In lieu of an abstract, here is a brief excerpt of the content:

Reviewed by:
  • Kayardild morphology and syntax by Erich R. Round
  • Gregory Stump
Kayardild morphology and syntax. By Erich R. Round. Oxford: Oxford University Press, 2013. Pp. xix, 295. ISBN 9780199654871. $135 (Hb).

Since its introduction by Aronoff (1994), the concept of the morphome has had profound effects on the field of morphology. Once one acknowledges that a language’s morphology may be sensitive to properties, categories, and patterns to which the rest of its grammar is simply blind, one is naturally inclined (or more inclined) to think of morphology as an autonomous grammatical component. Formerly tolerated as pesky anomalies, morphomic phenomena now invite closer examination for the insight they give into the organization of a language’s morphology—and they do so pervasively.

In Kayardild morphology and syntax, Erich Round discusses a system of morphosyntax that, in his words, ‘provides the clearest example yet of a linguistically significant morphomic level of representation, in the sense that a significant range of generalizations are accorded their simplest and most elegant expression in terms of the same, morphomically represented units’ (38). The analysis that R proposes for Kayardild morphosyntax provides convincing support for this claim; in particular, his analysis provides a systematic account of the fact that, repeatedly in Kayardild, a heterogeneous set of morphosyntactic feature values is realized by precisely the same inflectional morphology (for example, the locative case, the TAM (tense/aspect/modality) properties ‘instantiated’, ‘present’, and ‘immediate’, the clausal property of complementization, and the derivation of place names are all independently subsumed by a morphome ‘µloc’ whose realization is the suffix -ki).

Kayardild is doubly remarkable for the intricate system of syntax in which such morphomes do their work. R argues that the distribution of morphosyntactic feature values in Kayardild syntactic structure is most cleanly definable not with respect to sentences’ superficial constituent structure, [End Page 497] but with respect to richly layered ‘nonsurface’ structures from which superficial syntax may be projected. In his analysis, these layers are motivated primarily by the notion of concord—‘[t]he morphological realization, on multiple words dominated by a syntactic node n, of a morphosyntactic feature value associated with n’ (76). In accordance with this notion, nonsurface syntactic structures are postulated in which each phrasal node may serve as the initial attachment site from which a morphosyntactic feature value v percolates to word forms whose morphology expresses v; the nesting of a nonsurface structure’s phrasal constituents therefore reflects the nesting of morphosyntactic feature values’ domains of concord. The inflection of a given word form may therefore involve the morphological realization of morphosyntactic feature values inherited from several dominating nodes, of greater or lesser hierarchical proximity; indeed, a word form may even inflect for contrasting values of the same feature.

Kayardild has an unusually elaborate system of morphosyntactic features, whose values themselves vary with respect to their attachment sites. Thus, whether the word forms in a DP inflect for a morphosyntactic feature value v depends on whether that DP is dominated by v’s attachment site. Although the percolation of feature values cannot cross the maximal clausal node S″, a feature value originating in a matrix clause may, for example, percolate into an embedded VP. Thus, given a feature whose value v1 has a matrix attachment site X and whose value v2 has an embedded attachment site Y, the percolation of these values relative to a DP contained in an embedded VP node Z may have four logically possible outcomes: DP may inherit v1 but not v2 (if Z is dominated by X but not Y); v2 but not v1 (if Z is dominated by Y but not X); both v1 and v2 (if Z is dominated by both X and Y); or neither v1 nor v2 (if Z is dominated by neither X nor Y). This is precisely what happens in the morphosyntax of the feature tama (‘athematic tense/aspect/mood’), a fact predicted by the nonsurface structure that R proposes for Kayardild sentences (118ff.). Because of both the variety of attachment sites for values of tama and the complex ways in which the inflectional realization of these values is morphomically...

pdf

Share