Methodological questions concerning Chomsky’s generative approach to linguistics have been debated without consensus. The status of linguistics as psychology, the psychological reality of grammars, the character of tacit knowledge and the role of intuitions as data remain heatedly disputed today

I will assume that the study of language essentially falls under what Chomsky (1986) has called “Plato’s problem”: how do organisms form such rich cognitive structures from so little exposure to the environment? I take this to be the original and fairly classical motivation for cognitive science although, arguably, not everyone who currently works in cognitive science shares the motivation.1 Plato’s problem arises from what have come to be known as the “poverty-of-stimulus arguments ” which show that there is not enough information in the environment for the rich systems constructed by organisms (Chomsky 1957; Piatelli-Palmarini 1980; Wexler 1991; Crain and Pietrosky 2001, etc.). As Chomsky puts it, “We can check the experience available; we can look at it and see what it is. It’s immediately obvious that it’s just much too limited and fragmentary to do anything more than shape an already existing common form in limited fashions ” (Chomsky et al 2000a:6). Accordingly, the specific goal of Biolinguistics is to explain how sound-meaning correlations are established in languages within the constraints imposed by the poverty of the stimulus. The Government-Binding theory (Chomsky 1981) sought to explain those correlations from universal principles alone. The Minimalist Program went several steps ahead to show that such

this paper we consider how constraints on parsing can remove some of the burden from the theory of syntax. By making a minimal set of assumptions both with respect to syntax and parsing, it seems that certain languages should be unparsable. As a solution, we propose that long-range LF movement in general, and long-range LF wh-movement in particular, is rightward. This overcomes the apparent unparsability of certain languages, and makes certain syntactic and computational predictions, which will be discussed below. 2 Motivation & Background

Left-dislocations have been referred to by NT grammarians using various terms, among them cleft constructions,1 hanging nominatives,2 pendent nominatives,3 casus pendens,4 or independent nominatives.5 There have also been various claims made about their function: Wallace: either emotion or emphasis6 Young: emphasis7 Porter: highlighting8 or a topic shifter9

arguments of a head are stored on a single ordered list, the subcat list. However, Borsley (1989) argues that there are various deficiencies in this approach, and suggests that the unified list should be split into separate lists for subjects, complements, and specifiers. This proposal has been widely adopted in what is colloquially known as HPSG3 (Pollard and Sag (1994:Ch. 9) and other recent work in HPSG). Such a move provides in HPSG an analog of the external/internal argument distinction generally adopted in GB, solves certain technical problems such as allowing prepositions to take complements rather than things identical in subcat list position to subjects, and allows recognition of the special features of subjects which have been noted in the LFG literature, where keyword grammatical relations are used. In HPSG3, it is these valence features subj, comps and spr whose values are ‘cancelled off ’ (in a Categorial Grammar-like manner) as a head projects a phrase. A lexical head combines with its complements and subject or specifier (if any) according to the lexically inherited specification, as in (1). 1 1 This paper is based on part of a talk given at the Tübingen HPSG workshop in June 1995, and distributed as Manning and Sag (1995). However, it excludes much material presented there, which will now appear in other places (Manning et al. in press, Manning and Sag submitted, Sag and Manning forthcoming). The paper also has an updated analysis – one consistent with that of Manning et al. (in press) – which we believe avoids the flaws attributed to the analysis of Manning and Sag (1995) by Webelhuth (forthcoming). However, because of space limitations, causatives are no longer discussed here and the reader interested in this topic should

I fllflffl......lfl I fllflffloffl..

This paper proposes a new theory of the relationship between the sentence processing mechanism and the available computational resources. This theory -- the Syntactic Prediction Locality Theory (SPLT) -- has two components: an integration cost component and a component for the memory cost associated with keeping track of obligatory syntactic requirements. Memory cost is

efficient convergence to a correct grammar. We discuss implications for learning from overt data only, as well as other learning issues. We argue that Optimality Theory promotes confluence of the demands of more effective learnability and deeper linguistic explanation.

In this paper we explore a new theory of discourse structure that stresses the role of purpose and processing in discourse. In this theory, discourse structure is composed of three separate but interre-lated components: the structure of the sequence of utterances (called the linguistic structure

We demonstrate that an unlexicalized PCFG can parse much more accurately than previously shown, by making use of simple, linguistically motivated state splits, which break down false independence assumptions latent in a vanilla treebank grammar. Indeed, its