Studies in Text Grammar

360 Pages·1973·11.117 MB·Foundations of Language 19

by  János S. Petöfi, Hannes Rieser (auth.), J. S. Petöfi, H. Rieser (eds.)

STUDIES IN TEXT GRAMMAR
FOUNDATIONS OF LANGUAGE
SUPPLEMENTARY SERIES
Editors:
MORRIS HALLE, MIT
P. HARTMANN, Konstanz
K. KUNJUNNI RAJA, Madras
BENSON MATES, Univ. of California
J. F. STAAL, Univ. of California
PIETER A. VERBURG, Groningen
JOHN W. M. VERHAAR (Secretary), Jakarta
VOLUME 19
STUDIES
IN TEXT GRAMMAR
Edited by
J. S. PETOFI and H. RIESER
D. REIDEL PUBLISHING COMPANY
DORDRECHT-HOLLAND / BOSTON-U.S.A.
Library of Congress Catalog Card Number 73-75766
ISBN-13: 978-94-010-2638-3 e-ISBN-13: 978-94-010-2636-9
DOl: 10.1007/978-94-010-2636-9
Published by D. Reidel Publishing Company,
P.O. Box 17, Dordrecht, Holland
Sold and distributed in the U.S.A., Canada, and Mexico
by D. Reidel Publishing Company, Inc.
306 Dartmouth Street, Boston,
Mass. 02116, U.S.A.
All Rights Reserved
Copyright © 1973 by D. Reidel Publishing Company, Dordrecht-Holland
Softcover reprint ofthe hardcover 1st edition 1973
No part of this book may be reproduced in any form, by print, photoprint, microfilm,
or any other means, without written permission from the publisher
TABLE OF CONTENTS
PETER HAR TMANN / Foreword IX
JANOS S. PETOFI and HANNES RIESER / Overview 1
TEUN A. VAN DIJK / Text Grammar and Text Logic 17
1. Introduction 17
2. The Hypothetic Form of Text Grammar 19
3. Formal Logic and Natural Logic 21
4. Text Logic 28
4.1. Introduction 28
4.2. Proofs, Systems and Texts 31
4.3. The Basic Apparatus 33
4.4. Quantification and Identification 35
4.5. Identity 42
4.6. Partial Identity, Inclusion and Elements of Sets 50
4.7. Intensional Coherence 51
4.8. Intensional Identity, Predicates 53
4.9. Propositions, Presuppositions, Consequences, Connectives 56
4.10. A Provisional List of Derivational Principles of Text Logic 64
4.11. Example of Natural Derivation 66
5. Summary 73
Bibliography 74
IRENA BELLERT / On Various Solutions of the Problem of Presup-
positions 79
WERNER KUMMER / Pragmatic Implication 96
1. Elements of a Pragmatic Language 96
2. Truth Conditions for Formulas with Series of Epistemic Operators
with Alternating Subscripts 99
3. Pragmatic Implication 101
4. Other Types of Pragmatic Implications 107
5. Summary 112
WOLFRAMK. KOCK / Time and Text: Towards an Adequate Heuristics 113
1. Preliminaries 113
2. Note on the 'Meta-Theoretical Paradigm' 125
VI TABLE OF CONTENTS
3. Brief Sketch of a Model of Language Functioning 133
3.1. The Anthropo-Cybernetic Model of the System 'Man-World' 133
3.2. Language 146
3.3. Text 160
3.4. Time 170
4. TimeandText 176
5. Concluding Remarks 200
Bibliography 201
JANOS S. PETOFI / Towards an Empirically Motivated Grammatical
Theory of Verbal Texts 205
I. SENTENCE GRAMMARS AND TEXT GRAMMARS 205
O. Introduction 205
1. State of the Grammatical Theory of Verbal Texts 205
2. Some General Questions Concerning the Set-Up of Sentence
Grammars 207
3. Some General Questions Concerning the Set-Up of Text Grammars 216
II. A 'NOT FIXED LINEARITY TEXT GRAMMAR'. THE PRESENT
ST AGE OF ITS DEVELOPMENT 222
O. Introduction 222
1. The Formation Rule System 226
2. The Transformation Rule System 237
3. On the Structure of the Lexicon 243
4. The Algorithm for the Analysis of Texts 255
5. The Algorithm for the Synthesis of Texts 256
6. The Algorithm for the Comparison of Texts 268
7. Concluding Remarks 269
Bibliography 273
HANNES RIESER / Sentence Grammar, Text Grammar, and the Eval
uation Problem. Some Remarks Concerning the Theoretical Foun-
dation and the Possible Application of Text Grammars 276
1. Some Remarks on the M eta-Theoretical Postulates and Conventions
to Be Used 278
2. Some Informal Remarks on the Structure of Gd1, i.e. the Phrase
Structure Grammar (PSG) to Be Used 279
3. Specification of the Categories Used 281
3.1. Main Categories of VNT 281
3.2. Sub-Categories 281
3.3. Numerical and Other Indices 281
3.4. Semantic Relations 282
T ABLE OF CONTENTS VII
4. A Sentence Grammar G Generating a Set of Sentences (a Lan-
dt
guage L~ of which the Sentence SAT[IMP](1)~ is an element 282
4.1. Syntactic Rules 282
4.2. Semantic Rules 283
4.3. Lexical Entries with Specified Readings 284
4.4. An Applicability Condition for Rule (*14) 285
5. Some Arguments for the Delimitation ofa Sub-Grammar jG Based
dt
on Syntax and Semantics, i.e. a Grammar Enumerating only Sen
tences (Norms, Directives) Belonging to the German Language of
Jurisdiction 285
6. On the Derivation of Synonymous and Hyperonymous Sentences
by Grammars of the Type jG 291
dt
7. Some Remarks on the Evaluation of jG 295
dt
Bibliography 298
JENS IHWE / On the Validation of Text-Grammars in the 'Study of
Literature' 300
Abstract 300
O. Preface 300
1. The Validation ofT ext-Grammars in the Study ofL iterature 301
2. The Empirical Content of the Study of Literature 310
3. Reconstruction of the Text Concept 332
Bibliography 347
FOREWORD
If we consider how theoretical operations belonging to the methodological
inventory of linguistics are carried out (i.e. the way linguistic theories are
set up), three main criteria suggest themselves for classifying them:
(1) Both, nature and type of the aims of the scientific knowledge applied
which allow to specify the epistemological interests as well as the theoretical
impact constituting the purpose of linguistic operations;
(2) the nature of the intellectual procedures in connection with which a set
of intersubjectively acceptable operations should guarantee that current
postulates of the theory of science be maintained;
(3) the set of data serving as an empirical basis for the theories to be estab
lished on the one hand and as a correlate for the further development, the
testing and the evaluation of theories on the other hand.
It is to be considered a basic concept (as well as a motive) of current text
linguistic research that due to the linguistic analysis of discourses a further
development of linguistics has set in or is still to be achieved as regards the
three criteria mentioned above.
Therefore, if we want to estimate text-linguistic approaches (or concepts),
works (methods), or knowledge (results) we should take the view allowing
for the general valuation of the linguistic discipline or one of its sub-disciplines.
This should be done with respect to the contributions gathered in this volume
as well.
Considering that the different steps and research-programs should be
judged above all according to their global perspective, such a suggestion does
not seem to be out of place. The following passages try to make this perspec
tive clearer and to specify it by way of investigating the criteria mentioned
more closely, namely the aims, the set of theoretical operations, and the
problems of the empirical basis.
It has been mentioned already that the linguistic analysis of texts changes
the aims of linguistics as regards their nature and their type. Granted that
this assumption can be accepted, this implies the following: the type of
epistemological interest changes because the linguist now deals with objects
(i.e. discourses) which have to be considered embedding structures (frames)
for linguistic phenomena constituting the domain of the various sentence
grammatical approaches. The fact that discourses themselves may be de
scribed by means of theoretically determined complex embedding operations
x FOREWORD
is, of course, no valid objection against the textlinguistic approach. It is quite
clear, however, that the epistemological interest is now concentrated on
objects and relations between objects of a much more inclusive kind. There
fore the categories of the structural descriptions used, the rules of grammar,
and the corresponding diagrams try to mirror the structural properties of
these objects and the relations between them.
From this it follows that the type of epistemological interest changes as
well: objects (i.e. sets of discourses) have to be treated, the description of
which requires more dimensions than sentence-grammatical descriptions.
This is due to the fact that the linguistic construct 'text' denotes the structure
and the manifestation of linguistic objects observed in verbal communica
tion. Only those objects which are not reduced by the various abstracting or
idealizing methods of scientific analysis show all the various functions tied
up with language in its ordinary use. Precisely these functions may then be
considered as the legitimate objects of linguistic investigation and theoretical
reconstruction.
This 'multi-level-dimensionality' of the text-analytic tools to be developed
corresponds closely to the plurality of disciplines involved, if pragmatic and
other functions of discourses should receive an adequate theoretical treat
ment. Some characteristics of this development may be inferred from the fact
that logico-semantic languages are used for the description of the semantics
of discourses, that the semantic analysis of discourses is - at least some
times - linked up closely with the theory of literature and, that the interest
in rhetoric, psychology etc. is still growing. It should be taken into considera
tion, however, that the participation of the various disciplines mentioned
rests upon the delimitation of disciplines established and accepted thus far,
i.e. disciplines like linguistics, logics, psychology, theory of literature etc.
Especially two characteristics are closely linked up with this development
in linguistics as outlined above: In the first place it is more and more rec
ognized that elaborate linguistic operations play a basic role as regards
the semantic analysis of discourses, therefore the interest in developing
adequate heuristic procedures is still going on. Discourses are regarded as
objects the complex treatment of which has to be developed before the
process of theoretical idealization (reduction, formalization) can be carried
out successfully. In other words: Sound hypotheses about the complex prop
erties of the set of objects/phenomena in question have to be established
first, then it will be possible and reasonable to select subsets constituting the
main (although preliminary) domain of scientific investigation. In the second
place a growing interest for the different domains ofa pplication of text-gram
matical research can be observed. The reason for this is the fact mentioned
above that language is normally used in communication, discourses are there-
FOREWORD XI
fore to be considered as the primarily observable empirical data. Consequent
ly, their investigation will be of immediate relevance for solving the basic
theoretical questions of an empirically founded pragmatics.
The changes within the domain of operations used in linguistic methodolo
gy closely correspond to the current trends of the development of the lin
guistic intentions as it has been briefly outlined above. Roughly, intellectual
procedures of scientific importance may be classified into procedures belong
ing to the proper treatment of the objects in question and into procedures leading
to the controllable acquiring of empirically founded results, in other words,
into analytic (mainly inductive) procedures and (mainly deductive) procedures,
allowing for the set-up of hypotheses to be tested and evaluated such that some
decision as to the empirical validity, the equivalence of theories etc. can be
achieved. The first group is intended to serve the aims of structural analysis,
it contains operations such as segmentation and identification, the com
parison of objects, classification, the set-up of classes of relational properties
etc. The second group permits the description and systematization of the
outcome of the structural analysis by means of some well-defined descrip
tional or theoretical language, it contains operations such as symbolizing,
the proper set up of hypotheses, of definitions, the classification of hypotheses
according to the theoretical status of the descriptional language they are
formulated in, the rules of inference and deduction to be used etc. From the
methodological point of view the theoretical status of the operations enu
merated clearly remains the same, it should be noticed, however, that the
domain is changed: text-linguistic research requires classificatory procedures
and operations different from those applied in sentence-grammatical re
search. Above all, classes of elements and semantic functions of a different
kind have to be discovered and to be formulated, e.g. the semantic relations
between sentences, sequences of sentences and discourses. In addition to this,
operations originally belonging to the methodological inventory of other
disciplines are used, e.g. operations adopted from logics, psychology, sociol
ogy, communication theory, cybernetics etc. In this connection such ques
tions have to be solved as whether a specifically set up text grammar is
necessary, and if so, how it should be established, how textual coherence
should be described and explained, which classes (e.g. syntactic, semantic
referential, ontological) it is determined by etc.
The reason for the changes mentioned, which partly even necessitate
methodological revaluations and improvements, may be seen in the fact that
due to the linguistic analysis of discourses the empirical basis of linguistics
has been changed, more precisely, expanded. Therefore not only theoretical
statements, i.e. statements in some theoretical language based upon the ex
panded set of objects under investigation, have to be developed and evaluat-