Collective learning in a milieu approach: Conceptual elements and empirical evidence from Italy

The aim of the paper is twofold: a) it stresses the concept of collective learning from a theoretical point of view, and provides new insights into its definition; b) it provides a measurement of the presence of collective learning in three Italian milieux. An analysis of the definitions provided so far in the litterature on collective learning shows in fact an ambiguity between these two concepts. A parallel analysis of the two concepts is run, and similarities and differences underlined. One of the main distinguishing features of collective learning is embedded in the element of "club externality", while the elements of "continuity" and "dynamic synergies" represent the elements of similarities with the concept of learning. The paper explains these concepts in depth, and once the "public" nature of collective learning is emphasised, it moves into the problem of whether the public nature guarantees its exploitation. To reply to this question, the paper moves into the problem of which local spatial preconditions have to be present in a territorial setting to exploit collective learning. In this respect, the paper provides a logic framework of preconditions for a territorial system to move from specialised areas, through to "local districts", "milieux" and "milieux innovateurs", and provides a distinction between the concept of "milieu" and that of a "milieu innovateur", which is so far remained ambigous. The logic framework allows in fact to distinguish in a clearcut way among all these concepts. These reflections lead to some interesting empirical questions which are investigated in the empirical part of the paper. In particular, the empirical analysis replies to the following questions: ? is it true that collective learning is not the result of a cooperative behaviour, but of a collective process; ? if it is true that collective learning is the way of achieving new creative resources for SMEs in local areas. The empirical analysis is based on three Italian high-tech milieux. Descriptive and interpretative statistical instruments are used in this part of the paper.