Koopman, Hilda, and Anna Szabolcsi (2000) Verbal Complexes, Cambridge,
Mass.: MIT Press, paperback, Current Studies in Linguistics 34, ISBN:
0-262-61154-6.
Reviewed by: Michael Moss, Department of General and Celtic Linguistics,
University of Gdansk.
Synopsis
This book presents a generative analysis of complex verb structures in
Hungarian, Dutch and German using an extended model of functional
projections and what the authors describe as a 'minimalist' approach.
Complex verbs in these languages are understood to be the 'bunches' of verbs
that group together at the end of sentences classically known from German.
The analysis covers Hungarian in detail and uses essentially the same model
to cover German and Dutch without going into such great detail. The main
phenomenon being studied is presented in the fact that the following
sentences are all considered 'well-formed' in Hungarian:
1. [Nem] fogok be menni.
[not] I-will in to go
I will [not] go in.
2. a. [Nem] fogok akarni be menni.
[not] I-will to want in to go
I will [not] want to go in.
b. [Nem] fogok be menni akarni.
[not] I-will in to go to want
I will [not] want to go in.
3. a. [Nem] fogok kezdeni akarni be menni.
[not] I-will to begin to want in to go
I will [not] want to begin to go in.
b. [Nem] fogok kezdeni be menni akarni.
[not] I-will to begin in to go to want
I will [not] want to begin to go in.
c. [Nem] fogok be menni akarni kezdeni.
[not] I-will in to go to want to begin
I will [not] want to begin to go in.
While the following sentences are NOT 'well-formed':
4. *[Nem] fogok menni be.
[not] I-will to go in
I will [not] go in.
5. a. *[Nem] fogok kezdeni be akarni menni.
[not] I-will to begin in to want to go
I will [not] want to go in.
b. *[Nem] fogok akarni be menni kezdeni.
[not] I-will to want in to go to begin
I will [not] want to go in.
c. *[Nem] fogok akarni kezdeni be menni.
[not] I-will to want to begin in to go
I will [not] want to go in.
From the patterning seen in the data the authors conclude that there is a
kind of 'leap frog' effect occurring during the derivation of the sentence.
That is, the underlying form of sentence (3.a.) is:
6. a. [Nem] fogok kezdeni akarni menni be.
[not] I-will to begin to want to go in
I will [not] want to begin to go in.
The element 'be' (in) leap-frogs to a position above 'menni' forming the
well-formed sentence in (3.a.).
According to this analysis this model, 'be menni' is now a cluster that
cannot be broken up. As such, if it were now to move it would move to a
position above the next verb 'akarni' (to begin) forming sentence (3.b.),
and so on. Apparently the last step in the chain in which the whole cluster
'be menni akarni kezdeni' moves above 'fogok' (will) is invisible, because
'fogok' must move to Case, T and Agr positions beyond the 'reach' of the
verbal complex. The various word orders seen in (1), (2) and (3) are due to
the interaction of movement options during the derivation. This is achieved
by using two mechanisms: remnant movement and stacking positions, which will
be discussed in the critical part of this review.
German and Dutch are treated using the same underlying structure and
mechanisms. The difference being that both of these languages have filters
imposing a limit on the internal complexity of constituents allowed to move.
That is, only simple or in more traditional terminology 'light' elements can
move.
Critical Review
When I got this book, I was truly looking forward to reading about the
new developments in 'complex verb' structures. Both of the authors are well
known for introducing new and important analyses and procedures into the
model of generative grammar. My expectations were that this book would
provide insights into how multiple verb structures and their derivations are
analyzed. After reading the book however, I feel that the main questions
concerning word order in verbal complexes remain a challenge. While the
analysis proposed in this book is both interesting and innovative, several
technical and theoretical problems force me to question their adequacy. Two
main problems made the book questionable for me. First, the book was
technically difficult to read for the following reasons:
7. a. the argument is not developed in a linear form;
b. the concepts used are often partially or poorly
defined,
c. the diagrams are incomplete and as such are
difficult to analyze.
Secondly, and perhaps more importantly, I found the 'technology' used to
support the authors' arguments was not well defined and it's use not well
defended, this is a serious problem when introducing new elements into a
theory. I will go through these criticisms point by point.
Many times in the book, a point is made briefly, with the explanation
that more detail will follow in sections to come. The problem is that when
you finally get to that point, you have to go back to the earlier example to
clarify its significance. Furthermore, since the reader is unable to fully
understand the reference when it is introduced (before the full argument is
presented), it is hard to follow the train of thought. I find books in which
each conclusion is presented in a linear fashion, building on the
conclusions made before it are clearer and have greater impact.
This book uses quite a bit of 'technology' to explain the derivation of
word order. However, much of the technology such as remnant movement,
licensing positions and stacking positions is poorly defined. Furthermore,
many times the reader is simply referred to other works which reportedly
discuss this technology in detail. This is the case with VMs or verbal
modifiers. VMs are introduced on page 17, but are defined on page 20, and
there the definition is only vaguely referent: "All of these expressions
have been dubbed 'verbal modifiers' (VM) in recent Hungarian descriptive
literature. For ease of pre-theoretical reference, we adopt this cover
term." This is unfortunate. I feel that a work should be as self-contained
as possible. It is very difficult to really evaluate an analysis if the
definitions and concepts are not included in the book itself.
Next, the authors assume that there are many functional projections above
VP, and at the beginning of the volume there is a diagram showing what they
propose to be a 'standard' universal set of functional projections. However,
this standard set is not followed during the rest of the book, and almost
all derivations are diagrammed piece by piece, each diagram showing one or
two projections at most. Since the standard set of projections is modified
for the following analysis, it is difficult to follow how the derivation is
developing in the diagrams which only show one or two steps in the
derivation. This problem is increased when combined with the manner in which
these new projections are defined, as discussed above.
These three issues add up to make the book and its ideas unfortunately
difficult to follow.
Now I will discuss the technology used at the core of the authors'
analysis showing how it was defined and its use defended. The analysis
presented is heavily based on the assumption that Functional Projections
(generally non-lexical projections responsible for tense, agreement and Case
assignment which as a group dominate the VP structure) are 'cheap' meaning
that they do not cost much in terms of the derivation, and as a result many
of them can be and are used. The authors introduce several new universal
functional projections such as PredP, RefP, DistP, FP, LP, InfP as well as
language specific phrases such as IsP for Hungarian. Furthermore a new type
of '+' phrase (+P) such as VP+, InfP+ are introduced as elements which
immediately dominate VP and InfP+ and attract elements which would normally
have landed in the [Spec VP] or [Spec InfP] positions for Spec-head
agreement. The need for the +Ps (reminiscent of VP* from the Koopman and
Sportiche version of the VP internal subject hypothesis (Koopman and
Sportiche 1991)) arises mainly due to the adoption of the 'Generalized
Doubly Filled Comp Filter which reads:
8. No projection may have both an overt specifier and
an overt head at the end of the derivation (pg. 40).
As such, agreement cannot occur in the Spec-head configuration, because
one of these positions must be empty by spell-out. The answer proposed in
this book is to postulate a projection which occurs 'beyond' the XP node but
has (seemingly) the same categorical features. This really just seems to be
a way of producing Chomsky's (or Gazdar's) multiple-specifier configurations
without referring to Chomsky (1995) (or Gazdar (1982)). As such, the +Ps
create a position into which an element may move, but justification of this
type of phrase is far from obvious.
LPs or licensing positions are similar in this respect. They are
introduced on page 39, where we find: 'We are led to assume that both
arguments and adjuncts have their own licensing positions (to be notated as
LP(xp)) and move into the m as soon as possible.' Apparently, LPs are
'motivated by Case and other feature checking positions' (pg. 43). It is not
clear what these LPs do, or how they come into the derivation. No clear
definition is given, and no defense of their existence is offered. Yet they
are a central part of the book's analysis.
Next, we have 'stacking positions' which are also labeled (LP) to "avoid
a proliferation of labels" (pg. 43). Apparently stacking positions occur in
the derivation to preserve word order: "movement into them [stacking
position LPs] is constrained by the convention that it must replicate the
already existing linear order of the pertinent XPs" (pg. 44). I thought that
the derivation was supposed to explain linear order phenomena. How is it
that the linear order is now supposed to determine structural projections?
While the authors themselves are hesitant about stacking positions saying
that they do not seem to be in the spirit of the 'minimal analysis', they
offer the following defense for using them in the derivation: "^�since it is
possible to employ them [stacking positions] in a completely mindless,
mechanical fashion, we choose to live with them as a provisional solution
that we hope will give way to a more insightful one" (pg. 44). This does not
seem to be adequate defense for an element which is central to a derivation.
These weak points in the book's organization and argumentation made it
frustrating to read and difficult to evaluate.
Having said that, the book does present quite a large variety of data
with lengthy derivations of several types of commonly found sentences. If
one is familiar with the literature on Hungarian verb complexes it will no
doubt be of interest. It is also worth pointing out the value of an analysis
that can explain word order in Hungarian and Germanic sentences. Such a
generalization that shows similarities in seemingly very different languages
is surely a step in the right direction.
In summary, I found the book disappointing. MIT's Current Studies in
Linguistics series usually brings out titles that are of general interest to
linguists, and that can be read and appreciated by specialists and
non-specialists alike. The problems discussed above make the work not only
difficult to read, but also, difficult to evaluate in terms of validity.
References:
All page numbers refer to quotes taken from the book being reviewed.
Chomsky, Noam. 1995. The Minimalist Program. Cambridge, MA.: MIT Press.
Gazdar, Gerald. 1982 "Phrase structure grammar". In: Pauline Jacobson &
Geoffrey K. Pullum (eds.), The nature of syntactic representation.
Dordrecht: D. Reidel Publishing Company. Pages 131-186
Koopman, H. and D. Sportiche. 1991. "The position of subjects. Lingua 85,
211-258.
The reviewer:
My name is Michael Moss, I am currently writing a doctorate at the
University of Gdansk. My interests research interests include: theta-role
and case assignment, agreement phenomena, and subcategorization frames in
generative grammar.
|
