Puskas, Genoveva (2000) Word Order in Hungarian: the Syntax
of A' positions (Linguistik Aktuell / Linguistics Today, v.
33). 396 pp. Hardback. Amsterdam / Philadelphia: John
Benjamins
Phoevos Panagiotidis, Birkbeck College, University of London
This is a monograph dealing with issues pertaining to the
word order of Hungarian and, more specifically, the
properties of Focus, Topic, Wh-questions and Negation in
this language. The cross-linguistic perspective is present
throughout the book and comparisons are often made between
syntactic phenomena in Hungarian and other languages. Hence
it is a book of interest to the theoretical syntactician, or
graduate student of Syntax, to someone interest in language
typology or a scholar working on the structure of Hungarian.
Contents
The book opens with a long and meticulously laid out
Introduction (Chapter 1) which consists of three sections:
the first states the problem, namely what accounts for the
ordering of the various constituents in the left periphery
of the Hungarian clause. The second section presents the
theoretical background of the monograph, namely X' theory,
Pollock's (1989) split Infl(ection) study as well as Rizzi's
(1997) split C(omplementiser) ones, the central position
Spec(ifier)-Head relations play in the above analyses and
Brody's (1995) Lexico-Logical Form. The third section
presents basic facts about Hungarian (including its 'free'
word order, that it is a null subject language with rich
case morphology and indefinite vs. definite agreement on
verbs). The rest of this section reviews previous analyses
of Hungarian. An appendix listing the case morphemes and
functions of Hungarian closes the chapter.
Chapter 2 discusses Focus in Hungarian, which Puskas extends
to include all new information in this language. After
having discussed alternatives to the idea that there exists
a dedicated projection for Focus in the C(omplementiser)
P(hrase) field, she goes on to show that not only focused
constituents occupy the specifier of a dedicated F(ocus)
projection but also to show that this F(ocus) P(hrase) is
part of the 'former' CP. Moreover, in languages like
Hungarian, a 'Focus Criterion' holds: all focused
constituents must appear at the specifier of FP and, at the
same time, all Focus heads must have a focused constituent
in their specifier. The obligatory adjacency of a focused
constituent (in the specifier of FP) with the verb in
Hungarian, see (1) below, adapted from p. 68, is explained
as follows: the Hungarian Infl(ection) head contains a [+f]
feature that forces it to move to F. Given that Hungarian is
a verb raising language, the verb will hence appear always
adjacent to the focused constituent, as shown under (2),
adapted from p. 70. Focused constituents are in block
capitals.
(1) a. ATTILAT szereti Emoke
Attila-ACC loves Emoke-NOM
"It is Attila that Emoke loves"
b.*ATTILAT Emoke szereti
Attila-ACC Emoke-NOM loves
"It is Attila that Emoke loves"
(2) [[[V T] Agr] F] [[V T] Agr] [V T] V
Puskas then shows that focused constituents are quantified
expressions. The scope properties of 'focused' quantifiers
like 'everyone' and 'someone' are then dealt with and the
chapter closes with the discussion of two dialects of
Hungarian: one permitting long extraction of focused
constituents (the wh-word equivalent in English being "What
did you say that _ you'll fix _?") and one using an
expletive 'azt' strategy instead. Two appendices on focus in
infinitival clauses and the subject position close the
chapter.
Chapter 3 is on Topic. Topics in Hungarian always precede
focused constituents (3a) (from p. 175) and Puskas follows
Rizzi (1997) in postulating a separate projection inside a
split CP. After topicalisation in languages like Italian and
English is reviewed, she spells out the properties of
topicalised constituents which, although in an A' (non-
argument) position, are not quantificational. It is shown
that there can be no topicalisation without focusing in
Hungarian (3b) (also from p. 175) and, in the final
sections, extraction of topics is discussed.
(3) a. Attilat EMOKE szereti
Attila-ACC Emoke-NOM loves
"It is Emoke that loves Attila"
b.*Attilat Emoke latta az ekuvo elott
Attila-ACC Emoke-NOM saw the wedding before
"Attila, Emoke saw before the wedding"
Chapter 4 discusses Hungarian wh-questions. The comparative
aspect of this chapter is quite prominent, so wh-movement,
multiple wh-questions, the landing site of wh-words, the
lack of superiority phenomena and partial wh-movement in
Hungarian are compared with similar phenomena in Slavic
languages, Italian, English, Romanian, Japanese and others.
Puskas shows that, although in Hungarian wh-words seem to
appear in the specifier of the Focus projection - and hence
focused constituents and wh-words are mutually exclusive (4)
(from p. 230) - this does not seem to be the case in other
languages. An appendix on superiority and Weak Crossover
concludes the chapter.
(4) *Kirol EMOKEVEL beszelt?
Who-about Emoke-with spoke
"Who did s/he speak with Emoke about?"
The final chapter discusses Negation in Hungarian after,
once more, having reviewed the theoretical background and
recent research on the topic. Sentential negation is also
shown to abide by a Spec-Head configuration requirement.
Subsequently, N-words (words like nobody, nothing, nowhere)
and their configurational properties are dealt with as well
as the phenomenon of negative concord. An appendix on
negation in the acquisition of Hungarian concludes the
chapter.
Evaluation
Starting from the title of the book, it is a quite accurate
summary of what the monograph is about. Indeed, the book
explores in a thorough, detailed and exhaustive fashion the
A' positions in the left periphery of Hungarian, not a
straightforward task.
On the one hand, the properties of focusing, topicalisation,
wh-movement and negation in Hungarian are dealt with very
carefully and convincingly. Although I am not exactly
familiar with the language, I was not given the impression
that problems and exceptions were swept under the carpet.
Related to this is how Puskas does not ignore or
condescendingly dismiss alternative accounts, even if they
are 'old-fashioned'. Illustrating this, consider the whole
section 2.1 in Chapter 2, dedicated to carefully refuting
accounts that take focused constituents to be adjoined to
the I(nflection)P(hrase). Such an idea feels totally out of
place after Brody's (1990) influential paper on FP, Rizzi's
(1997) split C and the recent (after Kayne, 1994) fall from
grace of XP-adjunction and, moreover, is an idea the whole
of Puskas' work is set out to refute. Hence, others would be
happy to state that adjunction is unacceptable in their
framework, as is the case here, and state that they are not
going to look into this particular analysis any further,
whereas Puskas goes on to show that the actual predictions
that accounts positing focused constituents to be IP
adjoined are empirically incorrect. This meticulous approach
to facts and analyses underlies the whole of this study;
together with its firm theoretical grounding and its
thoroughness, this establishes the monograph as, I feel, one
of the most important pieces of research on the fine
structure of the left periphery.
Having said that, the monograph has in times a feel of
being, in a sense, too explicit. This does not entail
necessarily that it ever wanders out of focus: the study,
its background, aims and its analyses are so clear and
sharp, that the reader can promptly see the relevance of,
say, discussing the extraction properties of topics (192-
209) to the bigger picture (roughly: whether topics are part
of the 'Scope field' or a different type of A'
dependencies). It is just that the reader might sometimes
get the feeling matters are discussed in certain passages in
too much detail. Examples of such passages, to name two, is
the exposition of the theoretical background and the
discussion of properties and analyses of the Hungarian
language, both in Chapter 1. As far as the first is
concerned: most of the matters presented therein (pp. 20-41)
are well known to theoretical syntacticians but, naturally,
inadequately exposed for the non-syntactically informed
scholar of Hungarian (as this is no handbook of syntactic
theory!). In particular the working framework of the
monograph is the 'representational' framework of Lexico-
Logical Form (LLF: Brody, 1995). This is an extremely
interesting framework, by all means, but not as well known
and widely applied as it should. Laying out the main aspects
of LLF in the book (p. 33-35) occupies almost as much space
as the presentation of the wh-criterion (p. 36-39), a matter
most syntacticians are much more familiar with.
Incidentally, in the same part of the book, the discussion
of the Minimalist Program (pp. 28-33), of which LLF consists
a radical version, is a little too condensed. Moving to the
section about facts and analyses of Hungarian (pp. 41-53),
someone familiar with them might have to skip it, although
no such caveat is provided clearly.
What is implied above is the following: this monograph is a
wonderfully complete discussion of the left periphery of the
Hungarian clause. As such, it targets both Hungarian
scholars perhaps not familiar with syntactic theory and
syntacticians perhaps not familiar with Hungarian. The long
- but somehow not long enough - discussion of concepts of
syntactic theory is aimed to the first group; the exposition
of facts and analyses of Hungarian to the second. While
syntacticians not familiar with Hungarian might find pp. 41-
53 a welcome setting of the scene, Hungarian scholars not
familiar with syntactic theory will still find pp. 20-41
confusing and this cannot be helped at all, anyway. I reckon
that the author should focus more on the 'syntactic'
readership in Chapter 1 - she does so in the following
chapters, of course; it would hence be more useful to
dedicate more space on LLF and less on, say, Spec-Head
relations (pp. 35-41), although Spec-Head relations are most
central in her study, because LLF is more of a 'terra
incognita'. Despite of the clarity and structuring of the
discussion throughout, I understand that the book remains a
technical monograph that will still have to be 'filtered
down' in order to reach a non-linguistically informed
audience or linguists not familiar with concepts and
analyses of generative grammar.
I understand the above sound very pedantic, but it is hard
to find fundamental flaws in such a well-written, detailed
and solidly argued study. So, it is now time to discuss the
many virtues of this book.
First, the book focuses on Hungarian in a fruitful fashion.
Although the analysis is fine-grained and addresses the many
(prima facie) oddities of this language (e.g. how come a
language can possess both multiple wh-adjunction and partial
wh-movement), the analyses never errs to the direction of
making crazy claims and ad hoc postulations just in order to
accommodate the facts. On the contrary, there are very few,
if any, stipulative claims in the book: Puskas works
parsimoniously and reduces most of these oddities to well-
studied mechanisms of natural languages: Spec-Head
relations, chain formation, quantificational vs. non-
quantificational chains. Only when facts appear
overwhelmingly impenetrable (e.g. the different ways
Hungarian, Serbo-Croatian type and Bulgarian type languages
'do' multiple wh-adjunction') does she propose parametric
variation in terms of availability of certain functional
categories in particular languages (anything but a 'crazy'
claim, of course!). At the same time, the monograph
beautifully illustrates how recent advances in syntactic
theory can account for facts mysterious before or without
them. Puskas shows once and for all that only a split-CP
analysis (Rizzi, 1997) can capture the facts about the Scope
Field (Focus, raised Quantifiers, wh-, Negation and negative
concord) and Topic Field in a way CP and IP adjunction ones
are unable to. Even the details of Rizzi's analysis, e.g.
that Force ('standard' complementisers like English 'that'
or Hungarian 'hogy') are higher than Focus is given further
confirmation.
Having said that, this monograph is anything but merely a
dry application of theoretical constructs to a particular
language. The author does not restrict herself to a bunch of
prominent phenomena of Hungarian that illustrate Spec-Head
relations and the existence of a split CP. Hence, it is not
just Focus movement, topic movement and the mutual
impossibility of wh and Focus movement and negative concord
we read about - all straightforward results of a Focus
Phrase (that has to have an overt constituent in its
specifier), a Topic Phrase, wh-words moving to SpecFP (at
least in Hungarian) and a Neg(ation) P(hrase). A range of
superficially complex related phenomena are dealt with and
resolved, reduced to familiar syntactic operations and / or
similar phenomena in other languages.
Conclusion
'Word Order in Hungarian' is an outstanding monograph that
avoids most of the pitfalls of theoretical studies analysing
a particular language. It is a focused and detailed study,
not a loose collection of related issues and convincingly
shows both that progress in linguistic theory is real and
how studying individual languages sheds light on broader
theoretical issues. As far as showing that in a language
like Hungarian the left periphery of the clause consists of
several functional heads is concerned, this book
successfully demonstrates that.
Bibliography
Brody, Michael 1995. Lexico-logical form: a radically
minimalist theory. Cambridge, Mass.: MIT Press
Brody, Michael. 1990. Some remarks on the Focus field in
Hungarian. UCL Working Papers in Linguistics 2: 201-225
Kayne, Richard 1994. The antisymmetry of Syntax. Cambridge,
Mass.: MIT Press
Pollock, Jean-Yves 1989. Verb movement, Universal grammar
and the structure of IP. Linguistic Inquiry 20: 365-424
Rizzi, Luigi 1997. The fine structure of the left periphery.
In: Haegeman, Liliane (ed.) Elements of Grammar. Dordrecht:
Kluwer. 281-337
The reviewer is part-time lecturer of Psychology of Language
at the School of Psychology, Birkbeck College, University of
London. He is the author of the forthcoming "Pronouns,
clitics and empty nouns: 'pronominality' and licensing in
Syntax'", to be published by John Benjamins in their
'Linguistics Today' series. His main interests include
Nominal Phrases, the categorial make-up of functional
categories, the Syntax of Greek, the relationship between
Syntax and morphology, the acquisition and mental
representation of syntactic knowledge.
|
