoring what we say to the recipient is
not only normative, but in fact determines the communicative occasion. Thus,
an utterance of the kind “If I had been writing in language X, of course, I
would have written it differently” is perfectly normative.

In other words, it must be emphasized again that the model provides for
the linguistic analyis, description and comparison of texts, linking them with
situational and cultural contexts, and (through the category of GENRE) with
other texts of the same communicative purpose. This is to be conducted within
the framework of a theory of differentiated translational purposes {overt ver-
sus covert). This type of description, interpretation and explanation, enabled
via the model and the metalanguage it provides is not to be confused with the
type of “good vs bad” evaluative judgment m^e on the basis of social, political,
ethical or moral norms or individual persuasion. In the present climate, in
which the criterion of scientific validity is in danger of being usurped by criteria
such as social acceptability, political correctenss, or simply emotional com-
mitment, it seems Important to stress this distinction.
As regards translation, many contemporary studies appear to confuse a
concern with translation as a phenomenon in its own right, i.e., as a linguistic-
textual operation, with issues such as what translation is for, or what it should,
might, or indeed must be for. A one-sided concern with the covert end of the
dine not only reduces the Importance of the source text in translation, but
also blurs the borders between translations and other multilingual textual pro-
cedures. It is one of the purposes of the model developed in this book to
provide means of conceptually separating a translation from a version, through
positing functional equivalence between source and translation text as a sine
qua non in translation.
At the same time, I have, in the above discussion, suggested that in critical
Instances (cf. the Goldhagen text), a distinction is to be maintained between
the descriptive-explanatory, scientific moments in translation evaluation, as
opposed to soclo-psychologlcally based value judgments, which may well sur-
face in any critical account. This does not mean that the attempt to develop a
model of translation quality assessment is pointless. On the contrary. A de-
tailed analysis of the “hows” and the “whys” of translated texts versus their
originals has to be the descriptive foundation for an argued assessment of
whether and to what degree a given translation may be seen to be adequate or
not. Acknowledging the inevitably subjective element in any value judgment
does not, then, invalidate the objective part of translation evaluation, it merely
reinforces its necessity. Like language itself, translation quality assessment has
two functional components, the ideational and the interpersonal. In other
words, two steps may be distinguished: the first relates to analysis, description
and explanation based on knowledge (of linguistic conventions), and empirical
research, while the second relates to judgments of values, to social and moral
questions of relevance and appropriateness and, of course, to personal prefe-
rence or taste. Both components are implicit in translation quality assessments.
The second is pointless without the first. To judge is easy: to understand is less
so. If we can make explicit the grounds of our judgements, on the basis of an

166 *
argued set of procedures such as those developed in this book, then, in the case
of disagreement, we can talk and discuss: if we do not, we can merely disagree.

2 Practical Relevance: Some Pedagogic Implications

Any attempt to teach translational competence is premissed on the assumption

that one knows what translational competence is. Therefore a theory of trans-
lation and of translation quality assessment must underlie any pedagogic train-
ing for translators. Transposing this logic, it seems justified to posit some pe-
dagogic principles, which derive from the argumentation and empirical
evidence brought forward in the foregoing pages. The following principles
and procedures are therefore suggested:

(1) The difference between overt and covert as essential choices in translation
should be explicitly taught.
(2) Model analyses of parallel texts and translations based on the evaluation
model can be used in the teaching of translation.
(3) Different translations of one and the same source text, possibly into dif-
ferent languages, can be analysed and compared using the framework of
the model and the metalanguage provided.
(4) Source texts can be first analysed with the model s categories and then
translated or rendered as overt or covert versions.
(5) Source Texts belonging to different GENRES can be translated such that
Individual register dimensions are deliberately changed.
(6) Conversions of overt into covert translations and vice versa can be profi-
tably undertaken as they highlight the changes necessary on the different
levels of analysis, and the resultant type of functional equivalence.
(7) Corpora of texts and their translations belonging to particular GENRES
can be analysed on the basis of the model with the resultant differences
in realizing generic choices in individual textual examples serving as im-
portant base-line data.
(8) One can usefully compare the translation strategies that suggest themselves
in different GENRES, with special attention given to the contrast between
ephemeral, authorless, or multiply-authored texts and texts that gain status
though having a recognized author.
(9) Translation students might well be presented with cross-cultural research
involving relevant language pairs.

These suggestions are, of course, very general, as can perhaps be excused given
the purpose of this book. I therefore leave the detailed practical realization
and empirical testing of the above suggestions to experienced translation

3 Concluding Remarks

At the beginning of this book I asked three questions that a theory of trans-
lation should try to answer. These questions concerned the relationship be-
tween source and target texts, the relationship between those texts and the
relevant set of their human users (readerships, author, translator), and the de-
limitation of translation from other cross-cultural uses of source texts.
I have tried to present my views on al^three questions by developing a
model which deals in detail with both text-internal issues of analysis and com-
parison, and also with text-external constraints concerning llnguacultural
norms and expectations. I have tried further to relate these two perspectives,
and to Integrate them into one theory, while retaining a basic distinction be-
tween two translational types, and making a sharp distinction between empi-
rically-motivated and socially-conditioned grounds forjudging a translation.
I have thus fulfilled my overt purposes in writing this book. About my covert
purposes the reader may merely speculate.

168 *

A: The following two texts (each in its original and in a translated

version) are analysed in chapter 2

II Commercial Text
ST: M. F. Meissner, President, Investors Overseas Services,
Letter to Shareholders:

December 27, 1971

I 1 Dear Shareholder,
II 1 The Board of Directors of I.O.S., Ltd. has declared; a pro-
rata dividend payable on and after December 20, 1971, to all
shareholders of record as of the close of business on December
2 17, 1971. / The dividend consists of shares of Value Capital Lim-
ited, a newly established Bahamian holding company, and will be
paid on the basis of one share of Value Capital Limited for each
3 whole ten shares held of I.O.S., Ltd. / Of course, each share-
holder continues ownership of any share that he now holds of lOS.
III 1 In organizing Value Capital Limited, lOS contributed to it
certain companies including IVM (the Dutch Insurance company),
IVM Invest Management Company Limited, ILI Luxembourg, ILI
Bermuda, lOS Real Estate Holdings, IPI Management Co., and
Resources Services Limited, together with certain other contrac-
2 tual rights and assets. / In return for its contribution, lOS re-
ceived 6.2 million shares of Value Capital Limited (the total of
the issued and outstanding shares of that Company), and, in turn,
is distributing to its shareholders all of these shares.
IV 1 The total stockholders’ equity of Value Capital Limited
2 is $1.3 million. / Since future earnings of Value Capital
Limited will not be controlled by lOS, historical earnings
performance would not be Indicative of expected future
V 1 The dividend will be represented by bearer certificates
2 which, as you know, are negotiable instruments. / That is,
they may be traded by anyone in possession of the

3 certificate. / In order to avoid the possibility of accidental
misdirection of your certificates, and to expedite the
4 distrubiton, your assistance is required. / We have enclosed
a “Dividend Instruction Form” for your completion; this
should be returned in the pre-addressed envelope.
VI 1 As you will note, we have asked that you designate a
bank (or broker) to which your dividend certificates will be
2 sent. / Your bank (or broker) should- Indicate its confirmation
of your signature by executing the bottom half of the
“Dividend Instruction Form” Including its official signature
and stamp (or seal).
VII 1 It is anticipated that your new Company will issue its
first report, covering its financial position at May 31, 1972, as
2 soon as possible following that date. / This report will
include full details on the Company’s organization,
3 management and plans for future development. / In the
interim period, the 15,000 to 20,000 shareholders of Value
Capital Limited can expect that public trading of their shares
4 will develop. / It is the present intention of Value Capital
Limited to secure the listing of its shares on a recognized
exchange at the earliest possible time.
VIII1 As a result of the dividend by lOS of its complete holdings
of Value Capital shares, there remains no equity ownership or
2 control of Value Capital in the hands of lOS. / Therefore the fu-
ture market value of Value Capital shares should in no way be
related to, or depend upon, the future development of lOS.
IX 1 The principal reason for the establishment of Value
Capital Limited, and the distribution of its ownership to the
lOS share holders, was to permit the continuation and
expansion of essential communication with the hundreds of
2 thousands of fund clients. / Recent Swiss legislation
precluded the maintenance of these operations from
Switzerland as in the past.
X 1 Value Capital’s client service functions will be
conducted from new facilities being established outside of
2 Switzerland. / The implementation of these client services
should result in a residual benefit to the business of the
principal operating subsidiaries, lOS Insurance Floldings and
Transglobal Financial Services, which are retained by lOS.
XI 1 Value Capital Limited additionally Intends to establish
an international Insurance operation based upon the three

2 insurance companies which lOS contributed to it. / Certain of
the other Value Capital operations were contributed by lOS
in order to provide an immediate income flow to the new
Company and thus insure stability throughout its formative
XII 1 It IS expected that the lOS shareholders will realize a
greater growth.potential through their direct Interest in the
new Value Capital Limited operations than would have been
possible had those operations remained within the lOS group.

XIII1 Very truly yours,

Milton M. Meissner

TT: M. F. Meissner, President, Investors Overseas Services,

Brief an die Aktionare:

27. Dezemher 1971

I 1 Sehr geehrter Aktionar,

II 1 Der Verwaltungsrat der I.O.S., Ltd. hat eine anteilige
Dividende beschlossen, die ab 20. Dezember 1971 an alle Aktionare
zur Ausschiittung gelangt, die zum GeschaftsschluE am 17. Dezem-
2 ber 1971 reglstriert sind. / Die Dividende besteht aus Aktien der
Value Capital Limited, einer nach dem Recht der Bahamas neu-
3 gegriindeten Gesellschaft./ Jeder Aktionar erhalt auf je voile zehn
Aktien der I.O.S., Ltd. eine Aktle der Value Capital Limited. /
4 Er bleibt natiirlich weiterhin Eigentiimer aller seiner bisherigen
Aktien der I.O.S., Ltd.
III 1 Bel der Griindung der Value Capital Limited iibertrug die
lOS auf dlese Gesellschaft bestimmte Gesellschaften, einschliefi-
llch der IVM (die nlederlandische Versicherungsgesellschaft), IVM
Invest Management Company Limited, ILI Luxembourg, ILI Ber-
muda, lOS Real Estate Holdings, IPI Management Co. und Re-
sources Services Limited, sowie bestimmte vertragliche Rechte
2 und Aktiva. / Als Gegenlelstung erhlelt die lOS 6.2 Millionen
Aktien der Value Capital Limited (die gesamte Zahl der von dieser
Gesellschaft ausgegebenen und in Umlauf gesetzten Aktien), die
alle von der lOS an ihre Aktionare verteilt werden.
IV 1 Das gesamte Elgenkapital der Value Capital Limited be-
2 tragt 1.3 Millionen Dollar./ Da die lOS keinen Einfluft auf die zu-
kiinftlge Gewlnnentwlcklung der Value Capital Limited haben wird.

wiirde die bisherige Ertragsleistung keinen Aufschlufi iiber die
Gewinnentwicklung geben.
V 1 Die Dividende wird durch Inhaberzertifikate verbrieft. /
2 Diese sind bekanntlich frei begebbare Urkunden, d.h., sie konnen
3 von jedem veraufiert werden, der in ihren Besitz gelangt. / Um
zu vermeiden, daft Ihre Zertifikate versehentlich fehlgeleitet
werden und um die Zustellung zu b^schleunigen, bitten wir Sie, das
beigefiigte Dividenden-Zustellungsformular (Dividend Instruction
Form) auszufiillen und in dem ebenfalls beigelegten adressierten
Umschlag zuriickzuschicken.
VI 1 Wie Sie feststellen werden, haben wir Sie gebeten, eine
Bank (oder einen Makler) zu benennen, an den die Aktienzertifikate
2 geschickt werden sollen. / Sie miissen die Bank (oder den Makler)
bitten, Ihre Unterschrift auf dem Dividenden-Zustelllmgsformular
3 zu bestatigen. /Hierfiir ist auf dem unteren Teil des Formulars eine
Stelle vorgesehen, wo die Betreffenden unterzeichnen und ihren
Stempel anbringen.
VII 1 Den ersten Bericht iiber ihre Finanziage zum 31. Mai 1972
wird die Value Capital Limited so bald wie moglich nach dem be-
2 sagten Datum veroffentllchen. / Der Bericht wird u.a. iiber den
Aufbau der Gesellschaft, ihre Verwaltung und Entwlcklungsplane
3 voile Auskunft geben. / In der Zwlschenzeit konnen die 15.000 bis
20000 Aktlonare der Value Capital Limited erwarten, daft slch
4 der offentliche Handel ihrer Aktlen entwlckeln wird. / Die Value
Capital Limited beabslchtlgt z.Z., die Zulassung ihrer Aktlen zum
Borsenhandel an einer anerkannten Borse moglichst bald zu erlangen.
VIII1 Durch die Dividendenausschiittung begibt slch die lOS aller
2 von Ihr gehaltenen Aktien der Value Capital Limited. /Infolgedessen
verfiigt sie in Zukunft weder iiber Antelle am Kapltal der Value
Capital Limited noch iiber einen beherrschenden ElnfluB auf diese
3 Gesellschaft. / Irgendein Zusammenhang zwlschen der welteren
Entwlcklung der lOS und dem kunftlgen Kurs der Value Capltal-
Aktlen sollte deshalb ausgeschlossen sein.
IX 1 Die Aufrechterhaltung und weltere Entwlcklung wesentllcher
Kommunlkatlonen mlt den Hunderttausenden von Kunden waren die
Hauptgriinde fiir die Errichtung der Value Capital Limited und fiir
die dlrekte Betelllgung der lOS-Aktlonare an dleser Gesellschaft./
2 Infolge neuer schwelzerlscher Gesetzesbestlmmung war die
Fortfiihrung des blsherlgen Betriebes von der Schweiz aus unmog-
llch geworden.
X 1 Die Dlenstleistungen der Value Capital Limited fur die

Kunden werden von neuen Einrichtungen aufierhalb der Schweiz
2 erbracht. / Aus diesen Dienstleistungen diirften sich fiir das Ge-
schaft der wlchtigsten im Besitz der lOS verbleibenden Tochterbe-
triebsgesellschaften, lOS Insurance Holdings und Transglobal Fi-
nancial Services, restliche Gewinne ergeben.
XI 1 Ausgehend von den drei Versicherungsgesellschaften, welche
die lOS auf die Value Capital Limited iibertragen hat, beabsich-
tlgt dlese aufierdem, eln Internationales Versicherungsunternehmen
2 aufzubauen. / Gewlsse andere Betriebe slnd von der lOS auf die
Value Capital Limited iibertragen worden, um zu gewahrleisten,
daB die neue Gesellschaft iiber sofortige Einnahmen verfiigt und
somit die Stabllltat in der Errlchtungsperiode gesichert ist.
XII 1 Durch ihre direkte Beteillgung an der neugegrundeten
Value Capital Limited wlrd sich fiir die lOS-Aktlonare vorausslcht-
llch ein grofieres Wachstumspotentlal ergeben, als ihnen die auf
die neue Gesellschaft iibertragenen Unternehmen hatten bleten
konnen, wenn sie in der lOS-Gruppe verblieben waren.

XIIIl Mit freundllchem Cruft

Milton F . Meissner

III Journalistic Article

ST: Excerpt from William W Howells, “Homo Sapiens: 20 million years in

the making”, in The UNESCO Courier. August-September 1972. 6-8.

I 1 It was out of Dryopithecus stock that man emerged, and,

in fact it was from among the fossils of Dryopithecus that our an-
2 cestor Ramapithecus became known. / G. E. Lewis of Yale in 1934
described the first upper jaw, found in India’s Slwalik Hills,
and pointed to some man-like features.
II 1 Your own mouth will show you these things, where you
2 can feel them with your finger. / Your dental archis short and
rounded in front, while that of apes has become increasingly
longer and broad across the front, with large canine teeth and
3 broad incisors. / Your molar teeth have the cusp and furrow pat-
4 tern of Dryopithecus, but are square; an ape’s are longer. / This
length makes an ape’s face projecting; yours is straighten
III 1 Approaches to the human shape could be seen in the small frag-
ment of Ramapithecus as though he had just set his foot on a path
diverging from Dryopithecus, although unfortunately we have

2 not found the foot, only the jaw. / So Lewis thought Ramapithe-
cus might belong in our ancestry.
IV 1 But the tide of scientific opinion - and such tides are apt to
influence, not facts, but the way we see facts - was against
Ramapithecus, and the fossil was put away in a drawer as sim-
2 ply one more kind of Dryopithecus. / After almost thirty years,
however, L.S.B. Leakey found a very similar fossil at Fort
Ternan in Kenya, which he could date as
being 14 million years old.
V 1 It happened that at the same time, Elwyn Simons at Yale
2 was looking once again at Ramapithecus. / He was Impressed
with what Lewis had pointed to, and saw the same features in
Leakey’s new specimen.
VI 1 Perhaps more important, Simons rescued other pieces of
2 Ramapithecus from burial in museum drawers. / He began ex-
amining old collections in various places from the U.S.A. to
India, and recognized a few more fragments with the same spe-
cial features, fragments which had previously been misnamed
and Ignored, but which he identified as fossils of Ramapithecus.
VII 1 This careful sorting out made it easier to see the slight
distinctions between Ramapithecus on one hand and Dryopithecus,
2 ancestor of the apes, on the other. / Thus we also see the be-
ginnings of the separating paths of human and ape evolution, or
between animals properly called pongids (apes) and those called
3 homlnids (anything on the human side of the same group). / So
palaeontology is not all looking for fossils in old river banks.
VIII1/2 What brought the split about?/ Evolution has “reasons”-
It follows lines of successful adaptatlon-but we know so little
about Ramapithecus having only his jaws and teeth, that we can-
3 not see the “reason”. / We cannot simply say that it is better or
more successful to be “human”, because that really means no-
thing, and Ramapithecus certainly resembled the ancestral apes
4 far more than he resembled man. / Like some chimpanzee popu-
lations he seems to have lived in an open wood and, again like
chimpanzees, it is probable that he was still a tree-user.

TT; Excerpt from William W. Howells, “Zwanzlg Mlllionen Jahre unterwegs

zum Menschen”, in UNESCO Kurier. August-September 1972. 5-7.

I I Der Mensch 1st aus der Art des Dryopithecus hervor-

gegangen, und uber die Dryopithecus-Fosslllen wurde unser

2 Vorfahr, der Ramapithecus bekannt. / G. E. Lewis aus Yale
beschrieb 1934 den ersten Fund eines Oberkiefers im indischen
Siwalik-Gebirge und wies dabei auf einige menschenartige
Merkmale bin. .
1 Wenn wir in unserem Mund mit dem Finger den Zahnboden
entlangfahren, stellen wir fest, daft er kurz ist und vorne
2 gebogen. / Derjenige des Affen ist lang und vorne breit, mit
3 groften Eck- und scharfen Schneidezahnen. / Die Molaren des
Menschen haben dieselben Flocker und dasselbe Furchenmuster
4 wie die des Dryopithecus, sind aber quadratisch. / Jene der
5 Affen sind langer. / Die Lange der Molaren bedingt, daft das
Gesicht der Affen - im Vergleich zum menschlichen - anders ist.
1 Annaherungen an eine menschliche Gesichtsform
konnten aus kleineren Fragmenten vom Ramapithecus heraus-
2 gelesen werden. / Es schien als ob er seinen Fuft in eine
vom Dryopithecus abweichende Richtung gesetzt habe-obschon
wir leider nicht seinen Fuft, sondern bloft seinen
3 Kiefer fanden. / Dies bewog Lewis, ihn in unsere Ahnenreihe
1 Doch die Meinungen der Wissenschaft sind beeinfluftbar
- nicht allein durch Tatsachen, sondern durch die Art und
2 Weise, wie die Tatsachen ausgelegt werden. / Damals war
3 man gegen den Ramapithecus. / Das Fossil, als eine weitere
Art von Dryopithecus abgetan, verschwand vorerst in einer
4 Fort Ternan, Kenia, einen ahnlichen Fund, dessen Alter
auf vierzehn Millionen Jahre geschatzt wird.
1 Der Zufall wollte es, daft sich Elwyn Simons in Yale
zur selben Zeit nochmals mit dem Ramapithecus beschaftigte. /
2 Lewis’ Bemerkungen leuchteten ihm eln, und er erkannte an
Leakey’s neuem Exemplar die glelchen Merkmale.
1 Was vielleicht noch wichtiger war: Simons verhinderte
damlt, daft weitere Funde vom Ramapithecus in Museen ver-
2 schwanden und vergessen wurden. / Von den Vereinigten Staaten
bis Indien schaute er sich zahlreiche alte Sammlungen an und
entdeckte einige weitere Fragmente mit denselben Merkmalen. /
3 Fragmente, die bis dahln falsch benannt oder Ignorlert worden
waren, die er aber als Fossilien des Ramapithecus identifl-
zieren konnte.
1 Dieses sorgfaltlge Aussortieren erleichterte das
Erkennen der felnen Unterschiede zwlschen dem Ramapithecus
einerseits und dem Dryopithecus, dem Vorlaufer der

2 Menschenaffen, andererseits. / An diesem Punkt der
Evolutionsgeschichte beginnt die Trennung von Menschenaffen
und Menschen oder, allgemeiner gefafit, von Pongiden
(Affen und Menschenaffen) und Hominiden (alle Arten mit
Ansatzen von menschenartigen Merkmalen).
VIII1/2 Weshalb aber diese Trennung? / Die Evolution folgt
3 einem Plan./ Ziel ist eine moglichst^orteilhafte Adaptation. /
4 Wir wissen aber so wenig fiber den Ramapithecus-wir haben
ja nur seinen Kiefer und seine Zahne - daft wir hier keinen
5 Plan erkennen konnen. / Es ware unsinnig, “menschlich” mit
6 “vorteilhafter” gleichsetzen zu wollen. / Bestimmt stand der
7 Ramapithecus dem Affen naher als dem Menschen. / Wie
gewisse Schimpansenarten, so vermutet man, lebte er in
lichten Waldern, tells noch immer auf Baumen.

B: The following four texts (each in its original and in a translated

version) are analysed in chapter 5.

TEXT 1: Children’s Book

ST: Jill Murphy Five Minutes Peace. London: Walker Books, 1986.

The children were having breakfast. This was not a pleasant sight.

Mrs Large took a tray from the cupboard. She set it with a teapot, a milk jug,
her favourite cup and saucer, a plate of marmalade toast and a leftover cake
from yesterday. She stuffed the morning paper into her pocket and sneaked
off towards the door.

“Where are you going with that tray. Mum?” asked Laura. “To the bathroom,”
said Mrs Large. “Why” asked the other two children. “Because I want five
minutes’ peace from you lot,” said Mrs Large. “That’s why.”

“Can we come?” asked Lester as they trailed up the stairs behind her. “No,”
said Mrs Large, “you can’t. ’’What shall we do then?“ asked Laura. ’’You can
play,“ said Mrs Large. ’’Downstairs. By yourselves. And keep an eye on the
baby.“ ’’I’m not a baby,“ muttered the little one.

Mrs Large ran a deep, hot bath. She emptied half a bottle of bath-foam into
the water, plonked on her bath-hat and got in. She poured herself a cup of tea
and lay back with her ey£s closed. It was heaven.

“Can I play my tune?” asked Lester. Mrs Large opened one eye. “Must you?”
she asked. “IVe been practising,” said Lester. “You told me to. Can I? Please,
just for one minute.” “Go on then,” sighed Mrs Large. So Lester played. He
played “Twinkle, Twinkle, Little Star” three and a half times.

In came Laura. “Can I read you a page from my reading book?” she asked.
“No, Laura,” said Mrs Large. “Go on, all of you, off downstairs.” “You let
Lester play his tune,” said Laura. “I heard. You like him better than me. It’s
not fair.” “Oh, don’t be silly, Laura,” said Mrs Large. “Go on then. Just one
page.” So Laura read. She read four and a half pages of “Little Red Riding

In came the little one with a trunkful of toys. “For you!” he beamed, flinging
them all into the bath water. “Thank you, dear,” said Mrs Large weakly.

“Can I see the cartoons in the paper?” asked Laura.
“Can I have the cake?” asked Lester.
“Can I get in with you?” asked the little one.
Mrs Large groaned.

In the end they all got in. The little one was in such a hurry that he forgot to
take off his pyjamas.

Mrs Large got out. She dried herself, put on her dressing-gown and headed
for the door. “Where are you going now, Mum?” asked Laura. “To the kit-
chen,” said Mrs Large. “Why?” asked Lester. “Because I want five minutes’
peace from you lot,” said Mrs Large. “That’s why.”

And off she went downstairs, where she had three minutes and forty-five
seconds of peace before they all came to join her.

TT: Jill M.UYip\iyNurFunfMinutenRuh. Miinchen: Annette Betz Verlag. 1987
Aus dem Englischen von Britta Groifi.

Wenn die Elefantenkinder friihstiickten, ging es meistens unordentlich und
laut zu.

Mutter Elefant holte ein Tablett und stellte ihr Fruhstiick drauf: Teekanne,
Milchkrug, ihre Lieblingstasse, einen Teller voll Toast mit Butter und Marme-
lade und einen Krapfen vom Vortag. Sie stopfte die Morgenzeitung in die
Tasche ihres Schlafmantels und schlich zur Kiichentiir.

“Wohin gehst du, Mama” fragte Elefantinchen. “In das Badezimmer” sagte
Mutter Elefant. “Warum?” fragten die Kinder. “Well Ich fiinf Minuten Rube
haben mochte”, sagte Mutter Elefant. “Ganz elnfach deshalb.”

“Diirfen wir mltkommen?” fragte Elefantchen. “Neln”, sagte Mutter Elefant
bestimmt. “Ihr diirft nlcht! ”. “Was sollen wir denn tun?” fragte Elefantinchen.
“Geht splelen”, sagte Mutter Elefant. “Und zwar ins Klnderzlmmer. Und pafit
auf das Baby auf.” “Ich will nlcht mehr das Baby seln”, murrte Baby Elefant.

Mutter Elefant machte sich ein gemiitliches helloes Bad. Sie leerte elne halbe
Flasche Badeschaum in das Wasser, setzte die Badehaube auf und stleg in die
Wanne. Sie gofi Tee in die Tasse und lehnte sich mit geschlossenen Augen
zuriick. Es war himmlisch!

“Darf ich dir was vorsplelen?” fragte ein Stlmmchen. Mutter Elefant offnete
ein Auge. “Mufi das seln?” fragte sie. “Du sagst Immer, ich soil flelfiig iiben”,
sagte Elefantchen. “Darf ich ? Nur elne Minute. ” “Also fang an” seufzte Mutter
Elefant. Elefantchen begann zu floten. Er splelte “ Alle Voglein slnd schon da”
drel und ein halbes Mai.

“Darf ich dir eine Geschlchte aus dem Lesebuch vorlesen?” fragte Elefantin-
chen. “Neln,” sagte Mutter Elefant. “Raus mit euch. Geht hlnunter ins Kin-
derzimmer.” “Du hast Elefantchen erlaubt, Flote zu splelen”, sagte Elefantin-
chen. “Ich habs gehort. Du magst ihn lieber als mlch. Das 1st nicht fair.” “Also
fang schon an. Aber nur eine Seite.” Elefantinchen begann zu lesen. Sie las
vierelnhalb Selten aus der Geschlchte von “Rotkiippchen”.

Baby Elefant schleppte soviel Spielsachen an, wie er nur tragen konnte. “Fiir
dich! ” strahlte er und warf alle in die Badewanne. “Danke, Liebes”, sagte Mut-
ter Elefant schwach.

“Darf ich mir die Bilder in der Zeitung ansehen?”
“Darf ich den Krapfen haben?”
“Darf ich zu dir in die Badewanne kommen?”
Mutter Elefant stohnte.

Schliefilich sprangen alle drei in die Badewanne. Baby Elefant war so aufgeregt,
daft er vergafi, den Pyjama auszuziehen.

Mutter Elefant stieg aus der Badewanne. Sie rieb sich trocken, zog den Schlaf-
mantel an und ging Richtung Ttir. “Wohin gehst du denn jetzt, Mama?” fragte
Elefantinchen. “In die Kiiche”, sagte Mutter Elefant. “Warum?” fragten die
Kinder wie aus einem Mund. “Weil ich fiinf Minuten Ruhe haben mochte”,
sagte Mutter Elefant. “Ganz einfach deshalb”.

Und sie hatte drei Minuten und fiinfundvierzig Sekunden Ruhe, bevor die
Kinder kamen, damit ihre Mutter nicht so allein ware.

TEXT 2: Autobiography
ST: Excerpt from Richard P. Feynman Surely You’re Joking, Mr. Feynman!
Adventures of a Curious Character, as told to Ralph Leighton, edited
by Edward Hutchings, New York: Bantam Books, 1985, pp. 149-150.

The Dignified Professor

I don’t believe I can really do without teaching. The reason is, I have to have
something so that when I don’t have any ideas and I’m not getting anywhere
I can say to myself, “At least I’m living; at least I’m doing something; I’m
making some contribution” - it’s just psychological.

When I was at Princeton m the 1940s I could see what happened to those great
minds at the Institute for Advanced Study, who had been specially selected
for their tremendous brains and were now given this opportunity to sit in this

lovely house by the woods there, with no classes to teach, with no obligations
whatsoever. These poor bastards could now sit and think clearly all by them-
selves OK? So they don’t get any ideas for a while: They have every oppor-
tunity to do something, and they’re not getting any ideas. I believe that in a
situation like this a kind of guilt or depression worms inside of you, and you
begin to worry about not getting any ideas. And nothing happens. Still no
ideas come.
Nothing happens because there’s not enough real activity and challenge:
You’re not in contact with the experimental guys. You don’t have to think
how to answer questions from the students. Nothing!
In any thinking process there are moments when everything is going good and
you’ve got wonderful ideas. Teaching is an interruption, and so it’s the greatest
pain in the neck in the world. And then there are the longer periods of time
when not much is coming to you. You’re not getting any ideas, and if you’re
doing nothing at all, it drives you nuts! You can’t even say I’m teaching my class.
If you’re teaching a class, you can think about the elementary things that you
know very well. These things are kind of fun and delightful. It doesn’t do any
harm to think them over again. Is there a better way to present them? Are
there any new problems associated with them? Are there any new thoughts
you can make about them? The elementary things are easy to think about; if
you can’t think of a new thought, no harm done; what you thought about it
before is good enough for the class. If you do think of something new, you’re
rather pleased that you have a new way of looking at it.
The questions of the students are often the source of new research. They often
ask profound questions that I’ve thought about at times and then given up on,
so to speak, for a while. It wouldn’t do me any harm to think about them again
and see if I can go any further now. The students may not be able to see the
thing I want to answer, or the subtleties I want to think about, but they remind
me of a problem by asking questions in the neighborhood of that problem.
It’s not so easy to remind yourself of these things.
So I find that teaching and the students keep life going, and I would never
accept any position in which somebody has Invented a happy situation for me
where I don’t have to teach. Never.

But once I was offered such a position ...

TT: Excerpt from Richard P. Feynman “Sie belieben wohl zu scherzen, Mr.
Feynman!” Abenteuer eines neugierigen Physikers, Gesammelt von
Ralph Leighton, Heraiisgegeben von Edward Hutchings.
Aus dem Amerikanischen von Hans-Joachlm Metzger. Miinchen: Piper,
1985, S. 219-221.

Der wurdevolle Professor

Ich glaube, ohne zu lehren kann ich iiberhaupt nicht auskommen. Der Grund
1st, ich muft etwas haben, so daft ich mir, wenn ich kelne Ideen habe und nicht
weiterkomme, sagen kann: “Zumindest lebe ich; zumindest tue ich etwas; ich
lelste Irgendeinen Beltrag” - das ist rein psychologlsch.

Als ich in den 40er Jahren in Princeton war, konnte ich sehen, was mlt den
groBen Gelstern am Institute for Advanced Study passierte, die speziell wegen
ihrer ungeheuren Gehlrne ausgewahlt worden waren und denen man nun die
Gelegenheit gab, in diesem schdnen Haus da am Wald zu sltzen, ohne unter-
richten zu miissen, ohne Irgendwelche Verpflichtungen. Diese armen Kerle
konnten jetzt sitzen und ganz ungestort nachdenken, o.k.? und dann fallt
ihnen eine Zeitlang nichts eln: Sie haben jede Mogllchkeit, etwas zu tun, und
es fallt ihnen nichts ein. Ich glaube, in so einer Situation beschleicht einen eln
Schuldgefiihl oder eine Depression, und man fangt an, sich Sorgen zu machen,
well einem nichts elnfallt. Und nichts tut sich. Es kommen Immer noch kelne

Es tut sich nichts, well es nicht genugend wirkliche Aktlvltat und Herausfor-
derung gibt: Man hat keinen Kontakt zu den Leuten, die Experlmente machen.
Man mu£ nicht dariiber nachdenken, wie man die Fragen der Studenten be-
antwortet. Nichts!

Bei jeder geistigen Arbeit gibt es Momente, in denen alles gut liiuft und man
tolle Einfalle hat. Unterrlchten zu miissen, bedeutet eine Unterbrechung, und
deshalb 1st das die groftte Geduldsprobe, die man sich vorstellen kann. Und
dann gibt es die langeren Phasen, in denen einem nicht vlel kommt. Man hat
kelne Einfalle, und wenn man nichts zu tun hat, macht einen das wahnsinnlg!
Man kann nicht elnmal sagen: “Ich habe ja melnen Unterrlcht.”

Wenn man unterrichtet, kann man iiber die elementaren Dlnge nachdenken,
die man sehr gut kennt. Das macht irgendwle Spa£ und befrledlgt einen sehr.

Es schadet nichts, wenn man sie noch einmal iiberdenkt. Kann man sie besser
darstellen? Gibt es irgendwelche neuen Probleme, die mit ihnen in Zusam-
menhang stehen? Kann man irgendwelche neuen Uberlegungen iiber sie an-
stellen? Es ist so leicht, iiber die elementaren Dinge nachzudenken; wenn ei-
nem nichts Neues dazu einfallt, so schadet das nichts; die Gedanken, die man
sich vorher dariiber gemacht hat, geniigen fiir den Unterricht. Wenn einem
aber tatsachlich etwas Neues einfallt, freut t^an sich sehr, daft man eine neue
Methode hat, die Dinge zu betrachten.

Die Fragen der Studenten sind oft die Quelle neuer Forschungen. Sie stellen
oft tiefgrundige Fragen, iiber die ich zu Zeiten nachgedacht und die ich dann
fiir eine Weile gewissermaften aufgegeben habe. Es wiirde mir nicht schaden,
wieder iiber sie nachzudenken und zu sehen, ob ich jetzt weiterkomme. Die
Studenten sehen vielleicht nicht, worauf ich eine Antwort finden mochte oder
iiber welche Feinheiten ich nachdenken mochte, aber sie erinnern mich an ein
Problem. Wenn sie Fragen stellen, die in der Nachbarschaft dieses Problems
llegen. Sich selbst an diese Dinge zu erinnern, ist nicht so einfach.

Ich finde also, daft der Unterricht und die Studenten dafiir sorgen, daft das
Leben weitergeht, und ich wiirde nie eine Position akzeptleren, bei der mir
jemand eine angenehme Stellung elngerichtet hat, wo ich nicht zu lehren brau-
che. Nlemals.

Aber einmal 1st mir tatsachlich eine solche Position angeboten worden...

TEXT 3: Philosophical Essay

ST: Excerpt from Walter Benjamin “Die Aufgabe des Ubersetzers” in:
W. Benjamin Gesammelte Schriften Bd. IV/1 WA Suhrkamp, Frankfurt/
Main, 1972. Vorwort des Ubersetzers (W. Benjamin) zu der Ubersetzung
Charles Baudelaire Tableaux Parisien. S. 9-21.

Nlrgends erwelst sich einem Kunstwerk oder elner Kunstform gegeniiber die
Riicksicht auf den Aufnehmenden fiir deren Erkenntnls fruchtbar. Nicht ge-
nug, daft jede Beziehung auf ein bestlmmtes Publlkum oder dessen Reprasen-
tanten vom Wege abfuhrt, 1st sogar der Begrlff elnes Idealen Aufnehmenden
in alien kunsttheoretlschen Erorterungen vom Ubel, well diese ledlglich ge-
halten sind, Daseln und Wesen des Menschen iiberhaupt vorauszusetzen. So
setzt auch die Kunst selbst dessen lelbllches und geistiges Wesen voraus - seine

Aufmerksamkeit aber in keinem ihrer Werke. Denn kein Gedlcht gilt dem
Leser, kein Bild dem Beschauer, keine Symphonic der Horerschaft.

Gilt eine Ubersetzung den Lesern, die das Original nicht verstehen? Das scheint
hinreichend den Rangunterschied im Bereiche der Kunst zwischen beiden zu
erkliiren. Uberdies scheint es der einzig mogliche Grund, “Dasselbe” wieder-
holt zu sagen. Was “sagt” denn eine Dlchtung? Was tellt sie mlt? Sehr wenlg
dem, der sie versteht. Ihr Wesentllches ist nicht Mitteilung, nicht Aussage.
Dennoch konnte diejenige Ubersetzung, welche vermltteln will, nichts ver-
mitteln als die Mitteilung - also Unwesentllches. Das 1st denn auch ein Erken-
nungszeichen der schlechten Ubersetzungen. Was aber aufier der Mitteilung
in einer Dlchtung steht - und auch der schlechte Ubersetzer gibt zu, daft es
das Wesentllche 1st - gilt es nicht allgemein als das Unfafibare, Gehelmnisvolle,
"Dichterische”? Das der Ubersetzer nur wiedergeben kann, Indem er auch
dlchtet? Daher riihrt in der Tat ein zweites Merkmal der schlechten Uberset-
zung, welche man demnach als eine ungenaue Ubermittlung eines unwesent-
lichen Inhalts definieren darf. Dabei bleibt es, solange die Ubersetzung sich
anheischig macht, dem Leser zu dienen. Ware sie aber fiir den Leser bestimmt,
so miifite es auch das Original sein. Besteht das Original nicht um dessentwil-
len, wie liefie sich dann die Ubersetzung aus dieser Bezlehung verstehen?

Ubersetzung ist eine Form. Sie als solche zu erfassen, gilt es zuriickzugehen
auf das Original. Denn in Ihm liegt deren Gesetz als in dessen Ubersetzbarkelt
beschlossen. Die Frage nach der Ubersetzbarkelt eines Werkes ist doppelsin-
nig. Sie kann bedeuten: ob es unter der Gesamthelt seiner Leser je selnen zu-
langlichen Ubersetzer finden werde? oder, und eigentlicher: ob es seinem We-
sen nach Ubersetzung zulasse und demnach - der Bedeutung dieser Form
gemafi - auch verlange. Grundsatzllch ist die erste Frage nur problematlsch,
die zwelte apodiktlsch zu entscheiden. Nur das oberflachllche Denken wird,
indem es den selbstandigen Sinn der letzten leugnet, beide fiir gleichbedeutend
erklaren. Ihm gegeniiber ist darauf hinzuweisen, daft gewisse Relationsbegriffe
Ihren guten, ja viellelcht besten Sinn behalten, wenn sie nicht von vorne herein
ausschliefilich auf den Menschen bezogen werden. So durfte von einem un-
vergel^lichen Leben oder Augenblick gesprochen werden, auch wenn alle
Menschen sie vergessen hatten. Wenn namlich deren Wesen es forderte, nicht
vergessen zu werden, so wiirde jenes Pradikat nichts Falsches, sondern nur
eine Forderung, der Menschen nicht entsprechen, und zugleich auch wohl den
Verweis auf einen Berelch enthalten, in dem ihr entsprochen ware: auf ein
Gedenken Gottes. Entsprechend bliebe die Ubersetzbarkelt sprachllcher Ge-
bllde auch dann zu erwagen, wenn diese fiir die Menschen uniibersetzbar wa-
ren. Und sollten sie das bel einem strengen Begriff von Ubersetzung nicht

wlrklich bis zu einem gewissen Grade sein? - In solcher Loslosung ist die
Frage zu stellen, ob Ubersetzung bestimmter Sprachgebilde zu fordern sei.
Denn es gilt der Satz: Wenn Ubersetzung eine Form ist, so muft Ubersetzbar-
keit gewissen Werken wesentlich sein.

Ubersetzbarkeit eignet gewissen Werken wesentlich - das heifit nicht, ihre
Ubersetzung ist wesentlich fiir sie selbst, sc^dern will besagen, daft eine be-
stimmte Bedeutung, die den Originalen innewohnt, sich in ihrer Obersetzbar-
keit auftere. Da£ eine Ubersetzung niemals, so gut sie auch sei, etwas fiir das
Original zu bedeuten vermag, leuchtet ein. Dennoch steht sie mit diesem kraft
seiner Ubersetzbarkeit im nachsten Zusammenhang. Ja, dieser Zusammen-
hang ist um so inniger, als er fiir das Original selbst nichts mehr bedeutet. Er
darf ein natiirlicher genannt werden und zwar genauer ein Zusammenhang
des Lebens. So wie die Aufterungen des Lebens innigst mit dem Lebendigen
zusammenhangen, ohne ihm etwas zu bedeuten, geht die Ubersetzung aus
dem Original hervor. Zwar nicht aus seinem Leben so sehr denn aus seinem
“Uberleben”. Ist doch die Ubersetzung spater als das Original und bezeichnet
sie doch bei den bedeutenden Werken, die da ihre erwahlten Ubersetzer nie-
mals Im Zeitalter ihrer Entstehung finden, das Stadium ihres Fortlebens. In
volllg unmetaphorischer Sachlichkelt ist der Gedanke vom Leben und Fort-
leben der Kunstwerke zu erfassen. Daft man nicht der organlschen Lelblichkeit
allein Leben zusprechen diirfe, 1st selbst in Zelten des befangensten Denkens
vermutet worden. Aber nicht darum kann es sich handeln, unter dem schwa-
chen Szepter der Seele dessen Herrschaft auszudehnen, wie es Fechner ver-
suchte; geschwelge daft Leben aus den noch weniger maftgebllchen Momenten
des Anlmalischen deflnlert werden konnte, wie aus Empfindung, die es nur
gelegentllch kennzelchnen kann. Vielmehr nur wenn allem demjenlgen, wo-
von es Geschlchte gibt und was nicht allein ihr Schauplatz 1st, Leben zuerkannt
wlrd, kommt dessen Begrlff zu seinem Recht. Denn von der Geschlchte, nicht
von der Natur aus, geschwelge von so schwankender wie Empfindung und
Seele, ist zuletzt der Umkrels des Lebens zu bestimmen. Daher entsteht dem
Phllosophen die Aufgabe, alles natiirliche Leben aus dem umfassenderen der
Geschlchte zu verstehen. Und 1st nicht wenlgstens das Fortleben der Werke
unverglelchllch viel leichter zu erkennen als dasjenlge der Geschopfe? Die
Geschlchte der groften Kunstwerke kennt ihre Deszendenz aus den Quellen,
ihre Gestaltung im Zeitalter des Kiinstlers und die Perlode ihres grundsatzllch
ewlgen Fortlebens bei den nachfolgenden Generatlonen. Dieses letzte helftt,
wo es zutage trltt, Ruhm.

Ubersetzungen, die mehr als Vermlttlungen slnd, entstehen, wenn im Fortle-
ben ein Werk das Zeitalter seines Ruhmes erreicht hat. Sie dlenen daher nicht

sowohl diesem, wie schlechte Ubersetzer es fiir ihre Arbeit zu beanspruchen
pflegen, als daft sie ihm ihr Dasein verdanken. In ihnen erreicht das Leben des
Originals seine stets erneute spateste und umfassendste Entfaltung.

TT: Excerpt from Walter Benjamin “The task of the translator” in: Andrew
Chesterman ed. Readings in Translation Theory. Oy Finn Lectura Ab,
1989. Translated by Harry Zohn, pp. 14-17.

In the appreciation of a work of art or an art form, consideration of the receiver
never proves fruitful. Not only is any reference to a certain public or its rep-
resentatives misleading, but even the concept of an “ideal” receiver is detri-
mental in the theoretical consideration of art, since all it posits is the existence
and nature of man as such. Art, in the same way, posits man’s physical and
spiritual existence, but in none of its works is it concerned with his response.
No poem is intended for the reader, no picture for the beholder, no symphony
for the listener.

Is a translation meant for readers who do not understand the original? This
would seem to explain adequately the divergence of their standing in the realm
of art. Moreover, it seems to be the only conceivable reason for saying “the
same thing” repeatedly. For what does a literary work “say”? What does it
communicate? It “tells” very little to those who understand it. Its essential
quality is not statement or the imparting of information. Yet any translation
which intends to perform a transmitting function cannot transmit anything
but Information - hence, something inessential. This is the hallmark of bad
translations. But do we not generally regard as the essential substance of a
literary work that it contains in addition to Information - as even a poor trans-
lator will admit - the unfathomable, the mysterious, the “poetic”, something
that a translator can reproduce only if he is also a poet? This, actually is the
cause of another characteristic of inferior translation, which consequently we
may define as the inaccurate transmission of an inessential content. This will
be true whenever a translation undertakes to serve the reader. However, if it
were Intended for the reader, the same would have to apply to the original. If
the original does not exist for the reader’s sake, how could the translation be
understood on the basis of this premise?

Translation is a mode. To comprehend it as mode one must go back to the
original, for that contains the law governing the translation: its translatabllity.
The question of whether a work is translatable has a dual meaning. Either:

Will an adequate translator ever be found among the totality of its readers?
Or, more pertinently: Does its nature lend itself to translation and, therefore,
in view of the significance of the mode, call for it? In principle, the first question
can be decided only contingently; the second, however, apodictlcally. Only
superficial thinking will deny the Independent meaning of the latter and declare
both questions to be of equal significance ... It should be pointed out that
certain correlative concepts retain their meaning, and possibly their foremost
significance, if they are referred exclusively to man. One might, for example,
speak of an unforgettable life or moment even if all men had forgotten it. If
the nature of such a life or moment required that it be unforgotten, that pre-
dicate would not imply a falsehood but merely a claim not fulfilled by men,
and probably also a reference to a realm in which it is fulfilled; God’s remem-
brance. Analogously, the translatabillty of linguistic creations ought to be
considered even if men should prove unable to translate them. Given a strict
concept of translation, would they not really be translatable to some degree?
The question as to whether the translation of certain linguistic creations is
called for ought to be posted in this sense. For this thought is valid here: If
translation is a mode, translatabillty must be an essential feature of certain

Translatabillty is an essential quality of certain works, which is not to say that
it is essential that they be translated; it means rather that a specific significance
Inherent in the original manifests itself in its translatabillty. It is plausible that
no translation, however good it may be, can have any significance as regards
the original. Yet, by virtue of its translatabillty the original is closely connected
with the translation; in fact, this connection is all the closer since it is no longer
of importance to the original. We may call this connection a natural one, or,
more specifically, a vital connection. Just as the manifestations of life are inti-
mately connected with the phenomenon of life without being of importance
to it, a translation Issues from the original - not so much from its life as from
its afterlife. For a translation comes later than the original, and since the im-
portant works of world literature never find their chosen translators at the
time of their origin, their translation marks their stage of continued life. The
idea of life and afterlife in works of art should be regarded with an entirely
unmetaphorlcal objectivity. Even in times of narrowly prejudiced thought
there was an inkling that life was not limited to organic corporeality. But it
cannot be a matter of extending its dominion under the feeble scepter of the
soul, as Fechner tried to do, or, conversely, of basing its definition on the even
less conclusive factors of animality, such as sensation, which characterize life
only occasionally. The concept of life is given its due only if everything that
has a history of its own, and is not merely the setting for history, is credited

with life. In the final analysis, the range of life must be determined by history
rather than by nature, least of all by such tenuous factors as sensation and soul.
The philosopher s task consists in comprehending all of natural life through
the more encompassing-life of history. And indeed, is not the continued life
of works of art far easier to recognize than the continual life of animal species?
The history of the great works of art tells us about their antecedents, their
realization In the age of the artist, their potentially eternal afterlife In succeeding
generations. Where this last manifests Itself, It Is called fame.

Translations that are more than transmissions of subject matter come Into
being when In the course of Its survival a work has reached the age of Its fame.
Contrary, therefore, to the claims of bad translators, such translations do not
so much serve the work as owe their existence to It. The life of the originals
attains In them to Its ever-renewed latest and most abundant flowering.

TEXT 4: History Text

ST: Excerpt from Daniel Jonah Goldhagen Hitler's Willing Executioners.
Ordinary Germans and the Holocaust.
London: Little, Brown, 1996, pp. 444-454.

Even though the elmlnatlonlst antlsemltlc Ideology was multlpotentlal In ac-
tion, It strongly tended, given the Germans’ twentieth-century conception of
the Jews, to metastasize Into Its most extreme, extermlnatlonlst variant, pro-
mising a commensurate political “solution” to the putative “problem.” The
elective affinity between a person subscribing to a racially based, virulent ell-
mlnatlonlst antisemitism and a person concluding that an extermlnatlonlst
“solution” was desirable could already be seen In the latter part of the pre-ge-
nocldal nineteenth century. Fully two thirds of the prominent antlsemltlc po-
lemicists proposing “solutions” to the “Jewish Problem”, who were examined
In one study, agitated during this period explicitly for a genocldal assault
against the Jews. The belief that promoted the Nazis’ steps to eliminate Ger-
man Jewish citizens first from Influence In German society and then from the
society Itself - aside from some cases where the material Interests of Germans
were severely harmed - produced enthusiastic support among the German
population for the ellmlnatlonlst measures. Indeed, every major feature of the
evolving ellmlnatlonlst program, from verbal violence to ghettolzatlon to the
killing Itself, was willingly abetted by an enormous number of ordinary Ger-
mans and failed to produce significant dissatisfaction and (principled) dissent
within the general German populace. The dire diagnosis and prognosis for

Germany - should the Germans not succeed in eradicating the putative Jewish
illness from the German body social engendered both the exclusionary mea-
sures, eventually seen as but temporary and insufficient, and the exterminatory
Impulse. These beliefs justified extermination as the Germans’ appropriate
final treatment for the putative social pathology of Jewry. As one physician
who worked for a time in Auschwitz explains, the nexus between belief and
action - between Germans’ antisemitism a^d their willing slaughter of the
Jews, who were considered by Germans to be, in his words, the “arch enemies
of Germany” - was exceedingly close. “The step,” as this man trenchantly
puts it, “from the monstrous accusations leveled at the Jews, to their annihi-
lation is only a millimeter long.”

Genocide was Immanent in the conversation of Germany society. It was im-
manent in its language and emotion. It was Immanent in the structure of co-
gnition. And it was immanent in the society’s proto-genocidal practice of the
1930s. Under the proper circumstances, eliminationist antisemitism metasta-
sized into its most virulent exterminationlst form, and ordinary Germans be-
came willing genocidal killers. The autonomous power of the eliminationist
antisemitism, once given free rein, to shape the Germans’ actions, to Induce
Germans voluntarily on their own initiative to act barbarously towards Jews,
was such that Germans who were not even formally engaged in the persecution
and extermination of the Jews routinely assaulted Jews physically, not to men-
tion verbally.

So profound and near universal was the antisemitism during the Nazi period
that to the Jewish victims it appeared as if its hold on Germans could be cap-
tured and conveyed only in organic terms: “A poison of diseased hatred per-
meates the blood of the Nazis.” Once activated, Germans’ profound hatred
of Jews, which had in the 1930s by necessity lain relatively dormant, so pos-
sessed them that it appeared to exude from their every pore. Kaplan, the keen
diarist of the Warsaw ghetto, observed many Germans from September 1939
until March 1940 when he penned his evaluation derived from their actions
and words.

“The gigantic catastrophe which has descended on Polish Jewry has no parallel,
even in the darkest period of Jewish history. First, in the depth of hatred. This
is not just hatred whose source is in a party platform, and which was invented
for political purposes. It is a hatred of emotion, whose source is some psycho-
pathic malady. In its outward manifestations it functions as physiological
hatred, which imagines the object of hatred to be unclean in body, a leper who

has no place within the camp. The (German) masses have absorbed this sort
of qualitative hatred

Significantly, this characterisation ... (is) based on the words and acts of Ger-
mans - of SS men, policemen, soldiers, administrators, and those working in
the economy - before the formal genocidal program of systematic killing had
begun. It is the masses, the ordinary Germans, not the Nazi ideologues and
theoreticians, whom Kaplan exposes. The causal link between the Germans’
beliefs and actions is palpable ...

These were the beliefs that prepared officers of Police Regiment 25 to boast,
like so many other Germans engaged in the slaughter, and to believe themselves
“to have accomplished feats of heroism by these killings”. These were the
beliefs that led so many ordinary Germans to kill for pleasure ... These same
beliefs moved the men of Police Battalion bl’s First Company, who guarded
the Warsaw ghetto and eagerly shot Jews attempting to sneak in or out of the
ghetto during 1941-1942, to create a recreational shrine to their slaughter of
Jews. These German reservists turned a room in their quarters into a bar ...
After successful kills, these Germans were in the habit of rewarding themselves
by holding special “victory celebrations” (Siegesfeiern).

The beliefs about Jews that governed the German people’s assent and contri-
butions to the elimlnationlst program of the 1930s were the beliefs that pre-
pared the men of Police Battalion 101 and so many other Germans to be eager
killers who volunteered again and again for their “Jew-hunts”.

These were the beliefs that led so many ordinary Germans who degraded,
brutalized, and tortured Jews m camps and elsewhere — the cruelty in the camps
having been near universal - to choose to do so. They did not choose (like the
tiny minority who showed that restraint was possible) not to hit, or, if under
supervision, to hit in a manner that would do the least damage, but instead
regularly chose to terrorize, to inflict pain, and to maim. These were the beliefs
that prepared the men of Police Battalion 309, ordinary Germans, not to hate,
but to esteem the captain who had led them in their orgy of killing and synago-
gue-burning in Blalystok in a manner similar to the glowing evaluations of
“Papa” Trapp given by the men of Police Battalion 101 ...

These were the beliefs that led Germans to take joy, make merry, and celebrate
their genocide of the Jews, such as with the party (Abschlufifeier) thrown

upon the closing down of the Chelmno extermination camp in April 1943 to
reward its German staff for a job well done. By then, the Germans had killed
over 145 000 Jews in Chelmno. The German perpetrators’ rejoicing proudly
in their mass annihilation of the Jews occurred also at the conclusion of the
more concentrated slaughter of twelve thousand Jews on the “Blood Sunday”
of October 12,1941, in Stanlslawow, where the Germans there threw a victory
celebration. Yet another such celebration '^as organized in August 1941, du-
ring the heady days in the midst of the Germans’ campaign of extermination
of Latvian Jewry. On the occasion of their slaughter of the Jews of Cesls, the
local German security police and members of the German military assembled
to eat and drink at what they dubbed a death banquet (Totenmahl) for the
Jews. During their festivities, the celebrants drank repeated toasts to the ex-
termination of the Jews.

The beliefs that were already the common property of the German people
upon Hitler’s assumption of power and which led the German people to assent
and contribute to the ellmlnatlonist measures of the 1930s were the beliefs that
prepared not just the Germans who by circumstances, chance, or choice ended
up as perpetrators but also the vast majority of the German people to under-
stand, assent to, and, when possible, do their part to further the extermination,
root and branch, of the Jewish people. The Inescapable truth is that, regarding
Jews, German political culture had evolved to the point where an enormous
number of ordinary, representative Germans became - and most of the rest of
their fellow Germans were fit to be - Hitler’s willing executioners.

TT: Excerpt from Daniel Jonah Goldhagen Hitlers willige Vollstrecker. Ganz
gewohnliche Deutsche und der Holocaust. Berlin; Sledler Verlag, 1996.
Aus dem Amerlkanischen von Klaus Kochmann. S. 520-531

Diese Auspriigung des Judenhasses llefi also vlele Handlungsmogllchkelten
offen. Aber sie tendlerte, gerade vor dem Hlntergrund der fiir das zwanzlgste
Jahrhundert typlschen Vorstellungen von den Juden, zur extremsten Varlante,
zur deflnitiven “Losung”. DaB zwischen rasslstlschen Antlsemlten, die fiir
elne gewaltsame Aussonderung der Juden pliidlerten, und solchen, die die “Lo-
sung” in der Vernlchtung sahen, eine Wahlverwandtschaft bestand, wurde be-
relts Ende des neurizehnten Jahrhunderts deutllch. Es glbt eine Untersuchung
fiber promlnente antisemltlsche Autoren, die “Losungen” fiir die “Judenfra-
ge” vorschlugen; zwei Drlttel von Ihnen pladlerten bereits Im letzten Jahrhun-
dert fiir die vollige Vernlchtung der Juden. Die Uberzeugungen, die die Na-
tionalsozialisten dazu brachten, den judischen Biirgern zunachst den

gesellschaftlichen Einfufi zu nehmen und sie dann ganz aus der Gesellschaft
auszuschlie^en, wurden von der deutschen Bevolkerung geteilt und fiihrten
zu entsprechendem Anklang der eliminatorischen Maftnahmen. Jeder Schritt
des Programms, von der yerbalen Gewalttatigkelt bis zur Ghettoisierung und
Vernichtung fand Zustimmung bei einer grofien Zahl von ganz gewohnlichen
Deutschen, losten jedenfalls weder Unruhe noch Dissens in der Bevolkerung
aus. Die erschreckenden Diagnosen und Prognosen - fiir den Fall, daft es den
Deutschen nicht gelingen sollte, die vermeintliche jiidische Krankheit aus dem
deutschen Gesellschaftskorper zu entfernen - forderten die Unterstiitzung,
die die eliminatorischen Mafinahmen fanden, und man sah diese schliefilich
als vorlaufig und unzureichend an, was wiederum den Drang zur Vernichtung
bestarkte. Was nicht erstaunen kann, wenn man die Juden tatsachlich als Krankheit
am Volkskorper betrachtete. Eln Arzt, der eine Zeltlang in Auschwitz Dienst
tat, hat den aul^erordentllch engen Zusammenhang zwlschen Uberzeugung
und Handeln - zwlschen dem Antlsemitismus der Deutschen und ihrer Be-
reitschaft, die als “Erzfeinde Deutschlands” betrachteten Juden zu ermorden
- genau erfafit: Der Schritt von den ungeheuerlichen Anschuldlgungen gegen
die Juden bis zum Wunsch, sie zu vernlchten, sei nicht langer “als ein Millimeter”.
Der Volkermord gehorte zum gesellschaftlichen “Gesprach” der Deutschen.
Er war in Ihre Sprache und Gefiihle elngelassen und gehorte zum kognitlven
Modell, war Bestandteil der Praktlken und Handlungen, mit denen in den
dreifiiger Jahren das spatere Programm vorbereitet wurde. Nachdem die Be-
dingungen relf, die autonome Kraft des eliminatorischen Antlsemitismus eln-
mal freigesetzt war, erwies slch dieser als so stark, daft selbst diejenlgen Deut-
schen, die offiziell nlchts mit der Verfolgung und Vernichtung der Juden zu
tun hatten, Juden wie selbstverstandlich korperlich angrlffen, von verbalen
Attacken ganz zu schwelgen.
Der Antlsemitismus war wahrend der NS-Zeit so allgemeln verbreltet, dafi
die jiidischen Opfer den Elndruck hatten, sie konnten die Macht, die er fiber
die Deutschen gewonnen hatte, nur in elnem gleichsam medlzlnlschen Bild
erfassen; “Das Blut der Nationalsozlallsten ist von einem krankhaften Hafi
verglftet”. Als der tiefe, in den dreifiiger Jahren aufgrund der aufieren Um-
stande schlummernde Hafi elnmal aktlviert war, hat er die Deutschen so ge-
packt, daI5 sie ihn aus jeder Pore auszuschwltzen schlenen. Kaplan, der lelden-
schaftllche Chronist, konnte von September 1939 bis Marz 1940 vlele
Deutsche bei ihrem Tun und Reden beobachten.
“Die glgantische Katastrophe, die fiber die polnlsche Judenhelt gekommen
war, hat keine Parallele, nicht elnmal in den dunkelsten Perioden der jfidlschen

Geschichte. Das betrifft zunachst die Tiefe des Hasses. Das ist nicht nur ein
Haft, dessen Quelle ein Parteiprogramm ist und der zu politischen Zwecken
erfunden wurde. Es ist ein Hafi des Gefiihls, dessen Quelle eine seelische Er-
krankung 1st. Nach seinen aufterlichen Anzeichen verlauft er als physiologi-
scher Hal^, der slch das Hafiobjekt als korperlich unrein vorstellt, als elnen
Aussatzigen, der aufierhalb des Lagers wohnen muft. Die Massen haben diese
Art qualltativen Hasses in slch absorbiert

Wohlgemerkt, diese Charakterlslerung stiitzt slch ... auf Taten und Worte von
SS-Mannern, Pollzlsten, Soldaten, Verwaltungsbedlensteten und Wirtschafts-
leuten, bevor das offizlelle, systematische Morden begann. Kaplan beobachtet
ganz gewohnllche Deutsche und kelne NS-ldeologen. Der kausale Zusam-
menhang zwlschen den Auffassungen und den Handlungen der Deutschen 1st

Es waren Immer die glelchen Vorstellungen und Bllder von den Juden. Sle
brachten die Offizlere des Polizelregiments 25 dazu, slch wle so vlele Beteillgte
zu riihmen, “durch Erschlel^ungen ... Heldentaten vollbracht zu haben”. Sie
liefien vlele ganz gewohnllche Deutsche an Ihren Mordeinsatzen Gefallen fin-
den ... Die gleichen Vorstellungen und Beweggrtinde liefien die Manner des
Polizeibatalllons 61, die das Warschauer Ghetto bewachten, pfllchtbesessen
auf Juden schiefien, die sich in den Jahren 1941/42 als Kurlere aus dem Ghetto
hlnaus- und wleder zuriickschllchen. Diese Polizelreservlsten richteten slch
eine Art Heldengedenkstatte ein, in der sle slch von Ihrer morderlschen Arbeit
erholen konnten: sle verwandelten elnen Raum in ihrem Quartier in eine Bar
... Nach erfolgreichen Mordeinsatzen belohnten sie sich gern selbst mit be-
sonderen “Slegesfelern”.

Es waren immer die gleichen Vorstellungen und Bllder von den Juden, die
Deutsche dazu bewogen, dem elimlnatorlschen Programm der drelfilger Jahre
zuzustlmmen, sich daran zu betelllgen und spater auch wle die Manner des
Polizeibatalllons 101 am Vernlchtungsprogramm tellzuhaben; jene wurden
wle viele andere zu eifrlgen Mordern, die slch immer wleder freiwillig zu “Ju-
denjagden” meldeten.

Es waren immer die .gleichen Vorstellungen und Bllder von den Juden, aus
denen heraus slch vlele gewohnllche Deutsche dazu entschieden, in Lagern
und anderswo Juden zu ernledrigen, brutal zu behandeln und zu quiilen. Sle
entschieden slch nicht - wle eine klelne Mlnderheit, die Zuriickhaltung iibte -,
nur dann zu schlagen, wenn sie unter Aufslcht standen, nur so zu schlagen.

daft sie moglichst wenig Schaden annchteten, sondern sie entschlossen slch,
ihre Opfer regelmafiig zu terrorisieren, ihnen Schmerz zuzufiigen und sie zu
Kriippeln zu machen. Die immer gleichen Vorstellungen und Bilder bewogen
auch die ganz gewohnlichen Manner des Polizeibataillons 307, den Haupt-
mann nicht etwa zu hassen, der sie in diese Orgie aus Mord und Brandschat-
zung in Bialystok gefiihrt hatte, nein, sie schatzten ihn ahnlich wie die Manner
des Polizeibataillons 101 ihren “Papa” Trapp.

Es waren immer die gleichen Vorstellungen und Bilder von den Juden. Sie
sorgten dafiir, daft die Deutschen an ihren Morden Genugtuung fanden, daft
die Tater ihre Taten frohllch felerten; so etwa mlt jener “Abschluf^feier” an-
lafilich der Schliefiung des Vernlchtungslagers Chelmno im April 1943, mit
der die deutsche Mannschaft dafiir belohnt wurde, dafi sie ihre Arbeit so ta-
dellos verrichtet hatte. Bis April 1943 hatten die Deutschen in Chelmno mehr
als 145 000 Juden umgebracht. Auch nach der Mordorgie, bei der am 12. Ok-
tober 1941, dem “Blutsonntag” von Stanislawow, 12 000 Juden starben, ver-
anstalteten die Tater voller Stolz eine Slegesfeier. Ein weiteres Fest dleser Art
wurde im August 1941, als der Vernichtungsfeldzug gegen die lettlschen Juden
in vollem Gang war, organlsiert. Nach der Ermordung der Juden von Cesis
versammelten sich Angehorige der Sicherheitspolizel und der Wehrmacht zu
einem, wie sie es nannten, “Totenmahl fiir die Juden” Die Feiernden erhoben
mehrfach ihre Glaser, um auf die Vernlchtung der Juden zu trinken.

Es waren also die immer gleichen Vorstellungen und Bilder von den Juden,
die berelts zum Zeitpunkt der Machtiibernahme Hitlers den Deutschen eigen
waren und diese dazu brachten, den antisemitlschen MaEnahmen der drelfilger
Jahre zuzustlmmen und sie zu unterstiitzen. Mehr noch: Sie berelteten nicht
nur all jene, die durch die Umstande. durch Zufall oder in freier Entscheidung
zu Tatern wurden, auf ihre Aufgabe vor, sondern sie veranlafiten auch die
grofte Mehrhelt der Deutschen, die totale Vernlchtung des judischen Volkes
zu verstehen, Ihr belzupflichten und sie nach Moglichkelt zu fordern. Man
mufi den Tatsachen ins Auge sehen: Die deutsche Politlk und Kultur hatte slch
bis zu einem Punkt entwickelt, an dem die melsten Deutschen hatten werden
konnen, was eine ungeheure Zahl ganz gewohnlicher Deutscher tatsiichllch
wurde: Hitlers willige Vollstrecker.


Agar, M. 1992 “Review of Werner Holly 1990. Politikersprache”. Language in Society

21, 158-160.
Agar, M. 1996 “The Slippery Surface: Shifting between Language and Culture”. Plenary
talk given at the LAUD Symposium on “The Cultural Context in Communication
Across Languages”, Duisburg, March 1996.
Albrecht,]. 1987 “ Wissenschaftstheoretischer Status und praktischer Nutzen der Uber-
setzungswissenschaft” in R. Ehnert and W. Schleyer eds. Ubersetzen im Fremd-
sprachenunterricht. Regensburg: AKDAF, 9-23.
Albrecht,]. 1990 “Invarianz, Aquivalenz, Adaquatheit” in R. Arntz and G. Thome eds.
Ubersetzungswissenschaft. Ergebnisse und Perspektiven. Festschrift fur Wolfram
Wilss. Tubingen: Narr, 71-81.
Austin, ].L. 1962 Flow to do Things with Words. London: Oxford University Press.
Baker, M. 1992 In Other Words. A coursebook on translation. London: Routledge.
Bateson, G. 1972 Steps to an Ecology of Mind. New York: Ballantine Books.
Bell, R. 1991 Translation and Translating: theory and practice. London: Longman.
Benjamin, A. 1989 Translation and the Nature of Philosophy. A New Theory of Words.
London: Routledge.
Benjamin, W. 1972 “Die Aufgabe des Ubersetzers” in W. Benjamin Gesammelte Schrif-
ten Bd. IV/1 WA. Frankfurt/Main: Suhrkamp. Vorwort des Ubersetzers (WBen-
jamin) zu der Ubersetzung Charles Baudelaire Tableaux Parisien, 9-21.
Besnier, N. 1988 “The Linguistic Relationships of Spoken and Written Nukulaelae Re-
gisters”. Language 64:4, 707-736.
Bhatia, V. 1993 Analysing Genre. Language Use in Professsional Settings. London:
Biber, D. 1988 Variation across Speech and Writing. Cambridge: Cambridge University
Biber, D. 1994 “An Analytic Framework for Register Studies” in D. Biber and E. Fi-
negan eds., 31-58.
Biber, D. and E. Finegan 1994 eds. Sociolinguistic Perspectives on Register. Oxford:
Oxford University Press.
Bielick, B. 1991 “Aufforderungen in englischen und deutschen Dialogen”. Paper given
at the 22. Annual GAL Congress, Mainz 1991.
Blum-Kulka, S. and ]. House 1989 “Cross-Cultural and Situational Variation in Reque-
sting Behavior” in Blum-Kulka, House, Kasper eds. 1989, 123-154.
Blum-Kulka, S.,]. House and G. Kasper eds. 1989 Cross-Cultural Pragmatics: requests
and apologies. Norwood, N.].: Ablex.
Broeck, R. van den 1985 “Second Thoughts on Translation Criticism. A Model of its
Analytic Function” in T Hermans ed. The Manipulation of Literature. London:
Croom Helm, 54-62.

Broeck, R. van den 1986 Contrastive Discourse Analysis as aTool for the Interpretation
of Shifts in Translated Texts” in House and Blum-Kulka ed. 37-47.
Brotherton, A. 1981 Review of Translation Quality Assessment by Juliane House”
van Taalto Taal 25:\, 16-19.
Brown, G. and G. Yule 1983 Discourse Analysis. Cambridge: Cambridge University Press.
Biihler, K. 1965 Sprachtheorie. 2nd.ed. Stuttgart: Fischer.
Busch-Lauer, I. 1995 Kontrastive Fachtextanalyse am Beispiel englischer und deut-
scher Zeitschriftenartikel der Medizin” in B. Spillner ed. Fachkommunikation.
Frankfurt/Main: Lang, 71-73.
Byrnes, H. 1986 Interactional style in German and American Conversation”. Text 6:2,
Carroll, J.B. 1966 “An Experiment in Evaluating the Quality of Translation”. Mecha-
nical Translation 9, 55-66.
Cary, E. and W. Jumpelt eds. 1963 equality in Translation. New York: Macmillan.
Catford, J.C. 1965 A Linguistic Theory of Translation. London: Oxford University
Chafe, W.L. 1982 “Integration and Involvement In Speaking” in D. Tannen ed. Spoken
and Written Language: Exploring Orality and Literacy. Norwood, N.J.: Ablex,
Clyne, M. 1981 “Culture and Discourse Structure”./o«rn^/ of Pragmatics 5, 61-66.
Clyne, M. 1987 “Cultural Differences in the Organization of Academic Texts: English
and German”.of Pragmatics 11, 211-247.
Clyne, M. 1994 Intercultural Communication at Work. Cultural Values in Discourse.
Cambridge: Cambridge Univeraity Press.
Connor, U. and R. Kaplan 1987 eds. Writing Across Languages: Analysis of L2 Texts.
Reading, Mass.: Addlson-Wesley.
Couture, B. 1986 ed. Functional Approaches to Writing: Research Perspectives. Nor-
wood, N.J.: Ablex.
Crisp, S. 1981 “Review of Juliane House A Model for Translation Quality Assessment”.
The Bible Translator 32:3, 346-348.
Crystal, D. and D. Davy 1969 Investigating English Style. London: Longman.
de Man, P. 1986 The Resistance to Theory. Minneapolis: University of Minnesota Press.
Derrida, J. 1985 “Roundtable on Translation” in C.V. MacDonald ed. The Ear of the
Other: Autobiography, Transference, Translation. New York: Schocken.
Derrida, J. 1992 “From Des Tours de Babel” in R. Schulte and J. Biguenet eds. Theories
of Translation. Chicago: University of Chicago Press, 218-227.
Diller, H.-J. and J. Kornellus 1978 Linguistische Probleme der Ubersetzung. Tubingen:
Duszak, A. 1994 “Academic Discourse and Intellectual Styles”. Journal of Pragmatics
Edmondson, W.J. 1981 Spoken Discourse. A Model for Analysis. London: Longman.
Edmondson, W.J. and J. House 1981 LePs Talk and Talk About It. A Pedagogic Inter-
actional Grammar of English. Miinchen: Urban und Schwarzenberg.
Edmondson, W.J. and J. House 1982 “Hoflichkelt als Lernziel im Engllschunterricht”
Neusprachliche Mitteilungen 4, 218-227.

Edmondson, W.J. and J. House 1993 Einfiihrung in die Sprachlehrforschung. Tubingen:
Edmondson, WJ., J. House, G. Kasper, B. Stemmer 1984 “Learning the Pragmatics of
Discourse. A Project Report”. Applied Linguistics 5, 113-127.
Enkvist, N.E. 1973 Linguistic Stylistics. The Hague: Mouton.
Faifi, K. 1981 “Review of Juliane House A Model for Translation Quality Assessment”.
IRAL 19, 79-80.
Ferguson, C.A. 1994 “Dialect, Register, and Gen'Fe: Working Assumptions About Con-
ventionalization” in D. Biber and E. Finegan eds., 15-30.
Fetzer, A. 1996 “Preference Organization und Sprechfertigkeit im Englisch-Deutschen
Kontext. Vom propositionalen Schlagabtausch zur interkulturellen Kompetenz”.
GAL Bulletin 24, 63-80.
Fiedler, S. 1994 “Linguistische Fachtexte im Englischen, Deutschen und Esperanto (ein
interkultureller Vergleich)” in B. Spillner ed. Fachkommunikation. Frankfurt:
Lang, 68-71.
Firth, J.R. 1959 Papers in Linguistics 1934-1951. London: Oxford University Press.
Foucault, M. 1977 Language, Counter-Memory, Practice. Ed. and transl. D.F. Bouchard
and S. Simon Ithaca: Cornell University Press.
Galtung, J. 1985 “Struktur, Kultur und intellektueller Stil. Ein vergleichender Essay
liber sachsonische, teutonische, gallische und nipponische Wissenschaft” in A.
Wierlacher ed. Das Eigene und das Fremde. Prolegomena zu einer interkulturellen
Germanistik. Miinchen: indicium, 151-193.
Gentzler, E. 1993 Contemporary Translation Theories. London: Routledge.
Gerzymisch-Arbogast, H. 1994 Ubersetzungswissenschaftliches Propadeutikum. Tu-
bingen: Francke.
Gleason, H.A. 1968 “Contrastive Analysis in Discourse Structure” 'm Monograph Series
on Languages and Linguistics, J.E. Alatis ed. no. 21. Washington, D.C.: George-
town University Press, 39-63.
Gnutzmann, C. and H. Oldenburg 1991 “Contrastive Text Linguistics in LSP-Research:
Theoretical Considerations and some Preliminary Findings” in H. Schroder ed.
Subject-oriented Texts. Berlin: de Gruyter, 103-136.
Goffman, E. 1974 Frame Analysis. New York: Harper and Row.
Gopferich, S. 1995 “Textsorten und Translationsprobleme: Ein Kanon fiir Uberset-
zungsiibungen zur Fachsprache der Technik” in W. Borner and K. Vogel eds. Der
Text im Fremdsprachen-unterricht. Bochum: AKS Verlag, 163-196.
Graham, J.F. ed. 1985 Difference in Translation. Ithaca: Cornell University Press.
Gregory, M. 1967 “Aspects of Varieties Differentiation”. Journal of Linguistics 3,
Gregory, M. and S. Carroll 1978 Language and Situation. Language Varieties and their
Social Contexts. London: Routledge.
Grice, H.P. 1975 “Logic and Conversation” in P. Cole and J.L. Morgan eds. Syntax and
Semantics vol 3. New York: Academic Press, 41-58.
Gutt, E.-A. 1991 Translation and Relevance. Cognition and Context. Oxford: Blackwell.
Hall, E.T. 1976 Beyond Culture. New York: Doubleday.
Hall, E.T. 1983 The Dance of Life. The Other Dimension of Time. New York: Double-

Hall, E.T and M.R. Hall 1983 Hidden Differences. Studies in International Communi-
cation. How to Communicate with the Germans. Hamburg: Gruner und Jahr.
Halliday, M.A.K. 1973 Explorations in the Functions of Language. London: Arnold.
Halliday, M.A.K. 1978 Language as Social Semiotic: the Social Interpretation of Lan-
guage and Meaning. London: Arnold.
Halliday, M.A.K. 1989 Spoken and Written Language. Oxford: Oxford University Press.
Halliday, M.A.K. and R. Hasan 1989 Language, Context and Text: Aspects of Language
in a Social-Semiotic Perspective. Oxford: Oxford University Press.
Halliday, M.A.K. and Martin 1993 Writing Science: Literacy and Discursive Power.
London: Palmer.
Halliday, M.A.K., A. McIntosh and P. Strevens 1964 The Linguistic Sciences and Lan-
guage Teaching. London: Longman.
Hasan, R. 1978 “Text in the Systemic-functional Model” in W.U. Dressier ed. Current
Trends in Text Linguistics. Berlin: de Gruyter.
Hatim, B. and 1. Mason 1990 Discourse and the Translator. London: Longman.
Havranek, B. 1964 “The Functional Differentiation of the Standard Language” in P.
Garvin ed. and transl. A Prague School Reader on Aesthetics, Literary Structure
and Style. Washington, D.C.: Georgetown University Press, 3-16.
Hinds, J. 1987 “Reader versus Writer Responsibility: A New Typology” in U. Connor
and R. Kaplan eds., 141-152.
Hofstede, G. 1984 Culture’s Consequences. New York: Sage.
Hofstede, G. 1991 Cultures and Organizations. London: McGraw-Hill.
Holz-Manttari, J. 1984 Translatorisches Handeln. Theorie und Methode. Helsinki: An-
nales Academicae Scientiarum Fennicae.
Honig, H. G. 1986 “Ubersetzen zwischen Reflex und Reflexion - ein Modell der iiber-
setzungsrelevanten Textanalyse” in M. Snell-Hornby ed. Ubersetzungswissen-
schaft. Fine Neuorientierung. Tubingen: Francke, 230-251.
Honig, H.G. 1995 Konstruktives Ubersetzen. Tubingen: Stauffenburg.
Honig, H.G. 1996 “Review of H. Gerzymisch-Arbogast Ubersetzungs- wissenschaft-
liches Propadeutikum” Target 8, 188-193.
Honig. H.G. and P. KuBmaul 1982 Strategie der Ubersetzung. Ein Lehr- und Arbeits-
buch. Tubingen: Narr.
House,]. 1977 (2nd. ed. 1981) A Model for Translation Quality Assessment. Tubingen:
House, J. 1979 “Interaktionsnormen in deutschen und englischen Alltagsdialogen”.
Linguistische Berichte 59, 76-90.
House, J. 1982a “Opening and Closing Phases in English and German Dialogues”.
Grazer Linguistische Studien 16, 52-83.
House, J. 1982b “Conversational Strategies in German and English Dialogues” in G.
Nickel and D. Nehls eds. Error Analysis, Contrastive Linguistics and Second Lan-
guage Learning. Heidelberg: Groos, 131-146.
House, J. 1982c “Gambits in deutschen und englischen Alltagsdialogen”. Grazer Lin-
guistische Studien 17/18, 110-132.
House,]. 1984 “Some Methodological Problems and Perspectives in Contrastive Dis-
course Analysis”. Applied Linguistics 5, 245-55.

House, J. 1985 “Contrastive Discourse Analysis and Universals in Language Usage”.
Papers and Studies in Contrastive Linguistics 20, 5-15.
House, J. 1989a “Politeness in English and German” in S.Blum-Kulka et al. eds., 96-122.
House, J. 1989b “Excuse me please: Apologizing in a Foreign Language” in B. Ket-
temann et al eds. Englisch als Zweitsprache. Tubingen: Narr, 303-328.
House, J. 1993 “Toward a Model for the Analysis of Inappropriate Responses in Na-
tive/Non-Native Interactions” in G. Kasper and S. Blum-Kulka eds. Interlanguage
Pragmatics. Oxford: Oxford University Pr'bss, 163-184.
House,]. 1994 “Text/Rhetoric” in The Encyclopedia of Language and Linguistics, Ox-
ford: Pergamon, 4581—4584.
House, J. 1996a “Contrastive Discourse Analysis and Misunderstanding: The Case of
German and English” in M. Hellinger and U. Ammon eds. Contrastive Sociolin-
guistics. Berlin: de Gruyter, 345-361.
House, J. 1996b “Developing Pragmatic Fluency in English as a Foreign Language”.
Studies in Second Language Acquisition 18, 225-252.
House, J. 1996c “The Role of Cross-Cultural Discourse Analysis in the Development
of Intercultual Competence”. Plenary Talk given at the LAUD Symposium on “The
Cultural Context in Communication Across Languages”. Duisburg, March, 1996.
House, J. 1997 “Politeness and Translation” in L. Hickey ed. The Pragmatics of Trans-
lation. Clevedon: Multilingual Matters.
House, J. forthcoming Cross-Cultural Pragmatics and Translation.
House,], and G. Kasper 1981 “Politeness Markers in English and German” in F. Coul-
mas ed. Conversational Routine. The Hague: Mouton, 157-185.
House, ]. and S. Blum-Kulka eds. 1986 Interlingual and Intercultural Communication.
Discourse and Cognition in Translation and Second Language Acquisition Studies.
Tubingen: Narr.
House, ]. and G. Kasper 1987 “Interlanguage Pragmatics: Requesting in a Foreign Lan-
guage” in W. Lorscher and R. Schulze eds. Perspectives on Language in Performan-
ce. Tubingen: Narr. 1250-1288.
Hymes, D. 1968 “The Ethnography of Speaking” in].A. Fishman ed. Readings in the
Sociology of Language. The Hague: Mouton, 99-138.
Hymes, D. 1974 Foundations in Sociolinguistics: an Ethnographic Approach. Philadel-
phia: University of Pennsylvania Press.
Inhoffen, N. 1991 “‘Skopos’ und ‘Kulturtransfer’. Zu zwei Schliisselbegriffen der neue-
ren iibersetzungswissenschaftlichen Diskussion” in C. Schmitt ed. Neue Methoden
der Sprachmittlung. Heidelberg: Egert, 21—47.
Ivir, V. 1996 “A Case for Linguistics in Translation Theory” Target 8, 149-157.
]ager; G. 1973 “Invarianz und Transferierbarkeit” in A. Neubert Neue Beitrdge zu
Grundfragen der Ubersetzungswissenschaft.
]acobs, C. 1975 “The Monstrosity of Translation”, Modern Language Notes 90,
]akobson, R. 1960 “Closing Statement: Linguistics and Poetics” inTA. Sebeoked. Style
in Language. Cambridge, Mass.; MIT Press, 350-377.
]akobson, R. 1966 “On Linguistic Aspects of Translation” in R.A. Brower ed. On
Translation. New York: Oxford University Press, 232-239.
]ohn5ton, ]. 1992 “Translation as Simulacrum” in L. Venuti ed., 42-56.


Joos, M. 1961 The Five Clocks. New York: Harcourt, Brace and World.
Kaplan, R.B. 1966 Cultural Thought Patterns in Inter-Cultural Education”. Language
Learning 16, 1-20.
Kaplan, R.B. 1987 “Culturaj^Thought Patterns Revisited” in U. Connor and R.B. Ka-
plan eds., 9-20.
Kaplan, R.B. 1996 “Is There a Problem in Reading and Writing Texts Across Languages?”
Working Paper. Duisburg: LAUD.
Kelletat, A.F. 1987 “Die Riickschritte der Ubersetzungstheorie” in R. Ehnert and W.
Schleyer eds. Uhersetzen im Fremdsprachenunterricht. Regensburg: AKADAF,
Klein, W. 1992 “Was kann sich die Ubersetzungswissenschaft von der Linguistik er-
warten?”. Zeitschrift fur Literaturwissenschaft und Linguistik 84, 104-123.
Kohlmayer, R. 1988 “Der Literaturiibersetzer zwischen Original und Markt. Fine Kri-
tik funktionalistischer Ubersetzungstheorien”. Lehende Sprachen 33, 145-156.
Koller, W. 1972 (4th ed. 1992) Grundprobleme der Ubersetzungstheorie. Bern: Francke.
Koller, W. 1987 “Zum Gegenstand der Ubersetzungswissenschaft” in R. Arntz and G.
Thome eds. Ubersetzungswissenschaft. Ergebnisse und Perspektiven. Tubingen:
Narr, 19-30.
Koller, W. 1995 “The Concept of Equivalence and the Object of Translation Studies”
Target 7, 191-222.
Konigs, F. 1981 “Ubersetzung und Fremdsprachenunterricht - vereinbar oder unver-
einbar?” in K.-R. Bausch and F.-R. Weller eds. Ubersetzen im Fremdsprachenun-
terricht. Frankfurt: Diesterweg, 203-216.
Kotthoff, FI. 1989 Pro und Kontra in der Fremdsprache. Frankfurt: Lang.
Kupsch-Losereit, S. 1988 “Die Ubersetzung als soziale Praxis. Ihre Abhangigkeit vom
Sinn-und Bedeutungshorizontdes Kez\p{et\ien”. Fremdsprachen Lehren und Ler-
nen 17,28-40.
Kupsch-Losereit, S. 1994 “Die Ubersetzung als Produkt hermeneutischer Verste-
hensprozesse” in M. Snell-Hornby et al eds. Translation Studies. An Interdiscipline.
Amsterdam: Benjamins, 45-54.
Kusch, M. and H. Schroder 1989 “Contrastive Discourse Analysis: The Case of Da-
vidson vs Habermas” in M. Kusch and H. Schroder eds. Text-Interpretation-Ar-
gumentation. Hamburg: Buske.
KuBmaul, P. 1986 “Ubersetzen als Entscheidungsprozeh. Die Rolle der Fehleranalyse
in der Ubersetzungsdidaktik” in M. Snell-Hornby ed. Ubersetzungswissenschaft.
Fine Neuorientierung. Tubingen: Francke, 206-230.
Kufimaul, P. 1995 Training the Translator. Amsterdam: Benjamins.
Leech, G. 1974 Semantics. Harmondsworth: Penguin.
Leech, G. 1983 Principles of Pragmatics. London: Longman.
Legrand, P. 1996 “European Legal Systems are not Converging”. International and
Comparative Law Quarterly 45, 52-81.
Levy, J. 1967 “Translation as a Decision Process” in To Honor Roman Jakobson on the
Occasion of his Seventieth Birthday, vol. 2. The Hague: Mouton. 1171—1182.
Levy, J. 1969 Die literarische Ubersetzung. Theorie einer Kunstgattung. Frankfurt/
Main: Athenaum.

Luchtenberg, S. 1994 "A Friendly Voice to Help You vs Working Through Your Manual:
Pragmatic Differences Between American and German Software Manuals”./o«r-
nal of Pragmatics 21, 315-319.
Lux, F. 1981 Text, Situation, Textsorte. Probleme der Textsortenanalyse, dargestellt am
Beispiel der britischen Registerlinguistik. Tubingen: Narr.
Lyons,]. 1969 Introduction to Theoretical Linguistics. Cambridge: Cambridge Univer-
sity Press.
MacNamara, J. 1967 "The Bilingual’s Linguistic'Terformance: A Psychological Over-
view”. The Journal of Social Issues 23, 58- 77.
Malinowski, B. 1923 “The Problem of Meaning in Primitive Languages” in C.K. Ogden
and LA. Richards The Meaning of Meaning. New York: Harcour, Brace and World,
Martin, J.R. 1989 Factual Writing. Oxford: Oxford University Press.
Martin, J.R. 1993 “A Contextual Theory of Language” in B. Cope and M. Kalantzis
eds. The Powers of Literacy. A Genre Approach to Teaching Writing. London:
Falmer, 116-136.
Martin, J.R. and J. Rotherby 1986 “What a Functional Approach to the Writing Task
Can Show Teachers About ’’Good Writing"” in B. Couture ed., 241-265.
Mathesius, V. 1971 “Die funktionale Linguistik” in E. Benes and J. Vachek eds. Stilistik
und Soziolinguistik. Miinchen: List, 1-18.
Mauranen, A. 1993 Cultural Differences in Academic Rhetoric. Frankfurt: Lang.
Mudersbach, K. 1992 “Modell zum Kulturvergleich”. Vortrag vor dem Pfalzer Dozen-
ten-Kolloquium am 4.12.1992 in Germershelm.
Mukarovsky, J. 1964 “Standard Language and Poetic Language” in PL. Garvin ed. and
transl. A Prague School Reader on Aesthetics, Literary Structure and Style. Wash-
ington, D.C.: Georgetown University Press, 17-30.
Neubert, A. 1968 “Pragmatische Aspekte der Ubersetzung” in idem ed. Grundfragen
der Ubersetzungwissenschaft. Leipzig: VEB Verlag Enzyklopadle, 21-33.
Neubert, A. 1985 Text and Translation. Leipzig: Enzyklopadle.
Neubert, A. 1994 “Theory and Practice in Translation and in Translation Studies: the
Textual Approach” in A. Kruger ed. New Perspectives on Teaching Translators and
Interpreters in South Africa. Pretoria: University of South Africa, 10-26.
Newmark, P. 1979 “Review of J. House A Model for Translation Quality Assessment”.
The Incorporated Linguist 18, 61-62.
Newmark, P. 1981 Approaches to Translation. Oxford: Pergamon.
Newmark, P. 1988 A Textbook of Translation. New York: Prentice Hall.
Newmark, P. 1991 “The Curse of Dogma in Translation Studies” Lebende Sprachen 3,
Nida, E.A. 1964 Toward a Science of Translation. Leiden: Brill.
Nida, E.A. and C.R. Taber 1969 The Theory and Practice of Translation. Leiden: Brill.
Nord, C. 1988 Textanalyse und Ubersetzen. Heidelberg: Groos.
Ochs, E. ed. 1989 “The Pragmatics of Affect”. Text 9, 1-123.
Ogden, C.K. and LA. Richards 1946 8th ed. The Meaning of Meaning London: Rout-
ledge and Kegan Paul.
Olson, D. 1996 “Language and Literacy: What Writing Does to Language and Mind”.
Ajinual Review of Applied Linguistics 16, 3-13.

Paepcke, R 1986 “Textverstehen-Textubersetzen-Ubersetzungskritik” inSnell-Hornby
ed. Ubersetzungswissenschaft. Eine Neuorientierung. Tubingen: Francke, 106-132.
Pike, K. 1967 2nd. ed. Language in Relation to a Unified Theory of Human Behavior.
The Hague: Mouton. -
Popper, K. 1972 Objective Knowledge: An Evolutionary Approach. Oxford: The Cla-
rendon Press.
Popper, K. 1976 Unended Quest. Glasgow: Collins.
Purves, A. 1988 Writing Across Cultures. Issues in Contrastive Rhetoric. Newbury Park:
Pym, A. 1992 “The Relations Between Translation and Material Text Transfer” Target
4, 171-189.
Rabassa, G, 1989 “No Two Snowflakes are Alike: Translation as Metaphor” in J. Bigu-
enet and R. Schulte eds. The Craft of Translation. Chicago: University of Chicago
Press, 1-12.
Reifi, K. 1971 Mdglichkeiten und Grenzen der Ubersetzungskritik. Miinchen: Hueber.
ReiB, K. 1989 “Ubersetzungstheorie und Praxis der Ubersetzungs-kritik” in R Konigs
ed. Ubersetzungswissenschaft und Fremdsprachenunterricht. Munchen: Goethe-
ReiB, K. and H.J. Vermeer 1984 Grundlegung einer allgemeinen Translationstheorie.
Tubingen: Niemeyer.
Savory, T. 1968 The Art of Translation. Boston: The Writer.
Schmitt, P.A. 1995 “Warnhinweise in deutschen und englischen Anleitungen: ein inter-
kultureller Vergleich” in Fremdsprachen Lehren und Lemen 24, 197-222.
Schreiber, M. 1993 Ubersetzung und Bearbeitung. Zur Differenzierung und Abgren-
zung des Ubersetzungsbegriffs. Tubingen: Narr.
Searle, J.R. 1969 Speech Acts. An Essay in the Philosophy of Language. Cambridge:
Cambridge University Press.
Slote, D. 1978 “Review of J. House A Model for Translation Quality Assessment”. Meta
23, 174-175.
Smith, V. and C. Klein-Braley 1997 “Advertising-a Five-Stage Strategy for Translation”
in M. Snell-Hornby, Z. Jettmarova, K. Kaindl eds. Translation as Intercultural
Communication. Selected Papers from the EST Congress, Prague 199^ Amster-
dam: Benjamins.
Snell-Hornby, M. 1986 “Ubersetzen, Sprache, Kultur” in M. Snell-Hornby ed. Uber-
setzungswissenschaft. Eine Neuorientierung. Tubingen: Francke, 9-29.
Soli, L. 1974 Gesprochenes und Geschriebenes Franzdsisch. Berlin: Schmidt.
Stalnaker, R.C. 1973 “Pragmatics” in D. Davidson and G. Harman eds. Semantics of
Natural Languages. Dordrecht: Reidel, 380-397.
Steiner, E. 1995 “An Extended Register Analysis as a Form of Text Analysis for Trans-
lation”. To appear in Festschrift fiir A. Neubert. Tubingen: Narr.
Stolze, R. 1992 Hermeneutisches Ubersetzen. Tubingen: Narr.
Swales, J.M. 1990 Genre Analysis. English in Academic and Research Settings. Cam-
bridge: Cambridge University Press.
Taavitsainen, I. 1994 “Subjectivity as a Text-Type Marker in Historical Stylistics” Lan-
guage and Literature 3, 197-212.

Thomas J. 1995 Meaning in Interaction. An Introduction to Pragmatics. London: Long-
Toury, G. 1985 In Search of a Theory of Translation. Tel Aviv: Tel Aviv University.
Toury, G. 1995 Descriptive Translation Studies and Beyond. Amsterdam: Benjamins.
Ventola, E. 1984 “Orientation to Social Semiotics in Foreign Language Teaching”.
Applied Linguistics 5, 275-286.
Ventola, E. 1995a “Englisch als lingua franca der schriftlichen Wissenschaftskommu-
nikation in Finnland und in Deutschland” Jh H. Kretzenbacher and H. Weinrich
eds. Linguistik der Wissenschaftssprache. Berlin: de Gruyter, 353-386.
Ventola, E. 1995b “What’s in an Academic Text” in B. Warvik et al. eds. Organization
in Discourse. Proceedings from the Turku Conference. Anglicana Turkuensia 14,
Venuti, L. ed. 1992 Rethinking Translation. Discourse, Sujectivity, Ideology. London:
New York.
Venuti, L. 1995 The Translator's Invisibility. A History of Translation. London: Routledge.
Vermeer, H.J. 1986 “Ubersetzen als kultureller Transfer” in M. Snell-Hornby ed. Uber-
setzungswissenschaft. Line Neuorientierung. Tubingen: Francke, 30-53.
Vermeer, H.J. 1987 “Literarische Ubersetzung als Versuch interkultureller Kommuni-
kation” in A. Wierlacher ed. Perspektiven und Verfahren interkultureller Germa-
nistik Miinchen: Gesellschaft fiir interkulturelle Germanistik, 541-549.
Vermeer, H.J. 1994 “Translation today: Old and new problems” in M. Snell-Hornby et
al eds. Translation Studies. An Interdiscipline. Amsterdam: Benjamins, 3-16.
Waard, J. de and E. Nida 1986 From One Language to Another. Functional Equivalence
in Bible Translating. Nashville: Nelson.
Wiesenthal, L. 1973 Zur Wissenschaftstheorie Walter Benjamins. Frankfurt: Athenaum.
Widdowson, H. 1973 “Directions in the Teaching of Discourse” in S.P. Corder and E.
Roulet eds. Theoretical Linguistic Models in Applied Linguistics. Paris: Didier, 65-76.
Widdowson, H. 1974 “Stylistics” in J. P. B. Allen and S. P. Corder eds. Techniques in
Applied Linguistics. London: Oxford University Press.
Wilss, W. 1974 “Probleme und Perspektiven der Ubersetzungskritik”. IRAL 12, 23-41.
Wilss, W. 1977 Ubersetzungswissenschaft. Probleme und Methoden. Stuttgart: Klett.
Wilss, W. 1980 Rezenslon von J. House A Model for Translation Quality Assessment
for K.-R. Bausch and F.-R. Weller eds. Ubersetzen und Fremdsprachenunter-
n'c/7f. Frankfurt: Diesterweg.
Wilss, W. 1995 “Die Rolle des Ubersetzers im Ubersetzungsprozel^” To appear in Fest-
schrift fur A. Neubert. Tubingen: Narr.
Wilss, W. 1996 Knowledge and Skills in Translator Behavior. Amsterdam: Benjamins.


Author Index

Agar,M. VIII (Fn), 92 Edmondson, W.]. 29, 30, 44, 80, 112, 162
Albrecht, J. 25 Enkvist, N.E. 44
Austin,]. 89
Austin, J.L. 31 FaiB, K. 104
Ferguson, C.A. 107
Baker, M. 19, 20 Fetzer, A. 86
Barth, K. 66, 67, 69 Feynman, R.P. 132, 179
Bateson, G. Ill Fiedler, S. 94
Bell, R. 19 Firth, ].R. 37
Benjamin, W, 9, 15, 121, 145 Foucault, M. 10
Besnier, N. 110
Biber, D. 108, 109, 110, 150 Gaining,]. 90,91
Bielick, B. 86 Gentzler, E. 9, 10, 15
Blum-Kulka, S. 79, 80, 82 Gerzymisch-Arbogast, H. 22, 23, 24
Broeck, R. van den 7 Gleason, H.A. 31
Brotherton, A. 102, 103 Gnutzmann, C. 91
Brown, G. 85 Coffman, E. Ill
Biihler, K. 17,33,35 Goldhagen, D.]. 121, 147, 155, 156, 157
Busch-Lauer, I. 94 Gopferich, S. 94
Byrnes, H. 86 Graham, ].F. 9
Gregory, M. 38, 40, 105, 110
Carroll, J.B. 5 Grice, H.P 85
Carroll, S. 38, 105 Gutt, E.-A. 19,20,21,22,112,118
Cary, E, 1
Catford, J.C 16, 25, 26, 31, 38, 68, 69, 70 Habermas,]. 91
Chafe, W.L. 109 Hall, E.T. 91,92
Churchill, W, 66, 67, 69 Hall,M.R. 91,92
Clyne, M. 89,90,91 Halliday, M.A.K. 20,29,32,34,35,36,37,
Connor, U. 89 38,47, 85, 105, 107, 109, 118
Couture, B, 106 Hasan, R. 32,34, 105
Crisp, S. 102 Hatim, B. 19, 20
Crystal, D. 38,39,40,41,42,44,105,107, Havranek, B. 43
108,109, no Hebei, ].P. 66,67, 104
Heidegger, M. 10
Davidson, D. 91 Hinds,]. 89
Davy, D. 38, 39, 40, 41, 42, 44, 105, 107, Hofstede, G. 91
108, 109, 110 Holly, W. 93
deMan, P. 9, 143, 145 Holz-Miinttari, ]. 12
Derrida,]. 9, 10 Honig, H.G. 12, 13, 14, 16,23
Diller, H.-]. 111 House, ]. 29, 48, 79, 80, 81, 82, 83, 84, 89,
Duszak, A. 94 92, 103, 104, 121, 131, 161, 162

Hymes, D. 34, 37, 107 Ochs, E. 109
Ogden, C.K. 32,33
Inhoffen, N. 13 Oldenburg, H. 91
Ivir, V. 24 Olson, D. 15

Jacobs, C. 143 Paepcke, F. 2

Jager, G. 18 Pike, K. 45
Jakobson, R. 33, 34, 48, 103 Plat<^ 33
Johnston,]. 143 Politzer, R. 103
Joos, M. 41, 109 Popper, K. 3, 15, 34, 35
Jumpelt, W. 1 Purves, A. 89
Pym, A. Ill, 114
Kade, O. 18
Kaplan, R.B. 89, 90 Rabassa, G. 11
Kasper, G. 79, 80, 82 Reifi, K. 11, 12, 13, 15, 16, 17, 19, 35, 66,
Kelletat, A.E 15 104
Klein, W. 16,21,22 Richards, LA. 32, 33
Klein-Baley, C. 97, 98 Rotherby, ]. 106
Kohlmayer, R. 13
Roller, W. 16,17,18,19,25 Savory, T. 2
Roller, W. 7 Schmitt, P.A. 93
Konigs, F. 102 Schreiber, M. 22, 24, 111
Kornelius,J. Ill Schroder, H. 91
Kotthoff, H, 86 Searle, J.R. 31
Kupsch-Losereit, S. 2, 3, 16 Slote, D. 102, 103
Kusch, M. 91 Smith, V. 98
Kufimaul, P. 12, 13, 16 Snell-Hornby, M. 16, 26
Soil, L. 44, 45
Leech, G. 30, 35 Stalnaker, R.C. 30
Levy,]. 26 Steiner, E. 22, 24
Luchtenberg, S. 93 Stolze, R. 2
Lyons, J. 36 Strevens, P. 47

MacNamara, J. 5 Taavitsainen, 1. 109

Malinowski, B. 32, 37 Taber, C.R. 4, 48
Martin, J.R. 105,106,107,108 Thomas,]. 30
Mason,!. 19,20 Toury, G. 6, 7, 8
Mathesius, V. 44
Mauranen, A. 94 Ventola, E. 93, 106
McIntosh, A. 47 Venuti, L. 9, 10, 11, 13, 114
Mukarovsky, J. 48 Vermeer, H.J. 11, 12,13,14, 15, 16,22,35,
Murphy,]. 122,178 104

Neubert, A. 18,19 Waard,].de 4

Newmark, P. 14, 19, 101,'104, 111 Wellek, R. 103
Nida, E.A. 4, 48 Widdowson, H. 30, 44, 67
Nord, Ch. 111 Wilss, W. 13, 17, 18, 19, 103, 118

O’Casey, S. 66, 67, 68 Yule, G. 85


Subject Index

ad-hoc formulation 84 cultural filtering 161,162

- action related 11 descriptive translation studies 6, 7
- anecdotal 1 dialect 38
- behavioural 2 dimension 38
- biographical 1 dimensions of cross-cultural difference 84
- deconstructionist 9 dimensions of language use 39
- functionalistic 11, 15 dimensions of language user 39
- linguistically oriented 16 directness 84, 88, 115
- literature-oriented 6 discourse phases 80
- neo-hermeneutic 1, 2 discourse strategies 81
- post-modernist 9 discourse world 112
- reception theory related 11
- relevance-theoretic 20 equivalence 12, 18, 24, 25, 30, 163
- response-oriented 2 - dynamic 4
- text-based 6 - functional 112,164
aspect of meaning explicitness 84, 87, 115
- pragmatic 30
- semantic 30 field 108
- textual 31 frame 111
clausal linkage 44 - conative 33
components - emotive-expressive 33
- functional 42 - ideational 35
- ideational 42 - interpersonal 34
- interpersonal 42 - representational 33
content focus 86,115 - textual 34
context of situation 37, 38 functions of language 32
contrastive discourse analyses 79 functions of text 32
contrastive pragmatic discourse re-
search 79 gambit 82
contrastive pragmatic studies 88 genre 105, 106, 159, 160
covert 29, 161, 163, 166
covert translation 69, 70,104,111,114,115 Iconic linkage 45
covert version 73,117 implicitness 84
covert-overt dine 78 indirectness 84
covert-overt continuum 77,164 individuality 38
covertly erroneous errors 45 interpersonal focus 88
cultural filter 29, 71, 74, 115, 117 Invariance 25

invisibility of translation 10 primary level function 115
province 39, 42
justification of method 46
quality in translation, judgment of trans-
language 159 lation 1, 119
- casual 41 reader - vs writer responsibility 89
- consultative 41 ref^nce
- formal 41 - explicit 109
- frozen 41 - situation-dependent 109
- intellectual 41 register 106, 159
- intimate 41 register analysis 105,107
routine reliance 115
medium 38
- complex 38, 40 second level 163
- simple 39, 40 secondary level function 15,46
method of analysis self-reference 115
- Neo Firthian 43 singularity 39
modality 39 situational-functional text analysis 39
mode 108 social attitude 41, 109
social role relationship 41
norm of usage 18 source texts
- overt historically-linked 66
orientation - overt timeless 66
- towards content 84 speech act 82
- towards other 84 status 39
- towards persons 84 style
- towards self 84 - casual 41
original analyses 49 - consultative 41
original model of quality assessment 29, - formal 41
101 - frozen 41
overt 29, 161, 163, 166 - intellectual 41
overt translation 66, 67, 68, 76, 104, 111, - Intimate 41
114, 115
overt version 73 tenor 108
overt-covert dichotomy 111 text 159
overtly erroneous errors 45 text production
- informational 109
participation 39 - involved 109
- complex 40 text transfer 111
- simple 40 text type 17
pedagogic implications 167 text
polysystem theory 6 - commercial 49
presentation of information - conative 17
- abstract 109 - content oriented 17
- non-abstract 109 - emic 45

- etic 45 - second-level-functional equivalence 163
- form-oriented 17 translation evaluation 4, 6, 118
- poetic-aesthetic 48 translation typology 65
textual profile 42
use of verbal routines 84
theme dynamics 44
theoretical aspects of the model 159 version 160
time 38
translation equivalence ”8, 115 World Three 15,34


3 9999 04232 784 9

Studies on Translation
Wolfram Wilss Veronica Smith
Ubersetzungsunterricht Thinking
Eine Einfiihrung in a Foreign Language
Begriffliche Grundlagen und 'Xn Investigation into
methodische Orientierungen Essay Writing and Translation
narr studienbiicher 1996, X, 229 S., by L2 Learners
DM 34,80/OS 254,-/SFr 34,80
Tiibinger Beitrage zur Linguistik 393,
ISBN 3-8233-4958-9
1994, XII, 217 Seiten,
DM 58,-/OS 423,-/SFr 55,-
ISBN 3-8233-5058-7
Radegundis Stolze
• •

Ubersetzungstheorien Franz Pochhacker

Eine Einfiihrung Simultandolmetschen
narr studienbiicher 1994, 253 Seiten,
DM 29,80/OS 218,-/SFr 29,80
als komplexes Handeln
ISBN 3-8233-4956-2 Language in Performance 10,1994, XII,
306 S., DM 78,-/OS 569,-/SFr 74,-
ISBN 3-8233-4079-4
Johann Strutz /
Peter V. Zima (Hrsg.) Elzbieta Tabakowska
Literarische Cognitive Linguistics
Polyphonic and Poetics
Ubersetzung und Mehr- of Translation
sprachigkeit in der Literatur Language in Performance 9, 1993,
1996, 249 Seiten, V, 146 S., DM 64,-/OS 467,-/SFr 61,-
DM 68,-/OS 496,-/SFr 65,- ISBN 3-8233-4078-6
ISBN 3-8233-5163-X
Sonja Tirkkonen-Condit (ed.)
Cornelia Lauber Empirical Research
Selbstportrats Studies in Translation
Zum soziologischen Profil and Intercultural Studies
von Literaturubersetzern Selected Papers of the TRANSIF
aus dem Franzosischen Seminar 4, Savonlinna
Transfer 10, 1996, VIII, 140 Seiten, Language in Performance 5, 1991,
DM 58,-/OS 423,-/SFr 55,- 184 S., DM 48,-/OS 350,-/SFr 46,-
ISBN 3-8233-4089-1 ISBN 3-8233-4075-1

Gunter Narr Verlag Tubingen


This is a compieteiy revised and updated version of one of
the standard works on transiation theory. The originai modei
was one of the first attempts to appiy pragmatic theories of
ianguage use to the assessment of transiation quaiity. This
pragmatic perspective is now enriched in the iight of re-
search into the interdependency of ianguage, situation
and cuiture. Specificaiiy, issues of cuiture-specific discourse
behaviour modes, vaiue-systems and perspectives are
confronted with the demands of intercuiturai communication
imposed by the transiationai task,
in the iight of detaiied contrastive studies of the ianguage-
pair German and Engiish, the author iiiustrates the work-
ings of a so-caiied “cuiturai fiiter” in different types of text
under different conditions, in the iight of these findings, the
notions of ‘covert’ and ‘overt’ transiation and their appro-
priacy under different sociai constraints are radicaiiy re-
vised. The new modei for the anaiysis of source text and
transiation is appiied to a corpus of German/Engiish/
Engiish/German transiations.
The book is therefore rich in both theory and iiiustrative data.

ISBN 3-8233-5075-7

Das könnte Ihnen auch gefallen