LINGUIST List 14.1485

Thu May 22 2003

Review: Computational Ling: Hasselgard, et al (2002)

Editor for this issue: Naomi Ogasawara <naomilinguistlist.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 Simin Karimi at siminlinguistlist.org.

Directory

  1. Anna Maria De Cesare, Information Structure in a Cross-Linguistic Perspective

Message 1: Information Structure in a Cross-Linguistic Perspective

Date: Thu, 22 May 2003 10:08:41 +0000
From: Anna Maria De Cesare <decesaremidway.uchicago.edu>
Subject: Information Structure in a Cross-Linguistic Perspective

Hasselgard, Hilde, Johansson, Stig, Behrens, Bergljot,
Fabricius-Hansen, Cathrine eds. (2002) Information Structure in a
Cross-Linguistic Perspective. Rodopi, xiii + 228 pp., Hardback ISBN 90
420 1469 5, 55.00 Eur, 61.00 USD

This book is vol. 39 in the collection ''Language and Computers:
Studies in Practical Linguistics'', edited by Jan Aarts and Willem
Mejis

Announced at http://linguistlist.org/issues/14/14-426.html


Anna-Maria De Cesare, University of Chicago

Purpose of the book

The book ''Information Structure in a Cross-Linguistic Perspective''
presents the results of the Symposium on Information Structure in a
Cross-linguistic Perspective held at the University of Oslo in
2000. Its intent is to discuss 'information structure', which is, as
the editors claim, a rather vague concept that can be applied to many
different areas of linguistic description. Information structure can
be studied at the level of the phrase, the clause, the text, or the
context. This concept is essential in text production, since ''it
involves the way in which a speaker/writer chooses to present a
message in terms of given/new information, focus, cohesion, and point
of view'' (ix).

With this book, the editors not only present current research in the
field of information structure but also hope that the readers ''will
be inspired to give more thought to this problematic, but
linguistically challenging area of cross-linguistic study'' (xiii).

Book Content

The book begins with a brief introduction by the editors in which the
goals and structure of the book are presented and defined
(ix-xiii). The book consists of 13 papers, which ''do not converge on
a common topic'' (xiii). The papers, however, have been organized
according to their topics and approaches. As the editors note
''Information structure has been approached in a variety of ways
... Some of the articles may be seen as direct attempts to define
text-linguistic constraints on sentence structuring across languages''
(x).

In the first paper, ''Interpreting concessive adverbial markers in
English and Norwegian discourse'' (1-19), Thorstein Fretheim compares
concessive markers in English and Norwegian. Specifically, Fretheim
considers the fact that Norwegian 'likevel' covers the whole range of
functions of English markers, including 'nevertheless/nonetheless',
'even so', 'after all'. In order to account for the polyfunctionnality
of 'likevel', Fretheim argues that 'likevel' has a univocal meaning
and that its context-dependent interpretation involves enrichment as
proposed in Sperber and Wilson's Relevance Theory. In the second part
of the paper, Fretheim discusses the scope of the concessive marker
'for det' and shows that, contrary to the established idea that
concessive markers take a wide scope, 'for det' does not. It is inside
the scope of the negation.

The goal of Bengt Altenberg's paper ''Concessive markers in English
and Swedish'' (21-41) is to define what relations can be identified in
a corpus-based study of concessive markers involving English and
Swedish. The paper compares the types and functions of adverbial
concessive connectors in the two languages on the basis of
'bidirectional translation data' from the English-Swedish Parallel
Corpus. The comparison proposed involves three steps, based on the
'Semantic Mirror approach' (55) devised by Helge Dyvik at the
University of Bergen (1998). Altenberg identifies and discusses the
Swedish translation of 'yet', the English equivalents of Swedish
'anda', and finally the Swedish equivalents of 'after all', 'anyway'
and 'at least'. This new strategy allows Altenberg to identify the
main paradigm of concessive connectors in both languages.

Bergljot Behrens and Cathrine Fabricius-Hansen, in their paper
''Connectives in contrast: A discourse semantic study of Elaboration
based on corpus research'' (46-61), present primary conclusions of a
research project of the theoretical as well as practical questions
related to translation studies, and specifically the concept of
'elaboration'. As in the previous article, the focus of interest is on
connectives and the way they are translated. This paper provides a
Corpus-based approach aiming at identifying some structural
differences between English, German and Norwegian. The 'Semantic
Mirror approach' is used again in this paper.

Carlota S. Smith's paper, ''Perspective and point of view: Accounting
for subjectivity'' (63-79), is theoretical in nature as it discusses
the fundamental categories that contribute to
'subjectivity'. Specifically, Smith attempts to sort out the different
uses of the terms 'point of view' and 'perspective'. Two main uses of
these concepts are distinguished: expressions of 'communication',
'content of mind' and 'evaluation', which convey 'point of view', and
expressions of 'perception' and 'perspective', which convey
'perspective'.

Ralph Salkie's contribution, ''Probability and necessity in English
and German'' (82-95), considers modality in English and German and
focuses specifically on modal verbs. Salkie begins by considering two
recent analyses of the English terms 'should' and 'must', and asks
whether they can be applied to German modals. His goal is to provide a
framework for modality, which combines ''the detailed empirical
coverage of monolingual analyses with the cross-linguistic validity of
broader studies'' (82). This paper employs a corpus-based approach to
compare translation equivalents of English and German.

Karin Aijmer's essay analyzes ''Modal adverbs of certainty and
uncertainty in an English-Swedish perspective'' (97-112). Her study
focuses on 'surely', 'certainly' and 'no doubt' and employs the
Parallel corpus of the Oslo Multilingual Corpus. Aijmer acknowledges
that expressions of certainty in English and Swedish can also express
uncertainty. The translations in the Oslo Multilingual Corpus are then
used as a tool to disambiguate the meanings of ''strong and week
certainty'' in the source language. The methodology employed is again
the 'Semantic Mirror approach' (first the Swedish equivalents of
'surely', 'certainly' and 'no doubt' are provided and discussed, then
the English equivalents of 'sakert'). Finally, Aijmer provides an
account of the polyfunctionnality of the modal adverbs analyzed in
terms of their grammaticalization.

Jeanette Gundel's article discusses ''Information structure and the
use of cleft sentences in English and Norwegian''
(113-128). Specifically, on the basis of the analysis of a section of
a Norwegian novel and its English translation, Gundel demonstrates
that cleft sentences are more commonly found in Norwegian than in
English. Gundel argues that this difference arises from a stronger
tendency to map information structure directly onto syntactic
structure in Norwegian. Consequently, she claims that this difference
cannot be accounted for by a difference in discourse distribution or
by structural properties of the cleft sentences in the two languages.

Geert-Jan Kruijff's paper, ''Formulating a category of informativity''
(129-145), is typological in nature. Its goal is to formulate the
first steps towards an account of information structure throughout
languages. Specifically, it formulates several typological hypotheses
that predict when languages use word order, tune, or a combination of
both to realize information structure. The paper provides examples of
a wide variety of languages and language groups, including English,
Turkish, Japanese and Hungarian.

In ''Contrast - from a contrastive perspective'' (147-161), Valeria
Molnar discusses the link between the concept of 'contrast' and two
major notions related to information structure, 'topic' and
'focus'. In this article, she argues that 'contrast' should be treated
as an autonomous concept of information structuring rather than as a
feature of topicality and focusing. According to Molnar, 'contrast'
must be established ''as a further category of information structure,
superimposed on topic and focus'' (160).

In her contribution ''Accent and the notion of contrast: A
cross-linguistic approach'' (163-178), Jorunn Hetland discusses
'contrast' from an intonational perspective. This paper demonstrates
how contrast is related to pitch accent, and in particular the
'fall-rise' accent. On the basis of English, German, Hungarian and
Korean data, Hetland argues that only the 'fall-rise accent' can
express a contrast in a context in which no trace of contrast is
present. The fall-rise accent alone signals that a constituent belongs
to a certain set from which it is singled out.

Christiane von Stutterheim, Ralf Nuese and Jorge Murcia-Serra's paper,
''Cross-linguistic differences in the conceptualization of events''
(179-198), reports on a series of empirical studies in which
language-specific patterns in the construal of events are
investigated. Specifically, this paper demonstrates how German,
English and Spanish, three languages varying with regard to the
category of 'aspect' (English and Spanish having grammaticised
aspectual categories), verbalize an event through different
elicitation tests. The authors argue that the differences observed in
the elicitation tests are a consequence of the different grammatical
properties of the languages investigated.

The paper ''Maintenance and convergence in covert translation
English-German'' (199-211), by Juliane House, presents results from a
research project currently under way at the University of
Hamburg. This project investigates whether German written textual
norms are adapting to American and British norms, following the
important impact on German of English lexical items borrowing. One
possible change entails ''Anglicization'' of the information structure
or word order. On the basis of the comparison of 30 German and English
translational pairs, House demonstrates that the structure of both
languages is still quite different with regard to information
structure. According to House, however, some changes, possibly related
to the interpersonal functional component, seem to be under way.

Finally, Erich Steiner's paper, ''Grammatical metaphor in translation
- some methods for corpus-based investigations'' (213-228), focuses on
the textual properties of translations, especially in the
English-German language pair. The paper first recognizes that an
additional property of translated texts is 'grammatical
de-metaphorization', which implies rewording and rendering explicit
implicit information. This unpacking of grammatical metaphor in turn
has an effect on the information structure of the translated text. In
the second part of the paper, Steiner suggests quantitative methods to
test the de-metaphorization property of translation texts.

Discussion

The book ''Information Structure in a Cross-Linguistic Perspective''
is an important contribution to the linguistic literature, as it
investigates the field of information structure through corpus-based
approaches. The book contains papers from some of the most important
researchers in the field of both Corpus Linguistics and Information
Structure and presents results of research projects that are mostly
under way.

The significance of the book stems from the cross-linguistic
perspective on information structure, and the various methodologies
that the authors employ. First, the authors use a variety of
materials. Some contributors employ authentic corpus material, and in
particular material from a parallel corpus (such as the Oslo
Multilingual Corpus), while others primarily analyze created examples
(Kruijff). The authors also use a variety of methods in discussing
information structure in a cross-linguistic perspective. The three
papers of Altenberg, Behrens / Fabricius-Hansen and Aijmer, for
instance, present groundbreaking methods based on a bidirectional
translation corpus. I find the 'Semantic Mirror approach', the
technique devised by Dyvik (1998), to be particularly fruitful. The
use of bidirectional translation corpus and the 'Semantic Mirror
approach' that such a corpus provides, ''open up new possibilities of
research'' (Altenberg) that will cast light on phenomena that are
related to information structure.

While the book ''Information Structure in a Cross-Linguistic
Perspective'' is significant and useful in a variety of ways, my main
reservation is that some of the papers address the question of
information structure only indirectly. I found this to be the case
with the papers of Altenberg, Smith, Aijmer, Salkie, and
Steiner. Additionally, although some of the other papers deal
extensively with information structure, they do not adopt a
corpus-based approach as they primarily use created examples (see for
instance Kruijff, Molnar). I was expecting more papers that do both.

Finally, it seems to me that the discussion of information structure
was sometimes quite insensitive to the differences between spoken and
written data. Although the papers included in the book were primarily
focused on the spoken language (see Gundel, p. 113), the analyses of
the papers rely upon either written or constructed data
(Hetland). With a few exceptions (such as House), the papers included
in the book do not make their choice of code explicit. It is only
through the examples given and the factors mentioned (intonation, for
instance), that I was able to understand which code the paper
analyzed. This distinction is important given the differences between
spoken and written texts. For example, the phonological means of
marking the focus is typical of spoken texts, and does not at all have
the same importance in written texts. One would therefore expect
spoken and written texts to show differences in information structure,
and thus in both word order and markers used to identify the topic,
focus, etc. of a sentence.

To conclude, I highly recommend the book ''Information Structure in a
Cross-Linguistic Perspective'' to anyone interested in information
structure, corpus linguistics and contrastive linguistics. It includes
some very interesting results of the latest linguistic research
comparing two or more languages. Moreover, the book presents some new
and very promising techniques for analyzing text, text structures, and
information structure, which are based on authentic corpus material.

References

Dyvik, H. (1998), A Translational basis for semantics, in S. Johansson
and S. Oksefjell (eds.), Corpora and Crosslinguistic Research: Theory,
Method, and Case Studies, Amsterdam: Rodopi, 51-86.

ABOUT THE REVIEWER

Anna-Maria De Cesare holds a Ph.D. in linguistics from the University
of Geneva, Switzerland and is currently a Visiting Scholar in the
Department of Romance Languages and Literatures at the University of
Chicago. Her academic interests include lexical semantics (adverbs and
particles), contrastive linguistics (Italian compared to French and
English) and corpus linguistics. She is currently working on a
specialized Italian-English dictionary of adverbs and particles.
Mail to author|Respond to list|Read more issues|LINGUIST home page|Top of issue