Learning Long-Distance Phonotactics

Abstract

This article shows that specific properties of long-distance phonotactic patterns derived from consonantal harmony patterns (Hansson 2001, Rose andWalker 2004) follow from a learner that generalizes only on the basis of the order of sounds, not the distance between them. The proposed learner is simple, efficient, and provably correct, and does not require an a priori notion of tier or projection (contra the model in Hayes and Wilson 2008); nor does it rely on the additional structure provided by Optimality Theory grammars (Prince and Smolensky 1993Smolensky , 2004 or grammars in the principles-and-parameters framework (Chomsky 1981, Dresher and Kaye 1990, Gibson and Wexler 1994. Not only does the noncounting nature of nonlocal dependencies automatically follow from the way the learner generalizes, it also explains the absence of blocking patterns from the typology. Finally, the learner lends support to the idea that long-distance phonotactic patterns are phenomenologically distinct from spreading patterns, contra the hypothesis of Strict Locality (Gafos 1999, et seq.).

Long-Distance Phonotactics as Tier-Based Strictly 2-Local Languages

“…However, by referring not to string-adjacency but instead to precedence relations, which are by definition blind to distance and intervening material, unbounded harmony can be described as a Strictly 2-Piecewise (SP 2 ) pattern that disallows certain subsequences of length 2 (e.g., *s…ʃ, *ʃ…s). Heinz (2010) demonstrates that nearly all cases of unbounded consonant harmony from two typological surveys , Rose & Walker 2004 can be described as members of the SP 2 class of formal languages. However, the potential extension of this approach to all types of long-distance dependencies is impeded by a number of attested phonotactic patterns that are not SP 2 .…”

“…Consequently, as Rabin & Scott (1959) show, all stringsets generated by these relations (the surface phonotactics) are members of the regular region of the Chomsky hierarchy (Chomsky 1956). While certain syntactic processes result in relatively complex stringsets (of words rather than segments; Culy 1985, Shieber 1985, Kobele 2006, it turns out that nearly all phonological patterns are indeed regular, including long-distance consonant agreement and disagreement (Heinz 2010, Heinz et al 2011, Payne 2014.…”

“…The unbounded sibilant harmony pattern of Aari in (1) above is a restriction on sequences of [-anterior][+anterior] sibilants on a tier that includes all and only [+strident] segments, which can be generated by a TSL 2 grammar with the tier T ={s, ʃ , tʃ', ʒ, …} and S̅ ={*ʃs, *tʃʼs, *ʒs, …}. Heinz (2010) hesitates to describe patterns of longdistance consonant agreement in terms of tiers, citing a lack of known systems that exhibit blocking. We argue however, that the widely attested locality type of transvocalic consonant harmony, such as the Koyra sibilant harmony shown in (2), can be recast as long-distance dependencies that are blocked by any intervening consonant.…”

Cognitive and Sub-regular Complexity

“…The question of the relative degree of difficulty of distinguishing (proto-) linguistic patterns has received a considerable amount of attention in recent Artificial Grammar Learning (AGL) research [1,2], as well as in current research in phonology [3,4]. In the AGL research, as in the phonological research, the complexity of the learning task has been central.…”

Logic and the Generative Power of Autosegmental Phonology

“…Since SPE is generally considered powerful enough to capture any phonological alternation, it is reasonable to presume that phonology is regular. More recent research has pushed this hypothesis further, arguing that phonology is sub-regular (Heinz, 2009(Heinz, , 2010Graf, 2010b;Heinz & Lai, 2013;Rogers et al, 2013).…”