東河賽夏語「並列」和「伴同」標記ki-

ki as a Marker of Coordination and Comitativity in Saisiyat

賽夏語為台灣南島語言之一，分布於台灣西北的新竹及苗栗縣境內，主要分為大隘和東河兩個方言。本文為正在進行中的東河賽夏語構詞研究之部分成果（請見Zeitoun et al. in preparation）。我們回應之前研究的分析（見Starosta 1974, Li 1978[2004], Yeh 1991），認為ki 應分析為伴同格位標記，並且進一步比較伴同格ki「和、跟」和並列標記 =o「跟、還有」的句法分布、用法和功能。根據人稱專有名詞的特徵，我們證明 ki 可標示複數 (ki-l)，=o卻不能 (*=o-l)。依以上發現，我們認為有必要修正賽夏語的格位標記系統，同時也必須建立新的代名詞組（也就是所謂的伴同格代名詞）。另外，我們也認同李壬癸教授 (Li 1978[2004:377]) 的假設，認為伴同ki 也可以呈現並列的功能。我們根據不同的測試（如詞序、否定詞、動詞分類等），分別出ki身為連結詞和伴同的不同用法。

The present paper is part of on-going research into the morphology of Tungho Saisiyat (see Zeitoun et al. in preparation), a Formosan language spoken in Miaoli County, northwestern Taiwan, and focuses on coordination and comitativity. We first compare the syntactic distribution, use, and functions of the comitative ki marker ‘and, with’ as opposed to the coordinator =o ‘and, as well’ by demonstrating that ki should be treated as a comitative case marker, as shown in previous analyses (cf. Starosta 1974, Li 1978[2004], Yeh 1991, and Hsieh 2009). We go a step further, however, by demonstrating that when followed by personal nouns, ki can be marked as plural (cf. ki-l) while =o cannot (cf. *=o-l). Based on these findings, we argue that a new set of pronouns—the “comitative” ki-set—should be recognized in Saisiyat, while the case marking system of this language needs to be further re-assessed. At the same time, we show that ki can also function as a coordinator, as first hypothesized by Li (1978[2004:377]). We provide a number of syntactic tests (word order, negation, and verb classification) to distinguish the conjunctive and comitative usages of ki.