17.718, Review: Syntax/Ling Theories: Culicover & Jackendoff(2005)

[linguist at LINGUISTLIST.ORG]

LINGUIST List: Vol-17-718. Wed Mar 08 2006. ISSN: 1068 - 4875.

Subject: 17.718, Review: Syntax/Ling Theories: Culicover & Jackendoff(2005)

Moderators: Anthony Aristar, Wayne State U <aristar at linguistlist.org>
            Helen Aristar-Dry, Eastern Michigan U <hdry at linguistlist.org>
 
Reviews (reviews at linguistlist.org) 
        Sheila Dooley, U of Arizona  
        Terry Langendoen, U of Arizona  

Homepage: 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: Lindsay Butler <lindsay 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 at dooley at linguistlist.org. 

===========================Directory==============================  

1)
Date: 02-Mar-2006
From: Edward McDonald < laomaa20023 at yahoo.com.cn >
Subject: Simpler Syntax 

	
-------------------------Message 1 ---------------------------------- 
Date: Wed, 08 Mar 2006 15:25:31
From: Edward McDonald < laomaa20023 at yahoo.com.cn >
Subject: Simpler Syntax 
 

AUTHORS: Culicover, Peter W.; Jackendoff, Ray 
TITLE: Simpler Syntax 
SERIES: Oxford Linguistics
PUBLISHER: Oxford University Press
YEAR: 2005
Announced at http://linguistlist.org/issues/16/16-2264.html 

Edward McDonald, School of Asian Studies, University of Auckland

SUMMARY

This book attempts a thorough historically-informed re-evaluation of 
the syntactic theorising of the past half century within the generative 
tradition. Culicover and Jackendoff (hereafter C & J), whose links go 
back to their student days at MIT in the 1960s, have collaborated on a 
number of previous joint projects, and the current study represents a 
reworking of some of their previous collaborations (dealt with in Parts 
III and IV of this book) set within an original approach to syntactic 
theorising and description which they have dubbed ''Simpler Syntax'' 
(Parts I and II). The present review will focus on the metatheoretical 
aspect of C & J's account, examining how they justify their approach in 
contrast to other generative theories as set out in Parts I and II and 
the final summary Chapter 15 of Part IV.

The origins of the Simpler Syntax approach lie back in the days of the 
Generative Semantics-Interpretive Semantics ''wars'' -- on which the 
two authors, ''as befit [their] position as Chomsky's students, naturally 
[took]  the Interpretive side'' (p. xiv) -  when it seemed to them that ''the 
Interpretive position was leading generative grammar toward a leaner 
syntax with less complex derivations, and that a great deal of the work 
of predicting grammatical distribution would be pushed into the 
lexicon, into semantics, and into what were then called ''projection 
rules''...i.e. the rules that mediate between syntax and meaning'' (p. 
xiv). They see the Simpler Syntax approach as a ''contemporary 
version of Interpretive Semantics'' (p. xiv) which follows an 
interpretation of Occam's Razor, that ''important criterion for 
theoretical success'' (p. 4), that is in contrast with other previous and 
current approaches. The authors identify the following possibilities In 
the area of syntactic theories for conforming to Occam's Razor -- ''Do 
not multiply (theoretical) entities beyond necessity'' (p. 4, relevant 
theories and theorists added in parentheses):
a. Minimize the distinct components of grammar (Postal: Generative 
Semantics)
b. Minimize the class of possible grammars (Chomsky: Interpretive 
Semantics)
c. Minimize the distinct principles of grammar (Chomsky: Principles 
and Parameters Theory, The Minimalist Program)
d. Minimize the amount of structure generated by the grammar 
(Simpler Syntax)

The authors claim that ''the most explanatory syntactic theory is one 
that imputes the minimum structure necessary to mediate between 
phonology and meaning'' (p. 5) and that such a theory provides ''a 
vision of the language faculty that better facilitates the integration of 
linguistic theory with concerns of processing, acquisition, and 
biological evolution'' (p. xiv). The basic thrust of their approach can be 
summed up as follows (p. 5):
''... given some phenomenon that has provided putative evidence for 
elaborate syntactic structure, there nevertheless exist numerous 
examples which demonstrably involve semantic or pragmatic factors... 
[and thus] given a suitable account of the syntax-semantics interface, 
all cases of the phenomenon in question are [to be] accounted for in 
terms of the relevant properties of semantics/pragmatics''.

C & J frame their call for a major realignment of the field very 
persuasively from both empirical and historical points of view, arguing 
on the one hand that such a model provides a more economical 
account of the facts and thus a more feasible object for learner 
acquisition, and on the other that what they are doing amounts to 
a ''recorrection'' of certain unhelpful and complexifying directions 
generative grammar has taken since the mid 1960s. They identify as 
the ''overall questions addressed by this book'' the following, which 
could in fact be taken as programmatic for ANY syntactic 
theory: ''What is the role of syntax in the grammar vis-à-vis semantics, 
and what are the consequences for syntactic structure?'' (p. xiv).  
They position Simpler Syntax (hereafter abbreviated SS) as one of a 
group of what they call ''alternative generative theories'' which have all 
developed out of and identify themselves as in opposition 
to ''mainstream generative grammar'' (MGG) (p. 3). The view within SS 
of the ''syntactic component'' is ''thoroughly within the generative 
tradition'' but at the same time ''markedly at odds'' with ''views of 
syntax'' that have developed within MGG (p. 3). 

C & J still subscribe to most of the ''first principles of generative 
grammar'', which they claim ''have stood the test of time and have 
received further confirmation through the flood of research in cognitive 
science in the past forty years'' (p. 9). These principles include a 
fundamental orientation towards a ''mentalistic account of language''; 
the positing of a distinction in the brain between a ''finite basis'', the 
lexicon, and a ''productive system'', the grammar; a further distinction 
between ''the user's knowledge of his or her language'', competence, 
and the ''processing strategies by which this knowledge is put to use'', 
performance; and a focus on ''how speakers acquire their grammar 
and lexicon'', acquisition (p. 10). While the SS model retains the 
mentalistic and acquisitional focus of its generative forebears, it 
replaces the categorial distinction between a ''regular'' grammar and 
an ''exceptional'' lexicon (stemming ultimately from Bloomfield) with a 
conception of the two as forming a continuum, alongside a greater 
emphasis on the peripheral aspects of grammar in terms of the 
analogous distinction between the ''periphery'' and the ''core'' of the 
grammar (p. 25). It also puts more focus on the cognitive aspects of 
language by requiring that a ''theory of competence should be 
embedded in a theory of performance -- including a theory of the 
neural realization of linguistic memory and processing'' (p. 10).

As the name ''Simpler Syntax'' would suggest, a recurring theme in the 
book is the notion of ''simplicity'' and how that has been interpreted 
differently in different theories. In carrying out this comparison, mostly 
between MGG and SS, but also on occasion between SS and 
other ''alternative generative theories'' such as Lexical Functional 
Grammar, Head-Driven Phrase Structure Grammar, Construction 
Grammar, Autolexical Syntax, and Role and Reference Grammar, C & 
J distinguish two major ''aspects'' of theorising which they name 
the ''technological'' and the ''conceptual''. The technological aspect of 
a theory refers to the ''formal devices...a theory adopt[s] for its 
description of language'' (p. 4); while the conceptual, ''deeper and 
more difficult to characterize precisely'', refers to the ''theory's vision of 
what language is ''like'' '' (pp. 3-4). C & J characterize the kind of 
simplification put forward in SS as a ''Toolkit Hypothesis'' whereby 
the ''characterization of syntactic structure requires a multitude of 
principles, of varying degrees of regularity'', contrasting this with 
the  '''Galilean vision' of an extremely simple Grand Unified Theory'' 
put forward in MGG's Minimalist Program (pp. 4-5). 

In putting forward the SS model, C & J reject two of the basic 
methodological principles of MGG: Interface Uniformity, and Structural 
Uniformity. The first, Interface Uniformity, is basically an extension of 
the Katz-Postal Hypothesis (Katz & Postal 1964 -- see discussion on 
p. 48), and requires the ''syntax-semantic interface'' to be ''maximally 
simple'', in that ''meaning maps transparently into syntactic structure'' 
and ''maximally uniform'', so that ''the same meaning always maps onto 
the same syntactic structure'' (p. 6) In order to account for the regular 
mismatches between syntactic structure and meaning, MGG is then 
required to ''introduce a ''hidden'' or ''underlying'' level of syntax'', 
known at different times in the history of MGG as ''deep structure'' 
or ''logical form'', that ''maps directly onto semantics and is related 
derivationally to surface form'' (p. 7). In contrast, the SS approach 
claims that there is ''no syntactic structure beyond that present at the 
surface'' and that the ''syntax-semantics interface...supplies the 
[necessary] details of interpretation'' in line with 
the ''semantic/pragmatic structure'' of utterances (p. 7). The second, 
Structural Uniformity, also posits ''a ''hidden'' or ''underlying'' level of 
syntax'', claiming that an ''apparently defective or misordered [surface] 
structure is regular in underlying structure and becomes distorted in 
the course of derivation'' (p. 7). In contrast, an SS approach again 
rejects the necessity for an underlying level of syntax, ''licensing'' 
syntactically ''ill-formed utterances'' in relation to ''semantic/pragmatic 
factors rather than syntactic ones'' (p. 8). From an SS point of view, 
MGG accounts of syntax that assume Interface Uniformity and 
Structural Uniformity end up ''increasing rather than decreasing the 
overall complexity of the grammar'' (p. 9).

SS also shares with MGG the identification of the ''most important 
goal'' of linguistic theory as the characterization of the first principles 
of the generative tradition -- i.e. mentalism, the relation of competence 
to performance, acquisition, and innateness -- under the rubric 
of ''Universal Grammar'' (UG) or the ''language capacity'', defined from 
an acquisitional point of view as ''guidelines along which to pursue 
generalization -- a pre-narrowing of the class of possible analyses of 
the [child learner's] input'' (p. 11). However they take issue with 
Chomsky's identification of ''real progress in linguistics'' as ''the 
discovery that certain features of given languages can be reduced to 
universal properties of language, and explained in terms of these 
deeper aspects of linguistic form'' (Chomsky 1965: 35), claiming 
that ''a theory of language stands a better chance of being learnable if 
its syntax can be shown to have less abstract machinery such as extra 
nodes, hidden elements and covert movements'' (pp. 11-12). They 
also identify another kind of ''real progress'' in the ''discovery of how 
certain features of given languages, for which there is no UG input, 
can nevertheless be learned by the child from the input'', the most 
obvious of these being the ''voluminous facts of vocabulary'' (p. 12). 
Such an approach obviously also has implications for explaining the 
evolution of language: C & J do not deal much with issues 
of ''processing, acquisition and evolution'' in the current study, 
referring the reader to their previous work on the Foundations of 
Language (Jackendoff 2002) and Dynamical Grammar (Culicover & 
Novak 2003).

C & J then go on to characterise the ''architecture of the grammar'' in 
SS, in another words, ''an articulation of the grammar into rule types'' 
and a delimitation of ''the significant levels of linguistic representation'' 
(p. 14) and contrast the SS model with MGG under four headings: 
constraints rather than derivations; no ''hidden'' levels of syntax; 
multiple sources of combinatoriality; and the representation of 
meaning formally as a level of Conceptual Structure. In terms of the 
distinction between technological and conceptual aspects of theory 
introduced above, the first feature relates to the ''technology'' of 
representing syntactic structure. MGG does this in terms of 
derivations whereby ''linguistic structures are constructed by applying 
a sequence of rules, each applying to the output of the previous step''; 
there is in this view an ''inherent directionality in the logic of sentence 
construction'' (p. 15). In contrast, SS works in terms of constraints 
whereby ''each constraint determines or licenses a small piece of 
linguistic structure or relation between two small pieces'', with the 
structure being ''acceptable'' overall ''if it conforms to all applicable 
constraints'': such an approach requires ''no logical ordering among 
constraints'' and therefore ''readily lends itself to interpretations in 
terms of performance'' (p. 15). The derivational versus constraint-
based distinction between MGG and SS relates to a further contrast 
between the ''hidden levels'' posited by MGG, and noted in the 
discussion of Interface Uniformity and Structural Uniformity above, as 
opposed to the ''monostratal'' model of SS, the latter providing a more 
realistic model for acquisition.

In terms of the overall architecture of the grammar, C & J characterize 
MGG as ''syntactocentric'', in other words, ''the combinatorial 
properties of phonology and semantics are characterized entirely in 
terms of the way they are derived from syntactic structure'' (p. 17). SS 
in contrast has a ''parallel architecture'' which consists of ''parallel 
generative components, stated in constraint based form'' of 
phonology, syntax and semantics, ''each of which creates its own type 
of combinatorial complexity''. The structures generated by these 
separate components are mapped onto each other by means 
of ''interface components'', and the ''lexicon'' cuts across all three, 
with ''lexical items...inserted simultaneously into the three structures'' 
or in an alternative conception ''licensing a connection between 
fragments of the three structures'' (pp. 18-19). Above all this is an 
overarching ''Conceptual Structure'' which is ''not part of language per 
se'' but is rather the ''mental structure which language encodes into 
communicable form'', encoding such concepts as ''the categories in 
terms of which the world is understood, and the relations among 
various individuals and categories'' (p. 20). If ''independently motivated 
distinctions in Conceptual Structure are sufficient to account for a 
linguistic structure...there is no reason to duplicate them in syntactic 
structure'' (p. 21): the SS model thus requires a ''theory of syntax with 
the minimum structure necessary to map between phonology and 
meaning'' (p. 22).

Having in Chapter 1 set out the basic tenets of the SS theory in 
contrast to the MGG tradition, C & J then go on in Chapter 2 and 3 to 
carry out an admirably clear and -- as far as I can judge at least -- 
even-handed historical reappraisal of the ''principles and history of 
mainstream syntax''. This account serves as a useful adjunct to the 
more detailed studies of the Generative/Interpretive Semantics 
debates given in Harris 1993 or Huck & Goldsmith 1995, and also as a 
generally sympathetic although highly critical take on more recent 
developments like the Minimalist Program, treated more harshly in 
works like Seuren 2004. A particularly useful feature of these 
chapters -- which I will not attempt even to summarise here -- are the 
regular summary diagrams, which show how particular theoretical 
assumptions led on to, or provided the basis for, later developments 
(for example, Fig 20 on p. 61 or Fig. 49 on p. 72).

Chapter 4 then lays out in detail the ''Flat Structure'' model of SS. To 
briefly summarize it here, syntactic structure in SS is taken to be 
a ''linearized hierarchical tree structure whose nodes consist of 
syntactic features'' (p. 108). In contrast to common practice in MGG 
and other theories, the ''terminal nodes'' in each tree are not ''full 
lexical items'' but rather ''the purely syntactic features of lexical items'', 
with the ''phonological and semantic features of words'' appearing only 
in the ''phonological and semantic structure respectively'' (pp. 108-
109). (This is in line with the ''combinatorial autonomy'' of these three 
main levels in SS whereby each generates its own combinatorial 
structures - cf pp. 17-18.) These terminal nodes are ''chosen from the 
set of Xo (or lexical) categories...plus various affixal categories that 
consist of complexes of grammatical features'' (p. 110). ''Major'' lexical 
categories are ''characteristically dominated by a phrasal category 
XP'', the Xo category being the ''head of the XP'', while ''minor'' lexical 
categories ''do not (generally) have associated phrasal nodes'' (p. 
110). Along with MGG's ''hidden levels'' of syntax, SS also ''give[s] up 
entirely the notion of movement in syntax'' with ''what MGG has 
treated as syntactic movement from position X to position Y [being] 
replaced by a principle in the interface'' which specifies that ''a 
constituent bearing such-and-such a semantic role may appear either 
in position X or position Y depending on various conditions'' (p. 111). 
Furthermore, in SS the ''phrase structure rules'' are divided 
into ''principles of constituency and principles of linear order'' thus 
giving the model the flexibility to account for ''some autonomous 
syntactic principles that determine order'' as well as ''other facts about 
order that depend on semantics'' (pp. 143-144).

EVALUATION

Simpler Syntax is a very rich book, both in its basic content -- over 500 
pages worth -- and in its generous provision of food for thought. It 
should prove thought-provoking not just for scholars working within 
generative linguistics, for whom it will provide many novel and 
insightful solutions to some very old questions within that paradigm, 
but also for linguists from outside the generative tradition, who will find 
in it one of the very few historically and applicationally contextualised 
accounts of the preoccupations of generative linguistics. It is this 
second readership, to which I personally belong, that I would like to 
address here, with some thoughts on the claims made by the Simpler 
Syntax (SS) model and its place in the wider linguistic(s) universe.

C & J put an understanding of historical developments in generative 
grammar right at the core of the SS ''enterprise''. In contrast to the 
partiality and polemics of Newmeyer's Linguistic Theory in America 
(1980), which treated the rise of what C & J call ''Mainstream 
Generative Grammar'' (MGG), and particularly the first main break in 
that tradition with the ''Generative Semantics Wars'' of the late 1960s 
and early 1970s, in terms of a confrontation between ''truth'' 
and ''error'', in an account replete with a seemingly unconscious 
reliance on religious and military metaphor, SS joins a number of 
historiographically more grounded and ideologically more balanced 
studies that have come out since then, particularly on that key moment 
of rupture represented by Generative Semantics (Harris 1993, Huck 
and Goldsmith 1996). 

C & J also hark back to that seminal moment, putting SS forward as 
a ''contemporary version of Interpretive Semantics'', and pointing out 
that not only has MGG taken over many of the key ideas of 
Generative Semantics, the proliferation of syntactic complexity most 
commonly associated with that school is now a prominent feature of 
MGG. For example, in their treatment of the ''early history of 
mainstream syntax'' (Chapter 2), C & J make the strong claim 
that  ''the application of Uniformity in the development of MGG has 
produced the consequence that the Minimalist Program is 
indistinguishable in its essential architecture from Generative 
Semantics, long since discredited'' (p. 45). 

But If C & J were simply judging such an approach as out-of-date, with 
the Minimalist Program dismissed as merely a warmed-over version of 
Generative Semantics, their critique would not have the force it does. 
Their main critique of the ''simplicity'' requirements for Universal 
Grammar (UG) within both Generative Semantics and the Minimalist 
Program -- in terms of minimizing the distinct components of grammar 
and the distinct principles of grammar, respectively -- is that such a 
requirement only leads to complexity elsewhere in the grammar. C & J 
identify as the ''primary goal of explanation'' in SS -- a goal which is 
shared by but perhaps not as pivotal in actual theorising in the other 
two approaches -- the question of ''how the child acquires a grammar 
with a minimum of UG'' (p. 43). As they remark earlier in their 
argument, this goal of ''accounting for language acquisition gives 
empirical teeth to the desire to minimize the cross-linguistically 
variable principles of grammar'' (p. 11), and almost all of the 
theoretical innovations of SS in comparison to MGG can be traced 
back to the desire to minimize the ''internal resources that the child 
brings to bear in the construction of a grammar'' (p. 11). This empirical 
focus, though itself open to criticism (see below), gives the arguments 
for the SS model a seriousness and a ''reality'' that more theoretically-
focused accounts often lack.

How the SS model itself measures up against the empirical facts, 
which are of course as ''facts'' - as opposed to raw data - constructs of 
the particular theoretical model adopted, is a harder question to 
answer. In this regard it may not be irrelevant to cite the two epigraphs 
C & J provide for their book. The first is the familiar quotation from 
T.S. Eliot's Four Quartets about the goal of exploration being ''...to 
arrive where we started / And know the place for the first time'', a 
standard experience for workers in the human sciences like linguistics 
that set out to characterize explicitly the faculties people use 
automatically and unthinkingly in everyday life. However, the second, 
simply sourced to ''Morris Halle'', and presumably representing a 
favourite saying of his recalled from the authors' student days at MIT, 
makes the rather larger claim: ''I'm not here to tell you the news, I'm 
here to tell you the truth''.

I must confess to being rather at a loss as to how to ''read'' this second 
epigraph -- or how C & J may have intended it to be read -- but must 
also admit to some unease about it, because it seems to resonate with 
at worst a degree of arrogance and at best a certain blindness 
demonstrated in the book proper. On the most charitable 
interpretation, the prioritizing of ''truth'' over ''news'' simply indexes C & 
J's concern for the empirical groundedness of their theory, and their 
consequent disinclination for the elaboration of theory on merely 
theory-internal grounds. But to take a more critical point of view, C & 
J's ''truth'' can only be awarded such a status if one accepts the key 
theoretical and empirical tenets of the generative tradition, most of 
which C & J certainly do, but which I would predict at least a significant 
minority of their readers, including myself, definitely would not.

How for example is one to make sense of a claim like the following that 
the ''first principles of generative grammar...have stood the test of time 
and have received further confirmation through the flood of research 
in cognitive science in the past forty years'' (p. 9)? Now this is certainly 
not in the same league as the extreme but not exactly rare claim by 
some generative linguists, for example Bickerton, in a recent argument 
addressed to musicologists, that ''''[w]e have found out more about 
human language in the last thirty years than we did in the preceding 
three millennia'' (2000: 154), but it exhibits a similar flavour of self-
fulfilling prophecy. Bickerton goes on to claim that ''[w]e can now be 
sure that all human languages share a number of nonobvious 
characteristics, and that these characteristics derive directly from 
human biology'' (ibid.), but these ''nonobvious characteristics'' turn out 
to look suspiciously like a list of the main features of the Minimalist 
Program. Similarly, since cognitive science as currently constituted 
has been profoundly influenced by generative linguistics, it would be 
surprising if it DIDN'T provide confirming evidence for that paradigm.

The sort of mind-set revealed by examples like this is one in which 
there seems to be little awareness of the crucial distinction between 
the phenomenon itself and the theoretical characterisation of that 
phenomenon, even though we have no access to the former except 
through the latter. As I said above, ''facts'' are created by ''theories'', 
and so in characterising any phenomenon we need to be aware of the 
crucial ontological gap between the two. To give just a few examples, 
C & J refer at different points in their argument to the ''complexity of 
achieved grammar, as discovered by investigation in linguistic theory'' 
(p. 11), another claim of the self-fulfilling prophecy kind; and just a bit 
later claim that ''a theory of language also stands a better chance of 
being learnable if its syntax can be shown to have less abstract 
machinery'' (pp. 11-12). It is, I think, worrying for the SS ''enterprise'' 
that these kinds of claims bear directly on the main empirical ''guide'' 
for the form syntactic theorising takes, that of the relationship of UG to 
language acquisition.

Such a conceptualization of language acquisition leaves itself open to 
critique on two main grounds. Firstly, it seems to assume that the 
process by which a child learner ''constructs'' a grammar for 
him/herself is analogous to that by which an adult linguist describes 
the grammar of a language: in other words, that in both cases we are 
dealing with a similar kind of ''knowledge''. However, as Hockett 
pointed out very early on (1968), there are in fact two distinct kinds of 
knowing involved in the different cases. The first, for which Hockett 
used the Chinese verb hui 'to know, to be able to', is the ''know how 
to'' of the philosophers -- referring to the kinds of skills or habits that 
are learned without conscious analysis. The second kind, for which 
Hockett used the Chinese verb zhidao 'to know', is the ''know that'' 
kind of knowledge, factual, explicit, analytical. To base an argument 
about ''know how to'' on the basis of ''know that'' is hard to justify, and 
it is surely only the fact that the generative tradition by and large 
ignores the ''community'' in favour of the ''mind/brain'' of the individual 
(p. 10) -- in other words, creates a stark and quite unnecessary 
dichotomy between the social and the cognitive -  that has led 
generative theorists to fudge this difference.

Secondly, on purely empirical grounds, treating the question of ''how 
the child becomes grammatically competent so rapidly and effortlessly'' 
(p. 11) as some sort of mystery calling for a complex cognitive 
explanation, again because the complex supportive social context in 
which the child learns language is largely ignored, must at best be 
rated as a rather dubious theoretical move. It may well be true that at 
the time of the foundation of the generative paradigm the relevant 
empirical studies had simply not been done, but this is no longer the 
case. To mention only one tradition within linguistics, the body of work 
that stems from Halliday's 1975 study Learning how to mean, including 
Painter 1984 Into the Mother Tongue, both dealing with 
protolanguage and the transition into the adult language, and 
continuing with an extended study of the speech of older children 
directed by Hasan (e.g. Hasan 1992), shows very clearly that an 
enormous amount of language learning can be explained in terms of 
the interaction of the child with its caregiver. 

At one point in his lively and strongly expressed view of modern 
linguistics, Sampson (1980) remarks on the fact that the students at 
MIT used to dub the course they took on non-generative linguistic 
theories ''the bad guys''. On reading C & J's views on contemporary 
linguistics, one senses that for them, it is not so much a case of ''the 
bad guys'' as ''the invisible guys''. For example, in their account of 
the ''later history of mainstream syntax''  (Chapter 3), C & J refer to 
Baker's attempt to defend what they characterise as 
the ''syntactocentric'' Principles and Parameters model against the 
sorts of monostratal approaches to which SS belongs. They comment 
as follows (p. 74):
''Baker offers no argument against the monostratal approaches, 
whose syntactic components are, as he admits, highly constrained. He 
simply believes that there *has* to be more to syntax than this, and 
lays out a program in which this view is central. He is asserting not 
only architectural syntactocentrism, which is a legitimate theoretical 
position, but also a more dogmatic, perhaps even imperialistic, 
syntactocentrism.''

For an ''outsider'' reading SS, it is often difficult not to apply the same 
sort of ''imperialistic'' label to C & J's project. Of course the ''insider'' 
view which C & J are so well qualified to give is one of the reasons 
why they can speak with such authority on developments in the 
generative tradition, and why their criticisms have the force they do. 
But when it comes to accepting SS as ''truth'' rather than 
simply ''news'', it is hard to take seriously claims that such-and-such a 
feature of the SS theory, for example the Grammatical Function 
Tier, ''captur[es] something deep and true about language'' (p. 539), 
when the authors have so obviously not even looked at any 
comparable models outside of the generative tradition. Perhaps they 
might want to acknowledge that there are levels of ''mainstreamness'', 
and that those in the ''mainstream linguistic tradition'' might benefit 
from at least an increased sense of perspective through a greater 
awareness of ''alternative linguistic traditions''.

But when it comes to their concluding call for ''some change in the way 
people do syntax'' (p. 546), this current reviewer finds himself lined up 
unequivocally on C & J's side. On what they earlier called 
the ''conceptual'' aspect of theorising, their ''wish list'' includes what 
seem to me, allowing for differences in the way different theories 
frame such questions, unarguable propositions such as ''syntax 
cannot be studied without simultaneously studying its interaction with 
semantics (not to mention prosody)'' and ''an adequate theory of 
syntax should connect in a natural way to an account of how humans 
produce and understand sentences'' (p. 546). And on 
the ''technological side'', they strike a note of sheer wisdom gained 
from long experience in their call for ''more investigation that compares 
frameworks dispassionately, for it is only by doing such comparisons 
that we pit them against each other scientifically rather than merely 
sociologically'' (p. 546).  

It is true that from a broader perspective I personally do not accept 
that ''most of the alternative frameworks conceive of themselves 
primarily in opposition to mainstream generative grammar'' (p. 546), 
and it seems to me that the ghost of that old chestnut ''notational 
variant'' floats over statements such as ''the goal of such comparison 
should be to distill out of each framework the essence of what it thinks 
language is like, free of the peculiarities of its formalism'' (pp. 546-
547). But these are quibbles in relation to the overall attempt to open 
up debate in SS. I hope that reactions to C & J's proposal will be 
characterised by a tone similar to their own that is measured, 
generous and above all honest about its own claims and 
preoccupations.

REFERENCES

Bickerton, D. 2000. Can Biomusicology Learn from Language 
Evolution Studies? In Wallin, N. L., B. Merker & S. Brown, eds. 2000. 
The Origins of Music. Cambridge MA: MIT Press, 153-163.

Chomsky, N. (1965) Aspects of the Theory of Syntax. Cambridge MA: 
MIT Press.

Culicover, P. W. & A. Nowak (2003) Dynamical Grammar. Oxford: 
Oxford University Press.

Halliday, M. A. K. (1975) Learning How to Mean: Explorations in the 
Development of Language. London: Arnold.

Harris, R. (1993) The Linguistics Wars. Oxford: Oxford University 
Press.

Hasan, R. (1992) Speech genre, semiotic mediation and the 
development of the higher mental functions. Language Sciences 14.4, 
489-528.

Hockett, C. F. (1968) The State of the Art. The Hague: Mouton.

Huck, G. J. & J. A. Goldsmith. (1995) Ideology and Linguistic Theory: 
Noam Chomsky and the Deep Structure Debates. London: Routledge.

Jackendoff, R. (2002) Foundations of Language. Oxford: Oxford 
University Press.

Katz, J. J. & P. M. Postal (1964) Toward an Integrated Theory of 
Linguistic Descriptions. Cambridge MA: MIT Press.

Newmeyer, F. (1980) Linguistic Theory in America: The First Quarter-
Century of Transformational Generative Linguistics. New York: 
Academic Press. (2nd ed. 1986).

Painter, C. (1984) Into the Mother Tongue: A Case Study in Early 
Language Development. London: Pinter.

Sampson, G. (1980) Schools of Linguistics. Stanford: Stanford 
University Press.

Seuren, P. (2004) Chomsky's Minimalism. New York: Oxford University 
Press. 

ABOUT THE REVIEWER

Edward McDonald has taught Chinese language, linguistics and 
semiotics at universities in Australia, Singapore and China, and 
currently New Zealand. His research interests are in the areas of the 
grammar and discourse of modern Chinese, theories and ideologies of 
language, and the semiotics of language and music. His book, 
Meaningful Arrangement: exploring the syntactic description of texts, 
will be published by Equinox in November 2006.





-----------------------------------------------------------
LINGUIST List: Vol-17-718