Grammars for natural languages need to specify at least two types of procedure: word combination procedures, which build hierarchically organized structures, and procedures that establish dependencies between structural positions, for example by establishing grammatical agreement relations, or by moving an element from one position to another. Both procedures can generate their own complexity factors.
As far as the construction of structure is concerned, I'd like to illustrate its complexity by presenting some results from cartographic studies, a line of research that focuses on the fine details of syntactic structures by drawing structural maps that are as precise as possible. Through the cartographic magnifying glass, we can show that structures are far more complex and fleshed out than previously thought. Nevertheless, they are built by extremely simple generative mechanisms: it is the recursive operation of the system applying to a very rich and differentiated functional lexicon that determines the complexity of the structures (Rizzi and Cinque 2016).
With regard to the second type of complexity, I'd like to analyze it from the window offered by language acquisition. Children acquire their first language remarkably quickly and efficiently. Nevertheless, certain grammatical constructions remain difficult for the learner for a long time: the passive, certain questions, relative and cleaved questions concerning the direct object, and still other constructions, pose problems in comprehension and are avoided in production. What determines the difficulty of these structures for children? Recent work at the frontier between theoretical linguistics and developmental psycholinguistics has identified a systematic source of complexity in "intervention" structures, where an element with certain structural features is interposed between the positions to be connected (Friedmann et al. 2009).
I would like to conclude my presentation by discussing the close relationship between the cartographic complexity of structures and intervention effects in adult grammar and language development.
