Roberts and Roussou (2003) analyse grammaticalization within Minimalism, and Ledgeway (2011a, 2011b) deals with grammaticalization in Latin/Romance, also within Minimalism. Neither of them analyses the grammaticalization of KPs (case-markers) and so this is the theme of this paper. The grammaticalization of two very important Latin/Romance KPs (de marking genitive, ad marking dative) indeed conforms to both Robert & Roussou’s and Ledgeway’s hypotheses, since they originate from Latin PPs (de denoting separation, ad denoting direction), and within X’-theory complements (e.g. KPs) are ‘simpler’ than adjuncts (e.g. PPs) in that the former require fewer feature place-holders than the latter (Robert & Roussou (2003:106)), and so by Roberts & Roussou’s (2003:200-201) ‘structural simplification’ (reduction of ‘feature syncretisms’) PPs are grammaticalized as KPs. Robert & Roussou’s ‘structural simplification’ assumes configurationality and can only occur in configurational syntax, and so configurationality is a prerequisite for grammaticalization in Minimalism, which conforms to Ledgeway’s argument (2011a:405-434)) that the key syntactic change from Latin to Romance is the rise of configurationality, which gives rise to functional categories in Romance (Ledgeway (2011a:409)). Finally, as configurationality is a controversial notion, alternative scenarios are considered in the appendix where configurationality no longer has explanatory value, and ‘re-analysis’ is argued to be the key to understanding grammaticalization, since it is in itself sufficient to explain grammaticalization, with or without configurationality.
Simpson and Wu (2002) and Wu’s (2004) ‘lateral grammaticalization’ is a Minimalist analysis of Chinese de, which has been re-analysed from a determiner (D) to a verbal suffix (T). Roberts and Roussou (2003) and van Gelderen (2011) deal with grammaticalization within Minimalism, though neither take ‘lateral grammaticalization’ into account (Vincent and Borjars (2010:293)). A comparison between these accounts reveals that Roberts and Roussou’s (2003) and van Gelderen’s (2011) ‘feature economy’ also accounts for the cross-linguistic distribution of ‘lateral grammaticalization’, which is the main theoretical thrust of their accounts (Roberts and Roussou (2003:2-7), van Gelderen (2011:4-17)). However, the lack of ‘upward feature analysis’ (Roberts and Roussou (2003:200)) in ‘lateral grammaticalization’ sets it formally apart from grammaticalization, and this ties in empirically with the lack of ‘phonological weakening’, ‘univerbation’ and ‘semantic bleaching’ in ‘lateral grammaticalization’ when these are the diagnostic traits of grammaticalization (Campbell (2001), Roberts and Roussou (2003:224-232)). All this entails some significant revisions to Minimalism as a model for grammaticalization and ‘lateral grammaticalization’, the former of which involves an upward shift of features while the latter involves a re-analysis of features from pragmatics.
