Speech and Language Impairments in Children

Causes, Characteristics, Intervention and Outcome

Dorothy V.M Bishop, Laurence B Leonard, Dorothy V.M Bishop, Lawrence Leonard

• Publisher: Psychology Press
• Year: 2014
• ISBN:
  9781317715818
• Edition: 1
• Topic: Psychology
• Subtopic: History & Theory in Psychology
• Index: Psychology

Delayed development of speech and/or language is one of the commonest reasons for parents of preschool children to seek the advice of a paediatrician.
Accessible to non-academic Speech and Language Impairments provides an overview of recent research developments in specific speech and language impairments, written by experts in the field. Topics include normal and disordered development of problems, crosslinguistic studies, pragmatic language impairments, early identification, educational and psychiatric outcomes, acquired epileptic aphasia and experimental studies of remediation. The book concludes with a chapter by Michael Rutter that gives guidelines for conducting and evaluating research in this field.

1

Acquiring syntax is not what you think

Michael Tomasello

Many developmental psycholinguists assume that young children operate with adult syntactic categories. This assumption has never had strong empirical support, but recently a number of new findings have emerged — both from systematic analyses of children’s spontaneous speech and from controlled experiments — that contradict it directly. In general, the key finding is that most of children’s early language is item based, and therefore their language development proceeds in a piecemeal fashion with no indication of any system-wide syntactic categories, schemas, or parameters. Since nativist theories of language acquisition rely explicitly on adult linguistic categories as their major analytic tools (i.e. as these are embodied in formal grammars), the implications of these new findings for nativist theories are discussed. Also discussed are the outlines of an alternative, constructivist theory of child language acquisition.

INTRODUCTION

Most approaches to the study of first language acquisition use adult-like grammatical categories and rules to describe children’s language. This is especially true of the dominant approach to the study of children’s syntactic development, namely, the theoretical paradigm based on Chomsky’s universal grammar (see Pinker, 1994). In this paradigm, it is hypothesised that children innately possess the abstract syntactic competence of adults, and all they must do in development is discover how that applies to the particular language they are learning. So, for example, all children are born with a “head-direction parameter”; that is, they know innately that their language is either head-first (as in the Spanish casa grande, where the noun is the head of the phrase and comes first) or head-last (as in the English big house, where the noun/head is last). Hearing a particular language simply sets the head-direction parameter in one way or the other. Once the requisite parameters are set, children have essentially adult-like syntactic competence. This has come to be known as the nativist view of language acquisition, because the child’s innate knowledge is seen as all-important, and processes of social learning and imitation are not thought to play any important role in acquisition.

My colleagues and I are currently developing a theoretical approach to child language acquisition that conflicts with this Chomskyan approach in a number of important ways. My goal in this chapter is to sketch out the broad outlines of that approach. It has at least four elements that are surprising from the point of view of nativist (i.e. Chomskyan) theorising about language development, especially with regard to the syntactic dimensions of the process.

• Imitative learning is a necessary and crucially important part of language acquisition — albeit imitative learning of a very special type described more broadly as “cultural learning” (Tomasello, Kruger, & Ratner, 1993).

• The language that children acquire initially is almost totally concrete, that is, it is based not on abstract linguistic entities such as noun phrases, verb phrases and transitivity, but rather on the particular words and phrases of a particular language. Children construct more abstract linguistic categories and schemas only gradually, and they do this in an unsystematic, piecemeal fashion — so that at any given developmental moment there may also be great variety in the abstractness of the linguistic units with which a child can operate.

• The linguistic units that children acquire via imitative learning are not only small things like words but also larger things like phrases, clauses and, indeed, whole speech acts. Thus, at any given developmental moment there may be great variety in the complexity of the linguistic units that a child can use.

• This more pluralistic way of looking at linguistic units implies that in many cases children’s creative linguistic combinations are a pastiche of linguistic units varying from one another in both complexity and abstractness.

We have been led to this new view of language acquisition first and foremost by some new observations of children’s early language that appear to be quite consistent and robust across languages — and across methodologies (naturalistic observation and experimentation) as well. In addition, this view both benefits from and is supported by some new approaches in theoretical linguistics, known broadly as cognitive and functional linguistics (e.g. Langacker, 1987, 1991; Lakoff, 1987; see papers in Tomasello, 1998a). These approaches provide a rigorous foundation for identifying the kinds of linguistic units (referred to as “linguistic constructions”) that children experience in the adult language around them. I shall first present the new data and then provide some theoretical reflections on their implications for the study of child language acquisition.

SOME RECENT DATA

The central issue of current concern is whether children are operating from the outset of development with adult-like linguistic categories and schemas (as espoused in one form or another by the different varieties of linguistic nativism) or, alternatively, whether children begin the acquisition process with only highly specific and concrete linguistic items and structures. To make the distinction as clearly as I can, the question is whether children’s earliest utterances are underlain by such things as noun phrases, verb phrases, prepositional phrases, head direction parameters, functional categories, and so forth, or whether, alternatively, they are underlain by less abstract units of varying sizes. As always, there are two basic methodological approaches to this question: (1) the observation and analysis of children’s spontaneous speech, and (2) experients with children’s linguistic comprehension and production.

Observational studies

Following in the footsteps of pioneers such as Braine (1976), in my 1992 diary study I documented virtually all of my English-speaking daughter’s earliest verbs and linguistic constructions from 15–24 months of age (Tomasello, 1992). The major findings of that study may be summarised as follows:

• Of the 162 verbs used, almost half were used in one and only one construction type, and over two-thirds were used in one or two construction types — where construction type means verb-participant configuration (e.g. Draw car and Draw tree are the same construction type, whereas Draw on paper (locative), Mommy draw (agent), and Draw with pencil (instrument) are three additional construction types).

• At any given developmental period, there was great unevenness in how different verbs, even those that were very close in meaning, were used. For example, at 23 months of age the verb cut was used in only one simple construction type (Cut + X) whereas the somewhat similar verb draw was used in many different construction types, some with much complexity (e.g. I draw on the man, Draw it by Santa Claus). Where information on adult usage was available for a given verb, there was a very good match with child usage.

• There was also great unevenness in the syntactic marking of the “same” participant across verbs such that, for example, at a given developmental period, one verb would have its instrument marked with with or by but another verb, even when used in utterances of the same length and complexity, would not have this marker. Some verbs were used with lexically expressed subjects whereas others at the same time were not, even though they were used in comparable construction types and in comparable pragmatic contexts. For instance, there were subjects for take and get (I take and you get it) but not for put (put it there).

• Morphological marking (e.g. past tense -ed) on verbs was also very uneven. Roughly two-thirds of all verbs were never marked morphologically for tense or aspect, one-sixth were marked for past tense only, one-sixth marked for present progressive only, and only four verbs (2%) marked for both of these functions at any time during the second year of life. o On the other hand, within any given verb’s development, there was great continuity such that new uses of a given verb almost always replicated previous uses and then made one small addition or modification (e.g. the marking of tense or the adding of one new participant). By far the best predictor of the use of a given verb on a given day was not the child’s use of other verbs on that same day, but rather her use of that same verb on immediately preceding days.

The resulting hypothesis, the Verb Island Hypothesis, was that children’s early language is organised and structured totally around individual verbs and other predicative terms, that is, the 2-year-old child’s syntactic competence is comprised totally of verb-specific constructions with open participant slots. Other than the categorisation of participants, nascent language learners possess no linguistic abstractions or forms of syntactic organisation. This means that the syntactic categories with which children are working are not such verb-general things as “subject” and “object”, or even “agent” and “patient”, but rather such verb-specific things as “hitter”, “hittee”, and “thing hit with”.

Using a combination of periodic sampling and maternal diaries, Lieven, Pine, and Baldwin (1997; see also Pine & Lieven, 1993; Pine, Lieven, & Rowland, 1998) have found similar results in a sample of 12 English-speaking children from 1–3 years of age. In particular, they found that virtually all children used most of their verbs and predicative terms in only one construction type early in language development. In addition, they examined the children’s subject-verb-object (SVO) utterances for evidence that they knew how to syntactically mark subjects and objects. Looking at those SVO sentences that used personal pronouns, they found basically no evidence that children knew that such forms as I and me, we and us, and they and them contrasted in their participant roles (nor was there any evidence of verb-general participant marking from SVO utterances using full noun phrases). Following along these same lines, Pine and Lieven (1997) found that when young English-speaking children begin to use the determiners a and the they do so with almost completely different sets of nouns (i.e. there is almost no overlap in the sets of nouns used with the two determiners), suggesting that children at this age do not have any kind of abstract category of determiner that includes both of these lexical items.

A number of systematic studies of children learning languages other than English have found very similar results. For example, Pizzuto and Caselli (1992, 1994) investigated the grammatical morphology used by three Italian-speaking children on their simple, finite, main verbs, from approximately 1.5–3 years of age. Although there are six forms possible for each verb root (first-person singular, second-person singular, etc.), 47% of all verbs used by these children were used in one form only, and an additional 40% were used with two or three forms. Of the 13% of verbs that appeared in four or more forms, approximately half of these were highly frequent, highly irregular forms that could be learned only by rote. The clear implication is that children do not master the whole verb paradigm (i.e. all six persons and numbers) for all their verbs at once, but rather they master some endings with some verbs — and often different ones with different verbs. Rubino and Pine (1998), Berman (1982), and Berman and Armon-Lotem (1995), have found very similar patterns for Brazilian Portuguese and Hebrew-speaking children, respectively.

Of special note are so-called overgeneralisation errors in spontaneous speech because, presumably, children have not heard such forms used in adult speech. In the context of a focus on syntax, the overgeneralisations of most interest are those involving sentence-level constructions (for example, She falled me down or Don’t giggle me), in which the child uses verbs in syntactic constructions in an “incorrect” way that seems to indicate that she has some abstract, verb-general schema for such things as a transitive SVO construction. Bowerman (1982, 1988) in particular has documented a number of such overgeneralisations in the speech of her two English-speaking children, and Pinker (1989) has compiled examples from other sources as well. The main result of interest in the current context is that Bowerman’s children and the other children produced very few sentence-level overgeneralisations before about 3 years of age and virtually none before 2.5 years of age — suggesting that very young children do not have verb-general syntactic constructions.

These data-intensive studies from a number of different languages together show a very clear pattern. Young children’s earliest linguistic productions revolve around concrete items and structures — particular verbs such as push, pull, cut, and draw, and the basic sentence-level constructions in which they participate. There is absolutely no evidence that young children are using abstract categories and schemas in their spontaneous linguistic productions, other than the participant slots in these verb island constructions. Rather, each of these items and structures undergoes its own development — presumably based on the child’s linguistic experience and other factors affecting learning — in relative independence of other items and structures. This pattern persists in most cases until around the third birthday. Those who argue that young children do possess abstract, adult-like categories such as “subject”, “object”, “agent”, or “patient”, must maintain that their naturally occurring language for some reason does not reflect their underlying syntactic competence.

Experimental studies

There is no question that young children comprehend and produce adult-like linguistic items and structures from early in development, even if this is only with some items and structures (Braine, 1971; Brown, 1973; Bloom, 1992; DeVilliers & DeVilliers, 1973). But, as noted earlier, the adultlike production and comprehension of language by children is not diagnostic of the underlying processes involved. Adult-like comprehension and production may emanate either from abstract, adult-like linguistic knowledge or from item- and structure-specific knowledge in which children are simply reproducing the words and sentences they have heard from adults.

The main way to test for underlying process is to introduce children to novel linguistic items that they have never heard before, and then see what they do with them — as in the famous wug-test of Berko (1958). For questions of syntax in particular, the method of choice is to introduce young children to a novel verb in one syntactic construction and then see whether and in what ways they use that verb in other, syntactic constructions — perhaps with some form of discourse enc...