Section 2 is devoted to the introduction of data. I point out that -iva and -Vj(V) are complex items with -i- and -j- having an unclear morphological status. Based on hypotheses proposed in Garde (1972), Coats (1974) and Feinberg (1980), I argue in Section 3 that the occurrence of -i- and -j- can be predicted from morphological organization. More specifically, focusing on the underlying representation of -iva, I suggest that -i- is conditioned by the doubling of a categorial head. In Section 4, I make a proposal concerning the derivation of -i- and -j-: I assume an identity avoidance constraint banning repetition of a node without a linking element. Accordingly, I propose that -i- and -j- are realizations of an expletive root that licenses such sequences of identical nodes.
Bogdan Szymanek Introduction To Morphological Analysis.pdf TOP
2ff7e9595c
Commentaires