On Word, Clitic and Affix Distinctions in Some North Eastern Italian Dialects

The research into the systematic correspondences between speech acts and grammatical forms, involving different levels of analysis (syntax, morphology, lexicon) and having a unifying explanation in function, has highlighted a series of contradictions in the categorisation of Question Markers (QMs), belonging to many Northern and central Italian dialects (Fava 1993, 1997). These peculiar illocutionary devices involve complex distinctions of gender (m., f.), number (s., pl.) and person (1st p., 2nd p., etc.). Attested as early as the sixteenth century, they are still very much alive and characterise both yes/no questions and WH question. Diachronically, QMs, as well as non-inverted Subject Clitics (SCs), derive from the non clitic nominative forms of ancient mediaeval dialects, continuing the forms of the Latin nominative (Renzi & Vanelli, 1984). This well-known diachronic relation between QMs and SCs is clearly perceived in (1), where the homophonous SC la and QM la are both derived from Latin ILLA. Synchronically, the grammatical descriptions of QMs are rather controversial both in pre-theoretical and in theoretical terms, oscillating between a consideration either as an interrogative conjugation (and sometimes as an interrogative mood or even particle) or (clitic or full) pronoun inversion. Moreover, the grammatical representations of pronominal inversion from a pre-verbal to a post-verbal position, as in (1) and (2), presuppose the synchronic identity of the SC in a and the QM in b and c in some proposal (Renzi & Vanelli 1984; Rizzi 1986), but not in others (Poletto 1993). The problems raised by the oscillation and contradictoriness in the categorisation of this single illocutionary force device, from affix to NP, do not simply reflect terminological problems, but lie in the theory of grammar. Affix (Rohlfs 1969; Fava 1993) > Clitic in INFL Position (Brandi & Cordin, 1989; Rizzi 1986) > Inverted Subject Clitic Pronouns (Renzi & Vanelli 1984) > NPs in Spec Subject Position (Poletto 1993) Reconsidering the proposals for some Northern Eastern Dialects (NED), the affix characterisation is defended. Central Veneto, Pagotto and Trentino QM varieties display idiosyncratic inflectional patterns that some authors have taken as evidence for syntactic inversion despite the absence of genuinely syntactic arguments for such a restructuring. Even though diachronically derived via the enclisis of Subject pronouns, QMs now belong, at a synchronic level, to the verbal form as a whole.