LINGUIST List 13.1852

Thu Jul 4 2002

Review: Linguistic Theories: Mel'cuk Igor (2001)

Editor for this issue: Dina Kapetangianni <dinalinguistlist.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. Andrea Sanso', Mel'cuk (2001) Linguistic Theory: Communicative Organization in Natural Language

Message 1: Mel'cuk (2001) Linguistic Theory: Communicative Organization in Natural Language

Date: Wed, 03 Jul 2002 14:41:51 +0000
From: Andrea Sanso' <sansoarno.humnet.unipi.it>
Subject: Mel'cuk (2001) Linguistic Theory: Communicative Organization in Natural Language

Mel'cuk, Igor (2001) Communicative Organization in Natural Language:
The Semantic-Communicative Structure of Sentences. 
John Benjamins Publishing Company, xii+393pp, hardback 
ISBN 90 2723060 9 (Eur) / 1 58811 101 6 (US), $86.00, 
Studies in Language Companion Series 57

Book Announcement on Linguist:
http://linguistlist.org/get-book.html?BookID=2280 

Andrea Sanso', University of Pavia, Italy

OVERVIEW

Mel'cuk's book addresses a long-standing problem in linguistic theory,
that of communicative organization, from a very specific
perspective. The author's intended aim is to show how actual
utterances are constructed from a representation of their meaning
within a Meaning-Text Theory framework (cf. Mel'cuk 1981). The use of
theory-specific terminology, however, is kept to a minimum, so that
the concepts presented throughout the book are accessible to a wide
audience.

According to the author, the main advantages of the Meaning-Text
framework are systematicity and terminological rigor. In Melcuk's own
words, "I do not try to develop new theories, or present new facts,
but rather propose some formal means (simply put, systems of symbols)
that, if properly used in the corresponding representations, ensure
the construction of the sentences I am interested in" (p.2).

CONTENTS 

Unfortunately, the subtleties of the notational system proposed by
Mel'cuk cannot be fully appreciated in a plain ASCII text. Therefore,
I will limit myself to a very general sketch of the contents of the
book, which should act as a stimulus to read the book.

The introduction is devoted to a general discussion of the book's
aims. A general problem faced within the domain of linguistic studies
known as "information packaging" is terminological idiosyncrasy: there
is little or no overlap between the terminology used by different
schools, and the result is that notions such as, say, Theme or Topic
are rather vague, or, to put it more positively, at least
multifaceted. Mel'cuk's primary concern is thus to get rid of these
terminological and notional mess. Two basic notions are thus
introduced, namely the Semantic Structure (henceforth SemS) and the
Deep-Syntactic Structure (henceforth DSyntS) of a sentence: the former
exclusively represents propositional meaning, and defines a family of
more or less synonymous utterances, whereas DSyntS represents the
lexico-grammatical organization of a sentence.

In order to describe all the properties of a sentence that have to do
with its communicative aspect, the notion of Semantic-Communicative
Structure (Sem-CommS) is introduced in Chapter I. Sem-CommS is called
upon in the process of synthesizing a sentence S from its
SemS. Sem-CommS thus characterizes S as a message (i.e., as an answer
to an underlying question), or, in other words, it organizes the
meaning of a SemS from the point of view of its transmission by the
speaker and its reception by the hearer. This operation is analogous
to the process of profiling in Cognitive Grammar, as Mel'cuk himself
points out. Sem-CommS is not an atomic entity. Rather, it can be
described by eight logically orthogonal axes, or oppositions
(a.k.a. discourse functions or pragmatic functions in the linguistic
tradition).

The eight dimensions of the communicative organization of a sentence
are discussed in detail in Chapter II. These oppositions are: (i)
thematicity; (ii) givenness; (iii) focalization; (iv) perspective; (v)
emphasis; (vi) presupposedness; (vii) unitariness; and (viii)
locutionality. They are thoroughly discussed along with the main
linguistic phenomena that instantiate them (dislocation, clefting,
etc.). I will limit myself to a sketchy presentation of the
aforementioned oppositions. I will try to use Mel'cuk's idiosyncratic
terminology as rarely as possible, for this would imply a detailed
presentation of the terminology itself, and this is beyond the scope
of the present review.

(i) Thematicity, or the THEME-RHEME opposition. This is the most
universal and relevant among the Sem-Comm oppositions. Whenever
speakers intend to transmit a piece of information, they cannot avoid
thematicity, i.e. the division of the initial meaning into Rheme and
Theme. In a sentence such as "John will bring the booze" 'John' is the
Theme and 'will bring the booze' is the Rheme. The Theme-Rheme
relationship is reversed in the passive sentence "The booze will be
brought by John".

(ii) Givenness, or the GIVEN-NEW opposition. This opposition is
relevant within the limits of a coherent discourse, because it
reflects the contextual boundness of sentential semantic
elements. This opposition is more "semantic" than "communicative":
when the speaker selects his Theme and Rheme, he does so because he
wants to communicate some information about some specific item. Thus,
while Thematicity is a speaker-oriented category, Givenness has to do
with the mental state of the Hearer at the moment when the sentence is
uttered: the Speaker singles out some items in his initial SemS which
he believes are in the active zone of the Hearer's consciousness. In
the sentence "The book is on the table", both "book" and "table" are
given, whereas in the sentence "There is a book on the table" "book"
is new.

(iii) Focalization, or the FOCALIZED-NON FOCALIZED opposition.
Focalized is that part of the SemS which the Speaker presents as being
logically prominent for him. A typical example of a focalized element
is 'John' in the sentence "It was John who brought the booze". John is
also the Rheme of this sentence, i.e. this is a typical case of a
focalized Rheme. On the other hand, in "The booze was brought by John"
'John' is the Rheme without being focalized.

(iv) Perspective or the FOREGROUNDED-BACKGROUNDED opposition. This
opposition deals with the psychologically primary/secondary character
of a chunk of meaning, again from the viewpoint of the Speaker. A good
illustration of backgrounded chunks are parenthetical
expressions. Possessor Raising is an instance of foregrounding, and
reflects the importance that natural languages attach to human
possessors.

(v) Emphasis, or the EMPHASIZED-NEUTRAL opposition. This opposition
is related to the emotional load the Speaker attaches to a given
meaning. It is responsible for marking contrast and insistence, or a
high level of the Speaker's feelings (irony, anger, amazement). 'John'
(pronounced with rising intonation) is the emphasized element of the
sentence "JOHN drinks gin, it is Peter who drinks whiskey". Note that
a focalized element needs not be also emphasized: one can pronounce a
focalized element in a neutral tone of voice, without any emphasis.

(vi) Presupposedness, or the PRESUPPOSED-NON PRESUPPOSED opposition.
Presupposed is that part of the SemS which the speaker presents as
taken for granted, in the sense that, if the whole SemS is negated or
questioned, that part remains affirmed. When I negate the whole
meaning of the sentence "This lazy guy is here" by saying "This lazy
guy is not here", I'm still saying that the guy is lazy.

(vii) Unitariness, or the UNITARY-ARTICULATED opposition. This
opposition concerns the way in which the speaker presents a complex
event: as one single phenomenon or as a sequence of several
phenomena. The decision to unitarize or to articulate the situation
being described mainly depends on the Speaker's choice. The more
typical a situation is, the higher the probability that the Speaker
will unitarize it.

(viii) Locutionality, or the SIGNALED-PERFORMED-COMMUNICATED
opposition. This opposition distinguishes between two types of
utterances: (i) utterances that are meant to communicate something and
(ii) utterances that are meant simply to signal something or to
perform an action. For instance, if we have the meaning 'it hurts me',
we can use a communicating utterance such as "it hurts" or a signaling
utterance such as "Ouch!".

All those linguistic examples that are not strictly necessary to the
discussion in Chapter 2 are dealt with in Chapter 3. This chapter is
by and large the most appealing part of the book, since the complexity
of the notation system proposed in Chapter II is fully exploited to
address a number of linguistic examples chosen among the most debated
cases in the literature.

CRITICAL EVALUATION

I decided to review Mel'cuk's book after coming across his "Cours de
Morphologie Generale" (Mel'cuk 1994; see also Mel'cuk 2001). In that
thought-provoking book, Mel'cuk provided a brilliant notational system
for morphological categories. This book is similar to Mel'cuk 1994 in
that both underscore the importance of terminological and notional
rigor. If one is willing to accept the author's rather idiosyncratic
terminology, Mel'cuk's approach to communicative organization is
another important contribution to our thinking about the need for
crystal-clear definitions when working with linguistic phenomena and
their dimensions of variation. It is rather difficult for a reviewer
to find points for negative criticism. To sum up, the book is to be
recommended to anyone working in the domain of Information Packaging
from any perspective. It will not fail to stimulate both
functionalists/typologists and formal linguists to get rid of
terminological and notional vagueness when addressing the problem
of what is communicated through natural language utterances.

REFERENCES

Mel'cuk Igor. 1981. "Meaning-Text Models: A recent trend in Soviet
linguistics". Annual Review of Anthropology 10: 27-62.

Mel'cuk Igor. 1994. Cours de morphologie generale (theorique et
descriptive). Montreal: Les Presses de l'Universite de Montreal, CNRS
editions.

Mel'cuk Igor. 2001. "A formal language for linguistic morphology
(toward a coherent notional system)". Paper presented at the 37th
meeting of the Chicago Linguistic Society, April 2001.

ABOUT THE REVIEWER

Andrea Sanso', after having graduated in Linguistics from the
University of Pisa, attended a PhD programme on Linguistics at the
University of Pavia. During this course, he carried out some research
about passive constructions in Italian and Spanish in a
cognitive-typological framework, which resulted in the Dissertation
"Passive and elaboration of events: A case atudy from Italian and
Spanish". His research interests cover Linguistic Typology, Semantics,
Pragmatics, Discourse, Cognitive Linguistics.








Mail to author|Respond to list|Read more issues|LINGUIST home page|Top of issue