Alsina, Alex and Boban Arsenijevic. 2012b. Miriam Butt – Tracy Holloway King (note), competing hierarchies and generalizations in the Serbo-crocokroatic hybrid agreement. Proceedings of the LFG12 Conference, StanfordCSLI Publications web.stanford.edu/group/cslipublications/cslipublications/LFG/17/papers/lfg12alsinaarsenijevic.pdf (released February 17, 2016). In the next section, we will detail our analysis as well as a discussion of the mechanism of pluralistic matching of the hybrid subtantifs that we accept, which plays a fairly central role in our account. A number of analyses were presented to explain these differences. They can be divided into two groups: (i) analyses in which RRCs and NRCS have the same syntax overt, with differences only for the SAU (Demirdache 1991). Kayne 1994; Bianchi 2000) and (ii) analyses that postulate the differences between RRCs and NRCs in their syntax overt (Jackendoff 1977; Safir 1986; De Vries 2002; 2006; Resi 2011) e.a.). 8An evaluator asks whether the pronoun ona in (14b) could actually be plural neutered instead of a female singular (both forms being syncretist in the nominative case). On the advice of the expert, we tested the form of sentences in which the pronoun appears with different case marks, as indicated in (i), and our informants deemed them well trained. We assume that the zero-definitive pronoun present in the structure of NRCs behaves like any other pronoun, i.e. it corresponds to the head-nome in morpho-syntaxic characteristics (EMFs. Sg.
in the case of the collective nouns concerned), it also reflects characteristics that correspond to the semantic characteristics of the speaker (MASC.PL.). The PC, which introduces the free relative component in the analysis of the Vries NRC, may aim for an agreement in both functions. As a result, the PC appears in the form that corresponds either to the morpho-syntaxic characteristics of the head-nomen (syntactic arrangement) or to the semantic characteristics of the head-nomen (semantic chord). Thus, although there are three phi-features in the structure of the NRCs: (the name, the plural pronoun without sex and the definitive pronoun), the jenider that must be agreed by rp is the definitive pronoun that is the next goal. This is what emerges from Figure 2, which is a simplified version of De Vries` RNA analysis (2006) (2006) (the objective of the rp agreement is circled). The crude generalization seems to be that internal DP, the concordance is dictated by the morphological characteristics of the name, while DP-external, it is determined by semantic characteristics, because the noun braa `brothers` and the noun d (j)eca `children` are semantically plural (male and neutered).