AUTHOR: De Cat, Cécile
TITLE: French Dislocation
SUBTITLE: Interpretation, Syntax, Acquisition
PUBLISHER: Oxford University Press
Thierry Ponchon, University of Reims (France).
According to the author Cécile de Cat, the aim of this book lies in studying the
phenomenon of spoken French dislocations, which is still ''poorly understood'' (p.
1). It's a revised and amplified version of her doctoral dissertation, with two
chapters previously published. All the examples are picked out of an empirical
basis issued from two corpora: the York corpus (detailing the language
development of three children (one from Belgium, one from Quebec, one from
France) was collected during eighteen months from 1997 to 1999, and the Cat
corpus (collected between 1998 and 2001) is composed of a longitudinal data
which charts the linguistic progresses of two Belgian children, and a
cross-sectional data from two children and four adults in Montreal (Quebec). The
transcriptions and coding methods follow the convention of CHILDES. (Appendix A,
pp. 217-242, and Appendix B, pp. 243-266.) These attested and spontaneous
utterances of spoken French material from Belgium, France, and Quebec,
indubitably contribute to increase and justify de Cat's analyses.
This work is divided into six parts. After a short introduction presenting the
objectives, the author proposes a diagnostic analysis. Three chapters follow
which study the three major features selected to characterize dislocation in
spoken French: interpretation, syntax, acquisition. The work is completed by
concluding remarks, followed of three appendices offering an overview of the
empirical basis of the research and the methodology used. Each chapter starts
off with an overview, contains conclusions, and is subdivided into subsections
which proceed step by step.
Chapter 2 (Diagnostics for dislocated elements) gives a definition of spoken
French, which is the field of de Cat's enquiry. It establishes the interpretive
and syntactic properties of dislocated structures. Thus, the nature of prosody
and the presence of a resumptive element are constituents that belong to
diagnose the dislocated structures in language. Following Blanche-Benveniste's
precepts (1997) concerning the differences between spoken and spontaneous French
and written and prescriptive French, de Cat starts with the obvious but
necessary evidence that ''Dislocations are rare in standard French, but very
frequent in spoken French'' (p. 4). Through the study of elements intervening
between subject clitics and the verb, object ''pronouns'' _le_, _les_, indirect
''pronouns'' _en_, _y_, and the negative ''adverb'' _ne_, she shows that analyzing
French subject clitics as affixes faces numerous problems and object and subject
clitics behave like syntactic entities rather than affixes or grammatical
agreement markers (pp. 9-29). Thus, she demonstrates that the morphological
analysis of French subject clitics is less efficient than the syntactic one
insofar as it cannot decisively explain the behavior of French subject clitics,
even if it seemingly reduces some difficulties. Whereas the morphological
analysis considers that French subject clitics are agreement morphemes inserted
pre-syntactically (Miller & Monachesi, 2003), the syntactic analysis regards
French subject clitics as entities available for syntactic operations, bearing a
theta-role (Kayne 1975). Her analysis finishes here with the conclusion that
French subject clitics behave much more like anaphoric pronouns than like
grammatical morphemes. She continues her demonstration by an analysis of the
prosodic characteristics of French dislocation, its variations in melody and
The corpus gives her a lot of examples showing the prosodic differences between
right- and left- dislocated elements (34-62): Les Belges sont les plus braves,
eux. / Les Belges, ils sont les plus braves. / 'Belgians are the bravest' (p.
44). On the one hand, she characterizes the right-dislocation either by a pitch
higher than most of the sentence nucleus, or a flat pitch when there is a tag
between the dislocated element and the sentence nucleus, or a clearly rise of
the pitch on dislocated element in interrogative sentences, with or without a
tag (vs Rossi 1999). On the other hand, she details arguments in favor of the
necessary presence of an intonation peak on the last syllable of the
left-dislocated element, and its perception, and, if there is a heavy subject,
the domination of the subject by the left-dislocated element in all its prosodic
characteristics (melody, intensity).
Her study leads one to consider that the dislocated elements are neither
syntactically nor prosodically important, for, in spoken French, a sentence can
be perfectly structured syntactically and prosodically without dislocation.
Chapter 3 (Interpretation) is based on the idea that the information structure
or packaging (i.e. encoding and decoding information) has a direct impact on
syntax and prosody. After having shown that several authors defend the idea that
the dislocated elements must be analyzed like topics and having clarified the
definition of this concept formalized by Reinhart (1981), she considers
topichood as a relation between an argument and a proposition relative to a
context (p. 64). In addition, she defines the focus as the major element of the
information of the sentence or as the element on which the speaker wants to
attract the hearer's attention. However, the examples chosen by de Cat enable
her to prove that some elements can be neither topic nor focus in a sentence.
Moreover, even if the fact that a topic can carry previous information is not a
necessary and sufficient condition for a referent to be a topic, she considers
that this common property of the topics results from the definition of topics
itself. Thus, she puts forward that previous information must not be seen as an
established property and a necessary requirement of topics, but as a property
derived from the salience requirement. So she distinguishes two kinds of topic:
stage topic, coindexed with the time and place of discourse, and aboutness topic
which needs to be salient to be recoverable by the hearer. Then the author
applies these principles to the examples of her corpora: the heavy subjects are
part of the focus of spoken French and dislocation is necessary to topicalize a
spoken French XP (p. 78). She analyzes the topic status of generic, specific,
and d-linked indefinites, previously discussed by Gundel (1975), Larsson (1979),
Zribi-Hertz (1994), and clearly shows that the contextual reference is necessary
for the anaphoric expressions dislocated in spoken French to be identifiable.
Topic is a primitive linguistic form, and can be neither reduced to a trace of a
previous information nor defined as a derivative. So, the function of the topic
is to determine the referential limit to the only elements of the predication
which can be considered by the hearer to be true, and in spoken French, the
non-pronominal elements have to be dislocated to be regarded as topics (p. 96).
Chapter 4 (Syntax) defends the principle that French dislocation (with either a
non-clitic or a (left or right) clitic element) is a unified syntactic
phenomenon. First, de Cat reminds readers that the dislocated elements are
widely recognized to be topics, and recursiveness is one of the characteristics
of the topic projection. Moreover, if most generativists consider them as
occupying the specifier of a TopicP to the left of the clause, not all
researchers give theoretical explanations for right-dislocated topics; even if
there are lots of commentaries for French (Kayne 1975; Ruwet 1990; etc.). After
having brought up the functional projection postulate, she pays special
attention to other alternatives: one approach maintains that topics are
adjunction structures and the other stands up for the pre-eminence of a syntax
movement to shift topics to peripheral position. Then, she focuses on
syntactical movements (topic moves and resumptive moves), their relations and
interferences, and especially the criterion of syntactic connectedness (p. 108).
She also defines three characteristics of what she considers uniform: the
necessary expression of the topic by the dislocated element, the possibility for
the resumptive element of being situated outside the clause where the dislocated
element is, the impossibility for dislocated constructions to display the key
properties of movement configurations, and to be interpreted via reconstruction.
Her analysis succeeds in considering the ''resumptive'' element as an anaphoric
pronoun of discourse. This overview is used to study the syntax of spoken French
dislocated elements. The analysis proposed by de Cat is that French dislocation
is essentially a root phenomenon. The spoken French corpora she uses clearly
indicate that the derivation of French dislocation does not involve movement.
Furthermore, it details separately what belongs to the syntax and what belongs
to the information structure. On one hand, she shows that, in French, a left
dislocated element expresses the topic of the sentence, is recursive, often
resumed by a non-clitic element, but always has a right-hand counterpart, and
generally without dependency markers. On the other hand, she defines syntactical
characteristics which distinguish very local righ dislocated elements (J'en ai
vu un, de cheval noir. / 'I have seen a black horse.') of right-dislocated
preposition phrases (On en sort vivant, de la prison. / 'One comes out of prison
alive.') (pp. 112-113). However, she explains and argues that French left
dislocation (LD) and right dislocation (RD) are using the same syntactic
mechanisms, and that French RD is not some kind of LD, neither a LD in lower
Topic position, nor an inverted LD in inflection phrase. Corroborated by native
speaker's judgments (p. 129), de Cat's base-generation analysis is backed up by
the fact that minimality doesn't constrain topic chains, that a dislocated
element cannot be interpreted in its reconstructed position, and that there is
lack of Principle C effects. The interpretative differences between LD and RD
can be derived from the properties of each periphery of the clause.
So, this chapter demonstrates that the intrinsic discursive properties of the
dislocated elements and those of certain clausal elements define their
distribution in spoken French and their combination is what characterizes
Chapter 5 (Acquisition) deals with the study of child language. De Cat's purpose
is to question if a better understanding of the development of language
structures could be due to the acquisition of dislocated topicalized elements,
if a basic form of predication would not exist in child language, and if
arguments for analyzing dislocated structures in the adult grammar could be
found in the acquisition data and confirmed.
At the start, she postulates the dynamics of acquisition of the linguistic
system. This evolutionary conception of the grammatical system implies obviously
the existence of a Universal Grammar, but especially the adoption of the
Continuity Hypothesis (Pinker 1984) and the principle of an upward spiral
progress. So, she extrapolates the evolutionary process of child language from
one of the essential characteristics of the adult grammar: the principle of
economy. To demonstrate it, the research leans on a longitudinal study of York
corpus and Cat corpus. Thereby, she considers that the analysis of dislocated
elements identification can be the same for adult and child, but only when the
internalized representation of a grammatical structure is well expressed and
productively used by this one. Shee resumes, therefore, the five criteria of
recognition developed in chapter 2: omission, resumption, word order, context
and prosody (p. 172). She explains that omissions of dislocated elements without
syntactic or prosodic alteration exist in child French, but this criterion must
be taken cautiously. There is a dislocation of the clause-peripheral phrase in
the child data with a resumptive element, when both are near each other.
Dislocation appears also when the canonical order is not subject-verb-object.
(One finds postposition of the subject and anteposition of the object.) or when
an expressed or understood clause fits in between the dislocated element and the
subject or object, or between the verb and another constituent. She shows that
context and discourse factors allow one to resolve latent ambiguities and to
analyze the types of dislocation at stake. The analysis of prosody confirms the
conclusions of previous works (Delattre 1966; de Cat 2002): The French children
use prosody to point out dislocated elements in the same way as adults (low and
flat pitch, decrescent intensity). In many instances, they use tags to separate
them from the sentence nucleus, as it's done in spontaneous adult speech.
Afterwards, she is interested in the emergence of dislocations. She shows that
dislocated structures appear from the beginning of the language acquisition, but
their emergence raises a lot of questions, both in the child grammar and the
adult grammar. Among the problems investigated are: Can the child rely on
evidence to acquire an adult-like representation of dislocated structures? Does
the basis of early data give evidence of such a representation? Does adult
representation of dislocated elements appear in (verbless) utterances of (early)
child French? The continuation of the study yields relevant answers to these
crucial questions (p. 180): The eight diagnostics used to identify the first
manifestations of Complementizer Phrase in the children's production give a
positive proof that dislocations are used productively by children well before
there is any concrete evidence of implementation of the CP layer. And the
argumentation shows clearly that no indication allows to assert that a French
child's underlying representation is different from that of an adult.
This is a very interesting and well-balanced book. One does appreciate that each
chapter starts with an overview of the literature, and all examples are
attested. Although it is intended for graduate students and specialists, this
book analyzes a major aspect of (spoken) French with an intelligent and relevant
method. Indeed, it begins with interpretation from spoken French dislocation,
then enters upon the problems of the acquisition of these structures in child
French, with detailed syntactic and prosodic studies of a varied corpus. The
approach de Cat takes is heuristic and shows measurable and verifiable analyses.
Thus, it presents jointly an innovative and judicious approach and has an
undeniable epistemological value.
However, one could be not agree with every detail of the author's analyses. The
minimalist conception of the child grammar development could be questioned. And
some derivation processes suggest by de Cat (pp. 146-147) could be disputed,
more exactly, the order of the steps established. For example, she appears
convinced of (1) an extraction of the nominal object in a left-dislocated
position, (2) a left-dislocation of the sentential subject, (3) a regressive
movement of the nominal object to sentence-initial position (sentence without
dislocation: Ça a surpris tout le monde qu'elle ait invité le juge. / 'It
surprised everybody that she'd invited the judge.'): (1) Le juge, qu'elle l'ait
invité a surpris tout le monde. (2) Qu'elle l'ait invité, le juge, ça a
surpris tout le monde. (3) Ça a surpris tout le monde, qu'elle l'ait invité,
le juge. On the one hand, nothing gives real evidence that this syntactic order
is justified; the more so as knowledge and practice of French would tend to show
that the last two sentences are indifferently used in spontaneous French while
the first is less frequent. On the other hand, there are other dislocated
sentences, which it is difficult to integrate into the author's demonstration:
Qu'elle ait invité le juge, ça a surpris tout le monde. / Le juge, ça a surpris
tout le monde, qu'elle l'ait invité. If the first belongs to common speech, the
second belongs to spoken French and consists of a LD of the nominal object
(resumed with a clitic) and of a RD of the sentential subject. One can also
regret that de Cat didn't have the opportunity to give bold outlines of a
lexical or semantic study. Sometimes, a wider and more explicit context would be
necessary for some examples (pp. 3, 113, 121, 183, ...). The explanation would
have been less disputable and the demonstration would have inevitably gained in
truthfulness. Finally and incidentally, the explanation given for _on_ (p. 11 n.
9) can't be share: if _on_ denotes second, third and fourth person referents, as
an indefinite pronoun it is omni-referential, and its denotation depends on the
context in which it is used.
Nevertheless, these criticisms must be minimized, for they hardly blemish the
intrinsic quality of the work.
Furthermore, this book provides grounds for further research: dislocation and
interaction between information structure and lexicon, contrastive and
diachronic perspectives (Pagani-Naudet 2005), dialectal differences, structural
interferences between English and French of Quebec, Flemish and Walloon, etc.
That is why not only the French synchronic linguists and the spoken
communication linguists, but also psycholinguists, linguists concerned by
language acquisition and links between thought and language will surely be
attracted by this scientific contribution.
Blanche-Benveniste, C. (1997) _Approches de la langue parlée en français._
De Cat, C. (2002) _French Dislocation._ PhD dissertation. University of York.
Delattre, P. (1966) ''Les dix intonations de base du français.'' _French Review_
Gundel, J. (1975) ''Left Dislocation and the role of topic-comment structures in
linguistic theory.'' _OSU WPL_ 18: 72-131.
Kayne, R. (1975) _The Transformational Cycle._ Cambridge (MA): MIT Press.
Larsson, E. (1979) _La dislocation en français: étude de syntaxe générative._
Miller, P. & P. Monachesi (2003) ''Les pronoms clitiques dans les langues
romanes.'' In Godard, D. (ed.) _Les langues romanes: problèmes de la phrase
simple._ Paris: Ed. CNRS. 67-123.
Pagani-Naudet, S. (2005) _Histoire d'un procédé de style. La dislocation
(XII-XVII siècles)._ Paris: Champion.
Pinker, S. (1984) _Language Learnability and Language Development._ Cambridge
(MA): Harvard UP.
Reinhart, T. (1981) ''Pragmatics and linguistics: An analysis of sentence
topics.'' _Philosophica_ 27: 53-94.
Rossi, M. (1999) _L'intonation. Le système du français: description et
modélisation._ Paris: Ophrys.
Ruwet, N. (1990) ''_En_ et _y_: deux clitiques pronominaux anti-logophoriques.''
_Langages_ 97: 51-81.
Zribi-Hertz, A. (1994) ''The syntax of nominative clitics in standard and
advanced French.'' In Cinque, G. & al. (eds) _Paths Towards Universal Grammar._
Washington (DC): Georgetown UP. 453-472.
ABOUT THE REVIEWER
Thierry Ponchon is a Lecturer in Linguistics (Department of French and Classical
Languages) at the University of Reims - Teaching University Institute (France).
He is also Visiting Lecturer at the University of Orel (Russia) and Publishing
director at ''L'Harmattan'' (France). His research interests lie in Psychomecanics
of the language, French historical linguistics, and Semantics. He has been
working on topics such as the verb _faire_ ('to do'), the expression of
consequence, and polysemy. He is now specially interested in all aspects of