Languages are replete with cases of lexical allomorphy. Their characteristic property is that the distribution of allomorphs is explicable on general phonological grounds, but no actual phonological rule exists in the grammar of the language that would derive both from the same underlying representation. In this note, we take up the two cases mentioned above, the historically matured allomorphy of the Latin noun-forming endings and the newly emerging allomorphy of the plurality marker in Japanese loanwords. From a variety of evidence characterized as 'prosodic trapping', Mester 1994 argues that the optimal foot structure of Latin is the bimoraic balanced trochee, ('LL) (two light syllables) or ('H) (one heavy syllable). Crucially, in a quantitative system, the unbalanced ('HL) and ('LH) do not qualify as trochees, and neither does ('L). In this restricted foot inventory, light syllables are often prosodically trapped initially: #L(H)..., and medially between heavy syllables: ...(H)L(H)....

This paper discusses the pathway, motivating factors, and mechanism of the historical development of the initial accent pattern in trimoraic nouns in Kyoto Japanese, tracing its development from being one of the least common patterns in Early Middle Japanese (ca. 12th century AD) to becoming the most common pattern for accented nouns in present-day Modern Kyoto Japanese.

Accent, defined as a pitch fall from high pitch to low pitch (Haraguchi 1999), if it is present, may in principle fall on any mora in a Kyoto Japanese word, as illustrated by the example in (1).

(1) Possible accent patterns in Modern Kyoto Japanese

a. I’noti ‘life’ H’LL initial accent

b. OTO’ko ‘man’ HH’L second accent

c. kaBU’to ‘helmet’ LH’L second accent

d. osaK'i ‘the future’ LLF final accent

Examination of the frequencies of each accent pattern reveals that possible accent locations are not equally prevalent. Instead, there is a clear preference for initial accent, with 68% of trimoraic accented native nouns having initial accent (Yoshida and Zamma 2001) and 73% of trimoraic accented nouns across all lexical strata having initial accent (calculated from raw data from Sugito 1995). Despite this present-day prevalence of the initial accent pattern, Kyoto Japanese accented trimoraic nouns did not always exhibit this preference. In the accentual system of Early Middle Japanese, the earliest recorded ancestral accentual system of the Kyoto Japanese branch, the initial accent pattern accounted for only about 6% of accented trimoraic nouns (calculated from Sugito 1995).

Previous investigations (e.g. Hata and Hasegawa 1988, Hasegawa and Hata 1995, Kawakami 1995 as cited by Nakai 2001, Matsumori 1999, 2008, and Shimabukuro 2007) have proposed accounts for what is likely to have happened in the developmental pathway from Early Middle Japanese to Modern Kyoto Japanese, suggesting that several accent classes merged into the initial accent pattern or otherwise merged into patterns which would eventually come to have initial accent. One pathway described in these accounts is as follows. A pitch rise enhancement process proposed by Kawakami (1995) occurred, whereby an L tone immediately preceding an H tone was lowered in order to enhance the contrast between them, causing any L tones preceding the lowered L tone to rise to H. Thus, for LLH, the following changes would occur: LLH > MLH > HLH. By this process, the Early Middle Japanese forms LLL, LLH, and LLF would gain the initial H tone needed to eventually merge into initial accent (HLL), becoming HHL, HLH, and HLF respectively. HLH and HLF became HLL due to a pressure for words to have only one peak (“culminativity”) (Nakai 2001, Shimabukuro 2007). HHL then underwent leftward kernel shift, shifting the lowering kernel on the second mora to the first mora, yielding HLL (Nakai 2001).

Although these accounts propose plausible changes to account for the differences between Early Middle Japanese and Modern Kyoto Japanese, these accounts generally do not focus on what factors were the seeds of change which contributed to the occurrence of the proposed changes, with the exception of Hata and Hasegawa (1988) and Hasegawa and Hata (1995), who attribute the rightward shift of accent to peak delay.

In this paper, I argue that the changes proposed in previous work can be attributed to pattern frequencies that cause learners to induce and rerank constraints in response to these frequencies, changing the grammar in learning generations and leading to the mergers of several accentual classes and thus the development of initial accent as the most common accented pattern in trimoraic nouns. Accordingly, in the development of Modern Kyoto Japanese, pattern frequencies are the seeds of change. Using the pitch rise enhancement process described above as a starting point, I propose four stages of development, with all but one (Stage 3) building on pattern frequencies created by the previous stage. The proposal is framed in the framework of Optimality Theory (Prince and Smolensky 1993/2004), which allows for the precise specification of the state of the grammar at every stage in terms of competing pressures whose influence changes as constraints are reranked through the generations.

Pattern frequencies in pre-pitch rise enhanced nouns in the accentual system of Early Middle Japanese already show a weak preference for word-initial H tones. This weak preference in conjunction with the increased frequencies of word-initial H from Kawakami’s pitch rise enhancement process causes learners to induce INITIAL-H, a constraint which requires words to begin with an H tone, in Stage 1 and promote it to a higher rank. In Stage 2, INITIAL-H will have consequences for culminativity, preferring HLH > HLL instead of HLH > LLH. The constraint CULMINATIVITY becomes active because of relatively low frequencies of words with two peaks, causing learners to promote the constraint. In Stage 3, peak delay causes the change LHH > LLH. While this change does not directly lead to an increase in initial accent, as the new LLH forms do not become initial-accented, it nonetheless plays a role in reducing the amount of multiply-linked H tones in the accentual system, serving as the catalyst for Stage 4. In Stage 4, the change HHL > HLL occurs due to the promotion of NOMULTILINK-H, a constraint prohibiting H tones from being linked to multiple moras (Ito and Mester 2018), on the basis of the reduced prevalence of multiply-linked H tones resulting from peak delay in the previous stage.

Finally, a test of the initial Early Middle Japanese data using Hayes and Wilson (2008)’s UCLA Phonotactic Learner induces a constraint similar to the INITIAL-H constraint proposed here, suggesting that learners can induce constraints and vary their strengths based on pattern frequencies in the input data.

The contribution of the present investigation, therefore, is twofold. First, the present study advances our understanding of the development of the modern-day prevalence of initial accent in Modern Kyoto Japanese trimoraic accented nouns. Second, the present study advances our understanding of what kinds of factors can influence the reranking and promotion of constraints, both in Japanese and beyond. As one such factor, the present study proposes that constraints can be reranked on the basis of the surface frequencies of patterns present in the input learners receive.

In a range of languages, the mapping from syntactic to prosodic structure produces "mismatches", where a prosodic constituent has no matching syntactic constituent. This is puzzling, since prosodic structures are clearly based on syntax, and the two are often isomorphic. Here, I examine the predictions of three theories of the syntax-phonology interface using Optimality Theory: Align/Wrap Theory, Match Theory, and a c-command based theory I call Command Theory. Command Theory is shown to be well suited to deal with the phrasing of ditransitive constructions. The types of matches and mismatches predicted by these theories are examined through the lens of formal OT, with careful attention to candidate generation and constraint definitions. This is accomplished using the JavaScript application SPOT (Bellik, Bellik, & Kalivoda 2016). Data is drawn from Bantu, Germanic, Romance, Japanese, and other languages and language families.

This work presents OT analyses of two Hindi dialects: Kelkar's Hindi and Eastern Standard. A metrical analysis of Eastern Standard is necessary since the stress pattern cannot be predicted without feet. A new constraint, S-WSP, must be added into the constraint set in order to predict the different behavior of superheavy and heavy syllables regarding non-finality and clash effects in Eastern Standard. S-WSP is similar to the familiar WSP in that it requires superheavy syllables to bear stress, but it does not require heavy syllables to bear stress, giving special status to superheavy syllables. In addition to the empirical argument from the Eastern Standard data, S-WSP is further argued for from a theoretical perspective. S-WSP and WSP should be viewed as the result of a decomposed scalar constraint on the weight of syllables in non-foot-head positions. The addition of S-WSP is not imprudent but in line with other scalar constraints.

Turkish vowel harmony is very systematic, but in a little-studied class of words, appears to break down. I propose that this apparent harmony failure is actually covert harmony, or exceptional transparency.

Background: The backness of vowels in Turkish suffixes is determined by the backness of the nearest vowel in the word they attach to (1).

(1) Nom. Dative Gloss

a. kadran → kadran-a, *kadran-e 'clockface'

b. beden → beden-e, *beden-a 'body'

Harmony within roots is often violated, but harmony between suffixes and the nearest root vowel is extremely robust (2). Nonetheless, in certain roots, /a/ exceptionally behaves as transparent and suffixes must surface with front vowels (3).

(2) Singular Plural Gloss

a. kitap → kitap-lar, *kitap-ler 'book'

b. kahve → kahve-ler, *kahve-lar 'coffee'

(3)

a. harf → harf-ler, *harf-lar 'letters'

b. dikkat → dikkat-ler, *dikkat-lar 'attentions'

To account for this exceptional transparency, I extend Smolensky's (2006) headed feature domains theory, and propose that the input to phonology contains headless feature domains.

Proposal: Existing theories of vowel harmony require a segmental harmony trigger. But harf contains no [-back] segment to head a front feature domain and select front suffixes, so harf-ler remains unexplainable. Extending Smolensky (2006), I propose that all features are feature domains, which replace traditional segmental feature specifications.

In Turkish, backness domains normally coincide with roots. Alternating suffixes lack their own backness domains and merge into the stem's domain. Exceptional transparency is driven by faithfulness considerations: the root as a whole is a front domain, but it contains /a/ in an embedded back domain. Suffixes are incorporated into the outermost backness domain as usual, resulting in transparency.

(4) (dik(ka)bt)f) + lEr → (dik(ka)btler)f)

Since vowels must realize the backness of their containing domains, high- and mid- vowels cannot occur in conflicting embedded domains. However, as a low vowel, /a/ is more central than the other back vowels, and unlike the other Turkish back vowels, it lacks an exactly height-matched front counterpart phoneme. Perceptual reasons, then, lead the Turkish constraint-ranking to allow only /a/ as an exceptionally transparent vowel. The feature-as-domain analysis of vowel harmony models the selection of front suffixes by harf and dikkat without the need to posit phonetically unrealized front features on consonants, or any other additional apparatus.

In Japanese native phonology, geminate consonants are contrastive (as in [kata] ‘shoulder’ vs. [katta] ‘win-PAST’), but geminates in loanwords can have differing sources and motivations (see Kubozono, Itô, Mester 2009, Kawagoe 2015, and references cited therein): we see gemination of singletons in loanwords from English, in which consonant length is not distinctive ([kæt] E ng ‘cat’ > [kjatto] Jp ), whereas we see geminate-preservation in loanwords from Italian ([espresso] It ‘espresso’ > [esupuresso] Jp), in which the length of most consonants is contrastive. In loanwords from Italian, however, not all geminates are preserved. This research addresses the cases of degemination, and captures the pattern as stress-based neutralization (Beckman 1998) of consonant length within the framework of Optimality Theory (Prince & Smolensky 1993). Through a database built from dictionaries and a nonce-adaptation survey conducted online, it confirms the preference towards geminates in penultimate position and the ban against geminates in other positions, especially for liquid geminates.

There has been a recurrent debate in generative phonology concerning the inclusion of hierarchical prosodic structure in phonological representations. On one side, there are those who argue that prosodic structure plays an indispensable role in the conditioning of phonological phenomena, especially stress, intonation, and segmental phonotactics. On the other side of the divide are researchers who suggest that all such phenomena yield to empirically adequate non-structural analyses, which are independently favored by criteria of theoretical parsimony.

This dissertation focuses on one aspect of the larger debate over prosodic organization: the existence of the metrical foot. In standard conceptions of phonological structure the foot is a prosodic constituent, falling between the syllable and the word, which mediates the assignment of word-level stress. The foot obviously has no role to play in non-structural theories of prosody. Such frameworks assume that stress assignment is not dependent on prosodic constituents, but is instead directly computed over segments or syllables on a metrical grid.

The evidence brought to bear on the choice of prosodic theory has, by-and-large, been drawn from the typology of attested stress patterns. Both structural and non-structural theories of stress appear capable of modelling roughly the same range of stress systems. Consequently, many of the arguments for or against foot structure have centered on the ability of each theory to express the typology of stress assignment in a compact, elegant, and predictive way. This dissertation expands the terms of the debate by examining foot-conditioned phonotactics, and to a lesser extent foot-conditioned morphology, as a window on the nature of prosodic constituency in natural language.

A major conclusion of this dissertation is that hierarchical prosodic structure must be admitted as part of phonological representations in order to capture the full range of prosodically-conditioned segmental phonotactics found in natural language. Three case studies form the heart of this claim. Specifically, I show that Huariapano, Uspanteko, and Irish all manifest foot-dependent phonotactics that cannot be insightfully analyzed without recourse to abstract metrical structure. Importantly, these arguments go beyond claims about relative theoretical elegance: non-structural analyses simply fail to account for the relevant phonological phenomena in an explanatory way.

In arguing for a foot-based theory of stress assignment, I also make the case for a fairly traditional conception of the metrical foot. First, I contend that stress is always assigned on the basis of foot structure, and only to foot heads, though feet may be unstressed or `covert' as well. Second, I argue that feet are always maximally binary, even in languages where stress assignment itself is ambiguous between binary and unbounded footing. Third, I propose that any given language makes use of at most one system or `tier' of metrical organization (the uniformity of footing hypothesis). In particular, I demonstrate that the rhythmic phonology of Huariapano can and should be modeled within a single, unified system of foot structure, despite previous claims to the contrary. Along the way, I identify a novel source of phonological prominence effects: the augmentation of foot-initial syllables.

The dissertation closes with an artificial grammar study exploring how language learners acquire foot-conditioned segmental phonotactics. Given the `hidden' character of abstract prosodic structure, foot-conditioned phonotactics pose an interesting learning problem. This is especially true given the recent rise of the view that all synchronic phonological knowledge is the result of inductive learning over phonetic regularities in the speech stream. The results of the artificial grammar study suggest that speakers of both English and Japanese were inclined to learn a stress-conditioned vowel phonotactic in terms of foot structure rather than stress per se. The experiment thus supports the claim that language learning is sensitive to a foot-based parsing bias that encourages the use of the foot as a general explanatory device in acquisition. These findings provide a possible explanation for why binary foot structure is found in languages lacking rhythmic, foot-based stress, and suggest that the foot may be a prosodic universal.