Canadian raising with language-specific weighted constraints

Abstract

The distribution of the raised variants of the Canadian English diphthongs is standardly analyzed as opaque allophony, with derivationally ordered processes of diphthong raising and of /t/ flapping. This paper provides an alternative positional contrast analysis in which the pre-flap raised diphthongs are licensed by a language-specific constraint. The basic distributional facts are captured with a weighted constraint grammar that lacks the intermediate level of representation of the standard analysis. The paper also provides a proposal for how the constraints are learned, and shows how correct weights can be found with a simple, widely used learning algorithm. *

Formalizing the connection between opaque and exceptionful generalizations

“…As mentioned before, Mielke et al (2003) propose that Raising is opaque because it is not in the grammar; by allowing a contrast between /aɪ/ and /ʌɪ/ as well as between /aʊ/ and /ʌʊ/, Raising becomes a pattern that is exclusively encoded in the lexicon (see Sanders 2003and Jurgec and Torres-Tamarit 2008 for similar proposals for other opaque patterns). This is similar to the solution in Pater (2014), who proposes a Harmonic Grammar analysis where the interaction between a latent /aɪ/~/ʌɪ/ and /aʊ/~/ʌʊ/ contrast (encoded by a particular weighting of Faithfulness constraints) and a constraint that promotes Raising before [ɾ] leads to the emergence of an /aɪ/~/ʌɪ/ and /aʊ/~/ʌʊ/ contrast only before /ɾ/. Mielke at al.…”

“…In particular, this paper focussed on the opaque pattern of Canadian Raising (Joos 1942, Chomsky 1964, Chambers 1973. This fits in a tradition of lexicon-oriented analyses of Canadian Raising (Joos 1942, Mielke et al 2003, Pater 2014, as well as the literature that analyzes other cases of opacity in terms of lexical storage (Sanders 2003(Sanders , 2006 or contrast preservation (Lubowicz 2003). At the same time, differently from most of these approaches, the current account incorporates the insight that Canadian Raising is a productive process (Idsardi 2006).…”

“…The empirical case study in this paper is the opaque process of Canadian Raising (Chambers 1973). This process was chosen because it is well-studied in the literature on phonological opacity (see, for instance, Chomsky 1964;Bermúdez-Otero 2003;Mielke et al 2003;Hayes 2004;Idsardi 2006;Pater 2014), and specifically because analyses of it have been suggested that make use of elements of exceptionality (see below).…”

“…Mielke et al (2003) and Sanders (2003Sanders ( , 2006 hypothesize that opaque processes are not a part of the grammar, but, rather, lexicalized vestiges of former phonological processes. Pater (2014) shows that Canadian Raising (to be discussed in section 3.1) can be derived from a combination of universal and language-specific constraints in a Harmonic Grammar framework. In addition, Lubowicz (2003) provides an account of opacity explicitly motivated by the preservation of contrast.…”

Markedness and Lexical Typicality in Mandarin Acceptability Judgments

“…Clearly, feature co-occurrence constraints like * [+F, +G] Hayes and Wilson (2008) are forced to use various tricks to search efficiently through all logically possible feature combinations, which they estimate at around 3.5 million for the English lexicon (p. 392, fn. 8); Pater (2014) admits that he avoids this problem only by artificially limiting his feature set. Learning conjunctively coordinated constraints faces the same computational challenge.…”