Simpson and Wu (2002a) analyse Chinese shi-de constructions and propose a new type of grammaticalization in the Minimalist framework called ‘lateral’ grammaticalization, which conforms to Roberts and Roussou (2003) and van Gelderen’s (2004a, 2011) Minimalist analysis of grammaticalization in that it displays ‘structural simplification’ but differs from it in that it does not show Roberts and Roussou’s (2003:200) ‘upward feature analysis’, since it consists of a ‘lateral’ reanalysis from one functional category (D) to another (T). This entails empirical differences, namely the lack of phonological, morphological and semantic weakening in the grammaticalizing elements (Chinese shi and de) when it is traditionally assumed that weakening is a diagnostic trait of grammaticalization and is entailed by the rise in frequency of the grammaticalizing element (Bybee (2003, 2011)). This dissertation proposes a direct comparison between ‘standard’ grammaticalization (Roberts and Roussou (2003), van Gelderen (2004, 2011), e.g. the Romance future (Latin habere)) and ‘lateral’ grammaticalization (Simpson and Wu (2002a), Wu (2004), e.g. Chinese shi and de) and argues that the different formal properties (‘upward feature analysis’/’lateral feature analysis’) entails differences in collostructional frequencies (cf Stefanowitsch and Gries (2003, 2004)) which can be correlated with the differences in morphophonological weakening in grammaticalization. The traditional assumption that functional categories are necessarily weak (see e.g. Roberts and Roussou (2003:217-229)) is hence deemed as simplistic (if not mistaken), and a new mechanism in generative syntax-phonology interface is proposed where the empirical properties of functional categories are argued to be derivable from their frequencies in grammaticalization, which will be known as ‘Functional Spell-Out’.
展开▼
机译:Simpson和Wu(2002a)分析了中国的释义结构,并在极简主义框架中提出了一种新型的语法化,称为“横向”语法化,与Roberts and Roussou(2003)和van Gelderen(2004a,2011)的极简主义分析相符。因为它显示了“结构简化”,但与之不同之处在于它没有显示罗伯茨和鲁索(2003:200)的“向上特征分析”,因为它包括从一个功能类别(D)到另一个功能类别的“横向”重新分析(T)。这必然存在经验上的差异,即当传统上认为弱化是语法化的诊断特征并且由语法化要素的频率升高而引起时,语法化要素(汉语shi和de)缺少语音,形态和语义的削弱。 (Bybee(2003,2011))。本文提出了“标准”语法化(Roberts and Roussou(2003),van Gelderen(2004,2011),例如浪漫未来(拉丁哈贝))和“横向”语法化(Simpson and Wu(2002a),Wu)的直接比较。 (2004),例如中国的shi and de),并认为不同的形式属性(“向上特征分析” /“侧向特征分析”)带来了共构频率的差异(cf Stefanowitsch and Gries(2003,2004)),这可以相互关联。在语法化方面,形态学上的弱点有所不同。因此,传统的假设认为功能类别一定是弱的(例如参见Roberts and Roussou(2003:217-229)),这被认为是简单化的(如果没有记错的话),并且提出了一种在生成语法-语音学接口中的新机制,其中经验性在语法化中,功能类别的频率被认为是从它们的频率派生的,这将被称为“功能性拼出”。
展开▼