Clitic Particles and the Typology of 2P Languages

This paper was prepared after the Particle Workshop (Cambridge, 30-31 October 2008) for the prospective conference volume. I am grateful to the audience of the Particle Workshop in Cambrigde, October 30-31, 2008 and personally to Anders Holmberg, Anna Roussou and Sten Vikner for the valuable comments and criticism. I am also grateful to Natalia Serdobolskaya for the consultation concerning Dargwa examples in section 1. I am indebted to Bostjan Dvořák for the valuable comments on Slovene clitic fronting. Finally, I would like to thank Tomas Riad for alerting me about the relevance of Agbayani & Golston’s paper. All responsibility for the statements made in this paper is my own. This paper is devoted to the interactions of particle typology and clitic typology. I assume that the notion of 'particle' belongs to the parts-of-speech classification and define particles as function words with discourse or connective functions which are not specified as conjunctions or Ps. The notion of 'clitics' comes from a classification of word forms: clitics are deficient expressions lacking features of phonological words and taking an intermediate position on the way from autonomous word forms to affixes. The class of clitics includes pronouns, auxiliaries and particles. Particles may be either clitic-like or behave as phonological words. Clitic particles are prosodically deficient discourse or connective markers. I am focusing on the distribution of clitic particles. The paper is divided into two parts consisting of five sections each. In the first part I am assuming a lexicalist approach towards defining a class of discourse markers. In the second part I am working out a syntactic approach to clitic particles in 2P-languages, where clitic clusters occur clause-internally. I argue that 2P-languages constitute a syntactic type and share a number of type-specific constraints; I retrieve typological implications from the distribution of clitic particles and principles of clitic ordering in some 2P-languages.本文系2008年10月30日至31日于剑桥举办的小品词研讨会（Particle Workshop）后，为拟出版的会议论文集撰写。感谢2008年10月30日至31日剑桥小品词研讨会的全体听众，尤其感谢Anders Holmberg、Anna Roussou与Sten Vikner提出的宝贵意见与批评指正。此外，感谢Natalia Serdobolskaya就第1节中的达尔瓜语（Dargwa）示例提供的咨询协助。谨此感谢Bostjan Dvořák就斯洛文尼亚语（Slovene）附着词前置现象提出的宝贵意见。最后，感谢Tomas Riad提醒我关注Agbayani与Golston的论文的相关参考价值。本文所述观点的全部责任由作者本人承担。 本文聚焦小品词（particle）类型学与附着词（clitic）类型学之间的交互关系。本文认为，“小品词”这一概念隶属于词类分类体系，将其定义为具有话语功能或连接功能，但未被归类为连词或介词（原文缩写为Ps）的功能词。“附着词”的概念则源自词形分类体系：附着词是存在形式缺陷的表达式，缺乏音系词的特征，且在独立词形到词缀的连续统中处于中间位置。附着词范畴涵盖代词、助动词与小品词。小品词既可能具有附着词属性，也可作为独立的音系词存在。附着式小品词是存在音系缺陷的话语或连接标记。本文重点探讨附着式小品词的分布特征。本文分为两个部分，各含5个小节。第一部分采用词汇主义路径，对一类话语标记词进行界定。第二部分针对2P语言中的附着式小品词构建句法分析框架，此类语言的附着词组均出现在从句内部。本文主张，2P语言构成一类句法类型，且共享若干类型专属限制条件；本文还从部分2P语言的附着式小品词分布特征与附着词排序原则中，推导出类型学层面的蕴含关系。