The syntactic nature of negative concord (NC) is still under debate. On the one hand, some theoretical accounts connect the emergence of NC to the high position of negation in sentence structure. On the other hand, some studies establish a relationship between NC and the internal syntactic properties of Negative Concord Items (NCI) in a language. In particular, it is still an open question whether NCIs in NC-systems behave like negative quantifiers (as German niemand) or Negative Polarity Items (NPI, as English anybody). The goal of the present project is to contribute to this debate by studying the acquisition of NC in Italian as a second language.
We investigate how the acquisition of NC in L2-Italian is affected by cross-linguistic influence as determined by different types of L1s by considering learners with L1-Spanish, L1-English and L1-German. These three L1s differ from each other with respect to the internal syntax of negative elements: Spanish exhibits a NC system like Italian except for minor differences. Our hypothesis is then that they are the quickest L2 learners. English has a non-negative concord system in which negation appears relatively low in the sentence structure, and features both negative quantifiers (which do not need to be licensed by an overt negation) and NPIs. German also has a non-negative concord system, with negation appearing in the same position as in English but no NPIs corresponding to English anything, anybody etc. German only has negative quantifiers in the contexts analyzed in this project. We expect the cross-linguistic differences between the three L1s and Italian to affect the paths and outcomes of L2 acquisition. More specifically, the complexity of the learning task experienced by the three groups of L2 speakers will provide an indication of the nature of NCIs in Italian and settle a theoretical debate that has been going on since the 90s. If Italian NCIs are more similar to NPIs in not being real negative elements, English speakers should be able to acquire them more easily than German speakers and show a similar acquisition path as Spanish speakers. This is due to the fact that their L1 English allows for NPI in addition to negative quantifiers. If, on the contrary, Italian NCIs are more similar to negative quantifiers, learners with German or English as L1 should exhibit the same acquisition path and crucially differ from learners with Spanish as L1. Our analysis will also consider the type of NCIs – NC-pronouns (like niente ‘nothing’) versus determiners in NC-phrases (e.g., nessun uomo ‘no man’) – and their syntactic function (direct vs. indirect objects) as additional factors modulating L2 acquisition. We expect that the acquisition of NC will be fully in place only once the whole NCIs has been mastered. The investigation presented in this project will pave the way to the next funding periods of the CRC, in which we plan to analyse two related topics. First, we will look at the acquisition of a non-negative concord system (like German) by speakers of NC languages or languages featuring NPIs. Second, we will analyse the acquisition of NC in L3 by learners whose L1 or L2 is NC or not. In this way, we will examine to what extent learners’ L1s and L2s are sources of transfer of NC-related properties to the L3.
