LingLunch : Chris Reintges

Chris Reintges (LLF) en collaboration avec Edith Doron (Hebrew University) et Caterina Donati (LLF)
The parametric syntax of participial modifiers

Participles have been studied intensively in traditional grammar and generative syntax alike. However, as of yet the categorial status and internal syntax of these adnominal modifiers is not very well understood. Due to their ambivalent nominal and verbal behavior, some researchers treat them as adjectives, while others analyze them as reduced relative clauses, but there are problems with each of these views. Part of the problem is that the different types of participial modifiers do not correspond to a single syntactic configuration, but rather to three distinct ones that differ with respect to the amount of functional superstructure being involved and the correlated property of licensing nominal arguments.

Crosslinguistically, participial modifiers differ along two dimensions, namely (i) whether participles can express tense distinctions alongside with aspect, mood, and grammatical voice (active, passive, middle) or rather assume an invariant form, and (ii) whether positions other than the subject can be relativized on or not.The two dimensions of variation can be related to two parameters, which operate relatively independently from each other. The presence of tense inflection on the participle itself depends on the internal organization of verbal conjugational paradigms. Languages in which tense and aspect morphology is detachable from the person, number (and possibly gender) marking agreement inflection allow tense morphemes to surface on both finite verbs and participles. This option is excluded in languages in which the Tense Detachability Parameter is set negatively and tense inflection and subject agreement are fused into an unsegmentable morph.

When a non-subject position is relativized, the clausal modifier contains within its extended verbal projection an overt subject. Consequently, the participle enters into an agreement relation with the subject and a concord relation with the modified noun. Agreement and concord share the features of number and gender. These features must be morphologically realized in those language that have the appropriate morphology to begin with, but they can only be realized once, either as part of agreement or as part of concord. The possibility of reconciling concord with agreement therefore hinges on the possibility of the language to relax the conditions for the morphological expression of number and gender features either in participle subject agreement or in participle head noun concord.