We are being told that ‘we are who we know’, and convincing research highlights the effects of our relations and network embeddedness on health, opinions and almost all elements of our social behavior. Following this Social Network Analysis (SNA) perspective, it is relevant and necessary to ask how this embeddedness affects our identities, and is it even relevant to talk about an identity without taking a relational perspective?”

The most valuable knowledge is found in the weak ties on the outskirts of the knowledge worker's social network (Levin & Cross, 2004). Yet according to Wenger (1998), people find meaning and build identity and share knowledge within the close, long term communities of practice in which they participate; i.e. within the strong ties in their network. A paradox emerges for the knowledge organization, as the most valuable knowledge will potentially be created by combining expertise from across the organization in virtual teams of individuals with weak relational ties brought together through the use of information and communication technology (ICT); yet Wenger’s community of practice theories suggest that effective learning, and hence knowledge sharing, happens in close knit communities where a shared domain of interest brings people together in a process of participation and reification of a common practice. This process ultimately constitutes a mutual engagement in ongoing exchange and co-construction of knowledge. Wenger argues that such negotiated experience and community membership are the foundation of human identity, and that such identity is signaled in the community by a shared repertoire, i.e. community specific jargon or terminology, clearly signaling who is in and who is out. Neither a shared domain, a common practice, nor a shared repertoire are present in weak ties, thus emphasizing the challenge involved in building effective collaboration across the divisional, geographic and temporal borders of a global organization. Furthermore, ICT itself is often considered as a barrier to building trust and identity (e.g.Gannon-Leary & Fontainha, 2007). This presentation thus provides a theoretical perspective through which to understand how knowledge workers’ identities are being challenged by the increasing use of virtual teams in global organizations. Findings from an analysis of a corpus of case study research suggests that a corporate culture building on openness and flat hierarchies, in addition to an awareness of how ICT potentially shapes relations, may be ways of overcoming this paradox in knowledge worker identity.

Social network analysis can be characterized as a theoretical approach, an overall methodological approach and a hands-on tool to map and analyze networks (Scott, 2000).

One could argue that the nature of such social network analysis is instrumental and utilitarian, since the main purpose is to codify and measure relations and be able to explain x, or vice versa. In this sense the roots of SNA via a focus on methodological features can be traced back to Hobbes and in particular Mill and their approach to social relations. In his theory of recognition and the master-slave terminology, Hegel introduces a different conception of social relation. If one bases SNA on a Hegelian tradition, a social relation is not as dyadic as most traditional SNA would argue, but to a much bigger extent relying on the (historical) context of the social interaction.

After a number of years of constant growth and spread of SNA into a vast number of fields (see Freeman, 2004 for a review) some criticism has risen from within the ranks of social network analysts regarding e.g. ethics (Kadushin, 2005), and methodological commensuration (Kilduff et al., 2006; Zuckerman, 2010).

The aim of the present study is to root social network analysis in the history of philosophy. The methodological implications that can be drawn from this history of philosophy will be discussed, e.g. that SNA to a larger extent should pay attention to the individuals (as argued by Kilduff et al., 1994) and in particular include the institutional context (as argued by Owen-Smith et al., 2008).