Word Grammar New Perspectives on a Theory of Language Structure. docx

Word
Grammar
New
Perspectives
on a
Theory
of
Language Structure
This page intentionally left blank
Word
Grammar
New
Perspectives
on a
Theory
of
Language Structure
edited
by
Kensei Sugayama
and
Richard
Hudson
continuum
Continuum
The
Tower Building
15
East 26th Street
11

York Road
New
York
London
SE1 7NX NY
10010
All
rights reserved.
No
part
of
this publication
may be
reproduced
or
transmitted
in
any
form
or by any
means, electronic
or
mechanical, including photocopying,
recording,
or any
information storage
or
retrieval system, without
prior
permission

in
writing
from
the
publishers.
First
published
2006
British Library Cataloguing-in-Publication Data
A
catalogue
record
for
this
book
is
available
from
the
British Library.
ISBN:
0-8264-8645-2 (hardback)
Library
of
Congress Cataloguing-in-Publication Data
To
come
Typeset
by
BookEns

Ltd,
Royston,
Herts.
Printed
and
bound
in
Great
Britain
by MPG
Books
Ltd,
Bodmin,
Cornwall
© Kensei Sugayama
and
Richard Hudson 2005
The
problem
of the
word
has
worried general linguists
for the
best part
of a
century.
-P.
H.
Matthews

This page intentionally left blank
Contents
Contributors
Prefac e
Kensei
Sugayama
Introduction
1.
What
is
Word
Grammar?
Richard
Hudson
1.
A
Brief Overview
of the
Theory
2.
Historical Background
3. The
Cognitive Network
4.
Default
Inheritance
5. The
Language Network
6.
The

Utterance Network
7.
Morphology
8.
Syntax
9.
Semantics
10.
Processing
11.
Conclusions
Part I
Word
Grammar Approaches
to
Linguistic Analysis:
Its
explanatory power
and
applications
2.
Case Agreement
in
Ancient Greek:
Implications
for a
theory
of
covert
elements

Chet Creider
and
Richard Hudson
1.
Introduction
2.
The
Data
3. The
Analysis
of
Case Agreement
4.
Non-Existent Entities
in
Cognition
and in
Language
5.
Extensions
to
Other Parts
of
Grammar
6.
Comparison
with
PRO and pro
7.
Comparison

with
Other PRO-free Analyses
8.
Conclusions
3.
Understood Objects
in
English
and
Japanese with
Reference to Eat and Taberui A Word Grammar
accoun t
Kensei
Sugayama
1.
Introduction
2.
Word
Grammar
xi
xiii
1
3
3
5
7
12
13
15
18

21
24
27
28
33
35
35
35
41
42
46
49
50
52
54
54
56
viii
WORD
GRAMMAR:
PERSPECTIVES
ON
LANGUAGE STRUCTURE
3. Eat in
English
4.
Taberu
in
Japanese
5.

Conclusion
4.
The
Grammar
of Be To:
From
a
Word
Grammar
point
of
view
Kensei
Sugayama
1.
Introduction
and the
Problem
2.
Category
of Be
3.
Modal
Be in
Word
Grammar
4.
Morphological Aspects
5.
Syntactic Aspects

6.
Semantics
of the Be To
Construction
7.
Should
To be
Counted
as
Part
of the
Lexical Item?
8. A
Word
Grammar Analysis
of the Be To
Construction
9.
Conclusion
5.
Linking
in
Word
Grammar
Jasper
Holmes
1.
Linking
in
Word

Grammar:
The
syntax
semantics principle
2.
The
Event Type Hierarchy:
The
framework; event types;
roles
and
relations
3.
Conclusion
6.
Word
Grammar
and
Syntactic Code-Mixing Research
Eva
Eppler
1.
Introduction
2.
Constituent Structure Grammar Approaches
to
Intra-Sentential
Code-Mixing
3. A
Word

Grammar Approach
to
Code-Mixing
4.
Word
Order
in
Mixed
and
Monolingual 'Subordinate' Clauses
5.
Summary
and
Conclusion
7.
Word
Grammar
Surface
Structures
and
HPSG
Order
Domains
Takafumi
Maekawa
1.
Introduction
2.
A
Word

Grammar Approach
3. An
Approach
in
Constructional
HPSG:
Ginzburg
and Sag
2000
4.
A
Linearization
HPSG
Approach
5.
Concluding Remarks
Part
II
Towards
a
Better
Word
Grammar
8.
Structural
and
Distributional Heads
Andrew
Rosta
1.

Introduction
2.
Structural Heads
58
60
63
67
67
68
69
70
71
72
75
77
81
83
83
103
114
117
117
118
121
128
139
145
145
146
154

160
165
169
171
171
172
WORD
GRAMMAR:
PERSPECTIVES
ON
LANGUAGE
STRUCTURE
ix
3.
Distributional
Heads
4.
Thai-Clauses
5.
Extent Operators
6.
Surrogates versus Proxies
7.
Focusing Subjuncts: just,
only,
even
8.
Pied-piping
9.
Degree

Words
10.
Attributive
Adjectives
11.
Determiner Phrases
12. The type
of
Construction
13.
Inside-out Interrogatives
14.
'Empty Categories'
15.
Coordination
16.
Correlatives
17.
Dependency Types
18.
Conclusion
9.
Factoring
Out the
Subject Dependency
Nikolas
Gisborne
1.
Introduction
2.

Dimensions
of
Subjecthood
3.
The
Locative Inversion Data
4.
Factored
Out
Subjects
5.
Conclusions
Conclusion
Kensei Sugayama
Author Index
Subject Index
172
174
174
177
179
181
181
182
182
184
185
187
189
191

191
199
204
204
205
210
216
222
225
227
229
This page intentionally left blank
Contributors
RICHARD
HUDSON
is
Professor Emeritus
of
Linguistics
at
University College
London.
His
research
interest
is the
theory
of
language structure;
his

main
publications
in
this area
are
about
the
theory
of
Word
Grammar, including
Word
Grammar
(1984, Oxford: Blackwell);
English
Word
Grammar
(1990,
Oxford:
Blackwell)
and a
large number
of
more recent articles.
He has
also
taught
sociolinguistics
and has a
practical interest

in
educational linguistics.
Website:
www.
phon.
ucl.
ac.
uk/home/dick/home.
hrm
Email:
dick@linguistics.
ucl.
ac. uk
KENSEI
SUGAYAMA,
Professor
of
English Linguistics
at
Kobe
City
University
of
Foreign Studies. Research interests: English Syntax,
Word
Grammar, Lexical
Semantics
and
General Linguistics.
Major

publications:
'More
on
unaccusative
Sino-Japanese complex predicates
in
Japanese' (1991).
UCL
Working
Papers
in
Linguistics
3; 'A
Word-Grammatic account
of
complements
and
adjuncts
in
Japanese' (1994).
Proceedings
of the
15th
International
Congress
of
Linguists;
'Speculations
on
unsolved problems

in
Word
Grammar' (1999).
The
Kobe
City
University
Journal
50. 7;
Scope
of
Modern
Linguistics
(2000,
Tokyo:
Eihosha);
Studies
in
Word
Grammar
(2003, Kobe: Research Institute
of
Foreign Studies,
KCUFS).
Email:
ken@inst.
kobe-cufs.
ac.
jp
CHET

CREIDER,
Professor
and
Chair, Department
of
Anthropology, University
of
Western
Ontario,
London,
Ontario,
Canada. Research interests: morphol-
ogy,
syntax,
African
languages.
Major
publications:
Structural
and
Pragmatic
Factors
Influencing
the
Acceptability
of
Sentences
with
Extended
Dependencies

in
Norwegian
(1987,
University
of
Trondheim
Working
Papers
in
Linguistics
4); The
Syntax
of the
Nilotic
Languages:
Themes
and
variations
(1989, Berlin: Dietrich
Reimer);
A
Grammar
of
Nandi
(1989, with
J. T.
Creider, Hamburg:
Helmut
Buske);
A

Grammar
of
Kenya
Luo
(1993,
ed. ); A
Dictionary
of the
Nandi
Language
(2001,
with
J. T.
Creider, Koln: Riidiger Koppe).
Email:
creider@uwo.
ca
ANDREW
ROSTA,
Senior
Lecturer,
Department
of
Cultural Studies, University
of
Central Lancashire,
UK.
Research Interests:
all
aspects

of
English grammar.
Email:
a.
rosta@v21.
me. uk
NIKOLAS
GISBORNE
is a
lecturer
in the
Department
of
Linguistics
and
English
Language
at the
University
of
Edinburgh.
His
research interests
are in
lexical
semantics
and
syntax,
and
their interaction

in
argument structure.
xii
WORD
GRAMMAR:
PERSPECTIVES
ON
LANGUAGE
STRUCTURE
Website:
www.
englang.
ed. ac.
uk/people/nik. html
Email:
n.
gisborne@ed.
ac. uk
JASPERW.
HOLMES
is a
self-employed linguist
who has
worked
with
many large
organizations
on
projects
in

lexicography, education
and IT.
Teaching
and
research interests
include
syntax
and
semantics, lexical structure, corpuses
and
other
IT
applications
(linguistics
in
computing, computing
in
linguistics),
language
in
education
and in
society,
the
history
of
English
and
English
as a

world
language.
His
publications include 'Synonyms
and
syntax' (1996,
with
Richard Hudson,
And
Rosta,
Nik
Gisborne).
Journal
of
Linguistics
32;
'The
syntax
and
semantics
of
causative verbs' (1999).
UCL
Working
Papers
in
Linguistics
11;
'Re-cycling
in the

encyclopedia' (2000,
with
Richard Hudson),
in
B.
Peeters (ed.
)
The
Lexicon-Encyclopedia
Interface
(Amsterdam: Elsevier);
'Constructions
in
Word
Grammar' (2005,
with
Richard Hudson)
in
Jan-Ola
Ostman
and
Mirjam
Fried (eds)
Construction
Grammars:
Cognitive
Grounding
and
Theoretical
Extensions

(Amsterdam: Benjamins).
Email: jasper.
holmes@gmail.
com
EVA
EPPLER,
Senior Lecturer
in
English Language
and
Linguistics, School
of
Arts, University
of
Roehampton,
UK.
Research Interests: morpho-syntax
of
German
and
English, syntax-pragmatics interface, code-mixing, bilingual
processing
and
production, sociolinguistics
of
multilingual communities. Recent
main
publication:
'" because
dem

Computer
brauchst'
es ja
nicht
zeigen.
":
because
+
German main clause word
order'
International
Journal
of
Bilingualism
8. 2
(2004),
pp.
127-44.
Email:
evieppler@hotmail.
com
TAKAFUMI
MAEKAWA,
PhD
student, Department
of
Language
and
Linguistics,
University

of
Essex. Research Interests: Japanese
and
English syntax, Head-
Driven
Phrase
Structure Grammar
and
lexical semantics. Major publication:
'Constituency,
Word
Order
and
Focus Projection'
(2004).
The
Proceedings
of
the
llth
International
Conference
on
Head-Driven
Phrase
Structure
Grammar.
Center
for
Computational Linguistics, Katholieke Universiteit Leuven, August

3-6.
Email:
maekawa@btinternet.
com
Preface
Thi s
volume comes
from
a
three-year (April 2002-March 2005) research
project
on
Word
Grammar supported
by the
Japan
Society
for the
Promotion
of
Science,
the
goal
of
which
is to
bring together
Word
Grammar linguists
whose

research
has
been carried
out in
this framework
but
whose approaches
to
it
reflect
differing
perspectives
on
Word
Grammar (henceforth WG).
I
gratefully
acknowledge support
for my
work
in WG
from
the
Japan Society
for
the Promotion of Science (grant-in-aid Kiban-Kenkyu C (2), no. 14510533 from
April 2002-March 2005).
The
collection
of

papers
was
planned
so as to
introduce
the
readers into this theory
and to
include
a
diversity
of
languages,
to
which
the
theory
is
shown
to be
applicable, along
with
critique
from
different
theoretical orientations.
In
September 1994 Professor Richard Hudson,
the
founder

of
Word
Grammar, visited Kobe City University
of
Foreign Studies
to
give
a
lecture
in
WG on a
part
of his
lecturing trip
to
Japan.
His
talks were centred
on
advances
in
WG at
that
time,
which refreshed
our
understanding
of the
theory. Professor
Hudson

has
been writing
in a
very
engaging
and
informative
way for
about
two
quarters
of a
century
in the
world linguistics scene.
Word
Grammar
is a
theory
of
language structure which Richard Hudson,
now
Emeritus Professor
of
Linguistics
at
University College London,
has
been
building since

the
early 1980s.
It is
still
changing
and
improving
in
detail,
yet the
main
ideas remain
the
same.
These
ideas themselves developed
out of two
other theories that
he had
tried: Systemic Grammar (now known
as
Systemic
Functional
Grammar),
due to
Michael
Halliday,
and
then
Daughter-

Dependency Grammar,
his own
invention.
Word
Grammar
fills
a gap in the
study
of
dependency theory. Dependency
theory
may not
belong
to the
mainstream
in the
Western
World,
especially
not
in
America,
but it is
gaining more
and
more attention, which
it
certainly
deserves.
In

Europe, dependency
has
been better known since
the
French
linguist
Lucien Tesniere's
study
in the
1950s (cf. Hudson, this volume).
I
just
mention here France, Belgium, Germany
and
Finland. Dependency theory
now
also rules Japan
in the
shape
of WG.
Moreover,
the
notion
of
head,
the
central
idea
of
dependency,

has
been introduced into virtually
all
modern
linguistic
theories.
In
most grammars, dependency
and
constituency
are
used
simultaneously.
However, this adduces
the
risk
of
making these grammars
too
powerful.
WG's challenge
is to
eliminate constituency
from
grammar except
in
coordinate structures, although certain dependency grammars, especially
the
German ones,
refuse

to
accept constituency
for
coordination.
Richard Hudson's
first
book
was the first
attempt
to
write
a
generative
(explicit)
version
of
Systemic Grammar
(English
Complex
Sentences:
An
xiv
WORD
GRAMMAR:
PERSPECTIVES
ON
LANGUAGE
STRUCTURE
Introduction
to

Systemic
Grammar,
North
Holland,
1971);
and his
second book
was
about Daughter-Dependency Grammar
(Arguments
for a
Non-transforma-
tional
Grammar,
University
of
Chicago Press, 1976).
As the
latter
tide
indicates,
Chomsky's transformational grammar
was
very
much
'in the
air',
and
both
books accepted

his
goal
of
generative grammar
but
offered
other ideas about
sentence structure
as
alternatives
to his
mixture
of
function-free
phrase structure
plus
transformations.
In the
late 1970s when Transformational Grammar
was
immensely
influential,
Richard Hudson abandoned Daughter-Dependency
Grammar
(in
spite
of its
drawing
a
rave review

by
Paul Schachter
in
Language
54,
348-76).
His
exploration
of
various general ideas that
hadn't
come together
became
an
alternative coherent theory called
Word
Grammar,
first
described
in the
1984 book
Word
Grammar
and
subsequently improved
and
revised
in
the
1990 book

English
Word
Grammar.
Since then
the
details have been
worked
out
much better
and
there
is now a
workable notation
and an
encyclopedia available
on the
internet (cf.
Hudson
2004).
The
newest version
of
Word
Grammar
is now on its way
(Hudson
in
preparation).
The time
span between

the
publication
of
Richard
Hudson's
Word
Grammar
(1984)
and
this volume
is
more than
two
decades
(21
years
to be
precise).
The
intervening
years
have seen impressive developments
in
this theory
by the WG
grammarians
as
well
as
those

in
other competitive linguistic theories such
as
Minimalist Programme, Head-driven Phrase Structure Grammar (HPSG),
Generalized Phrase Structure Grammar (GPSG), Lexical Functional Grammar
(LFG),
Construction Grammar
and
Cognitive Grammar.
Here
are the
main ideas, most
of
which come
from
the
latest version
of the
WG
homepage (Hudson 2004), together
with
an
indication
of
where they came
from:
• It is monostratal - only one structure per sentence, no transformations.
(Fro m Systemic Grammar).
• It uses word-word dependencies - e. g. a noun is the subject of a verb.
(Fro m

John
Anderson
and
other users
of
Dependency Grammar,
via
Daughter-Dependency Grammar;
a
reaction against Systemic Grammar
where
word-word dependencies
are
mediated
by the
features
of the
mother
phrase.
)
• It does not use phrase structure - e. g. it does not recognize a noun phrase
as
the
subject
of a
clause, though these phrases
are
implicit
in the
dependency structure. (This

is the
main
difference
between Daughter-
Dependency
Grammar
and
Word
Grammar.
)
• It
shows grammatical relations/functions
by
explicit labels
-
e.
g.
'subject'
an d
'object'. (From Systemic Grammar).
• It uses features only for inflectional contrasts - e. g. tense, number but not
transitivity.
(A
reaction against excessive
use of
features
in
both Systemic
Grammar
and

current Transformational Grammar.
)
• It
uses default inheritance,
as a
very
general
way of
capturing
the
contrast
between
'basic'
or
'underlying' patterns
and
'exceptions'
or
'transforma-
tions' - e.
g.
by
default,
English words
follow
the
word they depend
on, but
exceptionall y
subjects

precede
it;
particular cases
'inherit'
the
default
pattern
WORD
GRAMMAR:
PERSPECTIVES
ON
LANGUAGE
STRUCTURE
xv
unless
it is
explicitly overridden
by a
contradictory rule. (From
Artificial
Intelligence).
• It views concepts as prototypes rather than 'classical' categories that can be
defined
by
necessary
and
sufficient
conditions.
All
characteristics

(i.
e. all
links
in the
network) have equal status, though some
may for
pragmatic
reasons
be
harder
to
override than others. (From Lakoff
and
early
Cognitive Linguistics, supported
by
work
in
sociolinguistics).
• It presents language as a network of knowledge, linking concepts about
words , their meanings, etc.
- e.
g.
twig
is
linked
to the
meaning 'twig',
to the
form

/twig/,
to the
word-class
'noun',
etc. (From Lamb's Srratificational
Grammar,
now
known
as
Neurocognitive Linguistics).
• In this network there are no clear boundaries between different areas of
knowledg e
- e.
g.
between 'lexicon'
and
'grammar',
or
between 'linguistic
meaning'
and
'encyclopedic knowledge'. (From early Cognitive Linguistics
- and the
facts).
• In particular, there is no clear boundary between 'internal' and 'external'
facts
about words,
so a
grammar should
be

able
to
incorporate
sociolinguisti
c
facts
- e.
g.
the
speaker
of
jazzed
is an
American. (From
Sociolinguistics).
In
this theory, word-word dependency
is a key
concept,
upon
which
the
syntax
and
semantics
of a
sentence
build.
Dependents
of a

word
are
subcategorized
into
two
types,
i. e.
complements
and
adjuncts.
These
two
types
of
dependents
play
an
important role
in
this theory
of
grammar.
Let me
give
you a
flavour
of the
syntax
and
semantics

in WG, as
shown
in
Figure
1:
Figure
1
xvi
WORD
GRAMMAR:
PERSPECTIVES
ON
LANGUAGE
STRUCTURE
Contributors
to
this volume
are
primarily
WG
grammarians across
the
world
who
participated
in the
research organized
by
myself,
and I am

also
grateful
for
being able
to
include critical work
by
Maekawa
of the
University
of
Essex,
who
is
working
in a
different
paradigm.
All
the
papers here
manifest
what
I
would characterize
as
theoretical
potentialities
of WG,
exploring

how
powerful
WG is to
offer
analyses
for
linguistic
phenomena
in
various languages.
The
papers
we
have collected come
from
varying
perspectives
(formal,
lexical-semantic, morphological, syntactic,
semantic)
and
include work
on a
number
of
languages, including English,
Ancient Greek, Japanese
and
German. Phenomena studied include verbal
inflection,

case agreement, extraction, construction, code-mixing, etc.
The
papers
in
this volume span
a
variety
of
topics,
but
there
is a
common
thread
running through
them:
the
claim that word-word
dependency
is
fundamental
to our
analysis
and
understanding
of
language.
The
collection
starts

with
a
chapter
on WG by
Richard
Hudson
which serves
to
introduce
the
newest
version
of WG. The
subsequent chapters
are
organized into
two
sections:
Part
I:
Word
Grammar Approaches
to
Linguistic Analysis:
its
explanatory
power
and
applications
Part

II:
Towards
a
Better
Word
Grammar
Part
I
contains seven chapters, which contribute
to
recent developments
in
WG and
explore
how
powerful
WG is to
analyze linguistic phenomena
in
various
languages. They deal
with
formal, lexical, morphological, syntactic
and
semantic
matters.
In
this way, these papers
give
a

varied picture
of the
possibilities
of WG.
In
Chapter
2,
Creider
and
Hudson
provide
a
theory
of
covert elements,
which
is a hot
issue
in
linguistics. Since
WG has
hitherto denied
the
existence
of
any
covert elements
in
syntax,
it has to

deal
with
claims such
as the one
that
covert case-bearing
subjects
are
possible
in
Ancient Greek.
As the
authors
say
themselves, their solution
is
tantamount
to an
acceptance
of
some covert
elements
in
syntax,
though
in
every case
the
covert element
can be

predicted
from
the
word
on
which
it
depends.
The
analysis
given
is
interesting because
the
argument
is
linked
to
dependency.
It is
more
sophisticated than
the
simple
and
undefined Chomskyan notion
of PRO
element.
In
Chapter

3,
Sugayama
joins Creider
and
Hudson
in
detailing
an
analysis
of
understood objects
in
English
and
Japanese, albeit
at the
level
of
semantics
rather than syntax.
He
studies
an
interesting contrast between English
and
Japanese concerning understood objects. Unlike English
and
most other
European languages, Japanese
is

quite unique
in
allowing
its
verbs
to
miss
out
their complements
on the
condition that
the
speaker assumes that they
are
known
to the
addressee.
The
reason seems
to be
that
in the
semantic structure
of
the
sentences, there
has to be a
semantic argument which should
be, but is
not, mapped onto

syntax
as a
syntactic complement.
The
author adduces
a WG
solution that
is an
improvement
on
Hudson's (1990) account.
Sugayama
shares
with
the
preceding chapter
an
in-depth lexical-semantic
WORD
GRAMMAR:
PERSPECTIVES
ON
LANGUAGE
STRUCTURE
xvii
analysis
in
order
to
address

the
relation between
a
word
and the
construction.
In
Chapter
4, he
attempts
to
characterize
the be to
construction within
the WG
framework.
He has
shown that
a
morphological, syntactic
and
semantic analysis
of be in the be to construction provides evidence for the category of be in this
construction. Namely,
be is an
instance
of
modal verb
in
terms

of
morphology
and
syntax,
while
the
sense
of the
whole construction
is
determined
by the
sense
of
'to'.
In
Chapter
5,
Holmes
in a
very
original approach develops
an
account
for
the
linking
of
syntactic
and

semantic arguments
in the WG
approach. Under
the
WG
account, both thematic
and
linking properties
are
determined
at
both
the
specific
and the
general level.
This
is
obviously
an
advantage.
In
Chapter
6,
Eppler
draws
on
experimental studies concerning
the
code-

mixing
and
successfully
extends
WG to an
original
and
interesting area
of
research. Constituent-based models have
difficulties
accounting
for
mixing
between
SVO and SOV
languages like English
and
German.
A
dependency
(WG) approach
is
imperative here.
A
word's requirements
do not
project
to
larger

units like phrasal constituents.
The
Null-Hypothesis, then, formulated
in
WG
terms, assumes that each word
in a
switched dependency
satisfies
the
constraints imposed
on it by its own
language.
The
material
is
taken
from
English/German conversations
of
Jewish refugees
in
London.
Maekawa
continues
the
sequence
in
this collection
towards

more
purely
theoretical studies.
In
Chapter
7, he
looks
at
three
different
approaches
to the
asymmetries
between main
and
embedded clauses
with
respect
to the
elements
in
the
left
periphery
of a
clause:
the
dependency-based approach within
WG,
the

Constructional
HPSG
approach,
and the
Linearization
HPSG
analysis.
Maekawa,
a
HPSG
linguist, argues that
the
approaches within
WG and the
Constructional
HPSG
have some problems
in
dealing
with
the
relevant
facts,
but
that Linearization
HPSG
provides
a
straightforward account
of

them.
Maekawa's analysis suggests that linear order should
be
independent
to a
considerable extent
from
combinatorial structure, such
as
dependency
or
phrase structure.
Following
these chapters
are
more theoretical chapters which help
to
improve
the
theory
and
clarify
what research questions must
be
undertaken
next.
Part
II
contains
two

chapters that examine
two
theoretical
key
concepts
in
WG,
head
and
dependency.
They
are
intended
to
help
us
progress
a
few
steps
forward
in
revising
and
improving
the
current
WG,
together
with

Hudson
(in
preparation).
The
notion
of
head
is a
central
one in
most grammars,
so it is
normal that
it
is
discussed
and
challenged
by WG and
other theorists.
In
Chapter
8,
Rosta
distinguishes between
two
kinds
of
head
and

claims that every phrase
has
both
a
distributional head
and a
structural head, although
he
agrees that normally
the
same
word
is
both distributional
and
structural head
of a
phrase. Finally,
Gisborne's Chapter
9
then challenges
Hudson's
classification
of
dependencies.
The
diversification
of
heads
(different

kinds
of
dependency) plays
a
role
in
WG as
well.
Gisborne
is in
favour
of a
more
fine-grained
account
of
xviii
WORD
GRAMMAR:
PERSPECTIVES
ON
LANGUAGE
STRUCTURE
dependencies
than
Hudson's
1990
model.
He
focuses

on a
review
of the
subject-of
dependency, distinguishing between
two
kinds
of
subjects,
which
seems promising.
Gisborne's
thesis
is
that word
order
is
governed
not
only
by
syntactic
information
but
also
by
discourse-presentational
facts.
I
hope this short overview

will
suggest
to the
prospective reader that
our
attempt
at
introducing
a
dependency-based grammar
was
successful.
By
the
means
of
this volume,
we
hope
to
contribute
to the
continuing
cooperation between linguists working
in WG and
those working
in
other
theoretical frameworks.
We

look
forward
to
future
volumes that
will
further
develop this cooperation.
The
editors
gratefully
acknowledge
the
work
and
assistance
of all
those
contributors whose papers
are
incorporated
in
this volume, including
one
non-
WG
linguist
who
contributed papers
from

his own
theoretical viewpoint
and
helped
shape
the
volume
you see
here.
Last
but not
least, neither
the
research
in WG nor the
present volume would
have
been possible without
the
general support
of
both
the
Japan Society
for
the
Promotion
of
Science
and the

Daiwa Anglo-Japanese Foundation, whose
assistance
we
gratefully
acknowledge here.
In
addition,
we owe a
special debt
of
gratitude
to
Jenny Lovel
for
assisting
with
preparation
of
this volume
in her
normal professional manner.
We
alone accept responsibility
for all
errors
in
the
presentation
of
data

and
analyses
in
this volume.
Kensei Sugayama
References
Hudson,
R. A.
(1971),
English Complex Sentences:
An
Introduction
to
Systemic Grammar.
Amsterdam:
North
Holland.
— (1976), Arguments for a Non-transformational Grammar. Chicago: University of
Chicago
Press.
—
(1984),
Word Grammar.
Oxford:
Blackwell.
—
(1990),
English Word Grammar.
Oxford:
Blackwell.

—
(2004,
July
1-last
update),
'Word
Grammar',
(Word Grammar),
Available:
www.
phon.
ucl.
ac.
uk/home/dick/wg.
htm
(Accessed:
18
April
2005).
—
(in
preparation),
Advances
in
Word Grammar.
Oxford:
Oxford
University
Press.
Pollard,

C. and
Sag,
LA.
(1987),
Information-Based Syntax
and
Semantics.
Stanford:
CSLI.
Schachter,
P.
(1978),
'Review
of
Arguments
for a
Non-Transformational
Grammar'.
Language,
17,
348-76.
Sugayama,
K.
(ed.
)
(2003),
Studies
in
Word Grammar.
Kobe:

Research
Institute
of
Foreign
Studies,
KCUFS.
Tesniere,
Lucien
(1959),
Elements
de
Syntaxe Structurale.
Paris:
Klincksieck.
Introduction
This page intentionally left blank
1
What
is
Word
Grammar?
RICHARD
HUDSON
Abstract
The
chapter
summarizes
the
Word
Grammar

(WG)
theory
of
language
structure
under
the
following
headings:
1. A
brief
overview
of the
theory;
2.
Historical
background;
3. The
cognitive
network:
3. 1
Language
as
part
of a
general
network;
3. 2
Labelled
links;

3. 3
Modularity;
4.
Default
inheritance;
5. The
language
network;
6. The
utterance
network;
7.
Morphology;
8.
Syntax;
9.
Semantics;
10.
Processing;
and 11.
Conclusions.
1.
A
Brief Overview
of the
Theory
Word
Grammar
(WG)
is a

general theory
of
language structure. Most
of the
work
to
date
has
dealt
with
syntax,
but
there
has
also been serious work
in
semantics
and
some more tentative explorations
of
morphology, sociolinguistics,
historical linguistics
and
language processing.
The
only areas
of
linguistics that
have
not

been addressed
at all are
phonology
and
language acquisition (but even
here
see van
Langendonck 1987).
The aim of
this article
is
breadth rather than
depth,
in the
hope
of
showing
how
far-reaching
the
theory's tenets
are.
Although
the
roots
of WG lie
firmly
in
linguistics,
and

more
specifically
in
grammar,
it can
also
be
seen
as a
contribution
to
cognitive psychology;
in
terms
of
a
widely used classification
of
linguistic theories,
it is a
branch
of
cognitive
linguistics
(Lakoff
1987;
Langacker
1987; 1990;
Taylor 1989).
The

theory
has
been developed
from
the
start
with
the aim of
integrating
all
aspects
of
language
into
a
single dieory which
is
also compatible
with
what
is
known about general
cognition.
This
may
turn
out not to be
possible,
but to the
extent that

it is
possible
it
will
have explained
the
general characteristics
of
language
as
'merely'
one
instantiation
of
more general cognitive characteristics.
The
overriding consideration,
of
course,
is the
same
as for any
other
linguistic
theory:
to be
true
to the
facts
of

language structure. However,
our
assumptions make
a
great deal
of
difference
when approaching these
facts,
so it
is
possible
to
arrive
at
radically
different
analyses according
to
whether
we
assume
that language
is a
unique module
of the
mind,
or
that
it is

similar
to
other parts
of
cognition.
The WG
assumption
is
that language
can be
analysed
and
explained
in the
same
way as
other kinds
of
knowledge
or
behaviour unless
there
is
clear evidence
to the
contrary.
So far
this strategy
has
proved productive

and
largely
successful,
as we
shall
see
below.
4
WORD
GRAMMAR:
PERSPECTIVES
ON
LANGUAGE
STRUCTURE
As
the
theory's name suggests,
the
central unit
of
analysis
is the
word, which
is
central
to all
kinds
of
analysis:
• Grammar. Words are the only units of syntax (section 8), as sentence

structur e
consists entirely
of
dependencies between individual words;
WG
is
thus clearly part
of the
tradition
of
dependency
grammar
dating
from
Tesniere
(1959; Fraser 1994). Phrases
are
implicit
in the
dependencies,
but
play
no
part
in the
grammar. Moreover, words
are not
only
the
largest units

of
syntax,
but
also
the
smallest.
In
contrast
with
Chomskyan linguistics,
syntactic
structures
do
not,
and
cannot, separate stems
and
inflections,
so
WG is an
example
of
morphology-free
syntax
(Zwicky
1992: 354).
Unlike
syntax,
morphology (section
7) is

based
on
constituent-structure,
and the
two
kinds
of
structure
are
different
in
others
ways
too.
• Semantics. As in other theories words are also the basic lexical units
wher e
sound meets
syntax
and
semantics,
but in the
absence
of
phrases,
words
also provide
the
only point
of
contact between

syntax
and
semantics,
giving
a
radically 'lexical' semantics.
As
will
appear
in
section
9, a
rather
unexpected
effect
of
basing semantic structure
on
single words
is a
kind
of
phrase structure
in the
semantics.
• Situation. We shall see in section 6 that words are the basic units for
contextua l analysis
(in
terms
of

deictic semantics, discourse
or
sociolinguistics).
Words,
in
short,
are the
nodes that hold
the
'language' part
of the
human
network
together.
This
is
illustrated
by the
word
cycled
in the
sentence
/
cycled
to
UCL, which is diagrammed in Figure 1.
Figur e
1
WHAT IS WORD GRAMMAR? 5
Table 1 Relationships in cycled

related
concept
G
relationship
of C to
notation
in
diagram
cycled
the
word
/
the
word
to
the
morpheme
{cycle}
the
word-form
{cycle+ed}
the
concept
'ride-bike'
the
concept
'event
e'
the
lexeme

CYCLE
the
inflection
'past'
me
now
subject
post-adjunct
stem
whole
sense
referent
cycled
isa
CYCLE
speaker
time
V
'>a'
straight
downward
line
curved
downward
line
straight
upward
line
curved
upward

line
triangle
resting
on
CYCLE
'speaker'
'time'
As can be seen in this diagram, cycled is the meeting point for ten
relationships which
are
detailed
in
Table
1.
These
relationships
are all
quite
traditional (syntactic, morphological, semantic, lexical
and
contextual),
and
traditional names
are
used where they exist,
but the
diagram uses notation
which
is
peculiar

to WG. It
should
be
easy
to
imagine
how
such relationships
can
multiply
to
produce
a rich
network
in
which words
are
related
to one
another
as
well
as to
other kinds
of
element including morphemes
and
various
kinds
of

meaning.
All
these elements, including
the
words themselves,
are
'concepts'
in the
standard sense; thus
a WG
diagram
is an
attempt
to
model
a
small
part
of the total
conceptual network
of a
typical speaker.
2.
Historical
Background
The
theory described
in
this article
is the

latest
in a
family
of
theories which have
been
called
'Word
Grammar' since
the
early 1980s (Hudson 1984).
The
present
theory
is
very
different
in
some respects
from
the
earliest one,
but the
continued
use of the
same name
is
justified
because
we

have preserved some
of the
most
fundamental
ideas
- the
central place
of the
word,
the
idea that language
is a
network,
the
role
of
default inheritance,
the
clear separation
of
syntax
and
semantics,
the
integration
of
sentence
and
utterance structure.
The

theory
is
still
changing
and
further
changes
are
already identifiable (Hudson,
in
preparation).
As
in
other theories,
the
changes have been driven
by
various forces
- new
data,
new
ideas,
new
alternative theories,
new
personal interests;
and by the
influence
of
teachers, colleagues

and
students.
The
following brief history
may
be
helpful
in
showing
how the
ideas that
are now
called
'Word
Grammar'
developed during
my
academic
life.
The
1960s.
My PhD
analysis
of
Beja used
the
theory being developed
by
Halliday (1961) under
the

name 'Scale-and-Category' grammar, which later
turned into Systemic Functional Grammar (Butler 1985; Halliday 1985).
I
spent
the
next
six
years working
with
Halliday, whose brilliantly wide-ranging
analyses
impressed
me a
lot. Under
the
influence
of
Chomsky's generative
6
WORD
GRAMMAR:
PERSPECTIVES
ON
LANGUAGE
STRUCTURE
grammar (1957, 1965), reinterpreted
by
McCawley (1968)
as
well-formedness

conditions,
I
published
the
first
generative version
of
Halliday's Systemic
Grammar
(Hudson
1970).
This
theory
has a
very large network (the 'system
network')
at its
heart,
and
networks also loomed large
at tihat time in the
Stratificational
Grammar
of
Lamb (1966; Bennett 1994). Another reason
why
stratificational
grammar
was
important

was
that
it
aimed
to be a
model
of
human language processing
- a
cognitive
model.
The
1970s.
Seeing
the
attractions
of
both valency theory
and
Chomsky's
subcategorization,
I
produced
a
hybrid theory which
was
basically Systemic
Grammar,
but
with

the
addition
of
word-word dependencies under
the
influence
of
Anderson (1971);
the
theory
was
called 'Daughter-Dependency
Grammar' (Hudson 1976). Meanwhile
I was
teaching sociolinguistics
and
becoming increasingly interested
in
cognitive science (especially
default
inheritance
systems
and
frames)
and the
closely related
field of
lexical
semantics (especially
Fillmore's

Frame Semantics 1975, 1976).
The
result
was a
very
'cognitive' textbook
on
sociolinguistics (Hudson 1980a, 1996a).
I was
also
deeply influenced
by
Chomsky's 'Remarks
on
nominalization' paper
(1970),
and in
exploring
the
possibilities
of a
radically lexicalist approach
I
toyed
with
the
idea
of
'pan-lexicalism'
(1980b,

1981):
everything
in the
grammar
is
'lexical'
in the
sense that
it is tied to
word-sized units (including word classes).
The
1980s.
All
these influences combined
in the first
version
of
Word
Grammar (Hudson 1984),
a
cognitive theory
of
language
as a
network
which
contains both 'the grammar'
and
'the lexicon'
and

which integrates language
with
title
rest
of
cognition.
The
semantics
follows
Lyons
(1977),
Halliday (1967-
8)
and
Fillmore
(1976)
rather than formal logic,
but
even more controversially,
the
syntax
no
longer uses phrase structure
at all in
describing sentence structure,
because everything that needs to be said can be said in terms of
dependencie s
between single words.
The
influence

of
continental depen-
dency theory
is
evident
but the
dependency structures were richer than those
allowed
in
'classical' dependency grammar (Robinson 1970)
-
more like
the
functional
structures
of
Lexical Functional Grammar (Kaplan
and
Bresnan
1982). Bresnan's earlier argument (1978) that grammar should
be
compatible
with
a
psychologically plausible parser also suggested
the
need
for a
parsing
algorithm, which

has led to a
number
of
modest Natural Language Processing
(NLP)
systems using
WG
(Fraser 1985, 1989, 1993; Hudson 1989; Shaumyan
1995).
These
developments provided
the
basis
for the
next book-length
description
of WG,
'English
Word
Grammar' (EWG, Hudson 1990).
This
attempts
to
provide
a
formal basis
for the
theory
as
well

as a
detailed application
to
large
areas
of
English morphology,
syntax
and
semantics.
The
1990s.
Since
the
publication
of EWG
there have been some important
changes
in the
theory, ranging
from
the
general theory
of
default
inheritance,
through matters
of
syntactic theory
(with

the
addition
of
'surface structure',
the
virtual
abolition
of
features
and the
acceptance
of
'unreal'
words)
and
morphological theory (where
'shape',
'whole'
and
'inflection'
are
new),
to
details
of
analysis, terminology
and
notation.
These
changes

will
be
described
below.
WG has
also
been
applied
to a
wider range
of
topics than previously: