TWO DIALECTS OF NGANASAN: WHAT THEY TELL US ABOUT PROSODY
The Problem The equivalence of long voweled syllables (CVV) and closed syllables (CVC), as opposed to open syllables containing short vowels (CV), is found in many languages under a variety of circumstances. Traditionally, this has been seen as a difference in syllable weight. CVV and CVC are heavy,...
| Main Author: | |
|---|---|
| Other Authors: | |
| Format: | Text |
| Language: | English |
| Subjects: | |
| Online Access: | http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.638.997 http://www.meertens.knaw.nl/glow2002/vaysman.pdf |
| Summary: | The Problem The equivalence of long voweled syllables (CVV) and closed syllables (CVC), as opposed to open syllables containing short vowels (CV), is found in many languages under a variety of circumstances. Traditionally, this has been seen as a difference in syllable weight. CVV and CVC are heavy, and CV are light. It has also been shown that in languages with a CV/CVV distinction, CVC syllables do not always pattern with CVV, but may count as light and pattern with CV (e.g. Zec 1988). Under Moraic Theory (Hyman 1985, Prince 1976, 1983, Hayes 1989, among others), the equivalence of CVV and CVC has been captured via bimoraicity. In contrast, in languages where closed syllables pattern with CV, both of these syllables are monomoraic. Recently, there has been several proposals (Rosenthall and van der Hulst 1999, Moren 1998) that closed syllables may vary in weight within one language depending on context, notably constraints on stress placement. According to this approach, contextually-dependent weight is a consequence of constraint interaction that determines the moraicity of coda consonants. This paper focuses on the two theories as applied to prosody-sensitive phenomena in a Uralic Samoyedic language, Nganasan. I present data from two close dialects of the language, Avam and Vadey, and focus on the difference in their treatement of consonant gradation. I argue that while the variable weight-type analysis is necessary, it is insufficient to account for all the relevant Nganasan data from one of the dialects. My proposal can be seen as an extension of the variable weight analysis. Data Source The data in this paper is partly taken from source grammars (Helimsky 1998, Tereshenko 1979, Prokofjev 1937) and subsequently checked with native speakers, and partly comes from field work |
|---|