This dissertation addresses the problem of the syntactic expression of change of state and causation and, more specifically, of the causative-unaccusative alternation at the syntax-semantics interface. In chapter 1 I adopt a constructionist view on the organization of grammar, where semantic meaning is syntactically represented, decomposed in functional structure, and not projected from the lexicon. After syntax, vocabulary items are inserted to spell-out terminal nodes, or even bigger chunks of structure (late insertion). Thus, verb meaning is not atomic: it is decomposed as (sub)event structure, and thematic relations among arguments are compositionally obtained. The causative-unaccusative is a puzzle for lexicalist models, since there is no one-to-one correspondence between lexical meaning and syntax: una single verb projects two different structure, the transitive and the unaccusative one. Under a constructionist approach, however, the causative-unaccusative alternation –that is, one verb being used either as transitive or unaccusative– can be accounted for in terms of late insertion or, more concretely, in terms of conditions of spell-out. Syntax may produce either a transitive (causative) or an unaccusative structure, the only puzzle concerns the vocabulary items available in the language to spell-out those outputs. This is one of the reasons why the morphosyntactic expression of the alternation will be one of the main concerns of this study. The causative-unaccusative alternation has been widely studied for the last decades. Nevertheless, even if our current understanding of the phenomenon is deep, there are still many unsolved problems keeping the discussion alive. Furthermore, the overwhelming literature on this topic makes it necessary to untangle and isolate the core, recurring ideas that must be captured by any analysis of alternating verbs and, most importantly, it makes it necessary to refine some controversial concepts whose specific definition is crucial to develop a coherent theory. Therefore, the first aim of this dissertation is to offer the reader a comprehensive overview on the matter, compiling and reviewing those proposals that I consider essential to the discussion, with special attention to the ones on which my proposal is built...