The limits of unification

Abstract

Current complex-feature based grammars use a single procedure-unification-for a multitude of purposes, among them, enforcing formal agreement between purely syntactic features. This paper presents evidence from several natural languages that unification-variable-matching combined with variable substitution-is the wrong mechanism for effecting agreement. The view of grammar developed here is one in which unification is used for semantic interpretation, while purely formal agreement involves only a check for non-distinctness-i.e, variable-matching without variable substitution.

Grammar Customization

“…in person, number or gender) as well as tense and aspect distinctions. More elaborate interactions, such as agreement with coordinated noun phrases, resolution of case requirements of coordinated verbs (Ingria 1990;Dalrymple and Kaplan 2000), and case marking which is sensitive to negation (e.g. Przepiórkowski 2000) remain unexplored.…”

Indeterminacy, complex features and underspecification

“…by structure-sharing): in (1a) Frauen is required to be both ACC and DAT simultaneously, and since the assignment of values to features is functional, it is impossible for a single instance of a form to bear multiple (competing) values. The treatment of such forms has in consequence generated large amounts of work in such frameworks, often involving some degree of departure from the simple assumptions outlined above (Zaenen and Karttunen 1984;Pullum and Zwicky 1986;Ingria 1990;Bayer 1996;Dalrymple and Kaplan 2000;Levy and Pollard 2001;Hinrichs and Nakazawa 2002;Blevins To appear). For work in other frameworks, see inter alia Vogel (2001) and Trommer (2006).…”

Morphology in Constraint-based Lexicalist Approaches to Grammar

“…The hypothesis of strong lexicalism has put the focus of much work in CBL frameworks on the interface between morphology and syntax, with two main areas of research. On the one hand, research on the relation between featural representations in syntax and morphology has focused on case stacking (Nordlinger, 1998;Malouf, 2000;Sadler and Nordlinger, 2004), deponency (Vincent and Börjars, 1996;Sadler and Spencer, 2001), agreement features (Pollard and Sag, 1994;Kathol, 1999;Wechsler and Zlatić, 2003), and syncretism (Ingria, 1990;Dalrymple and Kaplan, 2000;Daniels, 2002;Levy and Pollard, 2002;Sag, 2003;Crysmann, 2009;Dalrymple et al, 2009). On the other hand, much attention has been devoted to linguistic phenomena presenting apparent challenges to strong lexicalism, including pronominal affixes or clitics (Miller, 1992;Miller and Sag, 1997;Crysmann, 2003a;Monachesi, 1999Monachesi, , 2000Crysmann, 2003b;Bonami and Boyé, 2007;Penn, 1999;Samvelian and Tseng, 2010), portmanteau elements (Bender and Sag, 2000;Wescoat, 2002;Abeillé et al, 2003;Wescoat, 2007), particle verbs (Ackerman and Webelhuth, 1998;Kathol, 2000;Müller, 2003), and discontinuous affixation (Borsley, 1999;Kupść and Tseng, 2005;Crysmann, 1999Crysmann, , 2010aBroadwell, 2008;Fokkens et al, 2009).…”