Most consonants are free to appear in a singleton onset and coda position (see § 3 for complexonsets). We thus find, for example, l and k in onset and coda position (e.g. lak ‘nail polish
1 Introduction Complexonsets of one kind or another have been part of mainstream views of the structure of Old Chinese ( OC ) for nearly a century (Maspero 1920; Karlgren 1923). However, few consensus conclusions have been reached. The hypothesis that some OC words were sesquisyllabic, i
SR shows a typical acquisition pattern of first avoiding marked structures, then repairing them, then producing them faithfully. I propose a method for diagnosing and quantifying avoidance, and then analyze the avoidance using the null parse approach. I offer a computationally implemented Tessier 2007, 2008, 2009, Becker and Tessier 2012), modeling the avoidance of marked structure as selection of the null parse. The model predicts a realistic acquisition path using a persistent M > F > MParse bias and the relativization of MParse to markedness constraints.
Data obtained in a longitudinal study of three Hebrew-acquiring children indicate the existence of a stage in which target words with word-initial simple onsets are produced without an initial onset, while target words with an initial complex onset are produced with a simple onset. This is a case of chain shif, where the output of cluster simplification (w[CCV → w[CV) is the input of onset deletion (w[CV → w[V), but forms derived from cluster simplification do not result in onset deletion (w[CCV → *[V). Assuming the constraint-based framework of Optimality Theory (Prince and Smolensky 1993/2004), the main question to be addressed is how can we account for different outputs for input-simple onsets compared to input-complex onsets during the same stage of phonological development? In addition, given the principles of universal markedness and their role in language acquisition, how can we account for the production of onsetless syllables for targets with onsets? I argue that the underlying motivation for omission is an increase in prosodic complexity, and propose an analysis based on local constraint conjunction (Smolensky 1993) to provide a unified developmental account of simple and complex onsets.
will be acquired before TR (on the assumption that complex segments are less marked than complexonsets). Th is optionality of structure, it is argued, can account for the variation observed the acquisition of initial sT – initial TR (Barlow, 1997 ). However, this approach does not seem to be
complexonsets are also attested. The syllables found in the corpus are shown in Table 5 below. The vowel in onsetless syllables is glottalized; and some amount of closure is generally visible on a spectrogram, although it does not appear to be full closure in some examples (see Chapter 3, §3.3). In an
: Asherov and Bat-El lay out quantitative and qualitative details on MH syllable structure, with a special emphasis on word initial complexonsets; and Bat-El, Cohen, and Silber-Varod describe the distribution of MH stress and its phonetic manifestation.
The following three papers deal with the
unsyllabiﬁable consonant. For cases (a) and (b) above, C2 is lost as unsyllabiﬁable, since an obstruent does not form a legal complexonset with the following consonantalised and occlusivised /I/: * δ ón.tIa > δ ón.g´a , * o.mor.fIá > o.mor. k´á . In addition, complex codas are not allowed: * δ ónt.k´a , * omorf
“strap, strip” era > ia méra > mía “day” ara > aa skára > skáa “grill” ure > ue kurévo > kuévu “I cut someone’s hair” In complexonset clusters of the type CrVC, /r/ again deletes and lengthening oc- curs independently of the cluster’s position within the word. (4) /r/ in onset CrVC: deletion and
or *alpuro (as shown in Pagoni 1993a). These are environments justifying the S onority constraint, by which complexonsets rise in sonority and complex codas fall in sonority. Now, whereas word- medially in cases like: alburo only the Onset position is available, due to the presence of a consonant in