Clitics and affixes in some north-eastern italian dialects

Research on interrogative structures has been receiving increased attention in the framework of speech act theory. However, the description of the cluster of morphosyntactic features (clitics, particles, moods, word order variation, etc.) which, without any reliance on the extralinguistic context, permits the interpretation of an utterance as a question is very often unsatisfactory. In some cases, little theoretical consideration has been paid to the series of variations concerning the categories and strategies involved. In this paper I highlight a series of contradictions in the categorization of the Question Markers belonging to many North-Eastern Italian dialects and I propose a solution to this puzzling problem (Fava 1993, 1999). Already attested since the sixteenth century, North-Eastern Dialect Question Markers (NED QMs) are still very much alive, involving distinctions of gender, number and person which vary considerably from dialect to dialect, according to tenses, moods and other features. They have been described in different ways as noun phrase (NP) or clitic inversion, while others consider them variations in the morphological structure of the verb (affix of an interrogative conjugation or mood). These variations and contradictions in the categorization of this specific illocutionary device, from affix, to clitic inversion or to NP inversion deserve a theoretical approach that takes into account the different weight given to different levels of empirical generalizations. The aim of this paper is to defend the methodological importance of semantics and pragmatics research: starting from meanings and functions rather than forms involves a change in perspective in the evaluation of data even for independently observed phenomena, which forces us to unify them in order to offer a coherent grammatical description of them. Reconsidering some of the major proposals, I will try to bring to light some of their common features and pinpoint those properties that have a structural relevance which require reconsideration. Evaluations centred on the word, in its paradigmatic relations, support the thesis of the affixal character of the NED QMs, offering some evidence that the declarative and the interrogative paradigms have split up. Even though derived diachronically via enclisis of subject pronouns, NED QMs now belong, at a synchronic level, to the verbal form as a whole.