Review of The Acquisition of the DP in Modern Greek
|
|
|
|
|
Review:
|
Date: Tue, 14 Dec 2004 22:41:30 -0800
From: Dimitris Ntelitheos <dntelith@ucla.edu>
Subject: The Acquisition of the DP in Modern Greek
AUTHOR: Marinis, Theodoros
TITLE: The Acquisition of the DP in Modern Greek
SERIES: Language Acquisition and Language Disorders 31
PUBLISHER: John Benjamins
YEAR: 2004
INTRODUCTION
This book is a revision of Marinis' PhD dissertation at the University of
Potsdam. It investigates in detail the acquisition of the nominal domain
using Modern Greek (MG) as a case study. It addresses specific
developmental issues associated with the Determiner Phrase (DP),
discusses a number of different theoretical approaches within the first
language (L1) acquisition literature and the predictions that these
approaches make with respect to empirical data, and provides insights
into different aspects of language acquisition in general, always with
respect to how these aspects are realized in the nominal domain. The
book is divided into seven chapters.
CHAPTER 1, 'Acquisition theories and the acquisition of the DP (1-
34), discusses the widely accepted idea within generative grammar of
an initial state in L1 acquisition. In Principles and Parameters (PP)
theory (Chomsky, 1986) the initial state is assumed to be Universal
Grammar (UG), a finite set of universal principles and a set of
parameters that take language specific values also selected from a
finite set. Marinis discusses different ideas on the nature of
parameters and their possible values before moving onto later
approaches towards UG within the Minimalist Program (MP)
(Chomsky, 1995 and later work) in which UG is assumed to determine
a set of (different types of) features available for all languages.
Marinis adopts the minimalist idea that language variation is related
only to the phonological component & the lexicon. With respect to L1
acquisition he discusses the 'Maturation Hypothesis' (changes in child
grammar are caused by biological maturation of the mind/ brain, cf.
Borer and Wexler 1987, Radford 1990) and the 'Strong' and 'Weak'
versions of the 'Continuity Hypothesis' (children possess a UG-
constraint grammar from the early stages and development is not the
result of maturation but rather the result of interaction of the initial
grammar with the linguistic input, cf. Clahsen et al 1993). He evaluates
the different approaches and using some findings from the MG data
he shows a preference towards the 'Weak' version of the 'Continuity
Hypothesis' in which functional projections are gradually added to
child phrase structure when children learn properties of the heads of
these projections. These different approaches are further evaluated in
subsequent chapters.
CHAPTER 2, 'Methodology', (35-54), discusses the data and
methodology used in the research for the book. The data used in the
book is based on two longitudinal corpora, the Christofidou corpus
and the Stephany corpus. The latter is available in the CHILDES
database. The corpora consist of audio-recordings of five monolingual
Greek children. Details on data collection, transcription and coding
conventions and the treatment of formulaic expressions are provided
in subsequent sections. The data is quantitatively and qualitatively
analyzed using CLAN software programs available at the CHILDES
website. The analyses define three specific developmental stages
(mainly using Mean Length Utterance calculations) and attested
patterns of development.
CHAPTER 3, 'The DP in Modern Greek', (55-83) provides a syntactic
analysis of the MG DP. In particular it examines the distributional
properties of MG definite and indefinite determiners, the internal
structure of the DP, the syntax/semantics mapping focusing on
Chierchia's (1998) 'Nominal Mapping Parameter', and finally the
morphological marking of MG nominals and their satellites and in
particular the degree of syncretism observed in the nominal domain
and the existence of morphologically marked and unmarked forms
within each paradigm.
CHAPTER 4, 'Acquiring the DP in MG', (85-138), is the first chapter
dedicated to acquisition facts. The focus of the chapter is on the
syntactic, semantic and morphological aspects of 'simple' DPs
consisting of definite/ indefinite determiners followed by nominals. The
main finding of the chapter is that children seem to use indefinite
determiners at a later stage than definite determiners, a finding that
contrasts results in other languages. Marinis disassociates the
existence of definite determiners in early stages of child grammar from
the actual acquisition of the functional projection associated with these
determiners (FP). He treats these early occurrences as 'impostors',
i.e. used in a lexically based fashion. The assumption is supported by
a U-shaped curve in the production of definite determiners in the
speech of one child.
He further shows that the data are consistent with the predictions
made by the 'Nominal Mapping Hypothesis', namely that there are
three stages in the acquisition of definite determiners: an initial stage
in which the determiners are dropped completely, an intermediate
stage in which they are optionally used and a final stage at the outset
of productive use in which they appear preceding proper names and
kinship terms. As for the acquisition of morphological marking Marinis
provides convincing evidence that shows that children acquire plural
morphology from Stage II onwards while in the case domain
nominative case is used quite early as a default case substituting for
other more marked cases and in particular genitive. This is attributed
to a morphological gap rather than a syntactic one. The children seem
to be able to establish possessor relationships (a fact supported by
their productive use of possessive constructions involving two noun
phrases) but lack the morphological means (i.e. genitive case) to mark
overtly this relationship. This is maybe due to a lower frequency of
genitive forms in the input or to the morphological markedness of
genitive case. Marinis takes these facts to support the 'weak' version
of the 'Continuity Hypothesis' that assumes optional incremental
acquisition of the functional layers within the DP.
CHAPTER 5, 'The acquisition of the possessive construction', (139-
164), examines possessive structures in child speech. Marinis
provides a detailed syntactic analysis of the possessor structure in
MG DP and examines different types of possessor constructions
attested in child speech involving different word orders. He shows that
the developmental pattern in possessor constructions also seems to
support an incremental approach to phrase structure building in
language acquisition. More specifically, the unmarked word order
possessum>possessor emerges earlier than the pragmatically marked
order possessor>possessum even though the latter is assumed to be
formed as a base structure. Furthermore the definite article preceding
the possessor is acquired at Stage II and the genitive marking on the
possessor is acquired at Stage III.
CHAPTER 6, 'The acquisition of Determiner Spreading', (165-190),
investigates the acquisition of the language specific phenomenon of
multiple determiners in the nominal domain, termed 'Determiner
Spreading' (DS, Androutsopoulou, 1994). In MG multiple instances of
definite determiners can precede every adjective that modifies the
noun and when this happens nouns can also precede modifying
adjectives, an order not available without simultaneous occurrence of
multiple determiners. Marinis briefly discusses the properties of the
construction and two syntactic analyses of the phenomenon together
with the predictions they make for language acquisition. He assumes
that the 'Determiner Complementation' analysis (determiners take CP
complements, i.e. reduced relative clauses with adjectival predicates,
Alexiadou and Wilder, 1998) is on the right track when modified
slightly. The acquisition data shows that children use DS productively
from Stage II onwards (the stage when definite determiners are used
productively). Both possible orders with the determiner-adjective string
preceding or following the determiner-noun string appear
simultaneously. Marinis assumes that this fact provides further support
for minimalist views of acquisition in which there is no syntactic
difference between A movement and A' movement - the two are
triggered by the EPP on an uninterpretable feature.
CHAPTER 7, 'The acquisition of appositive constructions involving
kinship terms and proper names', (191-214) discusses the acquisition
of appositives. After a brief description of the syntactic properties of
these structures Marinis shows that their acquisition also seems to
follow an incremental path. Initially, definite determiners are omitted
and appositives appear in inverse order, a pattern that is not attested
in adult language. At Stage II, definite articles emerge and the target-
like word order is observed. This seems to support the assumption
that functional layers in the DP have been activated and can serve as
landing sites for moved elements.
Finally, CHAPTER 8, 'Summary and conclusion', (215-225) contains
Marinis' concluding remarks, including a brief comparison of the
acquisition of the DP cross-linguistically and discussion of the issues
addressed in the main text in relation to evaluating the 'Maturation' vs.
the 'Continuity' hypotheses, and the implication of the findings for
linguistic theory in general.
The book ends with a list of References (227-237), two appendices
(239-253) with tables referred to in the main text, and an Index (255-
259).
CRITICAL EVALUATION
This is another careful edition from this series of John Benjamins.
There were only a couple of typos (e.g. on page 80 the end of the
section should read 'marked forms' and not 'unmarked forms'). As far
as the content is concerned Marinis provides convincing arguments
for his theoretical assumptions and takes care to present in detail the
conflicting proposals for each of the issues discussed before
presenting the relevant data. This makes easy to follow the arguments
in the discussion sections and check whether the conclusions are
adequately supported by the data. I would like to discuss a couple of
issues associated with both the data used and the theoretical
discussion of the empirical findings in the book. First of all, as Marinis
repeatedly indicates, the Stephany corpus contains data drawn from
children that have advanced to some degree in the developmental
process and so the data is not particularly useful for some of the
issues discussed in the book. In fact the discussion of definite
determiners is solely based on Christos' data (Christofidou Corpus,
not available in CHILDES). This is problematic as one of the major
assumptions in Chapter 4 is that early definite determiners
are 'impostors'. Marinis supports this idea with the fact that Christos
presents a U-shaped curve in his use of definite determiners but this
may be only apparent (i.e. may be the result of the files that show low
frequencies of definite determiners being smaller, or containing
utterances that may not always require definite determiners) and it
would be nice if the assumption could be supported by data from at
least one of the other children.
The fact that the children do not have a functional projection
associated with the definite determiner at this age is also related to
the use of place holders by children at these stages. However, the
significance of place holders in the MG acquisition data is not
discussed at all in this section and is only mentioned briefly in a later
section where the acquisition of the morphological marking of nominal
satellites is discussed. Thus the existence of place holders is not
adequately explained in this system that assumes that FP (the host of
definiteness) is not activated. It seems to me that at least in this
section the 'Weak Continuity Hypothesis' is promoted as the best
option without much discussion of the relevant facts and that other
approaches can be shown to explain the facts adequately. A further
objection comes from the assumptions on the relation of activation of
functional layers and the possibility of movement to these projections.
Is lack of movement evidence that the functional projection is not
there? Hyams (1996) argues convincingly that the reason determiners
are dropped in early grammar has to do with the fact that D is the
projection that anchors the nominal domain into a discourse
representation in a similar fashion that T anchors the sentence in time.
If D/F is underspecified but still present optionality in the production of
definite determiners can be explained and no further stipulations are
needed. Presence of definite determiners anchor the noun phrase
(i.e. introduce a salient or novel referent) while absence of definite
determiners imply assignment of a default semantic interpretation for
the DP, that of a 'familiar' interpretation. Hyams (1996) based on
Schaeffer (1994) shows how such an approach can account for
scrambling patterns in the acquisition of the Dutch nominal domain
where movement depends on whether the moved element is specific
or not. Thus lack of movement can be assumed to stem from the
underspecification of a functional layer and not necessarily from the
fact that the layer does not exist at all (a similar assumption is put
forward to explain the fatc that non-finite verbs in Germanic languages
indicate an underspecified functional projection in the tense domain
and thus movement of V to TP and CP is blocked).
The relation of discourse factors to the acquisition of certain
structures is implied by Marinis in a number of contexts. Some of the
structures under investigation are pragmatically marked (i.e. focusing
of possessor in possessor>possessum orders). Thus, the fact that
these structures are not attested in early stages may be not because
the landing site is not available but because the children have not yet
acquired the pragmatic properties of the structure. Furthermore,
structural complexity expressed in movement operations is sometimes
used to explain late emergence of specific structures (e.g.
appositives) but does not seem to influence other structures (e.g. the
base structure possessor>possessum) is acquired after the derived
structure possessum>possessor).
Abstracting away from these minor objections to the problems the data
poses and the way some conclusions are forced, Marinis' work is a
valuable contribution to the language acquisition literature. In most
cases the conclusions are strongly supported by the empirical facts
and the discussion provides the base for further investigation into the
properties of the acquisition of the nominal domain. The book is
strongly recommended for people interested in the structural and
developmental properties of the DP and is an important contribution to
the advancement of MG language acquisition in general.
REFERENCES
Alexiadou, Artemis & Chris Wilder (1998) Adjectival Modification and
Multiple Determiners, in A. Alexiadou & C. Wilder (eds.) Possessors,
Predicates and Movement in the DP, Amsterdam: Benjamins, 303-332.
Androutsopoulou, Antonia (1994). The distribution of the definite
determiner and the syntax of Greek DPs, Proceedings of the 30th
Regional Meeting of the Chicago Linguistic Society.
Borer, Hagit and K. Wexler (1987) The Maturation of Syntax. In T.
Roeper & E. Williams (eds.) Parameter Setting. Dordrecht: Foris, 123-
172.
Chierchia, G. (1998) Reference to Kinds across Languages, Natural
Language Semantics 6, 339-405.
Chomsky, Noam (1986) Knowledge of Language: Its nature, origin
and use. New York: Praeger.
Chomsky, Noam (1995) The Minimalist Program, Cambridge,
Massachuesetts: MIT Press.
Clahsen, Harald, M. Penke, & T. Parodi (1993) Functional Categories
in Early Child German, Language Acquisition 3, 395-429.
Hyams, Nina. (1996) The Underspecification of Functional Categories
in Early Grammar, in Harald Clahsen (ed.) Generative Perspective on
Language Acquisition, Amsterdam, Philadelphia: John Benjamins
Publishing Company, 91-127.
Radford, Andrew (1990) Syntactic Theory and the Acquisition of
English Syntax, Cambridge, Mass.: Basil Blackwell.
Schaeffer, J. (1994) On the Acquisition of Scrambling in Dutch. In D.
MacLaughlin & S. McEwen (eds.) Proceedings of the Boston
University Conference on Language Development I, 521-532.
|
| |
ABOUT THE REVIEWER:
ABOUT THE REVIEWER
My name is Dimitrios Ntelitheos and I am a graduate student at the
Department of Linguistics, UCLA. I am interested in issues related to
syntax and L1 acquisition of the nominal domain. I have worked on the
synatx and L1 acquisition of nominal ellipsis in Modern Greek and on
the existence of a Root Infinitive stage and the acquisition of voice
morphology in Malagasy. I am currently working on the syntax of
Malagasy nominalizations for my PhD dissertation.
|
|
|
|
|
|
