This paper is designed to illustrate and consider the relations between three types of metarepresentational ability used in verbal comprehension: the ability to metarepresent attributed thoughts, the ability to metarepresent attributed utterances, and the ability to metarepresent abstract, non-attributed representations (e.g. sentence types, utterance types, propositions). Aspects of these abilities have been separately considered in the literatures on “theory of mind”, Gricean pragmatics and quotation. The aim of this paper is to show how the results of these separate strands of research might be integrated with an empirically plausible pragmatic theory. 1

ABSTRACT: In this article, we discuss the complex problem ofrelating genotype to phenotype and challenge the simple mapping ofgenes to higher level cognitive modules. We examine various methods that have been used to investigate this relation including quantitative genetics, molecular genetics, animal models, and in-depth psychological and computational studies ofdevelopmental disorders. Both single gene and multiple gene disorders indicate that the relationship between genotype and phenotype is very indirect and that, rather than identifying mere snapshots of developmental

This thesis is about the reconciliation of realistic views of rationality with inferential-intentional theories of communication. Grice (1957; 1975) argued that working out what a speaker meant by an utterance is a matter of inferring the speaker’s intentions on the presumption that she is acting rationally. This is abductive inference: inference to the best explanation for the utterance. Thus an utterance both rationalises and causes the interpretation the hearer constructs. Human rationality is bounded because of our ‘finitary predicament’: we have limited time and resources for computation (Simon, 1957b; Cherniak, 1981). This raises questions about the explanatory status of inferential-intentional pragmatic theories. Gricean derivations of speakers’ intentions seem costly, and generally hearers are not aware of performing explicit reasoning. Utterance interpretation is typically fast and automatic. Is utterance interpretation a species of reasoning, or does the hearer merely act as if reasoning? Within the framework of cognitive science, mental processing is understood as transitions between mental representations. I develop a traditional view of rationality as reasoning ability, where this is essentially the ability to make transitions that preserve rational acceptability. Following Grice (2001), I claim that there is a ‘hard way’ and a ‘quick way’ of reasoning. Work on bounded rationality suggests that much cognitive work is done by heuristics, processes that exploit environmental structure to solve problems at much lower cost than fully explicit calculations. I look at the properties of heuristics that find solutions to open-ended problems such as abductive inference, particularly sequential search heuristics with aspiration-level stopping rules. I draw on relevance theory’s view that the comprehension procedure is a heuristic which exploits environmental regularities due to utterances being offers of information (Sperber & Wilson, 1986). This kind of heuristic, I argue, is the ‘quick way’ that reasoning proceeds in utterance interpretation.

of at least 16 different foreign languages, and a talent for learning new ones that is clearly demonstrated even when the authors present him with the problem of learning Epun, an artificial language with peculiar properties that (according to the theory of grammar embraced by the authors) do not not exist in the real world and could not be acquired by any normal child. I was convinced by the end of the book that Christopher is indeed a fascinating young man, but the authors' agenda goes far beyond biography. They believe that they are describing a true savant; as we shall soon see, I am not sure that premise is correct. They also believe that they have provided incontrovertible evidence for the independence of language from cognition, for the modularity of the various subcomponents that make up the language faculty, and for the idea that Universal Grammar is an innate property of the human mind with tremendous explanatory value in the study of first- and second-language acquisition.

Two major mechanisms of encoding telicity across languages are either marking the object as exhaustively countable or measurable, or utilizing a specific prefix on the verbal form. English predominantly uses the first mechanism, while Russian mostly utilizes the second. The learning task of an English speaker acquiring Russian, then, is two-fold: to learn each individual verb with its subset of perfective prefixes, and to acquire knowledge of the fact that most prefixed verbs denote telic events. Sixty-six English-speaking learners of Russian as well as 45 controls took an on-line test of semantic interpretation. Results indicate that some low intermediate learners, and the majority of high intermediate and advanced learners are highly accurate in interpreting Russian telicity marking. It is argued that the difficulty in acquiring Russian aspect lies in learning the lexical items signaling telicity, but crucially NOT in learning the grammatical mechanism for telicity marking. 1.

The papers in this issue present a series of comparative studies of language processing and language development across clinical populations, including studies in the earliest stages of language development, as well as aspects of grammar, narrative, and verbal memory across the elementary school years. The populations covered include ‘‘late talkers,’ ’ children with congenital injuries to either the left or right hemisphere, children with Williams Syndrome, children with Down Syndrome, children with a diagnosis of specific language impairment (behaviorally defined), and a range of typically developing controls for each of these groups. As Holland points out in her commentary, two of the most surprising findings across these studies include the following. (1) Despite differences in rate of development, the sequences and error types observed are (with a few interesting exceptions) remarkably similar across these very different clinical groups. It appears that sequences and error types are determined primarily by the ‘‘problem space’ ’ posed by a particular language (in this case English), so that virtually all children who enter into this problem space end up behaving in much the same way, albeit at variable rates. Reilly et al. (this issue) suggest a metaphor for results like these: all of these children have undertaken a journey along the same highway, but some of them are in the slow lane, with occasional stops along the road. (2) Children with well-defined brain injuries of the kind that often lead to aphasia in adults perform within the normal range on most measures once they move into the elementary school years, with no trace of a selective disadvantage for children with left-hemisphere damage

In this paper we report on our attempt to teach the polyglot savant Christopher (‘C ’ hereinafter) British Sign Language (BSL). BSL presents C with a novel challenge in the use of hand-eye co-ordination, while at the same time offering him all the linguistic ingredients he is obsessed with. Despite deficits in key areas of intellectual ability, communication and visuo-spatial cognition, C has developed a working knowledge of BSL through processes of circumvention, adaptation and invention. We also taught BSL to a comparator group of talented second-language learners, though we do not discuss this comparison in depth here (see Morgan et al, in prep a) but merely refer to some of the test scores as a guide to how normal a sign learner C is. Results from formal tests of his linguistic knowledge and observational study of his developing communicative ability in BSL are analysed and described. These results reflect the structure and use of BSL, highlighting the important role of visuo-spatial cognition in its acquisition and manipulation.

interpretable and uninterpretable features.

Introduction The theme of this conference is "Generative Grammar since Syntactic Structures" (Chomsky, 1957). As Chomsky has noted, Syntactic Structures was based on "lecture notes for an undergraduate course at MIT:" This [Syntactic Structures] was a sketchy and informal outline of some of the material in LSLT [The Logical Structure of Linguistic Theory], along with some material on finitestate grammars and formal properties of grammars from 1956. (Chomsky, 1975:3) It is in LSLT, first circulated in 1955, that Chomsky sets out in more detail the theory of generative grammar. In this work "the 'realist' position" is assumed. The theory is "understood as a psychological theory that attempts to characterize the innate human 'language faculty'" (Chomsky 1975:37). Hence from its earliest beginnings the central motivation for the study of generative grammar (generative linguistics, I-linguistics, biolinguistics) has

Proponents of generative grammar sometimes refer to their particular theoretical framework as ‘the ’ theory of syntax (Chomsky, 1965; Grodzinsky, 1986; Smith & Tsimpli, 1995; Rice, 1997), implying that a consensus has been reached within linguistic theory. For child language researchers who do not find this framework useful, and�or are not willing to accept the epistemological baggage that it carries (i.e. radical claims about the innateness and autonomy of grammar), this is an unfortunate situation. The old alliance between linguistics and child language, forged with great optimism in the nineteen sixties, has turned into a partisan affair, practised almost exclusively by cognoscenti of MIT linguistics. In response, many developmental psycholinguists have abandoned linguistic theory altogether, basing their explanations on more general principles of representation and learning taken from developmental psychology and cognitive science. Although these fields also have a lot to offer, they do not provide the detail or the rigour that we once derived from a fruitful relationship with linguistics.