درک ایجاد اخبار در اکتشاف شبکه های ساختاری و رابطه ای مشترک در نظر گرفته شده

In exploration networks the key-organisational question is not how to organise a division of labour but instead how to create novelty. The aim of this paper is to develop an understanding of how such novelty in exploration networks is created. Based on an empirical analysis of the multimedia and biotechnology industries in the Netherlands, this paper shows that exploration networks face a trade-off between diversity and selection. Moreover, the findings indicate that depending on the type of exploration task, exploration networks need to make a combination of density and tie strength in such a way that diversity and selection are aligned. The paper concludes, among others, that the views of Burt, Coleman and Granovetter should not be seen as contradictory, but rather as proponents of complementary views.

The aim of this paper is to develop an understanding of how novelty is created in exploration networks. The focus of this paper is therefore on interfirm networks that engage in the exploration of novel combinations. We define exploration as a situation that can generally be characterised by breaking with an existing dominant design and a shift away from existing rules, norms, routines or activities, in view of novel combinations. Exploration is an inherent uncertain process that can hardly be planned for (March, 1991). There is growing consensus in the academic literature that in such a setting, strategic alliances can be an extremely effective organisational form as they bring together complementary actors from different technological backgrounds. In the context of learning and innovation, exploration thrives on a diversity of knowledge which yields a potential for Schumpeterian novel combinations to emerge (Nooteboom, 2000). It is in this heterogeneity that actors are able to combine and integrate complementary knowledge and capabilities (Porter, 1990; Hamel and Prahalad, 1994; Grabher, 1993; Hagedoorn, 1993; Hagedoorn and Schakenraad, 1994; Smith Ring and van de Ven, 1994; Grandori, 1997; Spekman et al., 1995; Uzzi, 1997; Nooteboom, 1999 and Nooteboom, 2004; Ahuja, 2000; Rowley et al., 2000). In spite of its noted importance, novelty creation forms an unaddressed topic in the innovation literature. In this literature, a distinction is made, following life-cycle theory, between two stages in the innovation process. The first initial stage is one of volatility, characterised by the creation of Schumpeterian novel combinations. The second stage is a stage of consolidation in which dominant designs emerge (Abernathy, 1978; Abernathy and Utterback, 1978; Abernathy and Clark, 1985) and in which production systems focus on efficiency, economies of scale and experience. The cycle is generally held to imply a shift from product to process innovations, as product forms settle down and competitive pressure shifts to efficient production. In other words, the literature is very clear about the change in focus from exploration to exploitation. However, the origins of exploration itself remain a mystery. The question of how exploitation ‘feeds’ future exploration is left unanswered (Nooteboom, 2000). This issue on ‘the origins of novelty’ forms the central question in this paper and will be studied in the context of interfirm networks engaging in exploration, further referred to as ‘exploration networks’. This focus on exploration networks is related to the literature on innovation systems (Carlsson and Stankiewicz, 1991; Malerba, 2004) and regional economics (e.g. Bathelt et al., 2004). Although in this literature the importance of relations and interaction among heterogeneous firms and actors is clearly acknowledged (Nelson and Winter, 1982; Dosi et al., 1988; Nieto and Santamaria, 2007), an in-depth understanding of the structure and role of interfirm networks is still underdeveloped (Pavitt, 2002; Malerba, 2004). For a deeper understanding of networks we need to turn to the social network literature. Social network theory enables us to describe and measure networks of relationships in great detail, providing us with several well-developed measures and techniques to assess the structure, ties and dynamics of relational networks.

The aim of this paper is to develop an understanding of how novelty is created in exploration networks. Our qualitative and tentative analysis of exploration networks in the multimedia and pharmaceutical biotechnology in the Netherlands indicates evidence that the combination of peripheral non-density through weak ties and a dense core through strong ties enables the alignment of diversity and selection mechanisms, needed for the creation of novelty. Moreover, the business exploration network in multimedia provided an illustration of our third hypothesis that abundant variety created through non-redundant ties may pre-empt selection with a negative effect on novelty creation. This implies that there is a trade-off between non-redundancy in networks on the one hand, in order to access cognitive variety, and redundancy in networks on the other hand, for triangulation and absorption. In other words, brokerage enables to access sources of potential added value, conform Burt's argument, network closure is however critical to capturing this value, conform Coleman's argument. Rephrased in Granovetter's terminology, weak ties cross social group boundaries, whereas absorption and triangulation are done by strong ties (strong in terms of frequency, openness not necessarily duration) within a group. So, when understanding novelty creation in exploration networks Burt, Coleman and Granovetter need not to be seen as contradictory, but rather as proponents of complementary views. Management should therefore try to build an optimal portfolio of alliances, featuring weak ties for the creation of novel combinations and strong ties that enable those companies to validate and assess the newly acquired knowledge. In contrast to our expectations were the findings on exploration of a systemic technology, as in multimedia. This required an overall dense, redundant network of ties that are strong in terms of frequency, openness but show low strength in terms of duration. As indicated, in this network no clear distinction could be made between a core and a periphery nor between strong and weak ties. It entailed a type of novelty creation that mirrored the more radical nature of exploring a systemic knowledge base, leading to the destruction of existing and creation of new technological architectures. As already mentioned, this points to a new insight. Apparently, novelty creation in exploration networks can be done in two ways. One is through the ‘classical way’ according to Granovetter's ‘strength of the weak tie’ argument. This is what we found in the peripheral network in network 1 in biotechnology, with its focus on technological exploration of a stand-alone technology. As analysed, this was done through a stable overall configuration in which novelty originates from new combinations of ties and where selection is endogenous to the network through a dense core of strong ties. This is in line with our theory and hypotheses. For management this requires the need to search for distinct partners that have unique capabilities or technological resources. Novelty creation in this respect is facilitated by teaming up with companies that are relatively far away in the network and that are dissimilar from the focal firm. Dissimilarity seems to breed novelty, albeit recent literature has shown that too much dissimilarity is decreasing the absorptive capacity of firms and therefore the ability to create novelty ( Hamel, 1991; Lane and Lubatkin, 1998; Mowery et al., 1996; Fleming and Sorenson, 2001). An interesting new insight, as it emerges from our empirical analysis, is that there is also another way of creating novelty, namely through a dense network in combination with a high volatility (entry/exit) of ties. This may be associated with a more radical level of exploration in which novelty originates from novel configurations of ties in combination with exogenous industry-level selection mechanisms. This implies the need for the creation of fast-to-build flexible alliances which are used to create a radar function that continuously monitors new windows of opportunities for the focal company (see Duysters and de Man, 2003). Firms should therefore try to make the most of the inherent flexibility and speed of non-equity alliances with a large number of partners at the same time. This allows companies to make use of the specific know how and competences of various individual partnerships rather than engaging in a few broad ranging partnerships with one specific partner. Our third case of business exploration in multimedia indicates that if both endogenous selection mechanism and exogenous selection mechanism are absent, the network falls prey to chaos and becomes ineffective in the creation of novelty (cf. Hypothesis 3). In sum, the conclusion is that a network structure should allow for the alignment of selection and diversity mechanisms in view of novelty creation. We find that this is the case for network 1 in pharmaceutical biotechnology. Although we do not find this in a similar way for network 1 in multimedia, the underlying logic is basically the same: selection and diversity mechanisms should be aligned. Here, an important new insight is that if the network structure does not allow for an endogenous selection mechanism, this can still be accomplished by exogenous selection mechanisms at the industry level. If no selection mechanisms are present—endogenous nor exogenous—there is no alignment with corresponding diversity and the network fails in its creation of novelty (network 2 in multimedia). Overall, the way in which novelty is created differs along the level of exploration, with consequences for aligning diversity and selection mechanisms and hence for the optimality of network structural properties. So, how structural and relational embeddedness in exploration networks (density and tie strength) combine in an optimal way, in view of novelty creation, is strongly conditioned by the nature of the knowledge base that is being explored. This is not only a far cry away from the underdeveloped notions on networks by innovation scholars but also from the universalistic tone of social network theorists.