Michael P Maratsos

Semantic restrictions on children's passives☆

Abstract Four studies are described in which children were asked to comprehend physical action passives such as Superman was held by Batman and mental verb passives such as Goofy was liked by Donald. Over a variety of comprehension methods, including questions in response to spoken sentences alone, or in choices of different pictures, children were found to be consistently poorer in comprehending the mental verb passives. Absolute competence could not be estimated securely because of variations in level of accuracy induced by different methods. But it appeared to be a reasonable interpretation that mental verb passives are understood very poorly at an absolute level by preschool children and many early grade school children, and are not understood as well as action verb passives until well into the grade school years. This is so despite the fact that both action and mental verb passives can be described uniformly at the level of the underlying grammatical relations, subject and object. Possibilities for explaining these limitations are discussed. It is concluded that it is not likely the children lack constructs such as subject, verb, and object. Rather, the limitations on the passive seem to arise from childrens active construal of input as indicating semantic conditions on the applicability of the passive.

The effects of stress on the understanding of pronominal co-reference in children

Children aged 3, 4, and 5 years old were given sentences to act out in which the presence or absence of stress on a pronoun was critical to the reference of the pronoun. They were then divided up into three groups based on their proficiency in imitating sentences. Performance on unstressed pronouns was uniformly high in all three groups. Stressed pronouns were acted out like unstressed pronouns by the least advanced group, and improvement was radical from the lowest to highest groups. In the sentences used, unstressed pronouns could be interpreted by a natural cognitive strategy, in which the least change is required in the roles of the sentence actors. Early performance can be accounted for by use of this strategy. Interpretation of stressed pronouns required reversal of the role of one actor, which the less advanced children evidently found difficult. Results are discussed for their significance for theories of pronominal reference.

Children who get worse at understanding the passive: A replication of Bever

Children 3 and 4 years old were tested for their competence in understanding three full passives as a pretest for another experimental task. The proportion of those failing two or more of three passives was found to rise significantly in the age periods 3, 8,3, and 11, replicating findings reported by Bever. The present results are discussed in this light, and a comment is made on Bevers outline of the general sequence of language acquisition in the child.

More overregularizations after all: New data and discussion on Marcus, Pinker, Ullman, Hollander, Rosen & Xu

Marcus, Pinker, Ullman, Hollander, Rosen & Xu (1992) claim that when the irregular past form of a verb is known, it is immediately known to be the correct form, such that over-regularizations only occur as speech errors, not as a genuine grammatical alternative; as a result, they argue, over-regularization rates are, when carefully inspected, very low. In the present paper: (1) it is shown that even if over-regularizations are a genuine grammatical alternative, overall rates in samples would still be low for most children; (2) careful analysis shows evidence for substantial over-regularization periods in three longitudinal subjects ages 2;5-5;2 (Abe), 2;3-5;2 (Adam) and 2;3-5;0 (Sarah); (3) Abes much higher rates follow from general developments in his past tense acquisition, in ways not consonant with Marcus et al.s formulations.

A Social-Technological Model for the Evolution of Language [and Comments and Reply]

This paper develops a model for the evolution of language that is consistent with semanticist and pragmaticist interpretations of the forms and functions of language and the processes of language acquisition and language change. The proximate aspect of the model emphasizes the social contexts of language acquisition and language change; its adaptive aspect emphasizes the sociobiological concept of communication as social manipulation. Both aspects emphasize the relationship between subsistence technology and social behavior. Specifically, the model suggests that the stages of evolution of the lexical and syntactical systems roughly parallel the stages of their acquisition; that primitive lexical forms of reference and request first arose among the earliest hominids for food location and food sharing in relation to extractive foraging on embedded foods; that simple syntax first arose among Homo erectus for encoding regulatory rules and procedures concerning recruitment, aggregation, and coordination of workers at resource sites in relation to big-game hunting; and that complex syntax arose among H. sapiens for encoding procedures for predicting resource distributions and constitutive rules for classifying relationships and performing ritual transformations of statuses and relationships, e.g., kinship terminological systems and rules of exogamy in relation to subsistence specialization.

A Developmental Study of Factivity and Negation in Complex Syntax.

Two groups of preschoolers and one group of young grade-schoolers were tested for their comprehension of presuppositions and negation in complex syntax. Four types of sentences were presented to each child: affirmative and negative versions of sentences with factive main predicates, which presuppose the truth of the proposition of the complement clause, and nonfactive main predicates, which do not. A forced-choice design was used: the children chose the agent subject mentioned in the complement clause, thus affirming the complement; or the unmentioned agent, thus denying the complement. Five factive predicates and five non-factive predicates were used so as to permit a comparison within each group. Competence increased into the early school years: the oldest group of participants showed a fair mastery of the syntax-semantics of the predicates of the study. The younger children showed errors of two different kinds, described as the Overextended Negation Tendency and the Overextended Affirmation Tendency. Both of these errors decreased markedly in the oldest group. The non-unitary nature of the acquired competence is discussed. In particular, it is pointed out that (1) factivity is not a grammatically marked operation and as such leads to what appears to be a gradual acquisition pattern; (2) the test of factivity comprehension employed here demanded a competence beyond that of normal use.

How fast does a child learn a word

This discussion argues that for many word meanings, the child has to assemble a new category, using relatively slow information-sifting processes. This does not cause high semantic errors, because children probably hold off using a word until much such sifting has occurred, rather than producing the new word as soon as they have any information on it.

Children's use of semantic analysis in the interpretation of missing subjects: Further evidence against the MDP

Preschool and many older children often have difficulty understanding who carries out the complement action (e.g., to go in sentences such asMary promised John to go; this is so, even though they easily understand this information in sentences such asJohn told Mary to go. C. Chomsky (1969) proposed that childrens errors arise from the overgeneral application of a purely structural Minimal Distance Principle. Maratsos (1974), however, hypothesized that children err by overgeneralizing a different principle which he called the Semantic Role Principle. According to this principle, the Goal-Recipient of the spoken message, not the closest noun phrase, is understood as the person who carries out the act referred to by the infinitival complement. The two studies reported here were designed to determine which of these principles children use. Preschool children were taught to understand a specially designed novel construction. The children then acted out sentences containing related but novel uses of the construction such that they would respond differentially according to which interpretive principle they used. The results strongly favored the Semantic Role Principle, rather than the MDP described by Chomsky (1969), or a related, more complex MDP described by Rosenbaum (1967). It is further discussed how childrens formulation of the Semantic Role Principle, rather than the MDP, might arise from their prior analyses of related constructions, such as the imperative construction, rather than following from an innate preference, as suggested in Maratsos (1974).

A sunny view of polarization.

There are a number of ways to react to the whole problem of polarization that Rispolis review brings up. As a first reaction, like Rispoli, personally, I deplore the polarization. I find neither side likely to be right enough about everything to be entitled to require the kind of all-or-nothing allegiance each side seems to demand. My complaints about the two sides are more basic than Rispolis rather specific criticisms of Elman et al . Unfortunately, going on to outline what seems generally useful and less useful in each approach in a short space like this basically consists of saying ‘I like this’ or claiming ‘this is obviously wrong’ with a little elaboration. This does not seem useful for such complicated problems, so I will leave my general complaint about the all-or-none polarization simply stated outright.

Book Reviews : Models of Language Development. Rosemary Stevenson (Milton Keynes: Open University Press, 1988). Pp viii + 125. Paperback £8.95. ISBN 0-335-09520-8

As is often true, the best introduction to the gist of this book lies not in the book jacket cover, which promises an ’up-to-date accessible treatment of major issues, theories and findings in language development’, and a treatment of ’new developments in linguistics, learnability theory, and computer simulation’, a book which because of its ’interdisciplinary content’, ’provides the only introductory text that examines language development in a way that would be useful for courses in cognitive science’. Rather, the key is the preface. Here the author writes that in 1981, ’I presented a talk to the Developmental Section of the British Psychological Society in which I was critical of the current models of language development.’ She notes that after being critical, she wished