TRANSLATING NEW GENRES BETWEEN SLOVENE AND ENGLISH: AN ANALYTICAL FRAMEWORK
David LimonDepartment of Translation, Faculty of Arts, University of Ljubljana
Aškerčeva 2, 1000 Ljubljana, SloveniaE-mail: [email protected]
Abstract: Due to processes of internationalisation and European integration, countries such as
Slovenia are having to accommodate many new textual genres, which are either introduced via
translation, or written in Slovene for later translation into English. This paper presents an analytical
framework developed with this situation in mind and describes how it was applied to a particular
translation of a political progress report. The aim was to support the development of a more
systematic approach to the production and translation of such texts. The model draws upon
discourse analysis, genre analysis and contrastive functional rhetoric, and is compatible with
functional approaches to translation. The top-down analysis begins with discussion of the
background to the report and its writing, the participants in the translation process, the training and
support provided, and the translation strategies employed. We then go on to consider the broader
linguistic and cultural background, including the relevant genre conventions. The more detailed
analysis of text profile, coherence, cohesion, information structure and register features is
summarised. The emphasis throughout is on the task facing the reader and whether the
communicative purpose of the text is being achieved.
Key words: context of situation, context of culture, discourse community, genre conventions,
communicative purpose, translation strategy
INTRODUCTION
Slovenia has to rely on translation into English in its dealings with international bodies of
which it is a member or a candidate country. For socio-historic reasons, many of the genres
involved in the process of communication with these organisations are new to the Slovene
environment and there is a clear need for a systematic approach to the production and
1
translation of such texts. An example of the kind of text under consideration is the annual
report submitted over recent years by Slovenia to the European Commission in Brussels,
summarising the progress made towards meeting EU membership requirements and
indicating what remains to be accomplished. Translated texts of this kind have more than
an informative function: they are intermediaries that help to establish a two-way dialogue
between Slovenia as an applicant (now an 'acceding country') and the EU Commission as
‘gatekeeper’. As such, their communicative effectiveness is of great importance, as is the
impression they create. However, current translation strategies produce target texts that are
too heavily influenced by the source text and language, with the result that reader
expectations with regard to cultural, textual and genre conventions are probably not being
met.
The compiling of the report in its current format began only in 1999; prior to that, officials
in the relevant ministries and government bodies had little experience of this kind of
writing and the relevant genre conventions were not available to guide them. The
translators responsible for producing the English version are native Slovene speakers
translating into a foreign language, who also have little experience of the genre concerned.
The support provided to the translators focuses primarily on the production of glossaries of
frequently translated lexical items or ‘terminology’, and standardisation in the naming of
institutions and legislation. In late 1999 and early 2000, some of the translators involved
also attended short seminars on translating into English, which dealt with a wide range of
genres including letters, speeches, memos, responses to requests for information or
clarification, and reports. An initial reading of samples of translations into English showed
that these were insufficiently distanced from the original, being heavily influenced by
Slovene word order, syntax and common lexical patterns, as well as by the stylistic features
of the original and its cultural assumptions. Further observations were that they were
frequently inconsistent in terms of register and seemed to make unnecessary demands on
the reader. In discussion, translators themselves expressed strong reservations about the
quality of the English translations they produced and frequently pointed, in mitigation, to
what they considered to be unclear and ‘badly written’ originals, as well as the immense
2
time and other pressures under which they had to work. Further analysis and discussion of
the translations pointed to the need for the development of clearer strategies for translating
texts from Slovene into English and for the revision of the resulting translations. It also
underlined the need for translation training in relevant genre conventions, as well as
possible training for the writers of the original texts.
THE FRAMEWORK – GENERAL CHARACTERISTICS
To look into this issue further, an analytical framework was developed, drawing upon a
number of fields – in particular discourse analysis, genre analysis and contrastive rhetoric.
The resulting approach to translation analysis is based on the functional comparison of texts
across languages; this is linked to a functional approach to translation, which looks at the
purpose of the target text and examines how successfully it has been realised through the
translation process. The model is an eclectic and flexible one that is not tied to any
particular linguistic model. It takes a top-down approach, starting with text in its situational
and cultural context, including the rhetorical traditions and genre conventions associated
with the two languages involved. It deals with contextualised meaning, considering
individual items only in terms of their function within the text: translator decisions, even at
word or phrase level, involve (albeit subconsciously) consideration of the wider context;
moreover, text strategies precede the syntactic formation of individual sentences – we do
not produce a sentence then give it 'textual fit' it after it is already there. The model focuses
on communicative rather than systemic factors and deals with text in context or socially-
situated language use. This means considering factors such as: setting, participants
(producer and receiver), roles (communicative and social), goals, social knowledge, norms
and values, and institutional or organisational constraints upon communication. The model
analyses the role played by translation within a very specific process of social interaction at
a particular point in history. It assesses language use in terms of markedness, which relates
to reader expectations, and deals with text as process as well as product, i.e. how it is
interpreted by the translator and by the final reader to construct a meaningful textual world.
Its point of departure is acceptability, rather than adequacy in relation to the source text (cf.
Toury 1995). Finally, at the practical level, it looks for similarity rather than identity, as
3
relativist notions are more suited to translation (cf. Chesterman 1998a: 39ff), and aims only
for explanatory adequacy – no translation analysis can be truly exhaustive, if for no other
reason than that the target system is in a constant state of flux.
GOALS OF THE MODEL
Analyse the context of situation
To understand the meaning of any text, including a translation, we need to consider both
the context of situation in which it occurs and the broader context of culture within which it
functions. Thus we have to look not simply at the completed translation but at the whole
process by which the translation was produced, the reader's interaction with it and the
context within which that interaction takes place. We start our analysis by describing the
translators and their linguistic and experiential background, the institutional environment in
which they operate, the strategies they employ, and the translation and revision procedures
involved in the translation process.
Analyse the context of culture
In order to understand the context of culture, we look first at the two languages involved in
the translation process – one of which, Slovene, is a 'smaller' language, while the other,
English, is a global one – as well as the rhetorical and textual conventions associated with
these languages. The special factors relating to the dominance of Anglo-Saxon cultural
values within many fields of communication and the use of English as one of the main
working languages within the European Union are also discussed.
Analyse in terms of genre
Genre is text defined in terms of social purpose. Genres are structured texts that unfold in a
particular way in order to achieve identifiable communicative purposes. Genre analysis sets
out to explain socio-cultural, institutional and organisational constraints upon
communication, as well as to identify conventionalised regularities in communicative
events (bearing in mind that such regularities are likely to vary between different languages
and cultures). For the analysis we are carrying out, which involves an assessment of how
4
well the translated text functions in comparison with similar texts within the target culture,
the concept of genre is more useful than that of text type (defined in terms of predominant
rhetorical purpose, e.g. instruction, exposition or argumentation) or register, which
represents the more general linguistic choices that are made in order to realise genres. Thus
we compare the translated text we are analysing with a 'control text', or parallel text
matched for genre, as well as with the source text; this comparison extends to sub-genres
identified within the text. In doing so, we try to identify the conventions by which the
reader – in particular, an expert member of the discourse community – is likely to accept
the translated text as an exponent of the genre.
Identify communicative purpose
Central to any genre analysis is the identification of the communicative purpose or the goal
of the text, as this constitutes the rationale for the genre. As such, it is crucial to any
evaluation of whether translation purpose is achieved. It is important also because it helps
determine the structure of the discourse and constrains both the content and how that
content is expressed.
Account for ideational, interpersonal and textual meaning
Text is about more than communicating information. Any textual analysis needs to take
into account the three strands of potential meaning that underlie language use: the
ideational, interpersonal and textual functions (cf. Halliday 1970: 143). In order to achieve
this, we borrow partly from the model of Contrastive Functional Rhetoric presented by
Chesterman (1998a) and partly from the methods of genre analysis as described by Swales
(1990) and Bhatia (1993). Of the 'text specifiers' relating to the ideational aspects of
messages described in the former we are most concerned with profile (Chesterman op.cit.
170ff), or the way in which ideas proceed through the text and the general structure of its
meaning. To analyse this profile to a greater degree of delicacy, we compare the surface
ordering of the text with its underlying rhetorical structure, describing the degree of fit
between them. In relation to interpersonal meaning, we rely on the generic concept of
communicative purpose, as well as the register variables, particularly tenor (cf. Halliday
5
and Hasan 1976: 23). Finally, as far as textual meaning is concerned, we look at different
aspects of coherence as described in Chesterman's (op.cit. 183ff) functional model. This
involves metatextual features – such as previewing, summarizing and signposting, which
we discuss under the heading of rhetorical structure and text ordering – as well as the
surface realisations of coherence, which we deal with under the heading of cohesion. It also
involves informational coherence or 'informativity', which we analyse in terms of
information structure – in particular, conformity with unmarked given-new patterns (cf.
Halliday 1967: 211 and 1994: 140; also Fries 1994: 233-234) – and intertextual coherence,
the implicit aspects of which we deal with in terms of degree of conformity to genre
conventions, while explicit intertextual references are noted when allusions are made to
related texts.
Analyse text-centred features
The key to understanding both the process of translation and to judging its product is to see
the meaning of the text as negotiated between producer and receiver (Hatim and Mason
1990: 64-65). Texts are not passively received: the reader is actively and creatively engaged
in a hermeneutic process drawing upon not only language and culture, but also experience
and perception (Stolze 2001). The meaning or function of a text is not inherent in the
linguistic signs of which it is composed, but rather a text is made meaningful by and for its
receiver (Nord 1997: 31). The model deals with the coherence of the text in terms of how
easy it is for the reader to process at particular junctures; the involvement of extra cognitive
cost at any point in the text is likely to relate back to the text's conformity to genre
conventions. There is a contrast here between cohesion, which is objective, and coherence,
which is subjective, so that judgements concerning it may vary from reader to reader (Hoey
1991). The most relevant definition for our purposes is that coherence is "a covert potential
meaning relationship among parts of a text, made overt by the reader or listener through
processes of interpretation" (Blum-Kulka's 1986).?? In order to maintain coherence, the
translator needs to strike a balance between what is effective, i.e. achieves the
communicative goal, and what is efficient, i.e. places fewest demands on user resources (cf.
Beaugrande and Dressler 1981: 11). The translator also needs to be aware that tolerances
6
and preferences with regard to coherence differ between languages and genres. However,
discussion of whether a text coheres or makes sense is hard to separate from analysis of the
surface features that signal the underlying connections between parts of the text, i.e.
cohesive features, and the way that information is distributed through the text, i.e. its
information structure. Although we discuss coherence, cohesion and information structure
separately, they clearly interact with and are mutually dependent on each other. Finally, to
hang together, texts need to display consistency of register: coherence of meaning is
dependent not only on content, but on selection from the semantic resources of the
language, which is the subject of the final stage of analysis within the model.
Analyse user-centred features
The intended meaning of a text emerges only when pragmatic factors are taken into
account, i.e. who is saying what to whom and for what purpose. However, discourse
analysis does not separate such user-centred features from the text-centred features just
discussed. The textual category of situationality covers the circumstances of the interaction,
including socio-cultural factors, and whether the text is relevant to this situation of
occurrence. This can be analysed in part by seeing how text users interact with register
variables such as field, mode and tenor. A useful distinction when assessing communicative
purpose is that between 'situation monitoring', i.e. when the main function of a text is to
provide a relatively unmediated account of the situation model, and 'situation management',
i.e. when the main purpose is to steer the situation towards the text producer's goals. This
category is best described in terms of dominances rather than either-or terms (Beaugrande
and Dressler 1981: 163ff). The category of intentionality relates to the writer's goals,
realised both globally and locally within the text, with varying degrees of explicitness,
which the translator seeks to convey to target readers in a manner appropriate to the context
of situation. One such global goal is to make the text acceptable to a particular discourse
community so that the text receiver accepts that the text has some relevance or use in terms
of acquiring information or taking action (i.e. the category of acceptability). These
categories enter into our analysis when we talk about communicative purpose and about
how the text matches up in terms of genre conventions, which in turn impacts on ease of
7
processing. In a similar way, the category of intertextuality, which is key to the translator's
work, is constantly present in the analysis due to comparison with the control text and
reference to reader expectations when assessing ease of processing. Finally, as already
noted, informativity of content, or the degree of givenness and certainty in the text, is
analysed under the heading of information structure.
STEPS IN THE ANALYSIS
Translation description can focus on three fundamental aspects of the translation and its
environment: the intralinguistic profile of the translation compared to a non-translated text
of the same genre in the target culture; the interlinguistic profile of the translated text in
relation to its source text; and the extralinguistic relations between the translation, the
situation in which it is produced and the socio-cultural context in which it is embedded
(Chesterman 1998b: 204). The analysis we carried out considers all three of these aspects,
placing more emphasis on the intralinguistic than the interlinguistic, but starting with
extralinguistic factors. The analysis comprises 9 main steps, as follows:
1. Describe the translation context and translator profile
2. Analyse the broader linguistic and cultural background
3. Identify communicative purpose
4. Identify relevant genre conventions
5. Compare text ordering with underlying rhetorical structure
6. Assess coherence of message and ease of processing
7. Analyse in terms of cohesion
8. Analyse in terms of information structure
9. Describe representative register and discourse features
In the present paper, we shall concentrate on the first four steps, as the remainder involve
detailed textual analysis and comparison.
THE CORPUS
8
The model was applied to one of the annual progress reports submitted by Slovenia as an
applicant country, the NPAA Report 2000, which is the English translation of the text
Državni program za prevzem pravnega reda EU – Poročilo 2000. The NPAA Report is
described by the Slovene Government as "the central document in the whole of the
integration process of the Republic of Slovenia into the EU" (NPAA Report May 1999, p.6)
and is regarded in Brussels as a document "of paramount importance" in establishing the
final position of the Commission with regard to the progress made by Slovenia over the
previous twelve-month period (personal communication from the Head of the EC
Translation Service, 6.12.2001). The report contains four main chapters, on political
criteria, economic criteria, the capacity to adopt the acquis and administrative capacity.
For purposes of genre comparison, we referred to the 2000 Regular Report from the
Commission on Slovenia’s Progress Towards Accession (hereafter Regular Report), which
was written in English and which could be said to represent a response to the report
submitted by Slovenia. It was not possible to obtain a control text in the form of a progress
report from an applicant country written in English, because all of the reports submitted are
translations. Many of those involved in the drafting of the Regular Reports in English are
non-native speakers, but an English translator is involved in the production of the final
version. In any case, we can assume that these reports represent genre models, as they are
produced by expert members of a particular discourse community – i.e. representatives of
the EU's bureaucratic structures. The first such report was submitted by the Commission to
the Council in October 1998 and they have since been produced on an annual basis. The
purpose of the reports is to review "the progress of each Central and Eastern European
applicant State towards accession in the light of the Copenhagen criteria, in particular the
rate at which it is adopting the Union acquis" (2000 Regular Report, p.5). The reports
represent the basis on which the Council takes decisions relating to accession negotiations.
Although the NPAA Report and the Regular Report share the same basic structure and
belong to the same genre (the political progress report), they differ subtly in terms of
communicative purpose. This is because of the different status and communicative role of
9
the (institutional) writers of the two texts – in the one case an applicant and in the other a
participant in the screening process. At the level of register, we might expect differences
particularly in terms functional tenor – specifically, to what extent the writer is involved in
informing, persuading, evaluating or exhorting. Thus, when making comparisons between
the two, we need to ask to what extent the differences can be ascribed to the differing
functions of the texts and to what extent they are due to the fact that one of them is a
translation.
CONTEXT OF SITUATION
The character of the translator, his/her cultural environment and knowledge base are an
important part of translation evaluation (cf. Wilss 1999: 146); it is also important to
understand the institutional constraints affecting the translator's work (Koskinen 2000).
According to information obtained from the body responsible, the Government Office for
European Affairs (SVEZ), the translation of the NPAA Report in 2000 involved 24 in-
house and freelance translators, none of whom were native speakers of English. Unlike the
document published in May 1999 (which, by comparison, involved 18 individual
translators and 2 translation agencies), the translated text was not subject to language
revision, due to time pressures and the unavailability of appropriately qualified personnel.
Moreover, the process of compiling and translating the report was not subject to any overall
co-ordination.
The traditional assumption within translation theory has been that translators work into
their first language, which is typically seen as “the only way you can translate naturally,
accurately and with maximum effectiveness” (Newmark 1988: 3); this conviction has no
doubt been perpetuated by the high value placed on the translation norms of fluency and
naturalness, particularly with regard to literary translation, so entrenched in the Anglo-
Saxon cultural sphere (cf. Venuti 1995). However, in recent years this insistence on
translation into the first language has come under increasing scrutiny, particularly in
relation to certain language pairs and certain genres (see, for instance, the papers in
Grosman et al eds. (2000)). For practical reasons, in many parts of the world a great deal of
10
translation takes place from the first language into a foreign language – most frequently
English. This is not only the case in smaller nations with less widely known languages,
such as Slovenia (where only a handful of members of the professional translator's
association are native speakers of English), but also in countries as widely different as
Spain, Japan and Australia, where for a variety of reasons there are insufficient translators
from the right background available (cf. Campbell 1998: 22ff). As long as there is
asymmetry between languages, with some having more de facto status than others, this
situation is likely to persist. Moreover, with regard to certain kinds of discourse, such as
scientific and technical texts, it can be argued that accuracy is more important than felicity
of style and that a thorough grounding in the subject matter, supported by close familiarity
with such texts in the TL, is more important to the translator than native-speaker
competence in the TL. In any case, both bilingual and translation competence are matters of
degree, and in the situation we are discussing, the translator's textual competence may well
be the most crucial factor.
The current reality is that native speakers of English able to translate from Slovene are
simply not available to cope with the volume of Slovene-English specialist translation now
required and the situation is unlikely to change in the very near future. One way of
ameliorating the influence of the SL on the TL texts produced by non-native speakers of the
TL is to employ native-speaker language revisers or rewriters. This is a strategy employed
by some translation agencies, companies and government bodies in Slovenia, but does not
seem to be applied consistently; furthermore, the level of knowledge of Slovene among
those carrying out revision work, many of whom are short-term employees with no formal
knowledge of the language, is extremely varied, as is the level of expertise in relation to the
target language. One result of this is that revision focuses largely on surface detail, which
may ensure that the translated text is largely free from obvious grammatical error, but does
not guarantee either accuracy, for which the translator has to be responsible, or
communicative effectiveness.
11
A good example is the original NPAA Report, from 1999, which was revised by a visiting
expert from Brussels (an Adviser on Enlargement from the European Commission
Translation Service), who admitted that his knowledge of the source language was limited.
The revision notes subsequently provided to assist future translations into English consist
of 19 points, which focus entirely on lexico-grammatical features and punctuation (e.g. use
of prepositions, articles, hyphenation) – the one more general point being the avoidance of
American English. Thus the only translation evaluation carried out in relation to this
ongoing translation project merely identified a small number of surface errors and failed to
assess the text in terms of genre or even register. This misses the important point that
translations free of lexico-grammatical errors can still be communicatively ineffective.
Mastering new genres and styles of discourse in a target language is not just part of the
translator's general competence in the language, it is something that has to be learned. If
native-speaker revisers cannot be relied upon to provide textual acceptability, then the
translators themselves should be provided with training in the area where text meets context
– i.e. in genre conventions and/or text-type conventions defined in terms of rhetorical
purpose. However, such training is not yet available and the support provided to
government translators tends to focus on terminological issues and lexical standardisation
through the use of translation software. Moreover, there seems to be a lack of any clear
guidance for the writers of the different parts of the original report, who are located in
many different ministries. As far as I was able to establish, the overall structure of the
NPAA Report was based on the list of negotiating chapters drawn up in Brussels, but no
detailed guidance was offered on issues such as content, form, length and layout. Similarly,
the approach taken to the writing and compilation of the report seems to have been an ad
hoc one: the purpose or goal of the report and the strategy to be used was not set down
anywhere and the writers did not receive any prior functional training in report writing.
A close reading of a range of texts translated from Slovene into English for SVEZ by both
in-house and freelance translators would seem to indicate that too much emphasis is placed
on faithfulness or adherence to the SL text, while insufficient emphasis is placed on the
12
communicative purpose of the TL text. What we frequently encounter is unwanted formal
interference brought about by the manipulation of source-text structures at the expense of
intended meaning. The largest part of the work carried out by SVEZ translators is the
translation of EU legislation from English into Slovene; in doing this, translators adhere as
closely as possible to the wording of the original, producing a fairly literal rendering of the
English text which is stylistically and rhetorically marked when compared to original
Slovene legislative texts. The reason for this is the nature of the texts involved and the
strategy deemed appropriate: because of the status of texts such as laws, supplementing
regulations, directives and decisions, a strategy based on "prudence" and "capitulation"
rather than "risk-taking" and "persistence" (cf. Campbell 1998: 104) is preferred. The fact
that such translations are passed on for detailed legal revision must also contribute to the
tendency to stay at the level of surface-level similarity. What then seems to happen is that
translators employ the same strategy when translating others kinds of texts from Slovene
into English – partly, perhaps, because they feel less confident translating into English. But
different genres place different demands on translators and one cannot assume that an
approach used when translating legal texts is going to be appropriate to the translation of a
progress report. With regard to the latter, we need to ask what matters most – faithfulness to
the SL reporting style or the function of the report within the target environment.
Through informal interviews with some of those involved and a written question
administered to 40 translators, I endeavoured to establish what strategy the translators saw
themselves as using. The application of the principles of legal translating is probably not a
conscious process and was mentioned by only one translator. In answer to the written
question In what circumstances do you tend to stay close to the original when translating?
the most frequently selected responses (respondents chose one or more options, or offered
their own) were: when the text deals with an unfamiliar topic (75%) and when the original
is difficult to understand or unclear (72.5%); a half also took this approach when the text
was seen to be an 'important' one and slightly less than half (45%) when they had been
criticised in the past for not 'sticking to the original' (usually by the original author).
Slovene translators of specialist texts into English, especially where those texts have high
13
status, find it 'safer' to adhere closely to the principles of lexical equivalence and surface-
level similarity. The degree of mediation, i.e. the extent to which translators intervene in the
transfer process, is very low with regard to EU texts; there is also no translator 'visibility' in
the form of a preface, commentary or footnotes. The translators are also invisible in another
sense: contact between them and other participants in the communication process is
extremely limited. The notion of the translator as an expert in cultural mediation (cf. Katan
1999; also Wilss 1999) has not gained any real currency within the Slovene administration
and translators are rarely asked to comment on the generic or textual appropriateness of SL
texts written specifically for translation into English – nor do most translators seem willing
to adopt a more interventionist strategy based on a knowledge of differences in textual and
genre conventions.
The problems we have described in this section are by no means restricted to the Slovene
environment. Koskinen (2000) describes how translation of EU legal texts into Finnish has
produced a new legal rhetoric, very different from that traditionally present in Finnish
legislation. She also points out how the literal mode of translation has gained a wide
currency, even beyond the field of legal texts, i.e. in areas where the translator presumably
has much more leeway for mediation or adaptation. One possible explanation offered is that
translating institutions tend to prefer a translation ethics of sameness that is not too
unsettling to existing norms (cf. Venuti 1998: 82). As is the case in Slovenia, the strategic
choices made by Finnish translators are not made on the basis of written guidelines but are
the result of assimilation into the prevailing 'climate' of the institution as well as the model
represented by previous translations. The collective nature of EU translating (i.e. the
number of translators that may be involved in the translation of a single text, which reflects
the way texts are written) and the level of intertextual allusion and reference involved (texts
frequently quote at length from other versions or texts) mean that individual translators are
unlikely to deviate radically from the general trend. A further constraining factor is the
"intertextual network" within which frequent redrafting and revising of both originals and
translations takes place: texts and their translations are collective products, which means
14
that responsibility for the relationship between them is also collective and thus "seems to lie
on no-one's shoulders" (Koskinen op.cit. 60).
LINGUISTIC AND CULTURAL CONTEXT
For small nations such as Slovenia there is an urgent need to communicate internationally
without being stigmatised for what Connor (1996: 47) refers to as "poor linguistic
manners". When a powerful TL such as English, which enjoys cultural hegemony and
prestige compared to Slovene, is involved one would expect motivated interventions by
translators (cf. Venuti 1995). Similarly, one might expect to find different strategies being
employed with regard to translation into and from more peripheral cultures (cf. Bassnett
1993: 142ff). When translating from English into Slovene, it does seem to be the case that
in many fields (academic and scientific writing, as well as computing, telecommunications
and marketing texts are obvious examples) Anglo-Saxon cultural values are widely seen as
universal or neutral: no cultural filter is employed, but rather rhetorical patterns and register
values are imported directly into Slovene, influencing a wide range of genres. And, as we
have already seen, when translating EU legal texts from English the same holds true.
However, in the case of the text we are analysing the opposite seems to be the case, i.e.
Slovene textual features are transferred into the target text in English. This is probably
because of the translation strategy employed, which we have already discussed, and also
because the translators involved are translating out of their mother tongue, which clearly
has a strong influence on what they produce in the (foreign) target language.
At this point it would be helpful to see what general differences can be said to prevail
between written discourse in Slovene and English; in the absence of relevant contrastive
studies, the claims made here are necessarily tentative ones. In comparison to English,
which is generally seen as 'writer responsible', Slovene could generally be characterised as
'reader responsible' – in other words, responsibility for effective communicative is seen to
lie primarily with the reader (cf. Hinds 1987). Thus, for instance, English texts tend to be
marked with clear transition statements, while transition devices are likely to be more
subtle in Slovene and demand more of the reader. Similarly, while English is generally
15
perceived as having 'addressee orientation' (cf. Katan 1999: 194), distinguished by low
information load (i.e. the speed at which information is introduced), reader friendliness,
simplicity and clarity, Slovene has more features of 'author orientation', such as high
information load, authority, explicit rhetorical skills and rich style. Another distinction that
could be made is that Slovene discursive writing relies more on 'knowledge telling', i.e.
retrieval from memory, than 'knowledge transforming' (cf. Grabe and Kaplan 1996: 124).
The greater tolerance for digression, recapitulation and repetition and the preference for
content over form that Clyne (1987) identified in German academic prose compared to
English also seems to be true of Slovene – not surprising, when one considers how much
influence the German educational and linguistic model has had on Slovenia over the
centuries. Indeed, much of what Čmejrková and Daneš (1997) say with regard to the
influence of 'Teutonic' style on Czech academic writing would also be applicable to the
Slovene context.
That written discourse in every language bears markings of cultural specificity at many
levels, and that this has to be taken into account in writing instruction that aims to equip
students with communicative, strategic, sociocultural and discoursal competence is
recognised by the Council of Europe's Common European Framework (2001) for language
learning, teaching and assessment. In the light of this, it is instructive to compare the
secondary school programmes for Slovene and English drawn up by the Slovene Ministry
of Education and Sport.1 The stress within the Slovene language syllabus is on creative
writing and the interpretation of literary texts, rather than the teaching of functional writing
and text types such as exposition and argument; moreover, there is a relative disregard for
text composing strategies, audience-awareness and text organisation. The lack of functional
writing instruction at secondary level and beyond means that Slovene students who later go
on to fill administrative posts are unlikely to have had any training which would help them
write the kind of texts we are analysing in this study.
1 http://www.mszs.si/slo/solstvo/ss/programi/gimnazija/programi.htm
16
So far, we have referred to the Anglo-Saxon cultural sphere, but we are concerned more
specifically in this study with English as a medium of communication between EU
institutions on the one hand, and member states and applicants on the other. The English
used within the EU is widely seen as a separate variant of the language with its own
distinguishing features, popularly known as Eurospeak but also referred to as Eurolect
(Goffin 1994). The process leading to the development of this variant is one of "interlingual
assimilation" (Wilss 1999: 106ff). arising from a multilingual administrative environment
in which texts are produced by groups of authors with different native languages. In the
popular imagination and the media, Eurospeak has become synonymous with the worst
kind of bureaucratese; EU institutions have responded by taking steps to improve the
accessibility of the texts they produce, as well as their own image. Resolutions have been
passed calling for guidelines for the quality of drafting of Community legislation and on the
use of clear language, culminating in a declaration on this issue within the Treaty of
Amsterdam. At the same time, some of those working within the institutions have initiated
a campaign against ‘Eurofog’ or 'brouillard linguistique', offering guidelines on clear
writing both for those involved in drafting texts in English and those translating. The Fight
the FOG booklet2 offers advice in the following key areas: putting the reader first (in
particular, members of the general public); preferring verbalisation to nominalisation,
concrete terms to abstract ones, and the active to the passive voice; putting given
information at the start of the sentence and new or complex information at the end; and the
benefits of keeping documents short and simple. One clear message of the campaign is that
there are inherent dangers in trying to say too much in too abstract a fashion: "English is a
notoriously blunt language. Too much abstract language (FOG) may make your reader
suspect that something real and unpalatable is being wrapped up in verbiage" (p.4). The
writers of this document draw a distinction between "Eurospeak", with which they would
like to refer to useful language coined for EU concepts that have no exact parallel at
national level (e.g. subsidiarity, convergence, cohesion), and "Eurojargon", which we might
describe as the specific lexis used by the discourse community made up of EU insiders.
2 http://europa.eu.int/comm/sdt/en/ftfog/index.htm
17
In talking about the context of culture, we have so far begged the question as to whether
this notion can legitimately be used in relation to a multinational (and multilingual)
organisation such as the EU. The concept of translation as intercultural communication has
traditionally been associated with communication across linguistic and national or ethnic
boundaries, which of course is not the situation with regard to the EU institutions. A
frequently quoted definition of culture is that given by Goodenough (1964: 36), who
defines it in terms of acceptability: "a society’s culture consists of whatever it is one has to
know or believe in order to operate in a manner acceptable to its members, and do so in any
role that they accept for any one of themselves”. This definition is reminiscent of the notion
of 'discourse community' within genre analysis. Swales (1990: 58) defines a genre as a class
of communicative events sharing a set of communicative purposes that are recognised by
the expert members of the parent discourse community, thereby constituting the rationale
for the genre. Discourse communities are 'sociorhetorical networks' that form in order to be
able to work towards a common sets of rhetorical goals. The language activities of such
communities are driven by communicative purpose and its primary determinants are
functional – aimed at particular objectives. The six defining characteristics of a discourse
community set out by Swales (op.cit. 24-27) could all be said to apply, in communicative
terms, to the European institutions as a collective body: the community has a broadly
agreed set of common public goals (written or tacit) and mechanisms for
intercommunication among its members (meetings, reports, etc.); it uses participatory
mechanisms primarily to provide information; it utilises one or more genres in the
communicative furtherance of its aims and has acquired some specific lexis (terminology,
acronyms, etc.); finally, there is a threshold level of members with a suitable degree of
relevant content and discoursal expertise. Thus, the EU institutions can at least be referred
to as a discourse community, or a secondary culture. We can also talk about a 'Euro-
rhetoric' that has come about through the interaction of the different linguistic communities
within the context of the EU (cf. Tirkkonen-Condit 2001: 262).
18
Koskinen (2001: 295) draws a useful distinction between two types of EU translation:
intracultural, which takes place within the 'a-national culture' of the EU institutions, and
intercultural, which entails communication between the national cultures of the member
states and the EU culture. The latter category can be further subdivided into inter-
administrative, involving the Commission and national authorities, NGOs or special interest
groups, and 'public', for direct communication between the Commission and the general
public. Although at the time of writing Slovenia is still not a member state, we can
legitimately place the text we are dealing with in the category of intercultural inter-
administrative translation as just defined; we are thus also referring to context of culture,
albeit in a slightly adapted form.
COMMUNICATIVE PURPOSE
Although the analysed text and the control text belong to the same genre, we can expect
them to differ in terms of communicative purpose, due to the different status and
communicative role of their (institutional) authors. They have certain general features in
common: multiple authorship, with authors unidentified (formally institutional);
institutional addressee; third person 'angle' (cf. Chesterman 1998a: 170); reference to the
same time frame (12 months); overall content determined by the same institution (European
Commission, as evaluator of the reports); a focus on action taken – legislation adopted,
international conventions ratified and other measures implemented – with the result that a
limited range of action verbs is used; and a shared structure.
Thus, in terms of institutional focus, there is little difference between the reports, other than
the level of detail (the NPAA Report consists of 269 pages or 96975 words; the Regular
Report of 101 pages or 43417 words). However, the communicative purpose of the two
texts does differ because of the role relationships involved: the author of the NPAA Report,
as an applicant, is mainly involved in informing, while the author of the Regular Report, as
an evaluator, is more involved in evaluating, notwithstanding the strong expository
element. A further point is that, in generic terms, the author of the Regular Report is
already an expert member of the discourse community, while it is not clear how far the
19
author of the Slovene text has been initiated into that community. Beyond that, we might
also say that the NPAA Report is persuasive, in that it seeks to present the information
provided in the best possible light, and mitigatory, when it tries to explain or justify failures
to meet targets, while the Regular Report is also hortatory, in that it makes
recommendations and sets goals (with varying degrees of obligation). The stated purpose of
both texts is to provide an ‘objective’ assessment of progress made and action taken within
the reporting period, but the report from Brussels displays a degree of overt situation
management (cf. Beaugrande and Dressler 1981: 169) in that the evaluator is systematically
applying the same criteria to 12 applicant countries (and thus 12 reports) and making
recommendations accordingly, the overt function of the Slovene report is situation
monitoring, with any management of the situation taking place covertly, i.e. subtly trying to
steer the addressee towards the text producer's goals. Hatim (2001: 178) points out that
texts vary along a continuum from detached and non-evaluative to involved and highly
evaluative. Another way of looking at this is the distinction between participant and
spectator: the author of the Slovene report remains a participant in the events described and
invites the reader to become one also.
Although reporting may be seen as a detached genre, the exposition that takes place within
it (either temporal or conceptual) will have differing degrees of detachment according to
rhetorical purpose, leading at times to significant shifts in function. The result of this will
be text type hybridisation, with a move away from the predominant focus (cf. Hatim 1997:
42; also Werlich 1976) to a subsidiary one: for example, in the NPAA Report towards
argumentation (i.e. where mitigatory circumstances are being cited) or in the Regular
Report towards instruction (i.e. where future action is recommended or prescribed). To
conclude: the main goal or communicative purpose of the two texts (and that which shapes
them as examples of the genre) is exposition, with persuasion and mitigation as the
subsidiary goals of the Slovene report, and evaluation and exhortation as subsidiary goals
of the report from Brussels.
GENRE CONVENTIONS
20
Writing is an attempt to communicate with the reader; the writer has intentions and a
purpose, as well as information to convey. Texts have a hierarchical structure that differs
due to the purpose, audience, status, author, information load and genre (Grabe and Kaplan
1996: 54ff). Beyond surface form, the text is organised by the writer's relation to it, by the
reader's assumed knowledge and by the subject matter. The information structure guides the
reader as to the writer's intent, showing what is presupposed and what thematic information
is highlighted. This structure is constrained by text ordering and by how rapidly and from
what perspective the author wants to present information: too much or too little information
can affect the text's coherence, making it difficult for the reader to process. It can also lead
to inferences being made by the reader that were not anticipated by the writer. In order to
write successfully, the writer needs to be familiar with rhetorical patterns in a language,
composing conventions, inter-sentential syntax, coherence creating patterns, writing
conventions, audience expectations and subject knowledge (Grabe and Kaplan op.cit.
171ff).
According to a guide in English on report writing (Gravett 1998: 27), the main ingredients
of a good report are that it is "user-friendly, written for its intended audience and achieves
the author's purpose". The writer needs a strong awareness of the reader's needs
(determined largely by experience and by feedback) in order to produce a report with a low
"cognitive cost", i.e. one that is easy to process and thus more likely to achieve its goals
(Gravett op.cit. 13). User-friendliness is achieved through: systematic organisation and
layout; clarity of expression and lack of verbosity; and clear development, with each point
leading naturally on to the next and with intentions clearly signposted (Gravett op.cit. 14).
The writer thus needs a clear set of objectives and a clear plan of writing – although there is
a heuristic element in all writing, less skilled writers generate content during composition
without sufficient attention to goals. As Grabe and Kaplan (op.cit. 116) put it: "good writers
have a richer sense of what they want to do when they write, and have a fully developed
image of the rhetorical problem". We tried to determine whether the writer and translator of
the NPAA Report had this defined sense of purpose, a sufficient audience knowledge, a
21
clear rhetorical perspective and an adequate control of genre conventions as we analysed
the original report and its translation.
DETAILED ANALYSIS (IN BRIEF)
Text profile
The next step in the analysis involves comparison of the underlying rhetorical structure of
the text with its surface ordering or organisation in order to determine the degree of fit
between them. This helps determine whether the message is being clearly communicated or
whether unnecessary demands are being made on the reader. The structure of the control
text is a frequent point of comparison during this process. Two representative chapters were
analysed in detail, followed by three sub-genres – introduction, evaluation and conclusion.
During this stage of the analysis, we evaluated the extent to which the lack of fit between
the rhetorical and surface organisational structures could be ascribed to the original writing
process and how much it was due to the translation process. We found that issues connected
to the way in which the text is ordered and how this relates to the underlying rhetorical
structure can be addressed to some extent through the translation process. However, less
localised problems – such as those concerning the structure of the sections relating to the
negotiating chapters and the information they contain – can only be dealt with through
liaison between the translator, as an expert in intercultural communication, and the author,
in an interactive process of review and amendment.
Coherence
The sixth step focuses on coherence, which we define in terms of reader interpretation and
ease of processing. Analysis here involves other aspects of the text such as cohesion and
information structure, metatextual features, text content, writer strategy and level of
information, as well as surface features such as paragraph organisation and the detailed
lexico-grammatical choices made by the writer and translator. We concluded that coherence
is dependent on a multiplicity of inter-related factors. These include the order in which
information is presented to the reader and the way in which that information is divided up,
22
as well as genre conventions such as clear development and signposting (the latter
including the use of topic sentences), clarity of expression, and lack of verbosity. In a
number of cases, the lack or omission of a cohesive link, the reliance on lexical cohesion in
the form of repetition, or unclear anaphoric reference, place unnecessary demands on the
reader. Finally, marked information structure patterns subvert reasonable reader
expectations. Thus the quality of reader-friendliness can be said to be distributed at all
levels of the text and cannot easily be separated out from other characteristics. The
following two steps in the analysis are concerned with aspects of the text closely linked its
coherence: cohesion, which might be described as the surface manifestation of the
underlying relations within the text, and information structure, or the way that information
flows or is distributed through the text.
Cohesion
Following a general discussion of cohesion in English and Slovene, we analysed how
lexical and grammatical cohesive features had been dealt with: in the former category most
attention was devoted to repetition, in the latter to conjunction and reference. The kind of
questions we dealt with were: whether the transfer of cohesive features from the ST to the
TT resulted in marked language use; whether there had been any loss of cohesion during
the translation process; and how the translator can compensate for the differences between
the SL and TL with regard to cohesive preferences. Our findings can be summarised as
follows: cohesion is most frequently achieved in the translated text through lexical means,
particularly repetition and the use of semantically-related terms. In a number of cases,
lexical cohesive features have been omitted by the translator, but not replaced by elements
of grammatical cohesion. Notwithstanding this, the level of repetition in the translation is
highly marked. At the same time, reader expectations are often not met, due to the
infrequent use of reference words, particularly determiners, and the low level of
conjunction; processing of the text is also made more difficult by the lack of cohesive links
between paragraphs, especially where enumeration is the mode of presentation.
Information structure
23
The strategic differences between English and Slovene word order were then noted and the
principles of information structure discussed in terms of given-new patterns, rising
information load and the principle of end-weight (cf. Biber et al 1999: 896ff). A wide range
of examples from across the text were then discussed, identifying instances of marked word
order in the TT that were unmotivated by communicative factors and violations of the
principles mentioned above. The conclusion we arrived at is that the translator's work had
not been sufficiently informed by an awareness of the different ways that information
structure is handled in the two languages. The approach taken in the translation is
unsystematic: sometimes, the original word order is retained, on occasion through the use
of the passive, with the result that the reader is presented with a marked but
communicatively unmotivated new-given pattern; in other instances the original word order
is dis-preferred, even though its use would have made processing of the text much easier.
Register features
Finally, we analysed a number of register and discourse features characteristic of the text.
The focus of this stage of the analysis was on how the author realised the subsidiary
communicative purpose of persuasion and whether the translator's work had supported or
undermined the achievement of this aim. We looked first at three techniques used in order
to persuade the reader that progress had been made: the accumulation of detail with regard
to action taken; the selection of adjectives and quantifiers to modify nouns; and the frequent
use of collocations involving the noun 'Europe' and the derived adjective. We then
discussed the marked use of the passive voice in the translation, the effect of the translator's
individual lexical choices on the tenor of the text and, briefly, the text's modality. This took
the analysis to a sufficient degree of delicacy for our purposes, so that we could then turn to
a general discussion of the results of our analysis and draw our research conclusions.
CONCLUSIONS
The analytical framework outlined here told us a great deal about the translator strategy
involved in the production of a particular text in a particular context, but it could also be
applied to any number of relevant translation situations. The analysis underlined the need
24
for translator training in the area where text meets context: i.e. in genre conventions and/or
text-type conventions defined in terms of rhetorical purpose. It also pointed to the need for
contrastive rhetorical research into the different textual norms and rhetorical preferences
between the two cultures, considering, for example: writer vs reader responsibility;
differences with regard to information load, reader friendliness, simplicity and clarity;
varying tolerances for digression, recapitulation and repetition; and so on. The translator's
textual competence should come into play from the start: the first steps are to identify the
main communicative purpose of the target text, as well as any subsidiary goals, and the
relevant genre conventions. These conventions are derived not only from parallel texts, but
also from relevant manuals or writing guides from the target environment on the genre in
question.
Comparing text ordering with underlying rhetorical structure in order to establish text
profile pointed in this instance to two main findings. First, that the translated text seemed to
employ far more moves than necessary to cover a topic, whereas the control text was
characterised by close fit, with approximately one rhetorical move per paragraph. Second,
the translation had insufficient metatextual elements, such as signposting, previewing and
transition statements. With regard to coherence, the translated text was judged to be less
user-friendly, while in cohesive terms it showed a preference for lexical cohesion and a
high tolerance for repetition; the control text, by contrast, made much more use of
grammatical devices, particularly reference and conjunction (and this, in turn, affects and is
affected by the general metatextual features of the text). As far as information structure is
concerned, the different distribution of given and new information proved to be the decisive
factor, and again had to be judged primarily in terms of reader expectations. Finally, with
regard to register and discourse features, the cumulative effect of the translator's individual
choices have to be considered in terms of whether they make the achievement of the overall
communicative purpose of the text more or less likely.
References
Bassnett, S. 1993. Comparative Literature: A Critical Introduction. Oxford: Blackwell.
25
Beaugrande, R. de and Dressler, W. 1981. Introduction to Text Linguistics. London: Longman.
Bhatia, Vijay K. 1993. Analysing Genre: Language Use in Professional Settings. London and New York: Longman.
Biber, D., Johansson, S., Leech, G., Conrad, S. and Finegan, E. 1999 Longman Grammar of Spoken and Written English. Harlow: Pearson Education.
Blum-Kulka, S. 1986. Shifts of cohesion and coherence in translation. In Blum-Kulka, S. and House, J. (eds.) Interlingual and Intercultural Communication: Discourse and Cognition in Translation and Second Language Acquisition Studies. Tubingen: Gunter Narr. 17-35.
Campbell, S. 1998. Translation into the Second Language. London and New York: Longman.
Chesterman, A. 1998a. Contrastive Functional Analysis. Amsterdam/Philadelphia: John Benjamins.
Chesterman, A. 1998b. Causes, Translations, Effects. Target Vol. 10 No. 2. 201-230.Clyne, M. 1987. Cultural differences in the organization of academic texts: English and
German. Journal of Pragmatics Vol. 11. 211-247.Connor, U. 1996. Contrastive Rhetoric. Cambridge: Cambridge University Press. Council of Europe, Modern Languages Division. 2001. Common European Framework of
Reference for Languges: Learning, teaching, assessment. Cambridge University Press. Čmejrková, S. and Daneš, F. 1997. Academic writing and cultural identity: the case of
Czech academic writing. In: Duszak, A. (ed.) Culture and Styles of Academic Discourse. Berlin and New York: Mouton de Gruyter. 41-61.
Fries, P.H. 1994. On Theme, Rheme and discourse goals. In: Coulthard, R. M. (ed.) Advances in Written Text Analysis. London: Routledge. 229-249.
Goffin, R. 1998. L'eurolecte: Oui, Jargon Communautaire: Non. Meta Vol. 39. No. 4. 636-642.
Goodenough, W.H. 1964. Cultural anthropology and linguistics. In: Hymes, D. (ed.) Language in Culture and Society: A Reader in Linguistics and Anthropology. New York: Harper and Row. 36-40.
Grabe, W. and Kaplan, R.B. 1996. Theory and Practice of Writing. An Applied Linguistic Perspective. London and New York: Longman.
Gravett, S. 1998. The Right Way to Write Reports. Kingswood Surrey: Elliott Right Way.Grosman, M., Kadric, M., Kovačič, I. and Snell-Hornby, M. (eds.). 2000. Translation into
non-mother tongues: in professional practice and training, (Studien zur Translation, Bd. 8). Tübingen: Stauffenburg.
Halliday, M.A.K. 1967. Notes on transitivity and theme in English: Part 2, Journal of Linguistics Vol. 3. 199-244.
Halliday, M.A.K. 1970. Language Structure and language Function. In: Lyons, J. (ed.) New Horizons in Linguistics. Harmondsworth: Penguin. 140-165.
Halliday, M.A.K. and Hasan, Ruqaiya 1989. Language, Context and Text: Aspects of language in a social semiotic perspective. Oxford and New York: Oxford University Press.
Hatim, B. 1997. Communication Across Cultures. Translation Theory and Contrastive Text Linguistics. Exeter: University of Exeter Press.
26
Hatim, B. 2001. Teaching and Researching Translation. Harlow: Pearson. Hatim, B. and Mason, B. 1990. Discourse and the Translator. London and New York:
Longman.Hinds, J. 1987. Reader versus writer responsibility: a new typology. In: Connor, U. and
Kaplan, R.B. (eds.) Writing across languages: Analysis of L2 text. Reading, Mass: Addison-Wesley. 141-152.
Hoey, M. 1991 Patterns of Lexis in Text. Oxford: Oxford University Press.Katan, D. 1999. Translating Cultures. An Introduction for Translators, Interpreters and
Mediators. Manchester: St Jerome.Koskinen, K. 2000. Institutional Illusions: Translating in the EU Commission, The
Translator Vol. 6. No. 1. 49-65.Koskinen, K. 2001. How to research EU translation?, Perspectives: Studies in
Translatology Vol. 9. No. 4. 293-300.Newmark, P. 1988. A Textbook of Translation. Hemel Hempstead: Prentice Hall.Nord, C. 1997. Translation as a Purposeful Activity. Manchester: St Jerome.Stolze, R. 2001. Translating legal texts in the EU, Perspectives: Studies in Translatology
Vol. 9. No. 4. 301-311.Swales, J.M. 1990. Genre Analysis. English in academic and research settings. Cambridge:
Cambridge University Press.Tirkkonen-Condit, S. 2001. EU Project Proposals as Hybrid Texts: Observations from a
Finnish Research Project. Across Languages and Cultures. Vol. 2. No. 2. 261-264.Toury, G. 1995. Descriptive Translation Studies and Beyond. Amsterdam/Philadelphia:
John Benjamins.Venuti, L. 1995. The Translator’s Invisibility: A history of translation. London and New
York: Routledge.Venuti, L. 1998. The Scandals of Translation: Towards an Ethics of Difference. London
and New York: Routledge.Werlich, E. 1976. A Text Grammar of English. Heidelberg: Quelle & Meyer. Wilss, W. 1999. Translation and Interpreting in the 20th Century: Focus on German.
Amsterdam/Philadelphia: John Benjamins.
Sources
Državni program za prevzem pravnega reda EU – Poročilo 2000. / NPAA Report 2000 http://www.gov.si/svez2000 Regular Report from the Commission on Slovenia’s Progress Towards Accession. EU Commission, November 2000.
27