15.2199, Review: Phonetics/Applied Ling: Teschner & Whitley

LINGUIST List linguist at linguistlist.org
Tue Aug 3 16:58:52 UTC 2004

LINGUIST List:  Vol-15-2203. Tue Aug 3 2004.

Subject: 15.2203, Review: Translation: Mauranen & Kujamki (2004)

Reviews (reviews at linguistlist.org):
        Sheila Collberg, U. of Arizona
        Terence Langendoen, U. of Arizona

Home Page:  http://linguistlist.org/

The LINGUIST List is funded by Eastern Michigan University, Wayne
State University, and donations from subscribers and publishers.

Editor for this issue: Naomi Ogasawara <naomi at linguistlist.org>
What follows is a review or discussion note contributed to our Book
Discussion Forum.  We expect discussions to be informal and
interactive; and the author of the book discussed is cordially invited
to join in.

If you are interested in leading a book discussion, look for books
announced on LINGUIST as "available for review." Then contact
Sheila Dooley Collberg at collberg at linguistlist.org.


Date:  Mon, 2 Aug 2004 17:11:14 -0400 (EDT)
From:  Gabriela Saldanha <Gabriela.Saldanha at dcu.ie>
Subject:  Translation Universals: Do they exist?

-------------------------------- Message 1 -------------------------------

Message 1: Translation Universals: Do they exist?
Date: Mon, 2 Aug 2004 17:11:14 -0400 (EDT)
From: Gabriela Saldanha <Gabriela.Saldanha at dcu.ie>
Subject: Translation Universals: Do they exist?

EDITORS: Mauranen, Anna; Kujamki, Pekka
TITLE: Translation Universals
SUBTITLE: Do they exist?
SERIES: Benjamins Translation Library 48
PUBLISHER: John Benjamins
YEAR: 2004
Announced at http://linguistlist.org/issues/15/15-548.html

Gabriela Saldanha, Centre for Translation and Textual Studies, School
of Applied Languages and Intercultural Studies, Dublin City


Mauranen and Kujamki's edited volume consists of selected papers from
a conference with the same title held in Savonlinna in 2001. It is
divided into four parts. The first one presents two theoretical
explorations of the concept of universals and one discussion of the
limitations of corpus-based methodologies for the study of translation
universals. The rest of the volume consists of empirical studies
testing potential universals. The second and third parts present
corpus-based studies and the last one describes classroom-based

The introduction specifically addresses the problematic aspects of the
notion of translation universals and the methodological difficulties
involved in finding empirical evidence. In particular Mauranen and
Kujamki point out that there are several potential explanations for
the patterns revealed in translated language: the cognitive process
involved translation, systemic differences between languages and
social and historical determinants. Distinguishing between the effects
of each of these factors is not easy, and the corpus-based methodology
that has been favoured - for good reasons - in studies of universals,
needs to be complemented by research into cognitive and social
factors. The editors also warn of the risks of drawing hasty
conclusions from small- scale studies with a limited scope.


Two of the papers in Part 1 (Toury's and Bernardini & Zanettin's)
prefer the term 'laws' to 'universals' on the grounds that a law can
be conditioned and has the possibility of exception built into it, and
exceptions can then be explained with resource to other laws operating
at a different level. Toury stresses that translational behaviour is
affected by a vast an heterogeneous array of factors which are present
all at once and affect each other apart from the observed behaviour.
This situation, argues Toury, requires explanations to be formulated
in probabilistic, rather than deterministic, terms, accounting for the
presence of more than one conditioning factor.

Chesterman compares three different ways in which Translation Studies
have attempted to go beyond the particular: from prescriptive,
critical and descriptive perspectives. Chesterman describes advantages
and problems in each of them, but focusing on the latter, which is the
one that relies on the notion of universals, or unrestricted
descriptive hypothesis as he also calls them. Still, Chesterman does
not object to the term universal, provided ''it is kept for claims
that are actually hypothesised to be universal, not specific to a
subset of translations'' (p43). He concludes by pointing out ways in
which the descriptive approach needs to proceed to move
forward. Chesterman mentions, among other things, the need to further
test and replicate work on restricted hypothesis, standardise concepts
and ways of operationalizing them, and work on testable explanatory
hypotheses to account for the evidence found.

Bernardini and Zanettin's concern is with evaluating the corpus-based
methodology typically applied for research in translation universals.
They highlight the difficulty of achieving representativeness in
translational corpora, with special emphasis on how to ensure the
comparability of the components in a parallel bi-directional corpora.
With reference to one such corpus, CEXI, they point out that the
different tendencies in terms of the text-types most commonly
translated from English into Italian and Italian into English can
impose a certain bias. In these cases there is a need to choose
whether to reflect the operation of different translation policies in
each culture or to prioritise the comparability of the components. In
their conclusion Bernardini and Zanettin stress the need to take into
account the social context in which translations are produced before
attempting to generalise beyond the relevant sample.

The three papers in Part 2 (Large-Scale Tendencies in Translated
Language) all deal with interference as a potential universal of
translation and present results from studies based in the Corpus of
Translated Finnish (CTF), compiled under the direction of Anna
Mauranen and held at the Savonlinna School of Translation Studies. The
first article, by Mauranen herself, discusses in some detail the
related concepts of interference and transfer, and also describes the
compilation of the CTF. Mauranen tests the hypothesis that tolerance
of interference is higher when the translation is from a
language/culture that is more prestigious than that of the target
text. According to this hypothesis, argues Mauranen, translations from
English into Finnish should deviate more from original Finnish than
translations form Russian into Finnish. Mauranen measures the distance
between components of the corpus (Finnish originals, Finnish
translations from English, Finnish translations from Russian and
Finnish translations from several languages) by comparing frequency
ranked wordlists. The subset of Finnish originals is taken as a
standard and then the wordlists are compared to this standard. Despite
this being a rather crude method, which apparently has not been tested
before, some interesting patterns are observed. The results indicate
that translation from English actually differ less from original
Finnish than translations from Russian. However, because only
differences in rank order are considered, this method can only show
that there is a source language effect, without any indication of what
causes this effect or what kind of shifts it brings about. Linguistic
interference could be one explanation, but the influence of factors
such as those highlighted by Bernardini and Zanettin in the previous
article could also have been considered.

Eskola adds to the variety of terms used to explain regularities in
translation by proposing to talk about 'local laws' (instead of norms)
and 'univeral laws' (instead of universals). Eskola looks at whether
the use of optional non-finite syntactic structures in translated
Finnish is affected by the existence (or not) of equivalent structures
in the source language. The results indicate that translations from
English and Russian into Finnish do tend to under-represent target-
language specific linguistic features, which supports Tirkkonen-Condit
unique-item hypothesis (see below). Eskola's findings, in line with
Mauranen's, also show translations from Russian differing more from
original Finnish than translations for English.

The aim of Jantunen's study is to demonstrate that translations show
untypical lexical patterns and that these are influenced by the source
language. Jantunen looks at the distribution, collocation and
colligation of three nearly synonymous words in non-translated
Finnish, translations into Finnish from English and translations into
Finnish from different source languages. In all cases the normed
frequencies for each of the lexical items were higher in the
translation components than in the corpus of non-translated Finnish. A
chi-square test revealed a significant difference between the
non-translated Finnish and the subset of translations from several
languages, but not between non-translated Finnish and translations
from English or between the two translation components. Jantunen
attributes these results to a source language effect. As in Mauranen's
study, we see translations from English differing less from
non-translated Finnish than from translations from several source
languages. Still, the fact that the frequencies are much higher for
all the items in both translational components remains without
explanation. The patterns emerging from comparisons of collocational
patterns show each of the lexical items behaving rather different, and
the colligational patterns are again less clear and more complex. In
any case, Jantunen offers a good example of the kind of detailed
analysis needed in order to account for the complexity of the data in
these types of study.

The first paper in Part 3 (Testing the basics) again demonstrates that
translations have different lexico-grammatical patterns than original
tests by comparing the frequency and collocations of a grammatical
word (av) in Swedish translated and non-translated fiction. Nilsson's
study is very limited in its scope. Although the most typical patterns
found in translated texts are shown to be straightforward translations
of English structures, there is no attempt to explain why they are
less common in original texts (are other alternative options being
used more in originals?).

paissss article explores instances
of explicitation in translations into Hungarian. First, explicitating
shifts in translations from English into Hungarian are identified
manually by comparing source and target texts. Then the overall
frequency of five of the linguistic features (for example,
conjunctions) used in the explicitating shifts identified in the
translated texts are compared with their respective frequencies in a
corpus of original Hungarian. The results are not without interest but
the article is poorly written (and edited), which makes it difficult
to follow.

Puurtinen's study is based on the hypothesis that frequent use of
clause connectives as explicit signals of clausal relations in
translation might be a manifestation of explicitation. The corpus used
is a monolingual comparable corpus of children's literature, part of
the Corpus of Translated Finnish. The results do not support the
hypothesis: Puurtinen found no clear overall tendency in either the
translational or the non-translational components. A closer look at
the data reveals that in some cases the higher occurrence of the
connectives in the translations could be explained by the existence of
more or less straightforward equivalents in the source language

Tirkkonen-Condit tests what she calls the Unique-items Hypothesis,
according to which target-language items that have no straightforward
equivalent in the source language tend to be less common in translated
texts. The corpus is again the Corpus of Translated Finnish and the
unique items are verbs of sufficiency and clitic particles typical of
Finnish. As predicted, they are considerably more common in original
than in translated Finnish. Tirkkonen-Condit then discusses the
results in the light of other studies that showed similar tendencies
and that offer possible explanations.

The last part of the book contains only two articles. Kujumki also
explores the unique-items hypothesis but in a classroom environment.
Kujumki wrote a short text in Finnish containing three lexical items
(hanki, kilos and keli) that have no straightforward equivalent in
English or German, then had this text translated into those languages
by native speakers. The student's task was to translate the text back
into Finnish. Few used the three lexical items present in the original
text. A control test was then designed to confirm that the students do
use these concepts when describing situations similar to that
presented by the text of the exercise. The results suggest that when
translating students tend to opt for lexical choices that are closer
to the source language. Kujumki claims that these results challenge
the translation students' belief that theory has little relevance when
it comes to practice, and their strong belief in their L1
competence. However, it seems to me that if the low frequency of
unique-items in translation is such a widespread phenomenon as some of
the papers in this volume suggest, then it cannot be considered an
indication of deficiencies in L1 competence.

Jskelinen deals with a different 'universal', the avoidance of
repetitions. In this case, the task set to students was to translate a
text were repetition was used as a stylistic device in order to
emphasise certain aspects of the message. Some groups were provided
with a set of general instructions that called attention to stylistic
patterns while others received no specific instructions. Students that
had received instructions generally showed more sensitivity to
repetitions. In the conclusion Jskelinen reflects on her
methodology and on how the student's background may also affect how
they approach translation.


It is interesting to note that despite the warnings in Chesterman's
and Bernardini and Zanettin's papers about the limitations of the
corpus- based methodology and the need to ensure replicability, few
authors reflect on the methodology applied or open the ground for
questioning representativeness. In Eskola's article, for example, the
methodology is not described in any detail.

In general, this book provides a good overview of the state of the art
in research on universals in translation studies: there are some
interesting findings, and evidence is accumulating that confirms the
existence of patterns that cut across translation cultures, genres and
individual language pairs as well as patterns that reflect the
influence of each of these factors. However, the research is still
patchy and very much at surface level. Identifying patterns (however
untypical) and differences is only worthwhile if they can tell us
something about the nature of translations or about the cultural and
social imperatives that shape them. In some areas results are fairly
consistent, and in others less, but in general there are missing links
between one and the other, links that could probably be explained by
the variables at work in each case, but that have not been explored in
depth. As some scholars in this volume point out, there is a need to
complement research on the translation product with research that
delves more deeply into the translation process and context. The
introduction and the first part of the book point in this direction,
but most of the empirical studies reported fail to engage with these
issues. Maybe a good idea would have been to add a conclusion as well
as an introduction, to provide an overall picture summing up and
making connections between the findings, discussing potential
explanations and implications, and refining the existing hypothesis.

Papers presenting empirical findings generally work on the assumption
that universals are indeed possible, but the question is far too broad
to be addressed in small-scale studies as the ones reported. Even
though the question in the title was probably intended to open a
discussion rather than elicit concrete answers, those authors that do
address it from a theoretical perspective suggest that it is not
accurately formulated, because it is not the existence of universals
that is at stake. In Toury's words: ''the whole question of
translation universals is not one of existence [?] but one of
explanatory power'' (p. 29). In any case, as three of the papers
suggest, it seems that the term 'law' is probably a better choice to
reflect the complexity of the data and it seems to be less


Gabriela Saldanha holds an MPhil in Translation Studies from UMIST, UK
and is currently doing a PhD in the same area at Dublin City
University, Ireland, where she has also lectured on Corpus Linguistics
and Translation Technology. Her research interests include
Corpus-based Translation Studies, Corpus Linguistics, Stylistics,
Translation Technology, and Gender and Translation.
 explicitation as a
potential universal in translation. P

More information about the Linguist mailing list