Search Results

Gradience has often been identified as an important factor of linguistic change. A lexical‐syntactic explanation of Sorace's (2000, 2004) gradients in auxiliary selection with intransitive verbs is provided. Of particular interest here are the gradience factors involved in the replacement of BE by HAVE in Old Catalan and Old Spanish.

This chapter develops a model that can handle all configurations of gradience and categoricalness. It believes that the solution lies in the trade-off between reliability and generality. It shows how the previous approach to the problem was not enough, and suggests a novel approach using the gradual learning algorithm (GLA), adapted to more general limitations.

This chapter begins by reviewing some of the key psycholinguistic evidence motivating the need for experience-based mechanisms, before turning to a discussion of recent models. It focuses on probabilistic models of human sentence processing, which attempt to assign a probability to a given sentence, as well as to alternative parse interpretations for that sentence. It discusses the association between probabilistic models of performance (gradient grammaticality). A vital result of the proposed view is that the likelihood of a partial structure is only significant relative to the likelihood of competing (partial) structures, and does not give an independently useful characterization of the grammaticality of the alternatives. Therefore, it argues that a probabilistic description of gradient grammaticality should differ from a probabilistic performance model.

This chapter outlines the linear optimality theory (LOT), a variant of optimality theory primarily suggested by Keller to model gradient linguistic data. LOT is a framework intended to account for gradient judgment data; gradience in processing data and in corpus data has different properties from gradience in judgment data, and it is unlikely that the two types of gradience can be accounted for in a single, unified framework. The chapter provides a summary of the empirical properties of gradient judgments that motivate the design of LOT. It defines the components of an LOT grammar, and introduces the LOT notions of constraint competition and optimality. It gives a comparison with other variants of OT, particularly with standard OT and with harmonic grammar. This chapter contains a survey of more recent developments, such as probabilistic OT and variants of OT on maximum entropy models.

This chapter shows that the source of graded acceptability judgments cannot be purely syntactic. Instead, such data area forecasted by information structure (IS) constraints. The chapter argues that extraction is completely determined by IS constraints, in particular that only focus domains are apparent for purposes of extraction. It concludes that any phenomenon which differs with context among and across speakers cannot have a syntactic account. An account in terms of IS is aided to predict this kind of variation. Thus, a syntactic constraint violation will be ungrammatical, a violation of an IS constraint will allow contextual variation and will thus result in gradience. There will be no weak syntactic constraints, only strong ones. A theory of IS, — f(ocus)-structure theory — geared to interact with syntax, phonology, and semantics, is introduced and viewed as an essential part of grammar.

This chapter begins by defining gradience in terms of an individual speaker's grammar. It also identifies gradient phenomena in various domains of grammar such as semantic ambiguities, syntactic blends, and in phonological entities including intensity and length, among others. It then discusses the theories of gradience in phonology and syntax.

A much closer connection between phonological grammar and mental lexicon is revealed by studies of gradience in phonology using linguistic and metalinguistic data. The discovery of these new dimensions of phonological variation would not have been possible without corpus methods and data from groups of participants in psycholinguistic experiments. The presence of gradient phonological constraints illustrates that phonological knowledge goes beyond a categorical symbolic representation of possible forms in a language. To accommodate the broader scope of phonological generalizations, models of grammar will have to become more like models of other cognitive domains, which have long recognized and debated the nature of frequency and similarity effects for mental representation and processing. Well-understood mechanisms of speech perception and speech production bind the scope of phonological variability. Phonological categories and phonological patterns give a sufficiently rich and intricate variety of alternatives to explore the full complexity of cognitive processes.

This chapter characterizes the critical phenomena from a behavioural perspective before turning to experimental methods, yielding more fine-grained data. It examines how linguistic judgments emerge from the real-time comprehension processes by drawing upon studies of word order variation in German. Based on a number of empirical observations, it argues that gradient data need not be interpreted as evidence against categorical grammars. Instead, gradience can come from a complex interaction between grammar-internal requirements, processing mechanisms, general cognitive constraints, and the environment within which the judgment task is performed.

This brief chapter summarizes the description of cognitive and semantic structures in Part I, and returns to the Single Event Condition from Chapter 1. That condition requires that a constituent containing A'-movement describes a single event. It therefore converts the theory elaborated in Part I into predictions concerning the distribution of A'-movement. Certain discrete patterns are expected, based on factors such as those governed by regular syntactic locality theory and the categorical well-formedness constraints on event structure from Part I, as well as syntactic height effects deriving from the distribution of Op (Chapter 4). However, there are also two expected sources of gradient acceptability. Firstly, world knowledge constrains the possibility of enrichment of noncontingent, temporal relations among events into contingent, macroevent-forming relations; secondly, aspectual coercion produces gradient effects in some cases.

This chapter elaborates on the theoretical underpinnings of usage-based perspectives on language and language change. The consequences of conceptualizing the language system as a network are discussed, with focus on growth, obsolescence, and reconfiguration of nodes and links in the network. The development of the way-construction is revisited using extensive electronic corpora and shown to illustrate the development of two subschemas, one intransitive motion, one transitive, followed by the further expansion of the latter.

This chapter addresses the question of whether gradience in acceptability should be considered evidence for gradience in grammar. Most current syntactic theories are based on a categorical division of grammatical versus ungrammatical sentences. In contrast, acceptability intuitions, that is, the data used to build those theories, have long been recognized to be gradient. The chapter presents two experiments collecting acceptability ratings for 100 sentences extracted from papers published in Linguistic Inquiry. The results show a gradient pattern. It is argued that this gradience in acceptability is highly unlikely to be due to methodological and other known extra-grammatical factors. Unless another factor can be identified, it seems reasonable to assume that the observed gradience comes (also) from the grammar. Furthermore, the chapter presents a proposal concerning diacritics, according to which the traditional asterisk is reserved for ungrammaticality only, and a new diacritic (“^”) indicates reduced acceptability.

This chapter discusses the literature dealing with the formal and functional features of the major grammatical categories noun and verb, focusing on the so-called “mixed” or “hybrid” structures known as deverbal nominalizations. It is argued that, to understand the peculiarities of these much-discussed constructions and, more generally, to investigate whether abstract linguistic concepts such as “noun” and “verb” are—as suggested in functionalist and cognitive linguistics—to a certain extent “iconic,” we should adopt an approach that devotes attention to structural as well as functional-semantic properties. The discussion will lead to the presentation of a theoretical model of functional-semantic nominality and verbality/clausality, which will serve as the core for further investigation of the functional-semantic organization of the English gerundive system.

The concluding chapter synthesizes the results of the preceding analyses. It highlights that the most important functional-semantic categorial shift that has taken place within the English gerundive system did not affect the morphosyntactically verbalizing component; instead, it affected the “original” nominal gerund, which started to functionally assimilate to more prototypical members of the nominal class. It is explained that in earlier stages, the English gerund exhibited functional hybridity, using an exclusively nominal form to realize more nominal as well as more clausal functions; but with the rise of the verbalized gerund, this functional hybridity started to be gradually sorted out. What emerges from the discussions of the case studies is that adopting a model of functional-semantic categoriality allows one to tackle the remaining lacunae in understanding this history of the English gerund, and perhaps, in the not-so-distant future, of “categoriality in language change” more generally.

Speech simultaneously conveys both propositional and indexical meaning. This is approximately the difference between ‘what you said’ and ‘the way you said it’. Some indexical meaning is conveyed by arbitrary language-specific devices such as lexical choice and sociophonetic variation, but some is conveyed by paralinguistic cues (‘tone of voice’, ‘body language’) that are broadly speaking universal. Describing the universal aspects in terms of gradient ‘modulation’ of the phonetic manifestations of phonological categories expresses their formal similarity to sociophonetic variation. However, in some languages modulation may also be categorical, in phenomena like ideophones and ablaut morphology, and in these it plays a more centrally linguistic or propositional role.

The main goal of this chapter is critically to address from a comparative perspective some of the most important issues in split intransitivity within the Romance context, covering such areas as conservation and loss of classical and late Latin patterns of split intransitivity; the effects and reflexes of Romance split intransitivity in relation to sentential word order, INDE-cliticization, absolute participle constructions, finite verb agreement, distribution of bare plural nominals, relative pronoun alternations, resultatives, adjectival adverb agreement, impersonal passive; syntactic, semantic, and lexical mismatches; gradience; stability and areal tendencies in the expansion/shrinking of overt reflexes of split intransitivity. Specific topics dealt with include: the advancement of active/inactive alignment; split intransitivity and the north-south divide; further split intransitivity diagnostics in Romance.