Page - (000342) - in Knowledge and Networks

340 them. The proximity approach, proposed originally by Boschma (2005), empha- sizes that similarity (conceptually the inverse of heterogeneity)—or, as he calls it, proximity—affects the ease of knowledge transfer between actors. He thereby dif- ferentiates between various dimensions of proximity whose prominence can differ from one type of alliance to another. In R&D alliances explicitly conceived to gen- erate novel ideas and innovations, cognitive proximity might predominate over other forms of proximity as the basis for potential knowledge flows, and social proximity (also called the strength of social ties between collaborators) might take precedence as the control mechanism for knowledge flows. Understood as the similarity of knowledge bases, cognitive proximity can deter- mine the degree of knowledge exchange between actors through two central charac- teristics representing a trade-off in collective learning: mutual understanding and learning potential. Mutual understanding is the degree to which different actors comprehend each other, and it increases with cognitive proximity. Potential partners therefore need to exhibit some minimum degree of cognitive proximity to warrant mutual understanding.1 Learning potential has to do with the amount of what can be mutually learned, and it decreases with cognitive proximity. The heterogeneity of firms in knowledge space is a source of learning effects because relatively great dis- similarity can increase learning potential and the exchange of knowledge (Nooteboom, 2005). The idea of combining the two dimensions of cognitive proximity—that of being a condition for mutual understanding and that of being a source of knowledge exchange—suggests the existence of an intermediate degree of proximity at which beneficial exchange of knowledge is maximized (Boschma, 2005; Gilsing et al., 2008; Nooteboom, 1999). A deviation from this level will lead either to increased potential for exchanging knowledge combined with lowered common understand- ing or to increased common understanding combined with lowered potential for novelty. Consequently, an actor conducting a strategic and rational search for a research partner should, at least theoretically, try to connect with a candidate who is similar in knowledge stocks and who partly complements his or her own so as to acquire the potential for creating novelty. Besides the relevance of an optimal degree of cognitive proximity for under- standing and learning, the second condition for effective collaboration to take place is the controllability of the knowledge-exchange-and-sharing relation. It is here that social proximity comes in. Social proximity accounts for familiarity and trust between cooperation partners, two facets that facilitate the transfer of tacit knowl- edge and reduce the occurrence of opportunistic behavior. Trust affects the effi- ciency of knowledge transfer, for familiar and trusting partners have internalized 1 The concept of cognitive proximity is closely related to that of absorptive capacity (the ability to assimilate external knowledge). Absorptive capacity is largely a function of the extent to which the knowledge bases of collaboration partners are related (Boschma, 2005; Cantner & Meder, 2007; Cohen & Levinthal, 1990). A lack of absorptive capacities tends to result in a sharing of knowledge rather than in its exchange, for the partners are not able to integrate the external knowledge into their own knowledge stock. U. Cantner et al.