~ ROA Version, 8/2002. Essentially identical to the Tech Report, with new pagination (but the same footnote and example numbering); correction of typos, oversights & outright errors; improved typography; and occasional small-scale clarificatory rewordings. Citation should include reference to this version.

T he distinction between marked and unmarked structures has played a role throughout this century in the development of phonology and of linguistics generally. Optimality Theory (Prince and Smolensky 1993) offers an approach to linguistic theory that aims to combine an empirically adequate theory of

This paper investigates the incorporation of a non-procedural theory of phonology into HPSG, based on the 'one-level' model of Bird & Ellison (1992). The standard rule-representation distinction is replaced by the description-object distinction which is more germane in the context of constraint-based grammar. Prosodic domains, which limit the applicability of phonological constraints, are expressed in a prosodic type hierarchy modelled on HPSG'S lexical type hierarchy. Interactions between phonology and morphology and between phonology and syntax are discussed and exemplified I

In this paper I present a novel Machine Learning approach to the acquisition of stochastic string transductions based on Pair Hidden Markov Models (PHMMs), a model used in computational biology. I show how these models can be used to learn morphological processes in a variety of languages, including English, German and Arabic. Previous techniques for learning morphology have been restricted to languages with essentially concatenative morphology.

Abstract not found

In this paper I present an algorithm for the unsupervised learning of morphology using stochastic finite state transducers, in particular Pair Hidden Markov Models. The task is viewed as an alignment problem between two sets of words. A supervised model of morphology acquisition is converted to an unsupervised model by treating the alignment as a further hidden variable. The use of the Expectation-Maximisation algorithm for this task is studied, which leads to calculations involving the permanent of a matrix of probabilities.

Abstract not found

A potential problem for connectionist accounts of inflectional morphology is the need to learn a `default' inflection (Prasada & Pinker, 1993). The early connectionist work of Rumelhart & McClelland (1986) might be interpreted as suggesting that a network can learn to treat a given inflection as the `elsewhere' case only if it applies to a much larger class of items than any other inflection. This claim is true of the Rumelhart & McClelland (1986) model, which was a two-layer network subject to the computational limitations on networks of that class (Minsky & Papert, 1969). However, it does not generalize to current models, which have available more sophisticated architectures and learning algorithms. In the current paper we explain the basis of the distinction, and demonstrate that given more appropriate architectural assumptions, connectionist models are perfectly capable of learning a default category and generalizing as required, even in the absence of superior type frequency.

Optimality-Theoretic learning algorithms are only guaranteed to be successful if the data fed to them contain full structural descriptions of the surface forms, i.e. descriptions that include hidden structure like metrical feet. This is not realistic as a model of acquisition, because children are only exposed to overt forms, e.g. unstructured strings of syllables. Optimality-Theoretic learning algorithms that learn solely from overt forms turn out to sometimes succeed and sometimes fail (Tesar & Smolensky 2000). This possibility of failure is a property of both on-line learning algorithms that have been proposed for OT, namely Error Driven Constraint Demotion (EDCD; Tesar 1995) and the Gradual Learning Algorithm (GLA; Boersma 1997). The possibility of failure is not necessarily bad: one would want an algorithm to fail for languages that do not exist, and to succeed for languages that do exist. Latin exists (or existed). This paper compares the performance of the two learning algorithms for the metrical stress system of Classical Latin. It turns out that EDCD cannot learn this system from overt forms only, and that the GLA can. This suggests that the GLA may be a better model of acquisition than EDCD. The results

In this squib I discuss an unusual type of reduplication in which the reduplicant varies not only in terms of its phonemic composition but also in terms of its prosodic shape. The variability in the shape of the reduplicant results from a grammar that does not impose any constraint particular to the shape of the reduplicant per se. Further, I demonstrate that even in cases where the reduplicant is shape invariant, this shape may also arise from a grammar that does not impose a constraint on the form of the reduplicant. In both cases all relevant aspects of the reduplicant’s realization arise from constraints that apply to the language in general. 1 Templatic Reduplication In the theory of word formation, the program of Prosodic Morphology (McCarthy and Prince 1986) has established that grammatical categories, usually in the domain of root-and-pattern and reduplicative morphology, are often expressed by invariant prosodic shapes or templates.