The handsome man
discussing freight from Missassagua
is loud.

He interrupts

the gentle creaking of the luggage bins,
the sweet steel-on-steel-swoosh-rumble.
The train cries: “I’m here, I’m here!”

He doesn’t hear;

nor see the gold glow broken
by streaking shadows of swallows diving down down between rows of corn
miniature from lack of rain.

What will his barge carry?

Containers of cream from the jerseys out to pasture?
Red barns filled with sweet hay?
A mountain of Queen-Anne’s Lace?
Blue flossy clouds painstakingly piled?

A hawk on a wire calls my attention, but

Covetous Voice

quickly restakes his claim.
Shop talk marbles plunk into blue upholstery behind my head,
together with certainty
we plunge predictably forward.

… inspired by responses to a series of haphazard library bookstack photos I took last month and a timely note from Mr. Tan.

This lit review (of sorts) was written as the final paper for a course on Urban Social Networks taught by Zak Neal in the Soc department at MSU (after this class, I asked Zak to join my dissertation committee and he agreed. Hooray!).

It teases out some of the ideas I’ve been trying to think through to relate this strange ambiguous concept of “social entrepreneurship” with social networks and social movements. Specifically, I ask how scholars of entrepreneurship versus scholars of social movements think about network capital, or  patterns of relationships. In the entrepreneurship context, social capital accrues to and benefits the individual. In the movement context, social capital accrues to and benefits the whole (even if sometimes the research methods focus on an individual organization as the unit of analysis).

What does this have to tell us about how networks of entrepreneurs who are interested in broad-scale social change (aka social entrepreneurs?) might function?

This is particularly interesting in the context of some of the practical work I’ve been doing recently on worker-owned cooperatives and other cooperative ownership models, and all of the deep thinking happening in Detroit around the commons… what are forms of network capital or “social architecture” as Marjorie Kelly calls it, that will replace current modes of organizing and organizations?

_____________________________________________________________

“As social entrepreneurs, we seek a balance between nurturing our individual enterprises, and understanding and acting according to our place within broader social and environmental systems.”

– FoodLab Detroit Charter, August 2011

FoodLab Detroit is a young organization that supports food-based social entrepreneurs[1]. FoodLab calls itself a network in the vein of other business networks that connect entrepreneurs to one another and to external stakeholders including small business development agencies, policy makers, and customers. In this sense, the network functions as a business incubator, connecting individual businesses with information, opportunities, and resources. At the same time, FoodLab also participates in Detroit’s “good food movement,” creating space for members to deepen awareness of problems within Detroit’s food system and encouraging and enabling collective action to move towards a more just, sustainable, and resilient alternative.

Social entrepreneurship is a relatively young field of inquiry which is still in the process of developing consistent definitions and terminology and lacks a history of empirical studies (Mair & Marti, 2006; Short, Moss & Lumpkin, 2009). As social enterprise grows in popularity as a strategy for social change (Deresiewicz 2011; Dees & Anderson 2006), national groups and local organizations like FoodLab are learning how to support this new breed of change agent[2]. Studies of social entrepreneurs suggest that the sum of their relationships, or social capital, is critical to emergence and persistence (Van Ryzin et. al. 2009; Sharir 2006). Others suggest that social capital is an outcome of social enterprise and an enterprise should be defined and measured by its capacity to build social capital within its sphere of influence (Thompson 2002; Smallbone 2001). Whether we consider it an input or outcome of social enterprise, what social capital actually looks like, and the process by which it is created and has an effect, is something of a black box. What is the role of relationships or social capital in social entrepreneurship? How does it (or does it) differ from the role and structure of relationships in conventional entrepreneurship or social activism? What does this tell us about the nature of social entrepreneurship?

Here, network analysis approach can help specify exactly what kind of ties that we mean when we talk about social capital (more friends? more advice? more funders? more customers?). Perhaps more significantly, a network approach re-imagines social capital as existing not only through the content, but also through the configuration or structure of relationships. This essay will synthesize network-based literature in the fields of entrepreneurship and social movements, focusing on structural definitions of social capital[3], which I will also refer to as network capital. The goals of the paper are (first) to help organizations like FoodLab better design and evaluate their network-building efforts, and (second) to inform the field of social entrepreneurship studies.

The first section sets up broad distinctions between bridging and bonding social capital and social capital that accrues to individuals versus groups, organizations, or movements. Next I review how the entrepreneurship literature operationalizes social capital as a network measure and look at studies that describe the particular benefits resulting from brokerage and closure in a network. I repeat the exercise with literature on social movements, and then summarize similarities and differences between the two strands. Finally, I close with directions for future research.

Types of Social Capital: Brokerage versus Closure and “Me” versus “We”

Social capital can be separated into two broad categories. Bridging capital (also known as brokerage or leverage capital) refers to relationships that connect sets of otherwise disconnected actors. People with bridging capital find themselves in positions as translators, agents, brokers, and gatekeepers between actors and groups of actors who do not interact directly. Bridging capital can provide access to a diversity of ideas and the potential for competitive advantage and innovation (Granovetter 1973, 1983; Burt 1992). On the other hand, bonding capital (also known as closure or support capital) refers to dense in-group relations. It can generate the internal cohesion, shared norms, and the trust necessary for collective action (Coleman 1988; Putnam 1995).

While early writing on social capital tended to emphasize one type of network capital or the other, current thinking generally acknowledges that a mix is essential to productive groups and organizations (Taube 2004; Burt 2001; Borgatti et. al. 1998). Whereas sparse networks offer the opportunity to connect to diverse ideas and can spur “out-of-the-box” thinking (or, more accurately, “out-of-the-network” thinking), denser networks make it easier to coordinate action by facilitating trust and shared language that lower the barriers to transactions and coordinated action (Obstfeld, 2004: 101). Figure 1, borrowed from Burt (2001), describes how bridging and bonding capital interact to affect a team’s performance. In networks where brokerage opportunities are similarly high for all actors, actors with more bonding capital are more successful. In networks where all actors are part of cohesive groups, those with more opportunities to bridge to dissimilar people or groups are more successful. A combination of  internal closure and external bridges leads to optimal performance.

Figure 1: The interaction of bridging and bonding capital, from Burt (2001)

Burt (2001) considers how patterns of brokerage and closure put a team and its members at an advantage relative to other external groups and teams. Entrepreneurship studies tend to take the same approach, asking how the benefits of social capital accrue to an individual entrepreneur or business. The focus in this case is on the returns to an entity via its web of immediate relationships, or ego-network. On the other hand, network studies of social movements are more interested in benefits that result from bridging and bonding in the whole movement. Generally the field is less interested in how a particular organization or activist gains individual advantage from its position within the network and more interested in how the movement is enabled or constrained by its overall pattern of relationships. This is particularly important when considering social entrepreneurship and the case of FoodLab since we are interested in how social capital affects the capacity and behavior of the individual entrepreneur and the group as a whole.

Figure 2 above represents four broad ways of thinking about social capital according to the network mechanism (bridging versus bonding) and the entity that benefits from the capital (the individual versus the group as a whole). In the upper left quadrant is “me-bridging” capital where an individual benefits from its position between two disconnected actors; below it is “me-bonding” capital where an individual benefits from being part of a cohesive group. A FoodLab entrepreneur who has a loosely connected set of advisors who do not know one other (a local leader, a peer in another town, a couple of fellow business owners, and a family member) has opportunity for new ideas and fresh perspectives. On the other-hand, an entrepreneur who has been involved in the food industry in Detroit for years, and has a dense set of relationships with farmers, suppliers, and activists (who all also know one another), has an advantage because she has access to specialized insider language and trust built up over time which makes for more potential for collaboration. In the upper right quadrant is “we-bridging” capital. An organization like FoodLab may benefit from internal bridging ties between a cluster group of gourmet foodies and a group of food justice advocates because this increases its legitimacy and power. In the lower right quadrant is “we-bonding” capital. Here, FoodLab benefits from a tight-knit steering committee or a dense subset (or subsets) of committed members. Again, a mix of “me-bridging” and “me-bonding” capital are necessary for entrepreneurial success, while a mix of “we-bridging” and “we-bonding” capital makes for a more effective movement.

(Click here or on chart above for larger view)

Entrepreneurship: Social Capital for “Me”

The previous section set up some broad distinctions between conceptions of network capital in entrepreneurship and social movement studies. The next two sections provide more detail on how social capital tends to be specified and measured in each field.  Figure two offers a general orientation to five aspects that define networks in each field: defining social capital, nodes, network boundaries, ties, structural measures. I start in this section elaborating on these five aspects in the entrepreneurship literature.

“Me” capital enables and constrains the individual entrepreneur

In network studies of entrepreneurs and social capital, individual entrepreneurs are generally the unit of interest. Social capital, even when it is a property of relationships, still belongs to the individual entrepreneur. For example, in Davidsson & Honig’s (2003) large sample of Swedish entrepreneurs, social capital is treated in the same way as other oft-studied entrepreneurial characteristics like innovativeness and human capital. Even when studies of entrepreneurs move beyond endogenous characteristics to consider the entrepreneur’s network position, the focus is still on how this position offers advantage to that individual business.

A counter example may clarify this point. Scholars who study entrepreneurship and regional development might be curious how the structure of connections between businesses in a region enhances or limits opportunities for the region as a whole, however, even this body of literature tends to focus on how network capital affects individual business success rather than the overall health of a sector or region (McQuaid 1996) . Assuming a pure market system, these two approaches might be one and the same, but empirical evidence makes it clear that even in contexts that we perceive to be “pure markets,” social dynamics often concentrate resources or advantage in ways that fail to maximize overall benefit (Baker 1990).

Social capital as independent variable

Network studies in entrepreneurship ask how an entrepreneur’s network affects her success by mitigating access to tangible and intangible resources (Hoang & Antoncic, 2003). Some examples of resources include advice, information, emotional support, and opportunity. Some studies examine the effect of social capital on the intervening resource and make assumptions about how access to resources relate to success; for example, Singh (2000) considers how network capital enhances or impedes entrepreneurial opportunity recognition, assuming that opportunity recognition is a key component of success. Hansen (1995) on the other hand looks at the effect of network capital on a direct measures of success: number of employees.

Nodes

In the case of network studies in entrepreneurship, nodes can be entrepreneurs or enterprises. Studies that focus on earlier stage and emerging entrepreneurs appear to use individuals (Davidsson & Honig 2003; Singh 2000) whereas studies that focus on later-stage businesses use businesses (Baum et. al. 2000; Uzzi 1996; Hansen 1995). This may be because in earlier stages, organizational boundaries are not as clearly defined.

Network boundaries

In all the studies that I reviewed, the individual entrepreneur or enterprise was the unit of analysis and data was limited to the entrepreneur’s personal ties (also known as alters) rather than a complete network. This relates to my first observation that network studies are primarily interested in how network position offers an individual advantage rather than characteristics of the network on the whole. Some data collection methods limit boundaries further by using name generators that only allow the respondent to name five or ten significant alters (Singh 2000). This compromises the power of the results since it offers an incomplete picture of an entrepreneur’s range of network capital.

Ties

Studies measure many different types of ties including friendship, advice giving and seeking, idea and opportunity sharing, reputation signaling, and business exchange ties. Some studies delineate between “strong” or “weak” ties based on the frequency or type of interaction (Hansen 1995; Davidsson & Honig 2003).

Structural measures

Most studies start with degree centrality or simply counting up the subject’s ties. Since most studies focus on an incomplete egocentric view of the network (Hoang & Antoncic 2003), they tend to use proxies like network density or network heterogeneity to understand the degree to which networks exhibit closure or possess structural holes.

Brokerage and Entrepreneurship

Students of entrepreneurship generally start with a proposition that network size and her centrality within it is central to her success. The more people she knows, the more likely she is to be able to find the resources and information she needs. Quickly, though, scholars shift to address not just the number of ties, but the degree to which these ties are connected to one another:

Whereas network size and centrality delimit the amount of resources an actor can access, the presence of structural holes in the network changes the ability of actors to gain access to a diversity of resources” (Hoang & Antoncic 2003: 166).

There are two ways to think about this type of bridging capital:  as a conduit of diverse, non-duplicative information that entrepreneurs can access, but don’t have to spend much time to maintain (what Granovetter (1973; 1983) calls weak ties), or as an opportunity for an entrepreneur to exploit structural holes between disconnected parties for competitive advantage (Burt 1992).

In their review of 70 articles that employed network analysis to study entrepreneurship, Hoang & Antoncic (2003) found “consistent support for Burt’s structural hole argument – advantage to occupying a bridging position” (p. 173). They cite McEvily and Zaheer’s (1999) study of competitiveness of manufacturing firms and Baum et al’s. (2000) analysis of Canadian biotechnology firms that both show a positive relationship between structural holes and firm performance and innovation.

Singh (2000), who was also included in this review, did not find a significant relationship between existence of structural holes (as measured by number of unrealized relationships between alters) and whether entrepreneurs saw and seized opportunities. However he did find a positive relationship between number of weak ties and entrepreneurial opportunity. This discrepancy highlights the danger of conflating “weak ties” and bridging capital. Singh only allowed respondents to name five alters on the survey’s name generator. Since it is likely that entrepreneurs learned of opportunities from many more than five sources, and were likely to name resources who they knew especially well (and who therefore likely also knew one another), this may have skewed the results.

Granovetter (1983) demonstrates that weak ties (defined as low frequency, low intensity, or low reciprocity relationships like casual acquaintances, old co-workers, etc.) tend to be bridges whereas strong ties (friends, family, and other close relationships) tend to create closure because of the law of transitivity: If Joe is friends with Bob and Greg, it’s likely that Bob and Greg will also know each other since Joe spends time with both of them and they probably share similar interests. However, the mapping of weak to bridging and strong to bonding ties is not one to one. Zaheer & McEvily (1999) for instance, found that weak ties (as measured by infrequency of contact) did not correspond to bridging ties and did not predict the firm’s adoption of  new strategies.

The brokerage advantage is not only in the mere existence of a weak or bridging tie, but in the non-redundant nature of the tie. Diverse ties are only useful insomuch as an entrepreneur is able and willing to act on this diverse information. In his recent book, Neighbor Networks Burt (2010) argues that the benefit of brokerage is not only access to opportunities, but the development of internal capabilities to understand, combine, and act on these opportunities in meaningful ways. Brokerage, he explains is a “forcing function” for the “cognitive skills of analogy and synthesis, and emotional skills for reading, engaging, and motivating” (Burt 2010: 10).

Closure and Entrepreneurship

Social capital associated with closure tends to get less emphasis in entrepreneurship literature. However two studies offer some insight into the benefit of density. Hansen (1995) studied the ego-networks[4] of 44 service-based start-ups in Tennessee. He compared the size, density, and frequency of interactions within an entrepreneur’s network of advisors to the organization’s performance, measured by payroll. He found that size and density of an entrepreneur’s network did correlate positively with payroll, but the analysis of the regression coefficients suggested a more complex interaction between the number of ties in an entrepreneur’s set, their degree of connectedness, and the frequency with which they interacted. Specifically, adding another member to the network did predict an increase in performance, but this increase would be offset by the decrease in closure unless the new member knew other members. The negative effect increased if the entrepreneur only interacted with the advisor infrequently.

Hansen’s (1995) findings suggests that at certain thresholds, new contacts can compromise performance unless there is some degree of closure. The theory he uses to explain this comes largely from organizational performance literature which suggests that a more integrated network structure leads to lower barriers to exchanging tacit information and making other transactions (consider, as an example, a team who has been working together for many years and has developed insider language, trust, and familiarity that make it easier to work together). He concludes by advising entrepreneurs to consider new relationships instrumentally: first consider what resources are required,  next seek out new network members who have these mission resources, and finally incorporate these new members into the group (Hansen 1995: 15).

Uzzi (1996) takes a slightly different approach, examining relationships between contractors and manufacturers in New York’s garment industry, and the relationship between social capital and the firm’s likelihood of failure. In this case, ties are based on “special relationships” or  “embedded exchanges” between a firm and its “preferred” trading partners (Uzzi 1996: 686). He calculates first-order coupling (the degree to which ego is sending work to a few preferred partners versus many random transactions) and second-order coupling (the degree to which a firms’ network partners are also transacting with a few preferred partners versus many random transactions). The second measure serves as a proxy for density or embeddedness, though it does not measure it directly. The results suggest that firms that operate in a more embedded context are less likely to fail up to a certain threshold point, after which the social capital associated with embeddedness or closure switches from asset to liability. This corresponds with Levitte’s (2004) qualitative study of how bonding capital can be an asset to small business development in Aboriginal communities, but can also be a barrier to business success.

Specifically, Uzzi (1996) points out, density in business relationships can support entrepreneurial success by “concentrating on cultivating long-term cooperative relationships that have both individual and collective level benefits for learning, risk-sharing, investment, and speeding products to market” (p. 693)[5]. However, it is possible to have too much of a good thing and it appears that dense structure of bonding capital can have negative implications at the extreme, in effect, “sealing-off firms in the network from new and novel information or opportunities that exist outside the network” (p. 675). Too much closure leads to a loss of “adaptive capacity,” and more susceptibility to shocks. In the same vein as Burt (1992; 2010), Uzzi (1996) concludes that, “optimal networks are not composed of either all embedded ties or all arm’s-length ties, but integrate the two” (p .694).

Social Movements: Social Capital for “We”

Snow, Soule and Kriesi (2003) define a social movements as:

Collectivities acting with some degree of organization and continuity outside of institutional or organizational channels for the purpose of challenging or defending extant authority, whether it is institutionally or culturally based, in the group, organization, society, culture, or world order of which they are a part. (P. 7)

Because they are not defined by institutional or organizational boundaries, but often include a mix of formal social movement organizations, informal organizing, and individuals activists, movements can be difficult to pin-down. A network analysis approach that looks at the relationships between movement entities can be useful to describe and understand the nature and social processes that make up a movement.

“We” capital enables and constrains the movement as a whole

Movement scholars use network analysis to examine the structure of movement networks to consider how bridging and bonding capital relates to capacity for collective action, identity formation, or political engagement. Granovetter (1973) for instance offers early insight into the role of “we” capital in movements when he describes how community social structure in Boston’s West End may have contributed to its inability to halt processes of “urban renewal” that ultimately destroyed the neighborhood. Though West-enders were organized in tight cliques, characterized by a high degree of bonding capital, they lacked bridging capital between groups that could have broadened their collective power, and lacked external ties to decision-making entities that ultimately influenced their fate.

One branch of network studies on social movements pays attention to how network capital affects how individuals are recruited and participate in a movement. Who do they know? Who are their friends and who do their friends know? What opportunities for participation exist in their social sphere? This line of inquiry asks how personal networks influence the choices of individual activists and rarely examines how overall network structure in a movement enables or inhibits outcomes (Diani 2003a: 8). The approach may look the same as studies on entrepreneurial “me-capital,” but is different. Though the unit of analysis is the individual and his network, studies like Kriesi et. al.’s (1995) study of gay movement subcultures are not as interested in how individual networks benefit the individual, but how they ultimately shape and constrain broader movement community.

Social capital as dependent variable

Movement scholars have had limited success linking network capital directly to movement outcomes. The connection between a movement’s structure and its effect on macro-level social or political change is complex and difficult to specify; on the other hand, micro-level analysis of a particular protest or organization is more precise, but offers less generalizable meaning (Diani 1997: 129). Rather than attempt to link network structures to specific outcomes, Diani (1997) suggests that researchers reverse the equation and consider specific network configurations as the outcome of social movement processes. Analysts can assess a movement’s efficacy indirectly in terms of the scope, density, heterogeneity, and overall pattern of its relationships. These relationships formed through various movement processes are assumed to constrain or enable future action,  and so on in a cycle over time:

According to this perspective, the central problem is no longer whether and how mobilization campaigns, and cycles of protest determine specific changes at different levels of the political and the social system. It becomes instead whether they facilitate the emergence of new networks, which in turn allow advocacy groups, citizens’ organizations, action committees, and alternative intellectuals and artists to be more influential in processes of political and cultural change.” (Pp. 130-135).

This view of social movements fits nicely with definitions of social movements as fluid “multi-organizational fields” (Evans 1997, Klandermans 1992; Curtis & Zurcher 1973) and with the new social movements perspective championed first by Melucci (1989) that conceives of social movements as ever-shifting processes of meaning and identity-construction that challenge existing cultural norms. The studies reviewed in this paper tend to take this approach to heart. Some, for example, explicitly examine correlations between a movement’s bridging and bonding capital and the nature and strength of collective identity frames (Ackland & O’Neill 2011, Carroll & Ratner 1996).

Nodes

When studies focus on measuring network capital within the movement as a whole, we see meso-level networks with nodes defined as social movement organizations (SMOs) (Ackland & O’Neill 2011, Ansell 2003, Diani 2003b, Phillips 1991, Rosenthal 1985). When studies focus on the role of network capital in activist recruitment, we see micro-level nodes: individual activists or would-be activists (Ohlemacher 1996, Della Porta 1988). Some movement studies also look at the relationship between sub-movements or entire movements. For example, Carroll & Ratner (1996) are interested in the relationship between progressive movements in the Toronto area; in this case, sub-movement and movement nodes are not structurally defined agglomerations of organizations (for example, a group of organizations defined by a modularity-Q score), but are defined according to the researcher’s qualitative analysis.

Network Boundaries

Scholars of activist recruitment generally use ego-networks, similar to scholars of entrepreneurship. this gets at questions of how a person’s position in her immediate web of contacts may affect her likelihood to engage, but doesn’t comment directly on the way bridging and bonding capital within a movement may provide differing opportunities for engagement. This decision may mostly have to do with expediency since it is more difficult and resource-intensive  to map a complete network of activists in a movement (Diani 2003a: 8).

Other social movement analysts set network boundaries based on topical (e.g. environmental), geographic (e.g. Bay Area), and chronological (e.g. current day) criteria. For example, Ansell (2003) studies environmental organizations in the San Francisco Bay Area  and Rosenthal (1985) looks at 19^th century women’s movements in New York. Defining a particular movement case makes data collection more manageable while setting up the possibility to draw inferences about the nature of movements generally.

Carroll & Ratner’s (1996) study spans multiple movement topics  in order to understand the development of more general “master” frames about social justice. No studies that I reviewed examined movements over time, but Diani (2003a) suggests that this could be a useful way to understand how network capital changes as a result of movement tactics.  Practically speaking, researchers use historical documents, on and off-line archives, and snowball surveys and interviews to identify activists and organizations to include in a study.

Ties

When it comes to specifying what ties constitute “social capital” within a movement network, some scholars often use two-mode analysis to identify where organizations or activists are connected. Generally this is done either through overlapping membership (e.g. if organization A and B share a member, they are “connected” and if member A and B belong to the same organization, they are “connected) (Rosenthal 1985; Carroll & Ratner 1996) or mutual participation in some action or coalition (Phillips 1991; Diani 2003b).

Again, it appears that this approach is mostly a matter of expediency. It is simpler and less resource intensive to collect member lists rather than survey organizations directly. It also may be a more viable option for scholars studying historical movements where member lists are available, but other good organizational documentation is sparse. The only drawback here is that these types of ties generally measure a likelihood of relationship rather than an actual relationship, depending on how the author interprets the tie: for example, two organizations with four members in common may not work together on any common campaigns or have any sort of “real” connection. The overlap may be happenstance.

Ansell (2003) is an exception to the two-mode trend; he surveyed organizations to ask directly if they had worked together in some capacity. This approach made sense in this research case because Ansell already needed to perform a survey to collect other information.

Structural measures

Network analysis in social movements employs similar structural measures of social capital as in entrepreneurship studies, but some studies are also able to get at brokerage measures like cliques and betweenness centrality scores more directly because they are mapping a complete field of relationships.

Brokerage and Social Movements

Social movement scholars borrow from the same seminal literature on social capital as scholars of entrepreneurship (Granovetter 1973; 1983; Coleman 1988; Burt 1992; 2010; Putnam 1995), and studies come to similar general conclusions about the necessity of a healthy mix between bridging and bonding capital[6].

The key difference between the concept of bridging capital in the social movements versus in entrepreneurship is an emphasis on the value of bridging to the overall movement, rather than to the individual actor. This relates to Obstfeld’s (2005) distinction between tertius gaudens and tertius iungens-style orientations to bridging positions. Whereas a tertius gaudens (“third who enjoys”) approach describes brokers who controls the relationship between partners for their individual benefit, a tertius iungens strategy (“third who joins”) relates to brokers that seek to be a bridge between otherwise disconnected alters for the purpose of “longer-term generativity and coordination” (p. 125). Both the tertius iungens and tertius gaudens orientations start from the same premise: a network full of structural holes or potential bridging capital. Whereas brokers with gaudens orientations attempt to maintain their unique position by acting as a gatekeeper, iungens-oriented brokers seek to introduce alters to one another and close structural holes, either maintaining an active coordinating role over time or disappearing from the triad once the connection is made. This strategy does not necessarily maximize direct and immediate returns to iungens; in fact, it often constitutes a loss of “me-bridging” capital, but it is key to understanding the nature of “we-bridging” capital in social movements. The mechanism behind bridging capital in the social movement literature is similar to cooperative game theory: organizations engage in relationships not necessarily because of their immediate transactional benefit, but because of how those relationships change the overall game (Obstfeld 2005: 125).

As Diani (2003) finds in his study on brokerage and centrality within Italian environmental organizations, whereas central SMOs tend to attract more resources and media, SMOs with high brokerage scores allow “the establishing of communication across different subgroups of the movement and in this sense facilitate[] the integration of otherwise heterogeneous organizations” (p. 113). While it may be the case that SMOs that operate as brokers enjoy some return on their advantage (whether in terms of exposure to more diverse ideas which in turn affects their capacity to innovate or in resulting increased capacity to occupy a liminal position as Burt (2010) suggests), the more significant point is that this form of bridging is crucial to building a broad base of support. For example, Diani (2003b) and Phillips (1991) studies both show that  removing key broker organizations would result in a fragmented and parochialized landscape of disparate sub-movements, each alone lacking the capacity to enact social, cultural, or political change. Carroll and Ratner’s (1996) analysis of 272 SMOs across 13 movements in greater Toronto comes to a similar conclusion and comments at length on a case in which one movement is not connected to the broader field by any bridges.

The Toronto example also suggests a relationship between we-bridging capital and movement framing, showing that SMOs with more cross-movement or bridging ties (as measured by overlapping memberships of activists) tend to employ a more inclusive master frame that focuses on the interconnectedness of multiple issues. Despite some methodological shortcomings[7] the authors’ main points seem to be corroborated by qualitative evidence. There is potential for more work to examine the relationship between the role of “me-bridging” capital in strategic framing processes including frame bridging, frame amplification, frame extension, and frame transformation, and in the resolution of frame disputes (Benford & Snow 2000).

Finally, Ansell (2003) comes to the same conclusion from the opposite end. In his study on environmental SMOs in the San Francisco Bay Area, relational embeddedness or closure (or relative lack of bridging capital) negatively affected SMOs attitudes towards collaboration with governmental organizations and groups with opposing interests.

What kinds of organizations make up bridging capital in movements? Diani (2003b) notes that bridges within movements are not necessarily characterized by a higher degree of ties to other SMOs, they are not necessarily more central. However, brokers tend to be more bureaucratic and neutral, opening up “opportunities for communication (and possibly collaboration) within the movement network which would not otherwise be available” (p. 119). Carroll & Ratner (1996) and Phillips (1991) come to similar conclusions when they note the high-degree of bridging capital tends to belong to organizations that are well resourced and more neutral in framing or explicitly designed to bridge movement boundaries. This, interestingly, is reminiscent of Burt’s (2010) description of how the individual broker exercises the advantage of her structural position: not necessarily by way of simply being connected to many diverse alters, but rather because of the characteristics and capacity to synthesize seemingly opposing views and more objectively weigh the risks and rewards of action and thus take advantage of these opportunities.

Closure and Social Movements

Broad conversations of bonding social capital within movements suggests that dense clustering and closure relates to trust and collective action, yet relatively few studies get at a causal relationship between bonding capital and collective action. Diani (1997) suggests that we flip the equation and consider how movement action contributes to differing degrees of density rather than assess the existence of social capital a priori and then use this to predict a cycle of action.

Rosenthal et. al.’s study (1985) shows high density within a network of organizations that comprised a 19th century women’s movement in New York, as compared to a network study of corporate interlocks. It also finds distinct relational clusters within the movement that corresponds to particular nodes of activity. As a result, the authors suggest that closure is a necessary component of action and that social movements on the whole tend to exhibit high density (closure) because of their stance in opposition to or apart from a dominant structure. They conclude that closure is both the result of a social movement’s position as a status quo alternative and a driver of the movement’s success in altering that status quo. If altering an extant social or cultural system requires connection with that system, then closure may indeed facilitate action, but at the same time, successful action will expand the movement and reduce closure. Thus as movements move from marginal to mainstream, we expect to witness less closure and more bridging (Diani 1997; Oliver & Marwell 1988). Taken to the extreme, a movement that is completely successful would eventually melt into broader society and nullify itself.

Deepening this line of thinking, Oliver & Marwell (1988, 1993) use mathematical modeling to examine the effect of movement size, heterogeneity, and density on the capacity for collective action. They find that larger, less dense groups are more effective than smaller, denser groups if the cost of action remains constant as the group increases in size. Larger groups become even more effective to the extent that ties are non-random and heterogeneous, which indicates selective linking.

Diani (2003), in the vein of new social movement scholars, points out that it is not just collective action, but also shared identity that distinguishes a movement network from a loose coalition network or random network of actors. Persistent clusters of dense ties within a movement are likely to indicate shared identities, values, and theories of change. In movement terms, these collective identities are called frames. Both Carroll & Ratner (1996) and Rosenthal et. al. (1985) find correspondence between dense clusters and shared identity or movement framing.

Social Capital, or the Goldilocks Story on Repeat

In the last few pages, I have described some of the ways that scholars think about network capital in entrepreneurship and social movements. For both the entrepreneur and the movement collective, a mix of bridging and bonding capital is associated with the most positive outcomes.. For both brokerage and closure, entrepreneurs and movement leaders can adopt a “Goldilocks” approach. Bridging capital is generally good, but  may be detrimental after a certain threshold especially when there is an associated decline in closure. Consider for example an entrepreneur with an overload of contacts, all of whom are pulling her in opposite directions or a movement full of randomly connected individuals each of whom has her own theory of change. On the other hand, closure can be useful to facilitate action, but too much closure can lead to stagnation and parochialism.  Too loose, too tight, or just right? The ideal mix depends on a variety of factors including the existence of other types of capital (human capital, political capital, financial resources, etc.), the goals of the entrepreneur or collective, and the stage of development.

In both cases, more work is needed to examine the process by which network capital changes over time. Both for the entrepreneur and for the movement, social capital should be conceived as a cycle rather than a static entity: brokerage and closure leads to outcomes which lead to new patterns of network capital. In reality, the Goldilocks story doesn’t end, but replays as her appetite and preferences change. Understanding these cycles through longitudinal approaches and action is key to accurately understanding the fluid, changing nature of entrepreneurs and movements.

Social Entrepreneurs: Collapsing “Me” and “We” Social Capital

Returning at last to our FoodLab example, from its inception, FoodLab members described the group as a network and used words like “connecting,” “networking,” “sharing,” “bringing together,” “building community,” and “building relationships” to describe our activities. FoodLab members expressed multiple reasons for engaging in networking exercises:

• Information, ideas, and opportunities: Documenting and exchanging information on best-practices; sharing information about events, opportunities for accessing education, funding or technical assistance;
• Reputation and legitimacy: Associating with a recognizable institution and connecting lesser-known businesses with well-connected or established institutions and individuals;
• Emotional support: Addressing day-to-day feelings of isolation (many entrepreneurs were sole-proprietors) and uncertainty (many entrepreneurs had never started a business before and had little-to-know experience);
• Developing shared resources: Shared kitchen or office space, cooperative purchasing, and sharing other tangible resources like equipment or even labor;
• Accountability to shared values: Engaging in ongoing dialogue and some sort of self-regulation of social, and environmental bottom lines;
• Advocacy and collective action: Working together to change broader political, economic, cultural and social structures.

These six goals for the network represent six different possible benefits of network capital. The first three relate to the individual entrepreneur’s “me-capital”. The next three relate to the “we-capital” of the group both internally and relative to the larger social sphere. In the case of FoodLab, we are interested not only in supporting the individual success of businesses in creating social, environmental and financial value, but also in fostering shared norms (e.g. a commitment to social and environmental values) and the capacity for collective action to effect broad-scale change. In both cases, we recognize the advantage of fostering closure within the group to facilitate trust, but also in bridging between otherwise disparate clusters of entrepreneurs (and entrepreneurial allies) to facilitate innovation and broaden our base of power. Students of food systems change in the US have commented on this bridging capacity as one of the major strengths of the movement (Hassanein, 2003; Starr, 2010). The US food movement is made up of a diverse array of loosely connected clusters of public health, environmental, sustainable agriculture, social justice, and other groups. Scholars have yet to undertake a comprehensive network mapping project; I hope to start with one chunk: How do social entrepreneurs fit within this multi-organizational field?

Scholars worry that market-based approaches within the food movement  are individualistic, unaccountable and depolicitized (Starr 2010; Donald 2008). They suggest that “entrepreneurship” as a tactic usually fails to address and sometimes exacerbate issues like food security for the most vulnerable and racial and cultural injustice (Allen et. al., 2003). A social entrepreneur’s tension between “me-capital” that facilitates individual action and “we-capital” that represents accountability to the collective may be a response to contemporary conceptions and criticisms of entrepreneurship in the food movement.

The literature seems to suggest that both the “me-capital” approach in entrepreneurship and “we-capital” in movements look similar structurally, but the “we-capital” approach optimizes bridging and bonding ties for the good of the whole whereas the “me-capital” approach attempts to optimize for the individual. Given what we know about scale-free networks and preferential attachment where actors rich in connections tend to continue to get richer over time (Neal 2012), the entrepreneur scholar’s approach to network capital means those who start with an advantage continue to accrue returns over time. Even in cases where the entrepreneur is benevolent, the unequal distribution of social capital has consequences because it often manifests along the lines of race and gender. Certain social, cultural and environmental values can come to dominate, marginalizing other interests. For example in Detroit, both bridging and bonding appears to be concentrating in a circle of (mostly) white entrepreneurs who tend to live in certain neighborhoods, who focus on “quality,” local,” and “artisanal” production and food as a means to attract more young educated people to the city. Building up the “me-capital” of these entrepreneurs may forward these goals (and the financial viability of these businesses) but at the expense of a focus on other neighborhoods, on increasing minority-owned businesses, on delivering culturally-appropriate and affordable food to neighborhoods.

The “we-capital” approach introduces an aspect of agency that may spread connections more equitably depending on which entrepreneurs are considered to be “within” the boundaries of the network. Social entrepreneurship may be the place where “me-bridging” meets “we-bridging”: entrepreneurs make decisions that weigh their own needs, but also the good of the collective.

Future research should combine these entrepreneurial and social movement perspectives on network capital to systematically study networks of social entrepreneurs. From this we can understand how and why network capital changes over time, how this affects the success of individual entrepreneurs, and how the structure of network capital enables or constrains directions for broader cultural, political and social change. The following are some specific propositions to test in the FoodLab case:

Me-Capital

Proposition 1: Entrepreneurs whose idea and opportunity networks (e.g. where have learned of new ideas or opportunities for your business?) contain more structural holes and are more heterogeneous will be more likely to have positive entrepreneurial outcomes (e.g. entrepreneurial emergence, entrepreneurial persistence, achieving financial sustainability, and an entrepreneur’s progress towards social, environmental, and financial goals) up to a threshold.

Proposition 2: The degree of closure in support (e.g. who encouraged or supported you in starting your business?) and advice networks (e.g. who did you go to for information or advice while starting your business?) will correlate with positive entrepreneurial outcomes up to a threshold point (after the threshold, more closure will lead to worse outcomes).

Proposition 3: The ratio of bridging and bonding capital among “successful” businesses will vary according to the stage of business development and personal characteristics (human capital, race, and gender).

We-Capital

Proposition 4: Entrepreneurs with high bridging capital who are strongly affiliated with FoodLab (and are assumed to be more enrolled in the food movement) will be more likely to adopt a tertius iungens orientation than new members and non-members.

Proposition 5: Network closure within FoodLab subgroups strengthens specific “good food frames”; bridges connect otherwise disparate groups with universal frames that synthesize multiple viewpoints.

Proposition 6: The mix of bridging and bonding capital changes over time within FoodLab and correlates to differing potential for collective social, political, and cultural action.

Works Cited

Ackland, R., and M. O’Neil. 2011. “Online collective identity: The case of the environmental movement.” Social Networks.

Ansell, Christopher. 2003. “Community Embeddedness and Collaborative Governance in the San Francisco Bay Area Environmental Movement.” Pp. 123-144 in Social Movements and Networks: Relational Approaches to Collective Action., edited by Mario Diani and Doug McAdam. Oxford University Press.

Baker, W. E. 1990. “Market networks and corporate behavior.” American Journal of Sociology. Pp. 589–625.

Baum, J.A.C., T. Calabrese, and B.S. Silverman. 2000. “Don’t go it alone: Alliance network composition and startups’ performance in Canadian biotechnology.” Strategic management journal 21(3):267–294.

Benford, R. D., and D. A. Snow. 2000. “Framing Processes and Social Movements: An Overview and Assessment.” Annual Review of Sociology, 639.

Borgatti, S.P., C. Jones, and M.G. Everett. 1998. “Network measures of social capital.” Connections 21(2):27–36.

Burt, R.S. 2010. Neighbor networks: Competitive advantage local and personal. Oxford University Press, USA.

Burt, Ronald. 2005. Brokerage and Closure: An Introduction to Social Capital. Oxford: Oxford University Press.

Carroll, W.K., and R.S. Ratner. 1996. “Master Framing and Cross-Movement Networking in Contemporary Social Movements.” The Sociological Quarterly 37(4):601–625.

Coleman, J.S. 1988. “Social capital in the creation of human capital.” American journal of sociology 95–120.

Davidsson, Per, and Benson Honig. 2003. “The role of social and human capital among nascent entrepreneurs.” Journal of Business Venturing 18(3):301-331. Retrieved February 12, 2012.

Dees, J. G., and B.B. Anderson. 2006. Framing a Theory of Social Entrepreneurship: Building on Two Schools of Practice and Thought.

Della Porta, D. 1988. “Recruitment processes in clandestine political organizations: Italian left-wing terrorism.” International Social Movement Research 1:155–169.

Deresiewicz, William. 2011. “Generation Sell.” New York Times, November 12, Online Retrieved November 12, 2011 (http://www.nytimes.com/2011/11/13/opinion/sunday/the-entrepreneurial-generation.html?pagewanted=all).

Diani, Mario. 1997. “Social movements and social capital: a network perspective on movement outcomes.” Mobilization: An International Quarterly 2(2):129–147.

Diani, Mario. 2003a. “Introduction: Social Movements, Contentious Actions, and Social Networks: ‘From Metaphor to Substance’?.”Pp. 1-18 in Social Movements and Networks: Relational Approaches to Collective Action., edited by Mario Diani and Doug McAdam. Oxford University Press.

Diani, Mario. 2003b. “‘Leaders’ or Brokers? Positions and Influence in Social Movement Networks.”Pp. 105-123 in Social Movements and Networks: Relational Approaches to Collective Action., edited by Mario Diani and Doug McAdam. Oxford University Press.

Doh, J.P., and N.M. Dahan. 2010. “Social Movements and Social Networks: An Evolutionary Perspecctive on Contemporary US Student Advocacy Campaigns.”

Granovetter, M. 1983. “The strength of weak ties: A network theory revisited.” Sociological theory 1(1):201–233.

Granovetter, M.S. 1973. “The strength of weak ties.” American journal of sociology 1360–1380.

Hansen, E.L. 1995. “Entrepreneurial networks and new organization growth.” Entrepreneurship theory and practice 19(4).

Hoang, Ha, and Bostjan Antoncic. 2003. “Network-based research in entrepreneurship: A critical review.” Journal of Business Venturing 18(2):165-187. Retrieved March 6, 2012.

Kriesi, Hanspeter. 1995. “Chapter 7: Gay Subcultures between Movement and Market.” in New Social Movements in Western Europe: A Comparative Analysis. University of Minnesota Press.

Levitte, Y. 2004. “Bonding Social Capital in Entrepreneurial Developing Communities-Survival Networks or Barriers?” Community Development 35(1):44–64.

Marwell, G., P. E Oliver, and R. Prahl. 1988. “Social networks and collective action: A theory of the critical mass. III.” American Journal of Sociology 502–534.

McQuaid, R. W. 1996. “Social networks, entrepreneurship and regional development.” Small firm formation and regional economic development 118–131.

Obstfeld, D. 2005. “Social networks, the tertius iungens orientation, and involvement in innovation.” Administrative science quarterly 50(1):100.

Ohlemacher, Thomas. 1996. “Bridging People and Protest: Social Relays of Protest Groups against Low-Flying Military Jets in West Germany.” Social Problems 43:197.

Oliver, P., and G. Marwell. 1988. “The Paradox of Group Size in Collective Action: A Theory of the Critical Mass. II.” American Sociological Review 1–8.

Oliver, P., and G. Marwell. 1993. The critical mass in collective action: a micro-social theory. New York: Cambridge University Press.

Phillips, S. 1991. “Meaning and structure in social movements: mapping the network of national Canadian women’s organizations.” Canadian Journal of Political Science 24(4).

Putnam, Robert. 1995. “Bowling Alone: America’s Declining Social Capital.” Journal of Democracy, 78.

Rosenthal, N., M. Fingrutd, M. Ethier, R. Karant, and D. McDonald. 1985. “Social movements and network analysis: A case study of nineteenth-century women’s reform in New York State.” American Journal of Sociology 1022–1054.

Van Ryzin, G.G., S. Grossman, L. DiPadova-Stocks, and E. Bergrud. 2009. “Portrait of the social entrepreneur: Statistical evidence from a US Panel.” Voluntas: International Journal of Voluntary and Nonprofit Organizations 20(2):129–140.

Sharir, M., and M. Lerner. 2006. “Gauging the success of social ventures initiated by individual social entrepreneurs.” Journal of world business 41(1):6–20.

Singh, R.P. 2000. “Entrepreneurial opportunity recognition through social networks.” New York, London.

Smallbone, D. 2001. Researching social enterprise: Final report to the Small Business Service. SBS Research Directorate.

Starr, Amory. 2010. “Local Food: A Social Movement?” Cultural Studies, Critical Methodologies, 490.

Täube, V.G. 2004. “Measuring the social capital of brokerage roles.” Connections 26(1):29–52.

Thekaekara, M.M., and S. Thekaekara. 2007. Social Justice and Social Entrepreneurship: Contradictory Or Complementary? Skoll Centre for Entrepreneurship, Saïd Business School.

Thompson, J.L. 2002. “The world of the social entrepreneur.” International Journal of Public Sector Management 15(5):412–431.

Uzzi, B. 1996. “The sources and consequences of embeddedness for the economic performance of organizations: The network effect.” American sociological review 674–698.

Zaheer, A., and B. McEvily. 1999. “Bridging ties: A source of firm heterogeneity in competitive capabilities.” Strategic management journal 20(12):1133.

---

[1] Academics and practitioners have yet to achieve consensus on a definition of social enterprise (Mair & Marti, 2006; Peredo & McLean, 2006; Short, Moss & Lumpkin, 2009; Weerawardena & Mort, 2006), but for the purposes of this paper, we conceive of it as a business (for-profit, non-profit, or hybrid) for whom the creation of social or environmental value is prominent to their mission – at least as important as profits.

[2] See, for example, the Green Garage in Detroit (http://greengaragedetroit.com/); Social Edge (http://socialedge.org,) run by the Skoll Foundation; Ashoka Changemakers (http://www.ashoka.org/); Business Alliance for Local Living Economies sustainable business networks (http://livingeconomies.org); Echoing Green network (http://www.echoinggreen.org/young-nonprofit-professionals-network); The Hub Bay Area (http://bayarea.the-hub.net/public/); and more.

[3] I  limit this review to the structural elements of social capital as opposed to the content or governance of relationships, though I know these three concepts are often related (Hoang & Antoncic 2003).

[4] To get at the ego-network, Hansen (1995) asked entrepreneurs to list “the people with whom you were interacting most during that six-month period that preceded the hiring of your first FTE employee] to secure the business information and the resources that were important to your business” (p. 11). He then asked entrepreneurs to specify the relationship between every pair of alters that the respondent listed.

[5] This is similar to Coleman’s (1988) famous example of diamond traders in New York City that demonstrate that “close ties, through family, community, an religious affiliation, provide the insurance that is necessary to facilitate the transactions in the market. […] The strength of these ties makes possible transactions in which trustworthiness is taken for granted and trade can occur with ease” (p. 97).

[6] As a quote from Rosenthal et. al.’s study on a 19^th century women’s rights movement (1985) illustrates, there is also a similar conflation of strong ties with bonding ties and weak ties with bridging ties: Weak ties are channels for communication with diverse publics. Strong ties, however, show more enduring bonds. They demonstrate the existence of alliances. It is through these channels that resources (both material and ideological) flow from one organization to another, exchange of leadership may occur, and influence and control can be exercised. (Rosenthal et. al. 1985: 1051).

[7] By choosing a purposive sample for their two-mode analysis rather than including the entire population of activists, the authors may have mischaracterized the structure of relationships between SMOs and between movements. This is especially true given their finding that a large degree of cross-movement ties were initiated by what they call “organic intellectuals,” or individuals who appear to specialize in making cross-movement connections. If the sampling missed just a handful of these connectors, then the structure presented in the paper  would be inaccurate. A more robust analysis might use a cluster sample: purposive sampling to identify organizations, and then collection of complete member lists for these organizations.