This article shows that specific properties of long-distance phonotactic patterns derived from consonantal harmony patterns (Hansson 2001, Rose andWalker 2004) follow from a learner that generalizes only on the basis of the order of sounds, not the distance between them. The proposed learner is simple, efficient, and provably correct, and does not require an a priori notion of tier or projection (contra the model in Hayes and Wilson 2008); nor does it rely on the additional structure provided by Optimality Theory grammars (Prince and Smolensky 1993Smolensky , 2004 or grammars in the principles-and-parameters framework (Chomsky 1981, Dresher and Kaye 1990, Gibson and Wexler 1994. Not only does the noncounting nature of nonlocal dependencies automatically follow from the way the learner generalizes, it also explains the absence of blocking patterns from the typology. Finally, the learner lends support to the idea that long-distance phonotactic patterns are phenomenologically distinct from spreading patterns, contra the hypothesis of Strict Locality (Gafos 1999, et seq.).

Abstract. We present a measure of cognitive complexity for subclasses of the regular languages that is based on model-theoretic complexity rather than on description length of particular classes of grammars or automata. Unlike description length approaches, this complexity measure is independent of the implementation details of the cognitive mechanism. Hence, it provides a basis for making inferences about cognitive mechanisms that are valid regardless of how those mechanisms are actually realized.

This paper presents a previously unnoticed universal property of stress patterns in the world's languages : they are, for small neighbourhoods, neighbourhooddistinct. Neighbourhood-distinctness is a locality condition defined in automatatheoretic terms. This universal is established by examining stress patterns contained in two typological studies. Strikingly, many logically possible -but unattested -patterns do not have this property. Not only does neighbourhooddistinctness unite the attested patterns in a non-trivial way, it also naturally provides an inductive principle allowing learners to generalise from limited data. A learning algorithm is presented which generalises by failing to distinguish sameneighbourhood environments perceived in the learner's linguistic input -hence learning neighbourhood-distinct patterns -as well as almost every stress pattern in the typology. In this way, this work lends support to the idea that properties of the learner can explain certain properties of the attested typology, an idea not straightforwardly available in optimality-theoretic and Principle and Parameter frameworks.

No abstract

How do infants find the words in the speech stream? Computational models help us understand this feat by revealing the advantages and disadvantages of different strategies that infants might use. Here, we outline a computational model of word segmentation that aims both to incorporate cues proposed by language acquisition researchers and to establish the contributions different cues can make to word segmentation. We present experimental results from modified versions of Venkataraman's (2001) segmentation model that examine the utility of: (1) language-universal phonotactic cues; (2) language-specific phonotactic cues which must be learned while segmenting utterances; and (3) their combination. We show that the language-specific cue improves segmentation performance overall, but the language-universal phonotactic cue does not, and that their combination results in the most improvement. Not only does this suggest that language-specific constraints can be learned simultaneously with speech segmentation, but it is also consistent with experimental research that shows that there are multiple phonotactic cues helpful to segmentation (e.g. Mattys, Jusczyk, Luce & Morgan, 1999; Mattys & Jusczyk, 2001). This result also compares favorably to other segmentation models (e.g. Brent, 1999; Fleck, 2008; Goldwater, 2007; Johnson & Goldwater, 2009; Venkataraman, 2001) and has implications for how infants learn to segment.

This paper characterizes a subclass of subsequential string-to-string functions called Output Strictly Local (OSL) and presents a learning algorithm which provably learns any OSL function in polynomial time and data. This algorithm is more efficient than other existing ones capable of learning this class. The OSL class is motivated by the study of the nature of string-to-string transformations, a cornerstone of modern phonological grammars.

This chapter studies the nature of the typology of phonological markedness constraints and the nature of the typology of the transformation from underlying to surface forms from a computational perspective. It argues that there are strong computational laws that constrain the form of these constraints and transformations. These laws are currently stated most clearly in terms of the so-called subregular hierarchies, which have been established for stringsets (for modeling constraints) and are currently being established for string-to-string maps (for modeling the transformations). It is anticipated that future research will reveal equally powerful laws applicable to non-string-based representations. Finally, this chapter argues that these laws arise as a natural consequence of how humans generalize from data.Wilhelm von Humboldt's phrase "language makes infinite use of finite means" (1836/1999) is oft-cited by Chomsky because not only does it encapsulate an important characteristic of natural language, but it also highlights why generative grammars play an important role in understanding this aspect of language. In brief, the generative grammars are the finite means, but the linguistic knowledge they represent can be applied to unboundedly many linguistic forms. The psychological reality of generative grammars is the powerful scientific hypothesis which underlies all work in generative linguistics.

An important distinction between phonology and syntax has been overlooked. All phonological patterns belong to the regular region of the Chomsky Hierarchy, but not all syntactic patterns do. We argue that the hypothesis that humans employ distinct learning mechanisms for phonology and syntax currently offers the best explanation for this difference.Keywords: Phonology; Syntax; Computational complexity; Language learning A role for phonology in cognitive scienceWhen it comes to the problem of how humans learn language, it appears that many computational learning theorists, cognitive scientists, and psychologists are primarily occupied with the problem of how humans learn to put words and morphemes together to form sentences. In this article, we argue that a further understanding of how sounds are put together to form words also bears directly on fundamental questions in cognitive science. In particular, we argue that computational analysis of the typology of patterns in phonology, when compared to the typology of patterns in syntax, reveals that cognitive learning mechanisms are likely multiple and modular in nature.The skew that many researchers exhibit toward morpho-syntax may really be a skew toward studying meaning. But we believe that it is because phonological systems impose different sound patterns in different languages without contributing to meaning that they are especially interesting. That is, phonology is about "Rules without Meaning" in Frits Staal's (1989) terms.Correspondence should be sent to