The present study investigated the universality of the early development of young children's understanding and representation of false beliefs, and specifically, the effect of language on Chinese-speaking children's performance in false belief tasks under three between-subjects conditions. The three conditions differed only in the belief verb that was used in probe questions regarding one's own or another person's beliefs, namely the Chinese verbs, xiang, yiwei, and dang. While the three words are all appropriate to false beliefs, they have different connotations regarding the likelihood of a belief being false, with xiang being more neutral than either yiwei or dang. Experiment i involved thirty-five Chinese-speaking adults who responded to false belief tasks to be used in Experiment 2 in order both to establish an adult comparison and to obtain empirical evidence regarding how Chinese-speaking adults use the three belief verbs to describe different false belief situations. In Experiment 2, 188 three-, four-, and five-year-old Chinese-speaking children participated in three false belief tasks. They were asked to report about an individual's false belief when either xiang, yiwei, or dang was used in the probe question. Results revealed a rapid developmental pattern in Chinese-speaking children's understanding of false belief, which is similar to that found with Western children. In addition, children performed significantly better when yiwei and dang, which connote that the belief referred to may be false, were used in belief questions than when xiang, the more neutral verb, was used. This finding suggests an important role of language in assessing children's understanding of belief and false belief.

A suggestion exists in the child language literature that the meanings of natural kind terms are acquired before the meanings of colour terms. Explanations have typically claimed that object terms are more salient than property terms. Such explanations, however, tend to ignore the fact that natural kind terms refer to categories with sharp, clear boundaries while colour terms refer to categories with unclear or variable boundaries. Nonetheless, there has been little evidence to show that the delay in the acquisition of colour terms arises from these semantic properties. This study compares natural kind and colour naming (and corresponding comprehension) by 48 children, ranging in age from 3;0 to 5;5. The results suggest that, contra the salience view, the apparent delay in colour naming may be explained on solely semantic grounds.

Twenty economically disadvantaged preschoolers (mean age 3;7) were randomly assigned to an intervention or a control group, and their mothers' styles of eliciting narratives from their children were assessed before and after intervention. Mothers of intervention children were encouraged to spend more time in narrative conversation, ask more open-ended and context-eliciting questions, and encourage longer narratives through back-channel responses. Children's narrative and vocabulary skills were assessed before and after the year-long intervention and 14 children participated in a follow-up assessment a year later. Narrative measures included the number and length of narratives as well as how decontextualized and informative they were. Intervention children showed significant vocabulary improvement immediately after intervention terminated, and a year later they showed overall improvements in narrative skill. In particular, intervention children produced more context-setting descriptions about where and especially when the described events took place. Such decontextualized language has been emphasized as important for literacy acquisition.

Cross-linguistic similarities and differences in early lexical and grammatical development are reported for 1001 English-speaking children and 386 Italian-speaking children between 1;6 and 2;6. Parents completed the English or Italian versions of the MacArthur Communicative Development Inventory: Words and Sentences, a parent report instrument that provides information about vocabulary size, vocabulary composition and grammatical complexity across this age range. The onset and subsequent growth of nouns, predicates, function words and social terms proved to be quite similar in both languages. No support was found for the prediction that verbs would emerge earlier in Italian, although Italians did produce a higher proportion of social terms, and there were small but intriguing differences in the shape of the growth curve for grammatical function words. A strikingly similar nonlinear relationship between grammatical complexity and vocabulary size was observed in both languages, and examination of the order in which function words are acquired also yielded more similarities than differences. However, a comparison of the longest sentences reported for a subset of children demonstrates large cross-linguistic differences in the amount of morphology that has been acquired in children matched for vocabulary size. Discussion revolves around the interplay between language-specific variations in the input to young children, and universal cognitive and social constraints on language development.

It is well known that mothers give their infants lessons in conversational competence from an early age. This study considered how maternal gestures and prosody contribute to this developing competence. It examines how mothers use ostensive marking to point out common references at different stages of development. The corpus consisted of longitudinal observations of four mother–infant dyads during free play (infants aged 0;4 to 1;1), at three stages of sensorimotor development (III, IV and V). Four dimensions of ostensive marking were considered: (1) the span of the marked utterance (holistic vs. local); (2) the communication channel used (gestural vs. prosodic); (3) the type of gestural marker (oriented, iconic, conventional, beats); and (4) the type of prosodic marker (emphasis, prosodic cliché, reinforced nuclear stress, focal accent). Although there was no clear change in the patterns of specific types of gestural or prosodic markers, the results showed that mothers adapt their gestures to the infant's processing level. Between stages III and V, they move from holistic to local and from gestural to prosodic marking. Stage IV appears to be an excellent period for observing the transition.

This study contributes to our understanding of sociocultural variation in children's early storytelling by comparing co-narrations produced by children and their families from two European-American communities, one working-class and one middle-class. Six children from each community were observed in their homes at 2;6 and 3;0 years of age, yielding a corpus of nearly 400 naturally-occurring co-narrations of past experience. Analyses of generic properties, content, and emotion talk revealed a complex configuration of similarities and differences. Working-class and middle-class families produced co-narrations that were similar in referential/evaluative functions and temporal structure, with a preponderance of positive content. Working-class families produced twice as many co-narrations as their middle-class counterparts, produced more negative emotion talk, and used more dramatic language for conveying negative emotional experience. These findings suggest that (1) differentiation between working-class and middle-class communities in the content of early narratives may occur primarily with respect to negative experience and (2) researchers need to go beyond emotion state terms in order to accurately represent sociocultural variation in personal storytelling.

In order to begin to learn a language, young children must be able to locate and distinguish linguistic units in the speech they hear. A number of cues in the speech stream may aid them in this task. Some cues, such as frequently occurring grammatical morphemes and prosodic changes at linguistic boundaries are inherent in the language. Other cues, such as short utterance length and placement of key words in utterance-final position, are not integral to the grammar of the language but are characteristically provided by caregivers. Although previous studies suggest that even infants are sensitive to many of these cues, it is not clear that young listeners actually use them in assigning structure to sentences. The experiments reported here asked whether 60 children aged 2;0 to 2;2 used grammatical and caregiver cues in sentence comprehension and how different types of cues interacted. Two findings are of note: children used all of the cues tested, and the presence of one type of cue did not diminish use of another.

Expressive vocabulary data gathered during a systematic diary study of one male child's early language development are compared to data that would have resulted from longitudinal administration of the MacArthur Communicative Development Inventories spoken vocabulary checklist (CDI). Comparisons are made for (1) the number of words at monthly intervals (9;10.15 to 2;0.15), (2) proportion of words by lexical class (i.e. noun, predicate, closed class, ‘other’), (3) growth curves. The CDI underestimates the number of words in the diary study, with the underestimation increasing as vocabulary size increases. The proportion of diary study words appearing on the CDI differed as a function of lexical class. Finally, despite the differences in vocabulary size, logistic curves proved to be the best fitting model to characterize vocabulary development as measured by both the diary study and the CDI. Implications for the longitudinal use of the CDI are discussed.

Review essay on: ELMAN, J., BATES, E., JOHNSON, M., KARMILOFF-SMITH, A., PARISI, D. & PLUNKETT, K. Rethinking innateness: a connectionist perspective on development. Cambridge, MA: MIT Press (1996). Pp. 447.

I believe that the field of developmental psycholinguistics suffers from two major weaknesses. The first is its impressionistic and inexact formulations. The second is its divisive polarizations. One can see the reasons for the first weakness. Developmental psycholinguistics is only about 30 years old (ignoring diary studies which preceded the linguistic and cognitive surge of the sixties). But speculation and hypothesizing on the basis of relatively little data and passing acquaintance with phenomena has reached the level of customary ‘business as usual’. We are skilful at hypothesis construction, yet we are regretfully delinquent at formulating clear tests of our hypotheses. We are fond of conjecture about causal relationships, but our empirical tests progress no further than weak forms of correlation.

With regard to our second major weakness, our knack for polarizing opinion regarding chimerical questions such as the innateness of language can also be understood. After all, are we not following the classical dialectic model of thesis, antithesis and eventual synthesis? I think this is an idealized view of ourselves. In fact, we are driven by hunch and bias far more often than we would like to admit. Following hunches may be a real sign of creativity and vitality in our thinking. However, polarization driven by biases is ultimately detrimental. At some point we must disentangle ourselves from customary dialogue and transcend our deeply rutted patterns of thought.

When I began Rethinking innateness, I had hoped that the book might help us overcome these two weaknesses. I believe that it contributes positively to the goal of increasing the precision of our hypotheses and their empirical substantiation. At the same time, I am afraid that it will have a negative impact by aggravating the degree of polarization in our field.

In his review of Rethinking innateness (Elman, Bates, Johnson, Karmiloff-Smith, Parisi & Plunkett, 1996; henceforth, RI), Rispoli (this volume) comments favourably on the dynamical change models presented in RI's Chapter 4. I think a more critical stance is warranted. In particular, I will argue that dynamical change models cannot in principle make reference to mental representation, that the models are stipulative, and that they fail as descriptions of behaviour. (For more extensive discussion, see Valian, in press.) The relation between dynamical change models and connectionist networks is not spelled out in RI, so it is not clear which of the criticisms that I direct at dynamical change models also hold for connectionist models.

Rethinking innateness is a timely volume which forcefully demonstrates the importance of modelling in understanding development, ‘innateness’, and the nature of change. It provides an inspiring vision of what developmental psychology could one day be like, linking behaviour and biology via connectionist models. However, Rispoli's worry about the book's potential for detrimental polarization does not seem unfounded. One aspect of the book that deserves comment in this respect is the focus on connectionism to the exclusion of other types of model. It is unclear from Rethinking innateness itself whether this exclusion of other approaches merely stems from the legitimate desire to write a focused book, or whether it possibly reflects an actively held view that connectionism is the one true approach to modelling development. Regardless of the authors' intentions, the role of connectionism in relation to other computational approaches is an issue which is particularly pertinent to the study of language acquisition.

To clarify straightaway, I not only concur with Elman et al. on the central role of modelling, but also strongly believe that the task of the cognitive scientist is not complete until one has an account of how a particular process is realised in a neural architecture. These two commitments give connectionist models a central role. But they do not make connectionism the exclusive modelling tool for the study of language acquisition, nor necessarily the best path to currently pursue. The reasons for this are twofold.

The first is a matter of research strategy. It is possible that more rapid progress and greater success might be made if one starts with high-level models which give less immediate regard to matters of implementation.

The debate between nativism and empiricism has been the central issue in developmental psychology for well over three centuries. Nowhere has the opposition between nature and nurture been felt more keenly than in the study of child language development. At times the debate has proven useful, or at least entertaining. But there are those, like Matthew Rispoli, who worry that a fixation on this issue may reinforce a polarization that will eventually ‘run developmental psycholinguistics into the ground’.

Rethinking innateness (RI) is a quixotic book, enthusiastic and visionary. It is hard not to be provoked as each author races full tilt towards their chosen combat. Some readers, like Fodor (1997), respond in kind. Rispoli is more measured. Nonetheless, every reader chooses their own windmills to elevate into giants, and each believes their battle righteous. Are Elman et al. ‘building castles in the air’, or is Rispoli ‘taking the wrong sow by the ear’?

This commentary concerns two issues: the contribution of connectionism and theories of learning to developmental psychology; and the status of connectionist models of syntax acquisition.

In his review, Rispoli's main concern is that Elman et al.'s book will aggravate the degree of polarization in developmental psycholinguistics. I cannot really comment on this worry, as developmental psycholinguistics is not my field. Instead, I will discuss some questions more related to my background – the role of computational modelling in Elman et al.'s approach.

Elman et al.'s ambitious goal is to propose theories of cognitive development that are grounded in our knowledge of biology. This is of course what the great Jean Piaget tried to achieve during his lifetime – unsuccessfully, as we know. Elman et al.'s advantage over Piaget is that they have a set of computational tools, connectionism, which both allows them to specify theories precisely and to study complex behaviours (such as epigenesis, where innate and environmental factors interact to create new levels of complexity) that are just beyond the (unaided) human mind. Even though I will highlight some of the weaknesses of their approach below, I should emphasize that reading their book was an exciting and enjoyable experience.

As noted by Rispoli, there are important problems with the simulations reported by Elman et al. Rispoli focuses on simulations of past tense acquisition and syntax acquisition, but the problems are by no means limited to these areas. I will briefly consider two recent developments in neural net research, one taken from the field of language acquisition, and one from elsewhere, which underscore some of the difficulties of the simulations discussed in the book.

There are a number of ways to react to the whole problem of polarization that Rispoli's 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 Rispoli's 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.