Abstracts Submitted: 464
Number of Users: 528
View Abstracts Submitted
Back to home Page
Most research on gemination, until fairly recently, has focused on intervocalic geminates, the most common variety, although research is expanding on edge geminates as well. Edge geminates, as their name indicates, are geminates which occur on the left or right edge of a word, either word-initially or word-finally; they have interesting typological characteristics (i.e., that the existence of word-final geminates in a language implies intervocalic geminates as well, but word-initial geminates do not imply intervocalic geminates), and their weight properties differ from intervocalic geminates, being either moraic or non-moraic depending on the language (Topintzi & Davis to appear). An even more interesting twist to this problem is the existence of derived, or fake, geminates as well, in languages that have phonemic geminates and even in languages that do not (like English, for example, which does have fake geminates in some cases of morphological concatenation, like the word “innumerable,” where the /n/ of the prefix /in-/ creates a fake geminate with the following /n/ of [numr̩əbl̩]) (Oh & Redford 2012). Fake geminates occur because of consonant assimilation, morphological concatenation, or identical consonants across word boundaries, and languages appear to differ in how they are realized phonetically. Lahiri & Hankamer (1988), for example, reported no acoustic differences in stop closure duration between fake and true geminates in either Bengali or Turkish. However, Jones (2016) found, through acoustic measurements of word-final fake geminates in Egyptian Arabic, that fake geminates were much shorter than what has been previously reported in the literature for Arabic geminates (a singleton to fake geminate stop closure duration ratio of 1:1.2 compared to most citations in the literature, including Al-Ani (1970), which give a 1:2 ratio for true geminates, although Al-Tamimi, Abu-Abbas, & Tarawnah (2010) found a 1:1.5 ratio for true geminates in Urban Jordanian Arabic). If fake and true geminates are acoustically different in Egyptian Arabic, what does this mean and what implications does it have for a theory of geminate weight? In order to investigate this question, an acoustic experiment was conducted which investigated phonetic length differences between word-final true and fake geminates in Egyptian Arabic. The results from two male speakers and two female speakers suggest that there may be a morphophonological difference underlying the phonetic instantiations of length, likely due to final consonant extrametricality, a process which operates in all varieties of Arabic and causes a word-final consonant, which would otherwise be assigned a mora and contribute, along with the nucleus, to a bimoraic, heavy, syllable, to be ignored, so that it does not contribute to syllable weight. In a similar way to consonant clusters, the phonology must ignore one of the consonants which compose a fake geminate when assigning stress, as fake geminates are phonetically shorter than true geminates in Egyptian Arabic.
© Copyright 2017 All Rights Reserved