Where is morphological complexity?

A number of approaches have been taken to defining complexity in language. The issue is important, since underlying some theoretical models has been an assumption, sometimes explicit, sometimes unconscious, that the simplest formal description of a language naturally matches speaker knowledge. But it is not clear that complexity is the same for the analyst, the speaker, and the learner. Here the issue is explored in two languages with relative morphological complexity, but of different types, Central Pomo and Mohawk. First the speech of bilinguals with varying degrees of English dominance is compared. Next, the development of morphological complexity is traced in children learning Mohawk as a first language. The results indicate that complexity is indeed not the same for analysts, speakers, and learners, findings more in tune with abstractive models of morphology than constructive ones.

Radical analyticity as a diagnostic of adult acquisition

Linguists have traditionally supposed that languages become radically analytic via two mechanisms: (1) pidginization and related processes of heavy and disruptive adult acquisition; and (2) ‘drift’ into analyticity due to withdrawal of stress from final syllables, or other grammar-internal processes sparked essentially by chance. However, theoretical economy directs us to suppose that radical analyticity emerges solely from adult acquisition, and that radically analytic languages such as Chinese and Yoruba can be identified as having experienced heavy second-language acquisition in their histories despite this not having been recorded historically. In support of this hypothesis this chapter notes that radically analytic languages are quite rare worldwide; that both older radically analytic languages and creoles bear the hallmark of eliminating contextual inflection rather than inherent; and that in this and other facets, radically analytic languages do not differ from synthetic ones only in degree of boundedness of morphemes. Rather, synthetic languages tend to mark a great deal of semantic distinctions that radically analytic ones do not mark with free morphemes, suggesting the operation of second-language acquisition rather than ‘drift’.

The complexity of grammatical gender and language ecology

This chapter investigates the evolution of grammatical gender agreement, taken as an instance of paradigmatic and syntagmatic morphological complexity, in a sample of thirty-six languages, organized per sets of closely related languages with different sociolinguistic profiles. Both loss and emergence of gender agreement occur in areas of intense language contact between diverse speech communities. However, given similar contact scenarios, asymmetries in the structure of the bilingual population and/or in the prestige dynamics between the languages in contact tend to favour one development over the other. Loss of gender agreement occurs when the demographically dominant and/or more prestigious language lacks grammatical gender. Conversely, borrowing of gender agreement is favoured when the demographically dominant and/or more prestigious language has grammatical gender. Finally, the data suggest that patterns of gender marking may have important ties to the way in which speakers construe their linguistic identity in opposition to that of their neighbours.

Irregularity, paradigmatic layers, and the complexity of inflection class systems

The complexity of an inflection class system is the average extent to which elements in the system inhibit motivated inferences about the realization of lexemes’ paradigm cells. Systems tend to exhibit relatively low complexity in this sense. However, representations of inflectional systems tend to include only affixal and regular patterns, leaving questions about how irregular patterns and non-affixal ‘layers’ of inflectional exponence affect the complexity of systems. We address these questions by exploring four layers of inflectional exponence of Russian nouns, including irregular patterns within each layer. We find that Russian nouns exhibit relatively low complexity even when irregular and non-affixal exponence are included. Implicative structure and the uneven distribution of lexemes across classes mitigate the uncertainty associated with irregular and non-affixal exponence. Low systemic inflectional complexity is, thus, an emergent property that may extend beyond affixal and/or regular inflectional patterns.

Demorphologization and deepening complexity in Murrinhpatha

Inflectional allomorphy is a prototypical form of morphological complexity, introducing unpredictability into the mapping of form to meaning. In this chapter, we examine a system of verb inflection allomorphy in the Murrinhpatha language of northern Australia, which shows a high level of complexity as measured by unpredictability of analogical relations in inflectional exponence. We argue that in this case the unpredictability is associated with incremental demorphologization, the process whereby morphology gradually dissolves into unanalysable lexical form. We present observations of analogical change in Murrinhpatha, comparing contemporary fieldwork documentation with data from forty years earlier, showing that a long-term process of demorphologization is still underway in recent generations, resulting in increasing complexity of the system.

Morphological complexity, autonomy, and areality in western Amazonia

This chapter investigates the relationship between morphological complexity and language contact and change across western Amazonia. We explore morphological proliferation in particular domains (nominal classification, tense, evidentiality, and valence-adjusting), and follow this with a more systematic exploration of morphological complexity in relation to wordhood status across a sample of eleven Amazonian languages. We argue that a large percentage of bound morphemes in these languages display ambiguity between morphotactic versus syntactic analyses, suggesting that morphological autonomy is best characterized as a matter of degree, and that different degrees of autonomy may apply on a regional scale. Since many accounts of word-internal morphological complexity implicitly rely on notions of autonomy, Amazonian languages invite a revision of our current conception of this domain.

Derivation and the morphological complexity of three French-based creoles

Creolistic research persistently asserts the simplicity of creoles, citing as evidence the claimed poverty of creole morphology. Yet, creoles not only exhibit morphology, but evince a surprising degree of morphological complexity. Drawing on the evidence of derivational morphology from three different French-based creoles − Mauritian (Indian Ocean), Haitian, and Guadeloupean (Caribbean) – the current contribution provides new evidence for this claim. It pursues a view of morphological complexity where the interaction of a lexeme's inventory of forms with its participation in deverbal derivation contributes to the integrative complexity of a language's morphology. Such a perspective is compatible with psycholinguistic approaches to language acquisition and language change.

Overabundance resulting from language contact

The reduction of morphological complexity, particularly in inflectional paradigms, is not uncommon in language contact. One area of morphological complexity which has received less attention is variation within the cells of a paradigm, e.g. ‘dived’ and ‘dove’ as different past tense word forms of {DIVE} in English. This type of morphological complexity, where multiple forms are realized in the same cell in a paradigm is termed ‘overabundance’. This chapter examines the development of overabundance in the subject-marking system of Gurindji Kriol, and claims that increasing complexity in this dimension is the result of language contact. We analyse new data from Gurindji children using generalized linear mixed models to determine whether the complexity in the case paradigm has stabilized or whether complexification is on-going. We show that overabundance in Gurindji Kriol is an example of a contact-induced change which involves the complexification of an inflectional paradigm rather than its simplification.

Introduction

This chapter overviews some of the foundational assumptions informing contemporary views on morphological complexity and raises some of the central questions to be addressed in the volume's chapters from different perspectives. We propose a new composite approach in terms of a set of complexities in morphology instead of a view of morphological complexity as a unified phenomenon. This, we argue, allows us to individuate different aspects (e.g., syntagmatic vs. paradigmatic; inflectional vs. derivational, etc.) as logically independent variables of crosslinguistic variation requiring their own measures and analyses. A synopsis of the volume and of the individual contributions is also provided.

Simplification and complexification in Wolof noun morphology and morphosyntax

This chapter addresses simplification and complexification in the morphology and morphosyntax of Wolof noun classes. Simplification, compared with its closest Atlantic cognates, is well known to have occurred in Wolof as a whole, ever since its earliest attestations. In addition, urban Wolof further simplifies noun classes which is partly due to the particular dynamics of linguistic prestige in the Wolophone community. What went unnoticed until recently is that at the same time also complexification took place locally in some spots of the grammatical system, with the rise of morphological irregularity (overabundance) in some noun paradigms and of defectiveness and other irregularities, for some noun classes, in the paradigm of the indefinite article.