LINGUIST List 15.1750

Wed Jun 9 2004

Review: Lang Acquistion/Disorders: Levy & Schaeffer

Editor for this issue: Naomi Ogasawara <>

What follows is a review or discussion note contributed to our Book Discussion Forum. We expect discussions to be informal and interactive; and the author of the book discussed is cordially invited to join in. If you are interested in leading a book discussion, look for books announced on LINGUIST as "available for review." Then contact Sheila Dooley Collberg at


  1. Susana Perales Haya, Language Competence Across Populations: Toward a Definition of SLI

Message 1: Language Competence Across Populations: Toward a Definition of SLI

Date: Tue, 8 Jun 2004 15:52:56 -0400 (EDT)
From: Susana Perales Haya <>
Subject: Language Competence Across Populations: Toward a Definition of SLI

EDITORS: Levy, Yonata; Schaeffer, Jeannette 
TITLE: Language Competence Across Populations
SUBTITLE: Toward a Definition of Specific Language Impairment
PUBLISHER: Lawrence Erlbaum Associates
YEAR: 2003
Announced at

Susana Perales, University of the Basque Country (Spain)


The editors of the book have gathered in this volume 17 articles
(chapters) that were presented at a workshop on Specific Language
Impairment (SLI) held in Jerusalem in 2000. The book is divided into
Part A, which comprises 15 chapters and whose title is ''Language
competence across populations'' and Part B, ''Toward a definition of
SLI?''. The first part is subsequently subdivided into three sections:
The first one is ''The characterization of Specific Language
Impairment'' (chapters 1-7), the second one is ''Methodological
Concerns'' (chapters 8-11) and finally, the third section is entitled
''Language competence in children with neurodevelopmental disorders''
(chapters 12-15).

As the editors themselves declare in the preface, the book's focus is
on ''the question of how much variability linguistic competence can
take in children as they develop'' (p. x). Thus, although the focal
theme is SLI, other language disorders (e.g. language in autistic
children or children with Williams Syndrome) are also portrayed and
discussed in the book.


The first chapter ''Lenneberg's dream: Learning, normal language
development, and specific language impairment'' (Ken Wexler) explores
the interrelation between the study of SLI and normal language
acquisition. Wexler first introduces the Optional Infinitive (OI)
stage in normally developing children, which is characterized by the
production of an infinitive form in contexts where the adult grammar
requieres an inflected verb. Wexler explains the OI stage as the
result of a limitation in the child's computational system which
prevents him from checking more than once against the D-feature of
noun phrases. The ''Unique Checking Constraint'' (UCC), as Wexler
calls it, ''is a developmental constraint on the computational system
of language'' (p. 33) and not only gives a linguistic account of the
OI stage, but also predicts that children acquiring null subject
languages (e.g. Spanish, Italian) do not go through this stage, as has
been reported in the literature (see Wexler 1998 for an overview). As
for how normally developing children recover from the OI stage, it is
proposed that ''it matures away, under genetic guidance'' (p. 40),
and Wexler reviews several arguments that support this proposal.

As far as SLI is concerned, Wexler explains how the proposal to
account for OIs in normal children can also be used for SLI
children. Thus, he claims that SLI children go through an Extended
Optional Infinitive (EOI) stage, implying that the maturation
responsible for the withering away of the UCC in normal children does
not operate or is delayed in SLI children. In this way, both normal
and impaired language development can be accounted for with the same
theoretical apparatus, something highly desirable in linguistic
theory. Moreover, as Wexler makes explicit all throughout the article,
linguistic-based accounts are preferable to other proposals that place
the child with no specific knowledge of linguistic categories but only
with a set of general cognitive learning mechanisms (e.g. Tomasello
2000, cited in the article), with the result that the phenomena
related to the OI stage are left unexplained.

In chapter 2 ''A unified model of specific and general language delay:
Grammatical tense as a clinical marker of unexpected variation'',
Mabel Rice deals with tense marking as a diagnosis of SLI and places
emphasis on distinguishing between a selective delay in language
acquisition, which is indicative of a deficit in an element of
grammar, and a general delay, which does not necessarily imply that a
particular aspect of grammar is affected.

Rice reports on various longitudinal studies of English- speaking SLI
children. First, she deals with grammatical tense marking (e.g. third
person ''-s'') and her results show that for the SLI group, the
variation expected at age 5 falls below that of normally developing
children. In order to examine whether the problem with grammatical
tense marking is a reflection of a more general deficit affecting low
salient morphemes, Rice compared performance on the plural marker
''-s'' and found that all groups of children used this marker
productively, which denotes that grammar, rather than phonology, is
involved in SLI. However, that wasn't the case in lexical acquisition
and mean length of utterance (MLU) measures, where SLI children seemed
to be lagging behind the normal controls, and no particular aspect of
grammar was affected.

When comparing SLI children versus children with Williams Syndrome,
Rice found that they performed similarly in MLU, plurals and
prepositions, but the SLI group performed significantly worse on tense
marking. This suggests that grammatical tense marking is a reliable
clinical marker. To conclude, Rice points out that, in order to decide
whether low performance in grammatical tense marking is indicative of
a general language delay or a selective language delay, other measures
(e.g. MLU, lexical development, etc.) must be taken into

In the third chapter (''Two of a kind? The importance of commonalities
and variation across languages and learners'') Martha Crago and
Johanne Paradis examine how SLI manifests itself with regard to verbal
morphology in three languages: English, French and Inuktitut, and in
three different populations: normally developing children, SLI
children and second language learners. Crago and Paradis raise several
issues regarding the Extended Optional Infinitive stage proposed by
Wexler (chapter 1). The matters they discuss concern the ultimate
attainment of SLI affected children, the fact that SLI does not
manifest in the same way across languages, and that SLI children do
not always behave like normal language-matched controls. They
highlight the importance of figuring out if SLI children reveal simply
a delay with respect to normal language acquisition or whether it
would be better to speak of a deviant language acquisition
process. Also, they cast doubt on the appropriateness of the term
''Optional infinitive'' because SLI errors across languages normally
involve the use of default rather than infinitive forms.

Chapter 4 ''Do heterogeneous deficits require heterogeneous theories?
SLI subgroups and the RDDR hypothesis'' (Heather K. J. van der Lely)
deals with the controversy surrounding the domain-specific versus
domain-general accounts of SLI. Some scholars regard evidence of SLI
children affected in phonological or pragmatic measures as indicative
that SLI is not a domain-specific language deficit (e.g. Karmiloff-
Smith 1998, cited in the article). However, after reviewing research
studies exploring different language abilities within homogeneous
subgroups of SLI children, Van der Lely argues that ''the simplest
explanation for the association between disorders is the propensity
for comorbidity of disorders in development'' (p. 123).

Van der Lely goes on to report the results of her study of a
homogeneous SLI sub-group: the Grammatical SLI subgroup, which is
characterized by deficits in tense and agreement marking along with
other problems with syntactic operations like, for instance,
theta-role assignment or embedded clauses. The author claims that
Grammatical SLI is caused by a deficit involving the obligatoriness of
movement operations. This proposal allows her to account for a wide
range of syntactic phenomena that characterize the Grammatical SLI
subgroup performance.

In chapter 5, Jeannette Schaeffer addresses the issue of modularity
within the human language faculty. More specifically, she concentrates
on the study of pragmatics as an independent module from the lexicon
and the computational system. For this purpose, she investigates the
performance of Dutch SLI and normally-developing children on object
scrambling, a syntactic phenomenon which is triggered by
pragmatics. Schaeffer entertains the hypothesis that SLI children are
impaired in their grammar, but not in their pragmatics and therefore,
the application of object scrambling will not be affected in these
children. Her results support this hypothesis as Dutch SLI children do
correctly scramble referential objects 96% of the time. Thus, she
concludes, the pragmatic module is not affected in SLI, which is
evidence that pragmatics and the computational system are independent
modules within the human language faculty.

In Chapter 6 (''Specific language impairment and linguistic
explanation'') Jan de Jong considers the issue of the different
predictions and explanations that stem from distinct theoretical
proposals. De Jong reports the results of an experimental study with
Dutch SLI children which shows that the developmental order of
acquisition of finiteness in English is not comparable to the Dutch
order. Then, he goes on to propose an alternative explanation of the
OI stage (Wexler, chapter 1), namely the ''Modal drop hypothesis''
(Ingram & Thompson 1996), which claims that children OI sentences
should be interpreted as containing a null modal. The chapter ends
with the observation that the same surface phenomenon can have diverse
theoretical interpretations, each of which differing in the degree of
knowledge the SLI child is credited with.

Chapter 7 is the last chapter of the first section and its title is
''The role of language typology in linguistic development:
Implications for the study of language disorders'' (Dorit Ravid, Ronit
Levie and Galit Avivi Ben- Zvi). These authors explore the knowledge
of Hebrew derivational morphology by SLI children and normally
developing controls. An experimental study was conducted on
comprehension and production of nouns and adjectives, which showed
that SLI children performed significantly lower than normal controls
in most cases. Thus, the authors assert that ''derivational morphology
was found to be diagnostic of the impaired population'' (p. 189). They
also underscore the importance of looking at typologically diverse
languages, and including measures like comprehension in order to gain
a greater understanding of SLI.

In chapter 8 (''Specific language impairment: characterizing the
deficits''), Laurence B. Leonard discusses several methodological
questions with regard to the study of SLI focusing on three broad
areas: definition of SLI, research design and hypothesis testing, and
data interpretation. These issues are central to a well known fact in
SLI research: children with SLI are far from being a homogeneous
group. This poses a problem for replicability, as two studies focusing
on the same linguistic aspect may obtain different results due to the
intrinsic characteristics of the sample they look at. Thus, Leonard
proposes the inclusion of descriptive data so as to facilitate subject
selection, replication and identification of subgroups. The author
also discusses the convenience of several research methods, such as
group versus individual case studies, group-matching criteria, the
importance of designing studies that take into account how SLI
children are affected by different types of treatment, as well as the
need for longitudinal studies. Finally, the chapter ends with the
discussion of how different hypotheses may lead to contrasting
conclusions, stressing that ''each discipline's basic assumptions
should be spelled out, measures should be compatible with these
assumptions, and null hypotheses should be framed in a uniform
manner'' (p. 228)

Chapter 9 (''Methodological issues in cross-group comparisons of
language and cognitive development'' by Carolyn B. Mervis and Byron
F. Robinson) discusses methodological questions regarding
group-matching procedures, chronological age confounds and other
problems stemming from cross-group comparisons. They also discuss an
alternative method: profiling, which implies, firstly, to determine
whether a given linguistic profile is characteristic of a particular
child and, secondly, to determine how characteristic/uncharacteristic
that profile is of the children who are/are not in the SLI group.

The main theme discussed in chapter 10 (''MLU-matching and the
production of morphosyntax in Dutch children with specific language
impairment'' by Gerard W. Bol) is the use of MLU measures in SLI
studies. The author illustrates some of the problems arising from the
use of MLU as a marker of productive morphosyntax as well as its use
as a means to match different groups of subjects, and he also puts
forward suggestions that may help to overcome these difficulties. In
the second part of the article, the author conducts a study on Dutch
children which attempts to figure out how Dutch SLI children
compensate for loss of utterance length. He identified seven variables
like, for instance, the use of conjunctions, prepositions and
adverbials, verb forms used for indicating past tense, etc. However,
none of the phenomena he looked at seemed to be used by Dutch SLI
children to compensate for loss of sentence length. He concludes that
MLU is not a valid measuring strategy for Dutch, and calls for
alternatives to MLU.

In chapter 11 (''Different methodologies yield incongruous results: A
study of the spontaneous use of verb forms in Hebrew'' by Esther
Dromi, Laurence B. Leonard & Anat Blass) spontaneous and elicited
production data from Hebrew SLI children and normally developing
controls are compared focusing on verb morphology. The first part of
the article is devoted to the review of published results obtained in
the investigation of verbal morphology using a series of tasks
specially designed to elicit specific verb forms. The use of tasks
allowed the researchers to manipulate variables and concentrate on
particular aspects regarding performance on verbal morphology.

In the second part of the article, the authors compare the results
obtained in the elicitation studies with spontaneous production
data. One of the strengths of the study is that the SLI children and
the controls were the same in both the elicited and the spontaneous
production studies. The authors resolve that both tasks are necessary
tools for the study of SLI and thus should be combined to get an
accurate picture of SLI children's performance.

In Chapter 12 (''Language impairment in children with complex
neurodevelopmental disorders'') Helen Tager-Flusberg examines the
linguistic differences and similarities between autistic and SLI
children. The author describes an experimental longitudinal study in
which two samples from the groups mentioned above were compared on a
series of phonological, lexical and grammatical tests. The results
point to the fact that within the group of autistic children there was
a subgroup which performed similarly to the SLI group, which leads the
author to highlight that autism and SLI are ''overlapping
populations'' (p. 311). The results of this study bear directly on
current research which is showing that there may be a shared gene in
autism and SLI.

Harald Clahsen and Christine Temple take up the issue of the
modularity of the mind in chapter 13 (''Words and rules in children
with Williams syndrome''). Their study measures performance of
children with Williams syndrome on comparative adjectives, receptive /
productive vocabulary and reading skills. They aim to show that
modular accounts are preferable to other non-modular connectionist
models, as they are able to explain a wider range of effects.

In chapter 14 (''Basic language skills in children with
neurodevelopmental disorders and the notion of brain plasticity'')
Yonata Levy studies the performance of several groups of children with
diverse neurodevelopmental disorders on a series of grammatical tasks,
as well as their response to requests for clarification. A first study
was conducted to examine the linguistic performance of 8 children with
neurodevelopmental disorders on 13 variables like syntactic agreement,
gender marking, etc. The language problems found in these children
reflected a delayed although not different linguistic profile with
respect to normally developing children.

The second study assessed children's response to clarification
requests. Roughly, it was found that the ability of children with
neurodevelopmental disorders to detect errors in their own speech as
response to clarification requests matches the pattern seen in
normally developing children. Results show that, up to an MLU of 3,
there is not much variation in the course of language development
irrespective of the population, and Levy suggests this may be due to
brain plasticity assuring a basic level of language performance.

Signed and spoken languages are compared in chapter 15 (''On the
complementarity of signed and spoken languages'') by Wendy
Sandler. She considers the idea that sign and spoken language are two
parts of a unique human language faculty. After reviewing some of the
similarities and differences between sign and spoken language, the
author shows evidence from gesture that supports her
proposal. Gesture, as the author claims, is a natural part of
communication, both for deaf signers (who gesture with the mouth) and
for oral speakers (who gesture with their hands), and its study should
lead to a deeper understanding of the human language faculty.

The second part of the book (''Toward a definition of SLI?'')
comprises the last two chapters. Chapter 16 (''Understanding SLI: A
neuropsychological perspective'' by Dorit Ben Shalom) considers how
the study of SLI can benefit from studies of other cognitive disorders
and vice versa. In the first place, the author compares SLI with
prosopagnosia (impaired face recognition) and claims that the fact
that there are specific grammatical deficits in SLI does not
necessarily imply that SLI should be defined in terms of exclusionary
criteria. Secondly, Shalom compares three phenomena related to SLI and
agrammatism: tense marking, reversible passives and nonword
repetition. She suggests that there may be two types of grammatical
processing, one phonological and one syntactic, and states that the
study of both should guide us to provide an accurate answer to the
general problem of grammatical processing in the brain.

In the last chapter (''Defining SLI: A linguistic perspective'') Jill
G. de Villiers takes up several issues that have been discussed in the
articles gathered in the book: (i) the question of whether SLI
children are a different group from other cognitively impaired
children, (ii) SLI as a reflection of a specific problem in the
language module, (iii) the heterogeneity of SLI and, lastly, (iv)
whether SLI is better described as a deficit or a delay. De Villiers
does an excellent job at summarizing and commenting the key points
raised by the articles in the book, as well as suggesting further
lines of investigation.


As a whole, the book provides an excellent overview of current issues
concerning research on SLI. For the most part, the authors have been
carrying out research on SLI for quite a number of years, which is an
extraordinary opportunity to get first-hand knowledge of the topics
that have been debated and the answers that have been advanced so far
with regard to SLI. Another outstanding issue regards the fact that
all contributors have read each other's articles and, thus, they refer
the reader to particular chapters where the same issues are discussed.
This contributes to maintain a coherent line and helps the reader to
situate arguments and positions within the SLI research framework.

The book is well suited for scholars doing research on SLI because
they will get an overall picture of the field as it stands nowadays,
as well as because there are specific proposals that suggest new lines
of investigation. This is particularly true of the section on
methodological problems, as the chapters herein included spot specific
methodological flaws that should be overcome in future
studies. Graduate students with a fair command of linguistic theory
will also find the book interesting, as well as those acquainted with
the literature on normal language acquisition.

To conclude, I consider this volume is an outstanding contribution not
only for the answers it provides, but also for the questions it leaves
open for further research.


Ingram, D. & Thompson, W. (1996) ''Early syntactic acquisition in
German: Evidence for the modal hypothesis'' Language 72: 97-120.

Karmiloff-Smith, A. (1998) ''Development itself is the key to
understanding developmental disorders''. Trends in Cognitive Sciences
2: 389-398.

Tomasello, M. (2000) ''Do young children have adult syntactic
competence?'' Cognition 74: 209-253.

Wexler, K. (1998) ''Very early parameter setting and the unique
checking constraint: A new explanation of the optional infinitive
stage'' Lingua 106: 23-79.


Susana Perales is about to obtain her Ph. D. at the University of the
Basque Country (Spain). Her interests include the study of first and
second language acquisition from a generative perspective.
Mail to author|Respond to list|Read more issues|LINGUIST home page|Top of issue