Pesetsky, David (2000) Phrasal Movement and Its Kin, MIT Press (Linguistic
Inquiry Monograph No. 37), ISBN 0-262-66166-7, 144 pp., $18.00/�12.50
Reviewed by Laura and Radu Daniliuc
School of Modern Languages, Department of Linguistics
The Australian National University
The study under review here investigates the types of movement and
movement-like relations that link positions in syntactic structure. David
Pesetsky argues that there are three such kind of relations. Besides the
Chomskian overt phrasal movement, Pesetsky brings evidence for other two
distinct types of movement without phonological effect: covert phrasal
movement, "a phenomenon in its own right", and feature movement. Focusing on
wh-questions, he shows how his classification of movement-like relations
allows for a better understanding of the syntactic behavior of wh-questions
in which an otherwise inviolable property of movement, "Attract Closest" in
Pesetsky's terms (in fact a restatement of Chomsky's Superiority Condition
for movement) appears to be violated. By demonstrating that more movement
takes place in such configurations than previously suspected, he shows that
Attract Closest is actually not violated at all in these cases. This
conclusion draws on recent research in both syntax and semantics,
particularly Beck (1996) and Richards (1997), and depends crucially on
Pesetsky's expanded repertoire of movement-like relations. His general view
of movement leads to a cross-linguistic explanation of the syntax of
Based on the three kinds of movement relations, the author concentrates on
what he calls "minimal triplets" in which movement to a particular head
shows up sometimes as overt phrasal movement, sometimes as covert phrasal
movement, as sometimes as feature movement. Identifying such a triplet,
Pesetsky believes, improves the ability to investigate the coexistence of
phrasal movement with feature movement in the grammar.
The main part of the book is represented by the analysis of such a triplet,
namely the interrogative wh-constructions. To support his view, Pesetsky
brings evidence from English and Bulgarian. The choice of Bulgarian, a
Slavic language, is motivated by the fact that, unlike English, a Germanic
language, Bulgarian shows multiple instances of overt phrasal wh-movement to
the left periphery of CP. Besides, if one takes into consideration the
feature movement proposal, the peculiarities of English multiple questions
turn out to faithfully reproduce the peculiarities of Bulgarian multiple
questions. The pronunciation of wh-phrase movement structures is shown to be
the only difference between English and Bulgarian multiple questions, whose
syntax is otherwise identical.
The book consists of five chapters that follow a logical succession of ideas
Chapter I - Introduction - defines some of the key words Pesetsky uses in
his study: overt and covert phrasal movement, feature movement and
superiority effects. In Pesetsky's framework, 'phrasal movement' is used as
a cover term for movement of any syntactic unit that is word-sized or
larger. Basically, overt phrasal movement refers to a word or phrase
pronounced in an "unexpected" position, while covert phrasal movement refers
to a moved element pronounced in a trace position.
If Chomsky claims that feature movement is the proper reanalysis of covert
phrasal movement and covert movement is the default, Pesetsky views feature
movement as a subcomponent of phrasal movement. To support his position, the
author brings evidence from such sources as ACD (antecedent-contained
deletion) and intervention effects and mostly from the typology of
The last subchapter of the introduction deals with the Superiority Effect,
which arises in a multiple question when more than one wh-phrase is relevant
to the answering patterns for the question. It also describes what Attract
Closest implies, namely movement viewed as triggered by particular features
of a "target" head K. Or, in Pesetsky's words, alpha can raise to target K
only if there is no legitimate operation Move beta targeting K, where beta
is closer to K. However, some minimal changes in a multiple question, such
as D-linking (see Pesetsky 1987), more than two wh-phrases, and translation
into German (issue mainly discussed in the last chapter), produce apparent
exceptions to the Superiority Effect.
Chapter 2 presents two observations about Bulgarian, namely a complementizer
that requires multiple specifiers (at least two wh-phrases) and superiority
effects and the principle of minimal compliance (for further reference, see
Richards 1997). The complementizer has a wh-feature that can and must be
deleted after attracting more than one instance of a corresponding
wh-feature to it, either by phrasal movement or feature movement. After
considering the order of wh-phrases in a Bulgarian multiple question,
Pesetsky is able to formulate the Bulgarian variant of the Superiority
Effect as follows: the leftmost wh-phrase in a Bulgarian multiple question
is the wh-phrase that was highest before movement (wh1 in Pesetsky's terms)
and the one that moves overtly in the corresponding English multiple
question. This leads to the formulation of a pronunciation rule for
Bulgarian stating that all wh-phrase movement to C is overt, in that wh is
pronounced in its new position and unpronounced in its trace positions.
Chapter 3 tries to answer the question if wh1-in-situ undergoes covert
phrasal movement. Based on examples from English, Pesetsky argues that in
D-linked exceptions to the Superiority effect wh1-in-situ does not undergo
covert phrasal movement. Further on, he considers some of the explanations
offered in the literature for ACD resolution through the eyes of covert
phrasal movement. The criterion for the classification of these explanations
points to the degree of limitation to special types of constituents that is
manifested by covert phrasal movement.
Chapter 4 investigates what happens to wh1-in-situ and examines several
apparent exceptions to the Superiority Effect. As Superiority effects
disappear in nonbinary multiple questions and in questions with D-linking,
Pesetsky points to the fact that the grammar of multiple questions contains
a multiple-specifier requirement and his conclusion is supported by the
similarity between English and Bulgarian with respect to multiple questions.
An interesting discussion is the question concerning the quality of "Feature
Movement" to be really FEATURE Movement or Feature MOVEMENT.
Chapter 5, "The Intervention Effect and the Typology of Interrogative
Complementizers", is the longest of this research. Its starting point is the
idea that multiple questions are introduced by a complementizer that
requires wh-movement to establish a multiple-specifier configuration. It
investigates the Intervention Effect (following Hagstrom's 1998 terminology)
in English Multiple Questions, in German Separation Constructions and in
German Multiple Questions. Pesetsky argues that the differences between the
intervention effect in German and English (and Bulgarian) multiple questions
are mainly due to a lexical difference, namely the complementizer requiring
a single specifier (in German) and the complementizer which must be
associated with more than one specifier (in English and Bulgarian). A
complementizer that tolerates no specifiers whatsoever is to be found in
Japanese and Korean.
This chapter also includes a discussion of Fanselow's scrambling proposal
(1997) and a typology of wh-specifiers according to the environments in
which intervention effects are found in wh-questions.
In the end of his study Pesetsky signals some unanswered questions about the
typology such as the repertoire of complementizers in a given language and
the possible existence of pronunciation patterns for multiple phrasal
wh-movement (beside those of English and those of Bulgarian described in the
In spite of the "loose ends" Pesetsky keeps mentioning throughout his study,
this study is a successful attempt to describe the types of movement and
movement-like relations that link positions in syntactic structures. The
focus is on wh-questions, seen through the eyes of movement-like relations,
a topic which would be interesting to investigate in less studied languages.
Pesetsky's theory is vigorously supported by evidence from several languages
and his ideas, though some of them may surprise at first sight, manage to
convince the reader through the light of the examples discussed in detail.
David Pesetsky is Ferrari P. Ward Professor of Linguistics at the
Massachusetts Institute of Technology. He is one of the editors of Is the
Best Good Enough? Optimality and Competition in Syntax (MIT Press, 1998) and
the author of Zero Syntax. Experiencers and Cascades (MIT Press, 1994). Is
the Best Good Enough? examines from a broad range of empirical and
theoretical perspectives the role of competition in syntax and in syntactic
interfaces with semantics, phonology, and pragmatics, as well as
implications for language acquisition and processing. Zero Syntax develops
the idea that the availability and syntactic positioning of arguments is not
a matter of chance but arises from laws governing the structure of lexical
entries and from laws governing syntactic structures themselves.
Beck, Sigrid. 1996. Quantified structures as barriers for LF movement.
Natural Language Processing 4, 1-56.
Fanselow, Gisbert. 1997. Minimal Link effects in German (and other
languages). Ms., University of Potsdam.
Hagstrom, Paul. 1998. Decomposing questions. Doctoral dissertation, MIT,
Pesetsky, David. 1987. Wh-in-situ: Movement and unselective binding. In Eric
Reuland and Alice ter Meulen, eds., The Representation of (in)definiteness.
Cambridge, Mass.: MIT Press
Richards, Norvin. 1997. What moves where in which language? Doctoral
dissertation, MIT, Cambridge, Mass.
Laura and Radu Daniliuc are the authors of the first Romanian translation of
Ferdinand de Saussure's Cours de linguistique g�n�rale (Curs de lingvistica
generala, Editura Cuv�ntul nostru, Suceava, 1998) and of Descriptive
Romanian Grammar. An Outline (Lincom Europe, Munich, 2000).