Review of Information Structure and the Dynamics of Language Acquisition
|
|
|
|
|
Review:
|
Date: Tue, 2 Mar 2004 15:48:30 +0100
From: Rolf Kreyer <rkreyer@uni-bonn.de>
Subject: Information Structure and the Dynamics of Language Acquisition
EDITOR: Dimroth, Christine; Starren, Marianne
TITLE: Information Structure and the Dynamics of Language Acquisition
SERIES: Studies in Bilingualism 26
PUBLISHER: John Benjamins
YEAR: 2003
ANNOUNCED IN: http://linguistlist.org/issues/14/14-2638.html
Rolf Kreyer, University of Bonn
The thirteen articles in this volume focus on the impact of information
structure on language acquisition, taking into account a vast range of
natural first and second language acquisition data. In particular, the
authors investigate into "the impact of the interplay between
principles of information structure and linguistic structure on the
functioning and development of the learner's system" (p. 1). The scope
of the volume is remarkable, examining acquisition data from different
languages, such as English, German, French and Italian, and different
first-language (L1) backgrounds of second- language (L2) learners (for
instance Turkish, Polish, or German). As a 'testing ground' for the
respective impact of communicative and structural forces on the
acquisition process, two empirical domains are considered, namely the
expression of finiteness and scope relations at the level of the
utterance and the expression of anaphoric relations at the level of
discourse.
SYNOPSIS
The papers in the volume fall into two groups based on the two
empirical domains named above. The first section, 'Finiteness and scope
relations', focuses on two main aspects: the precursors of verbal
morphology at early stages in L1 acquisition and their development at
later stages of the acquisition process, and the integration of scope
bearing elements into the utterance.
The first four articles in this section are concerned with the first
aspect. Schlyter's paper analyses the development of verb morphology
and finiteness in learners of French. Her data are drawn from ten
Swedish-speaking adult L2 learners and four bilingual Swedish-French
children who learn French as their first language. The author starts
off by discussing the question as to when a learner can be said to have
acquired finiteness and employs a set of criteria that includes
morphological as well as syntactic criteria, such as inflection of
copula and auxiliaries, or postverbal negation. Her analysis of the
bilingual children shows that these, exhibiting no evidence for
morphological and syntactic finiteness at a first stage of learning,
"acquire morphology and syntax simultaneously, and perhaps use local
cues like morphology to build up their syntactic structure [...]" (p.
40). Adult learners, in contrast, seem to have some access to syntactic
finiteness already at the first stage of learning. However, "their verb
morphology is strongly deficient, variable and irregular, until very
late in the development" (p. 40). Adults can rely on their L1 knowledge
of syntactic categories and structures but still need to learn the
specific forms of L2.
Haberzettl studies "'tinkering' with chunks" (p. 45) in the speech
production of two female six-year-old Turkish learners of German. She
focuses on the interlanguage phenomenon exemplified below:
Ein Junge ist die Fußball spielen.
A boy is the football play-INF
'A boy is playing (with the) football.' (p. 45)
The pattern combines the auxiliary or copula 'ist' and a lexical verb
which is usually used in the infinitive but may also be realized by a
participle; the lexical verb is preceded by a direct object. The author
claims that this pattern does not serve to mark imperfective or perfect
aspect but suggests a "form-oriented explanation" (p. 57): the pattern
NP-'ist'-NP-V is a result of a combination of two more basic patterns
that the learner is familiar with, namely NP-NP-V and NP-'ist'-NP. This
non-target pattern, in Haberzettl's view, serves as a precursor to
target-like constructions with split verb forms, which will be learned
later in the acquisition process. Furthermore, this pattern seems to
corroborate the claim that "L2 learners first concentrate on detecting
patterns in the input which are used as automatized routines, allowing
for fluent production of utterances" (p. 62). Such patterns, however,
are not merely reproduced but combined to construct new utterance
schemes. On the whole, L2 acquisition by children seems to show a
prevalence of form over function.
Dimroth et al. compare the development of finiteness marking in L1
learners and L2 learners of Dutch and German, with special
consideration of the semantic operation of "linking", which "validates
the state of affairs expressed in the utterance [..., i.e.] expresses
that this state of affairs is indeed true for the particular temporal-
spatial anchorpoint talked about" (p. 65). This validation, according
to the authors, is achieved by different means at different stages of
L2 acquisition. The first of these stages, the holistic stage, is
characterized by the use of Dutch 'ja' and 'niet', Dutch 'nee' and
German 'nein' and their positive counterparts as "modal operator with
scope over the clause structure as a whole" (p. 72). On the next stage,
conceptual ordering, validation is achieved through a closed class of
lexical items. These appear in a fixed position between the initial
topic and the final predicate and consist of either modal or assertive
expressions (e.g. Dutch 'kanwel', 'doettie' or German 'soll', 'kann')
or adverbial elements (e.g. Dutch 'eve', 'graag', or German 'auch',
'noch'). At the finite linking stage, the validation of the relation
between topic and predicate is grammaticalized: auxiliaries are used to
express aspect and illocutionary force and the grammatical relation
between auxiliary and predicate have been established, i.e. "the finite
verb has come to be used as a grammatical linking device functioning as
the head of a head- complement structure" (p. 90). Furthermore, the
"acquisition of morphological person/number agreement with an external
argument is evidence of the acquisition of a specifier-head relation
between the NP and the auxiliary" (p.86). The three stages of
validation are strikingly similar in L1 and L2 acquisition.
In the last article on the aspect of finiteness in language acquisition
Gretsch investigates into the means of expressing 'topic time' among L1
and L2 learners of German, in particular the exploitation of morphology
and adverbials. It is usually assumed that in expressing topic time L1
learners choose an 'early-morphology' strategy, i.e. temporal anchoring
at early stages of learning is expressed through tense and aspect
marking and is supplemented by the use of adverbials later in the
acquisition process. L2 learners, on the other hand, are generally
assumed to use adverbials at early stages of acquisition and only later
exploit morphological means. The author's analysis of data from the
European Science Foundation corpus of second language acquisition (ESF)
and three longitudinal corpora of child language and material from the
CHILDES database indicate that "this coarse picture of opposing
developments" (p. 95) needs to be refined: while L2 learners "exhibit a
sharp bias towards the adverbial option" (p. 115-16), both the
morphological and the adverbial option are available to children. The
two strategies, therefore, "do not constitute separate routes of
development but form an acquisitional continuum within which children
and to some extent also adults can find their individual paths [...]"
(p. 114).
In the first article on scope relations, Guilano investigates into the
acquisition of negation and verbal morphology, and how these two
domains interact in two Spanish L2 learners of French and two Italian
speaking learners of English. In regard to negation, the author finds
that the acquisition process consists of three major phases: a
nonverbal nominal stage, where the learner exploits the pattern
'negator + noun/adjective/adverb'; a non-finite verb stage, which is
marked by the appearance of uninflected lexical verbs and preverbal
negators, as expressed in the pattern 'negator + non-finite lexical
verb + (indirect object)'; and a finite verb stage with the replacing
of preverbal negation by postverbal/post- auxiliary or discontinuous
negation: 'auxiliary/copula + not'; 'don't/doesn't + verb'; '(ne) +
verb + pas'. Interestingly, the third stage is closely linked to the
development of verbal inflection: "finite (or relatively finite)
relational predicates always go along with postverbal negation" (148).
The author proposes a pragmatic explanation for this finding:
relational predicates ('be'/'être' and 'have'/'avoir') mark the topic
time of the utterance, i.e. the time span for which an assertion is
made. Accordingly, relational predicates may not fall within the scope
of the negator since "the non- validity of a prepositional content must
necessarily be asserted for a given time span" (150); the negator must
therefore follow the relational predicates. The lexical verb, however,
since it shows no clear reference to topic time may fall within the
scope of the negator and is therefore preceded by it.
The paper by Bernini discusses the use of the copula 'essere' in L2
Italian on the basis of acquisition data from one adult Eritrean
learner. The author finds that the development of copula functions
encompasses three stages: the copula is first used in an equational
function and then employed as an auxiliary together with a past
participle form in the formation of the compound past. In both these
functions the copula serves as "an explicit link to finiteness with
lexical elements which cannot incorporate finiteness as inflected verbs
in the target language do" (175). At a third stage the learner has
extended this use of the copula to a non-target auxiliary function
"with any other verbal elements which in the learner variety cannot yet
incorporate the expressions of finiteness [...]" (175). On the whole,
the copula is exploited by L2 learners of Italian to establish
finiteness at early stages in L2 acquisition when full verb morphology
is not yet available.
Benazzo explores the interaction between the development of verb
morphology and the acquisition of a particular class of temporal
adverbs, namely "temporal adverbs of contrast such as again, already,
still, yet [...]" (187). It is usually assumed that the use of this
particular kind of adverbs is a characteristic feature of the L2
production of advanced learners, and the supposedly late acquisition of
these adverbs is ascribed to their cognitive complexity. The author,
however, suggests a reassessment of this general statement due to the
findings of her analysis of longitudinal data of eight L2 learners of
English, French and German. While it is true that temporal adverbs of
contrast are not used at the prebasic state, iterative use of adverbs
(such as 'again' in 'at 10 John was sleeping again') is already
attested in L2 production at the basic stage of L2 acquisition. With
the emergence of finite verb morphology at the postbasic stage,
reference "to the actual time span of the event talked about" (207) can
be signalled by the verb phrase, which leaves temporal adverbs of
contrast free to "make reference to alternative (previous) time spans
of the same event" (207). Cognitive complexity, the author contends,
does not explain this order of acquisition. Rather, "[t]he reasons for
the observed acquisitional sequence are to be found in the constraints
governing the learner system at a given time and in the discourse
functioning of the items in question [...]: internal factors concerning
the grammaticalization process observed in learner production seem to
be better candidates than cognitive factors [...]" (208).
Hulk compares the acquisition and use of French 'aussi' and Dutch 'ook'
in the production of a bilingual French/Dutch girl to acquisition data
from one monolingual child for each of the two languages. In respect to
the semantics and pragmatics of 'ook' and 'aussi' the author does not
find any difference between the use of these particles in the bilingual
child and the monolingual children. Both 'ook' and 'aussi' are used as
topic- and focus particles, regardless of mono- or bilingual
acquisition. In contrast, slight differences in the syntax were
observed: the bilingual child, for instance, used 'ook' in ways that
are not found in the monolingual learner of Dutch, "they are "un-
Dutch", more "French- like"" (228). Similarly, the bilingual girl, in
comparison to the monolingual French leaner, used 'aussi' more
frequently at the end of the utterance and less frequently at the
beginning or utterance-internally. So while an intra-individual cross-
linguistic influence on the acquisition of 'ook' and 'aussi' can be
assumed for the level of syntax, there is no indication of such an
influence for the semantic and pragmatic level.
Becker and Veenstra in their paper on French-related Creole prototypes
as basic varieties address the question as "to what extent the
grammatical properties of Creole languages can plausibly be attributed
to what we know of the process of L2 acquisition of their lexifier
languages" (233). In particular, the authors focus on inflectional
morphology since it has been shown that both in Creole prototypes and
at a specific stage in L2 acquisition (the basic variety) inflectional
morphology gets marginalized and minimalized. In L2 acquisition of
French, the basic variety shows two formal variants of lexical verbs,
namely a short form 'verb-/0/' and a long form 'verb-/e/'. It is
important to note that this distinction does not have any functional
value, the two forms occur in free variation. The functional
differentiation of the two forms only emerges in the post-basic
variety. Interestingly, the formal distinction of long and short forms
is present in some French-related Creoles but serves functions that are
different from those that are found in French and may vary between
different Creoles. The authors therefore conclude that the genesis of
French-related Creoles proceeded in two distinct stages: at first, the
dominated group of African slaves had sufficient access to the target
language French, which led to a sequence of Creole development that is
similar to L2 acquisition. However, due to an increase in the number of
slaves in the colonies, the availability of the target language model
decreased. Because of this reduced access to the original target
language French, a target shift occurred. Henceforth, "the new target
was the Basic Variety" (255). This basic variety, in turn, was expanded
by first and second generation Creole speakers, which eventually led to
the Creoles as they are known today.
The second part of the volume focuses on the expression of anaphoric
relations in the discourse. Caroll and Lambert explore the factors that
determine information structure in narratives. In particular, they
focus on "the extent to which adult learners succeed in acquiring the
principles of information structure of the target language" (267). In
an analysis of data from a group of advanced and near-native French and
German learners that had to re-tell the content of a film, the authors
find that "[t]he barriers to near-native competence are not cultural
but grammatical in nature" (285). As far as coding of information
structure is concerned, L2 learners do not start from scratch. Thus,
universal principles such as 'assign topic status' are associated with
a particular set of grammatical means, which are, however, L1-specific.
L2-specific ways of coding information-structure still have to be
uncovered. That is why "native-speaker narratives sound native-like and
those of second language learners, though formally correct, do not"
(285).
Murcia-Serra explores how advanced Spanish learners of German acquire
the linkage between syntactic, semantic and informational roles in
narratives. His analysis, also based on data drawn from learners re-
telling a film, focuses on the way the learners use the subject in
their narrative. The author claims that German and Spanish exhibit
"differences of conceptualisations for the same state of affairs"
(300): while Spanish children learn to focus on Actor entities, German
children learn to pay particular attention to "the global topic entity
of a series of events" (301), i.e. the protagonist in a narrative. This
focus on the protagonist (and its coding as the subject of clauses)
contributes to the coherence of the discourse. Spanish speakers, on the
other hand, leave "the establishment of cohesion to communicate
inferences" (301). These differences in conceptualisation pose problems
for Spanish learners of German. Even at advanced stages of learning the
Spanish learners of German "keep to the conceptualisation patterns of
the Spanish language" (304), although formal means for a more target-
like way of conceptualising, namely the German passive construction,
are at the learners' disposal.
Gullberg's paper examines the use of gestures as a cohesive means in
learner discourse on the basis of data from five Swedish learners of
French and five French learners of Swedish. The author shows that
"spoken learner varieties come with particular gestural profiles that
are related to the characteristics of spoken varieties in non- trivial
ways" (312): learners, for instance, make an excessive use of full NPs
to refer to previously established referents. Similarly, referents are
usually accompanied with particular 'anaphoric gestures', disregarding
their referential status as given or new. Native speakers, in contrast,
use such gestures only when a referent is newly introduced or
reintroduced. Given referents, which would be referred to by pronouns
or other means of reference, are not accompanied by any gestures. Two
explanations for this and similar findings are discussed by the author.
Firstly, psycholinguistic evidence has shown that the over-explicitness
of referring expressions in learner language may lead to ambiguity
rather than clarity. Anaphoric gestures may be explained as a strategy
on the part of the learner to disambiguate speech through the use of
unambiguous gestures. Secondly, anaphoric gestures could be regarded
"as a reflection of speakers' (and learners') cognitive efforts to
construct utterances (323); gestures might be an reflection of idea
units and planning units, and the over-use of gestures could then be
interpreted as "a reflection of learners' L2 speech planning proceeding
by smaller units" (324).
Watorek, in the final paper of the volume, investigates into the
development of anaphoric means to refer to space and entities in 18
intermediate and advanced Polish learners of French on the basis of
data drawn from a spatial description task. Not surprisingly, it is
found that the advanced learners "use locative expressions which encode
more complex spatial concepts" (332). The author finds acquisitional
sequences of spatial reference that are akin to those found in previous
research on L2 and L1 acquisition: intermediate learners show a higher
proportion of topological relations, such as inclusion, exclusion, or
neighbouring, in spatial description. Advanced learners make more
frequent use of projective relations, i.e. spatial localisation on the
basis of a three-dimensional orthogonal axis system. The author's
explanation for this acquisitional path involves pragmatic complexity:
"[a] topological expression [...] is in principle less complex to use
than an expression that encodes the projective relations where the
speaker must make calculations based on his origio" (353).
CRITICAL EVALUATION
Christine Dimroth and Marianne Starren have compiled a very interesting
selection of papers for all those who are concerned with the
acquisition of language. By concentrating on the two domains of
finiteness/scope relations at the utterance and anaphoric relations at
the discourse level, the editors have chosen two important fields of
language acquisition, which allow them to explore learner production
from very early to very advanced stages. Furthermore, the number of
languages and the range of varieties that are analysed will make this
volume an interesting read for researcher from diverse backgrounds. It
has to be pointed out, however, that the number of articles that
explore L1 acquisition is low in comparison to those that are concerned
with L2 learning. This imbalance is most strongly felt in regard to the
second part of the book, which is concerned with the expression of
anaphoric relations: one or two articles on the development of this
aspect in child language would definitely have been welcomed by the
reader. Unfortunately, part 1 also emphasizes L2 acquisition, so that
this volume will not prove to be as stimulating for the researcher
interested in L1 acquisition as it will turn out to be for those whose
interest is with the acquisition of L2. A final remark may be made as
to the rather small number of informants that underlie most of the
analyses. Although the volume presents a considerable amount of
quantitative data, the question as to the representativeness of these
data arises. It would no doubt be helpful if some of the findings
presented in this volume were put on a statistically sounder basis in
future research. Nevertheless, the volume is highly stimulating. The
papers are, in general, of high quality, both in regard to content and
style. In addition, the proof-reading turns out to have been almost
perfect. Only a very few errata remains (for instance, "may provide
futher impetus for change" on p. 284, "or has at same time" on p. 290).
|
| |
ABOUT THE REVIEWER:
ABOUT THE REVIEWER
Rolf Kreyer is a Research Assistant in the English Department of the
University of Bonn/Germany. He holds a degree in English and
mathematics and is currently working on his PhD thesis, a corpus-based
analysis of inverted constructions in modern written English. His
research interests include syntax, text linguistics and corpus
linguistics.
|
|
|
|
|
|
