Laura J. Downing

On the Prosodic Misalignment of Onsetless Syllables

Cross-linguistically, onsetless syllables not only have limited distribution but also exhibit exceptional prosody: in some languages they are excluded from reduplication, or cannot bear main stress or a high tone. These exceptional properties are clearly linked to the relative ill-formedness of onsetless syllables, but previous analyses do not formalize this correlation in a way that generalizes straightforwardly to all cases of exceptional prosody. I argue in this paper that the theory of Generalized Alignment (McCarthy and Prince 1993a,b), developed within Optimality Theory, provides us with a unified way of accounting for onsetless syllable exceptionality. By constraining the domains for phonological processes to optimally begin with optimal syllables, the exceptional prosody of onsetless syllables can uniformly be analyzed as prosodically motivated constituent misalignment.

Canonical forms in prosodic morphology

1. Introduction 2. Prosodic Hierarchy-Based Templates 3. Morpheme-Based Templates 4. The Role of Phonology in Defining Canonical Form in MBT 5. Questions for Future Research and Conclusion References

Prosodic misalignment and reduplication

In KiHehe and IsiXhosa, the verb stem is the base for reduplication, and the reduplicant is prefixed to the stem ((la; 2a). As in other languages (McCarthy & Prince 1986, 1993a,b, 1995), the reduplicant (RED) is misaligned with vowel-initial stems: infixed after the initial vowel, in IsiXhosa (lb) and “exfixed” before a prefix which is syllabified with the initial vowel in KiHehe ((2b); notice the infinitive prefix ku- /kw-is also reduplicated):

Prosodic stems in Zezuru Shona

Abstract In the traditional Prosodic Hierarchy, the Prosodic Word is the lowest level of the hierarchy defined in terms of the interface between morphosyntactic and prosodic constituents. Recent work by Itô and Mester reaffirms this, defining Prosodic Word as matching the syntactic category X (N, V, and offering no sub-lexical constituents. In this article, we present empirical arguments in favour of the Prosodic Stem as a level of the Prosodic Hierarchy, immediately dominated by Prosodic Word. Minimality effects, tone and vowel hiatus resolution processes in Zezuru demonstrate that Prosodic Stem and Prosodic Word levels play distinct roles; they are domains for different phonological processes. Thus, we argue that more constituents of the Prosodic Hierarchy are needed, where the Prosodic Stem domain is distinct from the Prosodic Word.

Prosodic phrasing of Chichewa relative clauses

Abstract Earlier studies of Chichewa phrasal prosody – Kanerva (Focus and phrasing in Chichewa phonology, Garland, 1990) and Truckenbrodt (Phonological phrases: Their relation to syntax, focus and prominence, MIT, 1995, Linguistic Inquiry 30: 219–255, 1999, Linguistische Berichte 203: 273–296, 2005) – claim that Phonological Phrases in this language can potentially be quite large, as an entire maximal syntactic constituent – for example, a verb and all its complements – is parsed into a single Phonological Phrase in their analysis. This proposal predicts that even long and internally complex syntactic constituents will be parsed into a single Phonological Phrase. Relative clause constructions provide the ideal context for testing the proposal. However, neither Kanervas (Focus and phrasing in Chichewa phonology, Garland, 1990) study of Chichewa prosodic phrasing nor Watkinss (A grammar of Chichewa: A Bantu language of British Central Africa, University of Chicago, 1937) and Mchombos (The syntax of Chichewa, Cambridge University Press, 2004) grammars include sufficient examples to get a complete picture of their prosodic properties. Our study first presents a survey of prosodic phrasing in four relative clause constructions, then it provides an optimality theory (OT) analysis of the data. We show that Truckenbrodts (Phonological phrases: Their relation to syntax, focus and prominence, MIT, 1995) OT analysis must be revised to account for the correlation we find in this data between syntactic phase edges and prosodic phrasing. Following work like Chen (Phonology Yearbook 4: 109–149, 1987), we show that the complement/adjunct distinction is an additional factor conditioning prosodic phrasing.

All Depressors are Not Alike: A Comparison of Shanghai Chinese and Zulu

Shanghai Chinese and Nguni Bantu languages (including Zulu) are both tone languages, and both have a set of phonetically voiceless consonants which have a pitch lowering effect on the tone of a following vowel. In a recent paper, Jessen and Roux (2002) propose that depressor consonants in these two languages can be characterized by the same [slack voice] feature, implemented in a parallel fashion in the two languages, and with f0 lowering compensating in both languages for absence of phonetic voicing. This paper investigates their claims in some detail by comparing production studies of the effect of depressors on tone in Shanghai Chinese and in Zulu and by comparing the tone systems of the two languages. The production studies show that the phonetic implementation of the depressor effect is, in fact, quite different in the two languages. There is also no basis for claiming that f0 lowering compensates for lack of voicing. The comparison of the tone systems shows that the differences in phonetic implementation of the depressor effect follows from differences in their tonal phonologies. In the spirit of Kingston and Diehl (1994), [slack voice] can then be considered appropriate for both languages in spite of differences in phonetic implementation.

*NT revisited again: An approach to postnasal laryngeal alternations with perceptual Cue constraints

Phonological alternations in homorganic nasal-stop sequences provide a continuing topic of investigation for phonologists and phoneticians alike. Surveys like Herbert (1986), Rosenthal (1989), Steriade (1993) and Hyman (2001) demonstrate that cross-linguistically the most common process is for the postnasal stop to become voiced, as captured by Pater’s (1999) markedness constraint *NT. However, as observed since Hyman (2001), *NT alone does not account for all postnasal patterns of laryngeal alternation. In this paper, we focus on three problematic patterns. First, in some languages with a two-way laryngeal contrast, voiceless stops are aspirated postnasally, i.e. the contrast between NT and ND is enhanced, not neutralized. Second, in some languages with a three-way laryngeal contrast, the voicing contrast is maintained postnasally, while the aspiration contrast neutralizes in favour of aspiration. Third, in other languages with a three-way laryngeal contrast we find the opposite postnasal aspiration neutralization: aspiration is lost. We argue that an analysis based on perceptual cues provides the best account for this range of alternations. It demonstrates the crucial role of perceptual cues and laryngeal contrasts in a particular language while fitting the range of patterns into an Optimality Theoretic factorial typology that covers a wider range of postnasal laryngeal alternations than previous analyses.

How intonations interact with tones in Embosi (Bantu C25), a two-tone language without downdrift

This chapter presents a study of the intonational system of Embosi, a two-tone Bantu language without downstep (or downdrift) in which boundary tones are superimposed on tones and not inserted on the same line. To account for the interaction between lexical tones and boundary tones, a model with a dual register organization is proposed. It involves a basic register for the tone realizations and enlargements above and below this basic register, due to extra-high or extra-low boundary tones. Expansions were found in yes-no questions and in the expression of emphasis. Boundary tones are involved in the marking of assertions (L%), polar questions (HL%), wh-questions (L%), among others. Focus is not indicated by prosodic means. Embosi has a prosodic hierarchy with Phonological Words and Intonational Phrases but lacks Phonological Phrases.

(In)Direct Reference in the Phonology-Syntax Interface under Phase Theory: A Response to Modular PIC

Although in many interface theories, the domains of phrasal phonological processes are defined in terms of prosodic constituents, D’Alessandro and Scheer (2015) argue that their proposed modification of phase theory, Modular PIC, renders prosodic constituents superfluous. Phrasal phonological domains can instead be defined directly in the syntax. In this response, we argue that Modular PIC does not provide a convincing new approach to the syntax-phonology interface, as it is both too powerful and too restrictive. We show that the analysis offered of raddoppiamento fonosintattico in Eastern Abruzzese does not justify the loss of restrictiveness Modular PIC brings to phase theory. We also show that Modular PIC is too restrictive to account for phenomena, from Bantu languages and others, that have received satisfactory analyses within interface theories that appeal to prosodic constituents. We conclude that Modular PIC does not successfully replace prosodic constituent approaches to the interface.

Aspects of the intonational phonology of Bàsàá

Two major aspects of the intonational phonology of Bàsàá, a Northwest Bantu language spoken in Southern Cameroon with an underlying opposition between high, low and toneless tone bearing units, are presented in this chapter. First, two tonal processes, high tone spreading (HTS) and falling tone simplification (FTS), show sensitivities to prosodic domains: the phonological phrase and the intonational phrase, respectively. Second, tones do not seem to be affected by sentence modalities or information structure. Declarative sentences and yes-no questions show nearly identical intonation patterns, and varying the location of focus in a sentence does not significantly affect sentence prosody either. So far, Bàsàá is a language that shows little interaction between tone and intonation.