Where are syllables?

intro, what is this all about?

how Kahn (1976) builds a syllable and where this fails: Rule IIa precedes Rule IIb: onset maximization; Rule III creates ambisyllabicity (since words do not end in a short vowel, syllables are believed to not end in a short vowel either); the set of consonant clusters possible in an onset is somewhat arbitrary

binary features, why sounds are not atomic, scales defined by binary features (feature cooccurrence restrictions: *[+high,+low]; *[−son,−cons]); the autosegmental view of phonological representations; sonority as a scale, expressing natural classes in the sonority scale; sonority distance in clusters

different versions of the sonority hierarchy, very detailed sonority scales are also expected to reflect (consonant) lenition paths; Clements derives the sonority scale from binary features (unlike Selkirk) accompanied by feature coocurrence restrictions; the feature syllabic is not a binary feature linked to segments, but a property derived from the CV skeleton (C=[−syll], V=[+syll])

sonority dispersion: the lenght and the curve of the sonority plot iniinitial and final demisyllables; problems: how can French not have GV but have V initial demisyllables when the latter is more marked?; in final demisyllables: though VLO is supposed to be least marked, VNO is more common; underlying vs. surface representations

sonority “subsets” (all sounds > sonorants > vowels); restrictions on which subset may be syllabic, moraic; syllable weight; typology: CVV is always heavy, CVC may be light or heavy, CV is light; distinction between CVS and CVO; closed syllable shortening

Zec’s moraicity syllabicity scales can be complemented with a stressability scale (in English)

a recapitulation of previous readings: why we need syllables (natural classes); sonority in syllabification; sonority dispersion in initial and final demisyllables; superheavy syllables, degenerate syllables

the skeleton; segmenthood and problems; compensatory lengthening and problems; locality and problems

the maximal syllable that never occurs; what shows that s+C is not tautosyllabic; what shows that postconsonantal yod is not a vowel; the similarity of word-internal and word-final CC clusters

the untenability of the Uniform Boundary Principle, uncertainty about possible word-initial and -final clusters; syllabification is not required for most things it used for

test

read chapter 1 (pp. 20–67) of Kahn 1976

syllabify the following words using K’s rules: astronaut, structural, lambda, angry, anxious, anxiety (do not forget that you have to syllabify pronounced — not spelt — forms: look up and transcribe the words first), explain which rule provides which association line, mention problems if you encounter any

read Selkirk 1984 and answer the following questions

1. why is it problematic to exclude the feature [±syllabic] from among major class features?
2. find the typo in (9) (on p. 115)
3. what is minimum sonority difference? show examples
4. what is the English syllable template schema according to Selkirk? illustrate with examples
5. how does Selkirk analyse s+C clusters?
6. what is the difference between the sets of segments that may occupy R1 position in English and Spanish?
7. what inhibits an obstruent+glide+nasal syllable in Spanish?

read Clements 1990: 283–302 and answer the following questions

1. how is Jespersen’s sonority hierarchy different from that of Selkirk?
2. why does Clements use the feature [±vocoid] instead of the well-established [±consonantal]?
3. why does Clements use the feature [±approximant] instead of [±vocalic]?
4. what does a redundancy rule like [−sonorant] → [−approximant] entail by contraposition?
5. what is the condition for defining a hierachy in a binary-feature system?
6. compare Kahn’s rules I and II with Clements’ Core Syllabification Principle (CSP)

read Clements 1990: 302–333 and answer the following questions

1. what novelty does Clements formalize about the sonority relations at the two edges of the syllable?
2. what is a demisyllable?
3. how is the sonority dispersion of a demisyllable calculated? (don’t give the formula, explain it)
4. what is the problem with the V initial demisyllable?
5. how does the maximal onset principle follow from the ideal sonority dispersion within a syllable?
6. what is Clements’ argument against Selkirk-type sonority distance constraints?
7. what is Clements’ problem with Prince languages? (what are Prince languages anyway?)

read Zec 1995: 85–107 and answer the following questions

1. what features produce the sonority hierarchy according to Zec?
2. what are sonority classes? how are they different from a sonority scale?
3. why is the /l/ moraic in bell, but not in bellicose?
4. what is Osthoff’s law? what does it show us?

read Zec 1995: 108–124 and answer the following questions

1. give the syllabicity and the moraicity constraints for English, Lithuanian, and Lardil
2. what is the difference between lexical and postlexical vowel lengthening in Lithuanian?
3. how is the Kwakwala moraicity constraint different from that of Lithuanian?
4. what is an “edge constraint”? how is it different from a moraicity constraint?
5. what happens to a language without any syllabicity or moraicity constraint?

read Aoun 1979 and answer the following questions

1. what is a “superheavy” syllable and why is it problematic?
2. what is a “degenerate” syllable and why do you think it is problematic?
3. what do “ultima” and “penultima” and “antepenultima” mean?
4. what is a “foot”?
5. where do superheavy syllables occur in Cairene Arabic? and Lebanese Arabic?
6. is the last consonant in a superheavy syllable an onset or a coda in Arabic?

read Harris 1994: 32–52 and answer the following questions

1. what is the “skeletal tier” in a phonological representation?
2. what are the problems with the two-step linear analysis of compensatory lengthening?
3. why would Harris think compensatory lengthening like Old Hungarian [utu] ‘road’ > Modern Hungarian [uut] should be impossible? what principle does this change contravene?
4. how do English stress facts show that affricates are monosegmental? how do onsets suggest that they are not?
5. why does Harris deny that the syllable is a constituent?
6. what is the domain of phonotactic constraints?

read Harris 1994: 53–84

1. what facts argue that sC clusters are heterosyllabic?
2. what data suggest that postconsonantal [j] was earlier part of the following nucleus?
3. what data suggest that postconsonantal [j] is now part of an onset?
4. what facts argue for the constituenthood of the rhyme?
5. what facts argue against the constituenthood of the coda?
6. what conditions have to be satisfied in a superheavy rhyme in English?

read Szigetvári 2019

1. mention some cases where Wells’s syllabifications are inconsistent
2. why does Wells maximize codas (instead of onsets like others)?
3. the stressed vowel in Hudson [hə́dsən] is shorter than that in Pudsey [pə́dzij]; why is this a problem for Wells?
4. why is the Wellsian syllabification of atoll problematic?
5. what is the environment of aspiration in English if we adopt Lowenstamm’s proposal that “branching onsets” are single segments?

legend: shading = absence, tick = homework okay

Youssef Aoun. 1979. Is the syllable or the supersyllable a constituent? MIT Working Papers in Linguistics 1:140–148. get

George N. Clements. 1990. The role of the sonority cycle in core syllabification. In John Kingston and Mary E. Beckman (eds.), Papers in laboratory phonology I: Between the grammar and the physics of speech, Cambridge: CUP. 283–333. get

John Harris. 1994. English sound structure. Oxford: Blackwell, chapter 2. get

Daniel Kahn. 1976. Syllable-based generalizations in English phonology. PhD diss, MIT (Garland, 1980). get

Elisabeth O. Selkirk. 1984. On the major class features and syllable theory. In Mark Aronoff and Richard T. Oehrle (eds.), Language and sound structure, Cambridge, Mass: MIT Press. 107–136. get

Péter Szigetvári. 2019. Where are syllables? Radical: A Journal of Phonology 1: 92–125. get

Draga Zec. 1995. Sonority constraints on syllable structure. Phonology 12:85–129. https://doi.org/10.1017/S0952675700002396