Volume 32, Issue 3 -

We analyze three types of cases in which exceptional morphemes become regular in the presence of other morphemes ( regularization effects ). Vowel deletion in some Russian prepositions depends on the root that follows the preposition and also on the suffix that follows the root. In Japanese, dominant suffixes assign an accentual pattern to accented roots, but in Slovenian, dominance is conditional – revoked by another suffix. Finally, Tagalog and Dutch loanwords can contain non-native segments, except when certain affixes are present. We account for these phenomena in a new constraint-based framework, Lexical MaxEnt with regularization factors. In this framework, constraint weights are rescaled for exceptional morphemes, and some affixes carry regularization factors that reduce or cancel rescaling. We argue that regularization is a property of morphemes rather than whole words, and that it follows from how these morphemes are combined in the grammar rather than from whole-word storage in the lexicon.

In this paper, we analyze an unusual type of right dislocation in Zulu. We demonstrate that the construction in question combines A-movement of a DP out of an embedded finite CP into a matrix argument position (raising-to-object; RtO) with right dislocation of both the raised DP and the clause out of which it raises. The existence of such a construction is noteworthy for several reasons. First, the conclusion that Zulu allows A-movement out of a finite CP is at odds with standard Minimalist analyses of raising and also with the alternative analysis of RtO from finite clauses proposed in Bruening (2001, 2002). Second, on our proposed analysis, RtO can take place out of a CP-complement, which is subsequently dislocated and adjoined to a maximal projection, supporting Büring and Hartmann (1997), who argue that movement from an extraposed CP happens before CP-dislocation. Our findings thus provide a strong argument for a movement analysis of CP-extraposition. We finally conclude that these constructions present a number of complications for recent analyses of right dislocation as leftward movement plus clausal ellipsis (Ott and de Vries 2014, and Ott and de Vries (in press)), due to the interaction between A-movement and dislocation and the word order resulting from CP-extraposition.

This paper argues that the theory of phrase structure a certain linguistic approach assumes implies taking a stance on the formal nature of the computational procedures that generate that phrase structure. We will proceed by critically evaluating theories of phrase structure and labeling -which implies taking a structure as a unit for the purposes of further computations-, and building on and opposing to the proposals we review, we will claim that syntactic objects are not computationally uniform, and therefore the computational system in charge of establishing dependencies between symbolic objects within the mind is not uniform as well. We argue in favor of a linguistic-cognitive model which dynamically chooses different grammars based on the complexity of the input, and is capable of assigning a mixed phrase marker to an object that presents more than one computational pattern. Empirical evidence is provided in favor of our approach to phrase structure building, and further implications for a theory of labeling and predication are discussed as prolegomena to further research.

We argue that the cross-linguistic distribution of vowel systems is best accounted for by grammar-external forces of learnability operating in tandem with cognitive constraints on phonological computation, as argued for other phonological phenomena by Blevins (2004). On this view, the range of possible vowel systems is constrained only by what is computable and learnable; the range of attested vowel systems is a subset of this, constrained by relative learnability (Hale and Reiss 2000a, Hale and Reiss 2000b; Newmeyer 2005). A system that is easier to learn (e.g., one whose members are more dispersed in perceptual space) is predicted by our model to become more common cross-linguistically over evolutionary time than its less learnable competitors. This analysis efficiently accounts for both the typological patterns found in vowel systems and the existence of a non-trivial number of “unnatural” systems in the world’s languages. We compare this model with the leading forms of Dispersion Theory (notably Flemming’s (1995) implementation in Optimality Theory), which seek to explain sound patterns in terms of interaction between conflicting functional constraints on maximization of perceptual contrast and minimization of articulatory effort. Dispersion Theory is shown to be unable to generate the attested range of vowel systems or predict their interesting properties, such as the centralization typically found in two-vowel systems and the quality of epenthetic segments.