Getting New Things Done
Networks, Brokerage, and the Assembly of Innovative Action
David Obstfeld



All larger organizations were once small. This seemingly self-evident statement raises the question, how do small organizations get bigger?

The answer I offer, briefly stated, is by engaging in specific social processes through which they conceive of and pursue new outcomes. Schumpeter described the creative response underpinning innovation and entrepreneurship as “getting new things done” (1947, 151). Getting new things done, in the form of distinctly novel change, is of special interest for its capacity to generate outsized impact in various contexts.1 Of course, not all small new things lead to larger outcomes, but when they do, I will argue that it is frequently because skilled players mobilize both networks and knowledge to marshal support for markedly new initiatives.

In tracing organizational origins, we end up with variations on a pattern where a small number of strategic actors mobilize people and resources toward imagined future outcomes. Although every mobilization process is different, certain elements are common to the process of getting new things done across a wide variety of human endeavors, including business innovation, new venture growth, collective action, artistic movements, institutional emergence, and many other social domains. To be sure, there are often other external factors, such as natural disasters, institutional forces, legal or regulatory changes, or technological regimes, that may select, propel, and amplify these incipient efforts. However, in the end—that is, in the beginning—we need to explain how strategic actors drive growth along a predictable set of dimensions.

The Increasing Importance of Social Networks and Project-based Innovation

The ability to manage one’s relationships, and the resources that those relationships afford, has always been a central concern of strategic actors. It’s therefore worth examining how our networks and the means by which we orchestrate them have evolved over the past two decades. A series of digital tools for establishing, maintaining, and propagating ties, starting with mobile phones, e-mails, text messages, and more recently Facebook and LinkedIn, have dramatically altered the means by which we engage our networks. This was evident, for example, in the way that digital tools made the Arab Spring possible (Ghonim 2012). The importance of networking is also accelerated by the connectedness central to Friedman’s (2005) account of globalization, which allows greater freedom and speed in the combination of people, ideas, and means for production.

Making the case that the way we engage networks has changed, Boltanski and Chiapello (2005, 2007) argue that society has evolved a “new spirit of capitalism,” reflecting a shift from an industrial to a more project-based society, referred to as the “project-oriented city,” involving a fundamental shift in organizing and individual action. According to the authors, the new way of organizing involves a firm “featuring an organisation that is very flexible; organised by projects; works in a network; features few hierarchical levels; where a logic of transversal flows has replaced a more hierarchical one” (2005, 165). The corresponding new form of individual action involves activity aimed at generating projects and formulating “life conceived as a series of projects” (169), which involves continually assembling disparate people for relatively short periods of time. This new form of individual action involves networks, coordinating, connecting, locating new sources of information, inspiring trust among those being coordinated, flexibility, and adaptability. Boltanski and Chiapello view key players in the new project-oriented context as “mediators [who] . . . possess the art of reconciling opposites, and know how to bring very different people together, and put them in contact” (2007, 115).

In the shift from an industrial world to a project-based world, new skills come to the fore. The ability for strategic actors to read the social terrain has always been crucial, but the importance of forming interdependent projects comprising multiple, interdependent actors has further amplified that talent. In addition, as people tend toward working less within the stable assemblages of corporations and more within constellations of projects, the ability to conceive of worthy projects, identify and assemble the appropriate participants and resources, forge trust in temporary communities, and cultivate ideas within professional communities constitutes a newly evolving toolkit necessary in a project-oriented world.

Coordination, Brokerage, and Social Skill

Mobilizing action to get new things done revolves around coordination, first and foremost of people but also of the resources or ideas they bear and the organizations they represent. In the absence of coordinative action, those elements stand largely inert. Coordination addresses how new things first get started through the novel combination and recombination of elements (e.g., as found in emergent start-ups and social movements), how new things get bigger (i.e., how elements accrue to a growing project or initiative), and how combinatorial elements come together repeatedly (through sustained feats of coordination found in organizational routines). My focus is on this microsocial crucible of action, usually involving a small number of strategic actors and some form of collaborative action among them.

Coordination involves the integration of interdependent tasks, and is therefore at the crux of organizing (March and Simon 1958; Faraj and Xiao 2006; Okhuysen and Bechky 2009). Coordination accomplishes this integration through the mechanisms of accountability, predictability, and common understanding (Okhuysen and Bechky 2009). Okhuysen and Bechky (2009) emphasize coordination within a design tradition that takes place in a single organization, in which participation, tasks, and outcomes are well established and relatively predictable. This formalized design tradition contrasts with more emergent contexts, where organizing is more creative, boundaries are less defined, and tasks and goals are unfolding. This underscores the need for an approach that is able to address not only predictable settings but also less predictable ones like the project-based contexts described above.

Coordination takes many forms. Weick (1969) introduced the double interact, a communicative exchange between two individuals, as the core coordinative unit of action crucial to organizing. In the double interact, according to Weick, one person communicates a message to a second person, to which the second person responds, followed by the first person making an adjustment to his or her original message based on that response. This three-step exchange within the dyad speaks to a broad range of communicative exchanges, such as when new ideas are pitched, existing ideas are refined, interests are gauged, or gossip is exchanged.

The double interact concept, though inherently coordinative, is characterized by a certain inertness in scale; it assumes a preexisting pair of interlocutors, and considers primarily the possibilities for collaborative interchange within the pair, but it does not address larger numbers of actors or the potential for dynamic expansion beyond the two participating individuals. We could stretch Weick’s consideration of the dyadic interact to accommodate the exchange between one person and a group (e.g., to which the person is presenting), or between a leader and a larger audience. Nevertheless, the double interact still fails to accommodate social phenomena inherent in feats of coordination, aggregation, and growth that involve a larger or growing number of participants or more complex communication and response loops.

Imagine, as an alternative, that a member of that dyad introduced a third person into the conversation. Such an initiation of a new tie introduces an important potential dynamism into the dyadic interact. That triad might then be used as a proxy for a much wider range of dynamics associated with growth in the number of parties engaged. When we bring a new person or organization into a preexisting pair or examine an individual’s introduction or facilitation of two others, we invoke brokering activity—a central focus of this book. The move from two to three, as Simmel pointed out over a century ago, is profound in that it draws in a broad sweep of coordinative action of greater complexity, impact, and dynamism (Simmel and Wolff 1950). The move to three speaks to a number of social processes at the microsocial level, involving numbers far greater than three (e.g., an emerging venture or an incipient episode of collective action), and, correspondingly, among firms of three or more at a more macro level (e.g., alliance formation). In the simplest sense, the move from two to three involves the alignment of a greater range of interests, ideas, and resources.

To coordinate triads, a number of new social processes are implicated, including formation (e.g., the new introduction of nonacquainted others), inclusion (the invitation of a third into a preestablished group of two), and the strategy or gamesmanship involved in growing groups (i.e., the strategic choice to add or exclude a fourth or a fifth). This gamesmanship speaks to the fundamental dynamic of starting a social movement or organizing a dinner party. Because I am interested in the process of coordination that occurs among people or organizations to get new things done, I will emphasize the social dynamics in numbers of three and greater. This, I argue, is where an important dimension of mobilization on a social level begins.

The more fully elaborated theory of brokerage that I present in this book offers a theoretical frame within which to understand the coordination in emergent collaborative action. Brokerage, as it is currently employed in the sociological literature, involves an intermediary (the broker) who stands between two others (or alters) who do not have a tie to each other. This “open” triad conceptualization of brokerage has a long tradition (Marsden 1982; Fernandez and Gould 1994; Burt 1992). In effect, we might characterize the classic definition of brokerage as involving two links and a gap. The links connect the broker to two parties, between whom the broker stands. The gap refers to the absence of a tie between those two alters. This brokerage formulation is important both for what it captures as well as for what it overlooks. The structural arrangement where the broker stands in between two alters is a simple yet powerful encapsulation of the relational issues central to the coordination puzzles noted above, and is of particular importance when actors seek to get new things done. This formulation is also important because of its second defining feature: the gap or disconnect between the two alters. As noted, a basic assumption in current research is that brokerage occurs exclusively in “open triads,” causing scholars to overlook its equal relevance to closed networks, an assumption that unnecessarily obscures its connection to many forms of coordination and innovation.

This limitation unduly constricts many studies with a network focus. In an influential study of brokerage, Gould and Fernandez (1989) unpacked the brokerage phenomenon by identifying five empirically distinct “brokerage roles” distinguished by the different memberships (and associated interests) that the broker and each of the two unconnected alters might have. This approach allows for the evaluation of different structurally defined brokerage opportunities, but cannot empirically address the actual brokerage processes crucial to understanding coordinative phenomena (Spiro, Acton, and Butts 2013).2 The problem with the structural approach, however, is that it is less articulate about the process crucial to understanding coordinative phenomena. I seek to explain how the brokerage triad, when combined with brokerage process, may be used to better understand how cooperation and coordinative action are induced.

Strauss points out that at its core, brokerage involves a representational act: a triadic process of “(1) a representing action, (2) with respect to some social unit, which is (3) directed at another unit (audience)” (1993, 172). In the move from two to three, the broker no longer engages with one other, as with Weick’s double interact, but instead represents the position or viewpoint of one party (or World 1) to another individual or group (World 2). This, I will suggest, is far more than a special case of concern to social network theorists but the fundamental template for the coordinative act in organizing. A corporate manager advocating for an innovation within a firm, an entrepreneur mobilizing support for a new firm, or an activist initiating collective action on the street: each engages in different variations on these activities in triadic contexts involving numbers of three or greater.

The representational act described by Strauss inherent in triadic coordination is itself complex. Strauss observes that “the potential difficulty may be that the representing unit may not represent accurately (or honestly), or be judged not doing so by the represented” (1993, 172). That a strategic actor might intentionally represent inaccurately or adjust the knowledge ferried between alter parties has been anticipated by others (e.g., Burt 1992). There is, however, a raft of other communicative moves that the broker might make in selecting what might be extracted from one community and, once extracted, shaped for consumption by and communicated to another community. Strauss’s (1993) observation that the broker’s representation might not be judged as accurate suggests a range of cases whereby the broker continually shapes representations to maximize understanding, receptivity, or the perception of value—or to engender trust in the representation, the broker, or other combinatorial actors—as a prelude to collaborative effort. Taken together, these complex communicative moves suggest opportunities to leverage a structural orientation into a deeper understanding of the mobilization process.

Coordination between individuals or organizations becomes a more or less effective process in a social space due in part to a certain social dexterity. Such social dexterity might involve attributes like a command of the social hygiene specific to a given social context, the ability to forge trust between a broad range of familiar and less familiar actors, a talent for accurately assessing the character and interests of those whom you know less well, and a capacity to make effective appeals to enlist and connect people in the causes you wish to pursue. Examples of such dexterity can be found across recorded history. A combination of advantages arising from social network position and skill account, for example, for Cosimo de’ Medici’s ability to cultivate a position of power in fifteenth-century Florence (Padgett and Ansell 1993), Abraham Lincoln’s ability to assemble support around legislation to abolish slavery (Goodwin 2005), or Paul Sach’s creation of New York’s Museum of Modern Art by assembling support from museums, universities, and finance (DiMaggio 1992).

Fligstein and McAdam define social skill as “the ability to induce cooperation by appealing to and helping to create shared meanings and collective identities” (2012, 46). They argue that social skill is an important tool for strategic actors, whether those actors are operating within existing orders or endeavoring to create new ones. Some actors, Fligstein and McAdam point out, are more skillful than others. Social skill, they further assert, requires a capacity to forge linkages within and across contexts. It also requires a general and a context-specific dimension, the latter of which may evolve if the rules and mores associated with various contexts change.

Based on their in-depth study of several historical cases, Fligstein and McAdam (2012) identify several tactics that constitute social skill, including exercising direct authority, framing action, brokering (as noted, a core focus of this book), maintaining ambiguity, aggregating interests, and networking to outliers. While this list of tactics is certainly illuminating, it also suggests an opportunity to present a more parsimonious action framework consistent with the authors’ emphasis on the strategic inducement of cooperation.

A Model for Emergent Collaborative Action

This book presents a microfoundational account of the network-based social origins of mobilization to present a theory of how new things get done. This microfoundational emphasis on small-numbers social phenomena has received increased attention in a number of disciplines including organizational theory and strategy (e.g., Barney and Felin 2013; Felin and Foss 2005), entrepreneurship (Ruef 2010), and collective action and institutional theory (e.g., Fligstein and McAdam 2012; McAdam, Tarrow, and Tilly 2001). The model I present draws on and contributes to many of the theoretical perspectives found in the organizational, sociological, and strategic literatures over the past years. The crucible of microsocial action I examine in subsequent chapters leverages work from a wide range of disciplines including social network theory, innovation studies, the knowledge-based view of the firm, entrepreneurship, pragmatism and symbolic interactionism, organizational routines, collective action, and sensemaking. If in fact we have become, or always were, a project-oriented world, a well-elaborated theory of mobilization must draw on a broad set of inputs and correspondingly strive to present insight and implications associated with this crucible of action. In this book, I hope to link these differing literatures toward that effect.

The BKAP model—named for the model’s three most distinctive features: brokerage, knowledge articulation, and projects—that I will explore in this book revolves around five key dimensions that can loosely be understood as two pairs of explanatory variables and one dependent variable, which is the mobilization of one or more focal strategic actors for some form of innovation I refer to as the creative project. The explanatory variables include brokerage structure, understood as varying primarily along an open or closed dimension (consider the open or closed triad discussed earlier); brokerage process, introduced initially from a network perspective that concerns the action by which a strategic actor leverages his or her network; along with the strategic actor’s stock of knowledge, whether rooted in experience or education; and the strategic actor’s knowledge articulation skill with which he or she communicates or articulates that knowledge for the purposes of engaging or enlisting others. The simple path model presented below (see Figure I.1) shows the basic relationships that define the BKAP model. I will briefly consider each of these variables below, and offer more in-depth treatment in subsequent chapters.

Figure I.1. The BKAP Action Model: Brokerage, Knowledge Articulation, Projects
SOURCE: Original figure.

The Dependent Variable: Innovation in the Form of Organizational Routines or Creative Projects

In most cases, networks and knowledge are harnessed for more predictable, incremental, or routine innovation. I argue, however, that the act associated with the mobilizing of uniquely new action involves a creative project, a microsocial unit of analysis I define “as an emergent trajectory of interdependent action initiated and orchestrated by multiple actors to introduce change into a social context” (Obstfeld 2012, 1571). In the latter case, the network and knowledge processes noted above may mobilize collective action in support of an innovation, a trajectory of multiple innovations, and in certain cases a newly emerging organization (or movement) catalyzed or defined by that innovation.

To present a clear understanding of the mobilization of action to get new things done, we must begin with some broader framework within which we can locate how actors engage action and, in particular, new action. All action, whether repetitive or oriented toward creating entirely new outcomes, exhibits some degree of choice or agency. Emirbayer and Mische (1998) argue that agency can be decomposed into three elements: the iterational, repetitive, or routine; the improvisational (or what they refer to as practical-evaluative); and the projective, involving the imagining of new alternative possibilities and acting on them. Taken together, these three forms of agency suggest that action may be conceptualized along a continuum characterized by the degree of repetitiveness and orientation toward the future.

Most of the time, actors are replicating or extending existing social orders and patterns of behavior. Stark information-processing constraints alone dictate that we can deliberate only on a relatively small percentage of our activity. As a result, routines and habit compose the vast majority of individual and organizational action. Even a revolutionary has routines, having to commute daily to the revolution, whether to the barricades in the city or the ramparts on the edge of town. How the revolutionary gets to work and how she goes about meeting with fellow revolutionaries bent on disruption is subject to some relatively mundane and repetitive activities. It is therefore not a surprise that one important theme in social movement theory concerns the importance of collective action routines and repertoires (e.g., Tilly 1976, 1993). As Strauss indicates, “Even in the most revolutionary of actions, the repertoire of routines does not vanish; at least part of it becomes utilized in combination with the new” (1993, 195).

A large number of innovation cases are also fairly routine. Consider the choice to roll out a new product in a product line or a new course at a university. In comparatively rarer moments, we break more substantively from routines to create an entirely new product, initiate a new protest, or from a more individual perspective, quit our job in order to start a new company or, more mundanely, switch from one coffee shop to another. Each of these actions illustrates a projective element that punctuates everyday routines. My emphasis is on those projective episodes that lead to sustained mobilization among multiple actors in the form of creative projects. To periodically mobilize new action in the form of creative projects, the BKAP action model argues that strategic actors engage the world through brokerage network structure, brokerage process, and knowledge along with knowledge articulation, the four explanatory variables which I turn to now.

Brokerage Structure

The first explanatory variable is an actor’s social network structure, defined either by the pattern of ties that surround a given actor or that connect a network of actors together. Consistent with my brief treatment of brokerage above, networks can be either “open” or “closed.” Open networks offer an actor an opportunity structure that is ideal for accessing new information but present the challenge of mobilizing disparate actors with diverse interests—a challenge I have characterized elsewhere as the “action problem” (Obstfeld 2005). Closed or dense networks, on the other hand, offer a focal actor a more interconnected network that is, on average, more homogeneous with respect to attributes and interests or more normatively constrained (Coleman 1988; Burt 1992; Obstfeld 2005). Dense networks therefore pose an opportunity structure where the surrounding network is more aligned and therefore more conducive to mobilization and coordination. That homogeneous network, however, is far less likely to be importing novel ideas and thus poses an “idea problem” associated with the greater likelihood of redundant information circulating within the network (Obstfeld 2005).

Of particular importance in tracing the distinction between closed and open networks is Burt’s (1992) theory of structural holes. Consistent with the Marsden (1982) and Gould and Fernandez (1989) conceptualizations of brokerage as an open network, Burt defines a structural hole as “a separation between nonredundant contacts” (1992, 18). Burt offers an information and control argument for the structural hole advantage. Specifically, he suggests that brokers that stand between unconnected alters benefit both from the novel information that such a structure affords and from control benefits that allow the broker to leverage the disconnected actors. Over time, structural holes theory has become the predominant conceptualization of brokerage, because the structural holes-related measures have yielded compelling empirical evidence for the impact of structural holes on dependent variables at the individual and firm levels. The structural hole construct aligns with the properties of open networks. Burt equates brokerage with the open triad that defines the structural hole, and employs a contrasting term, “closure,” to refer to the closed triad and the more closed networks with which it corresponds.

A network’s properties are also determined by the content of the ties that connect actors, whether they involve friendship, advice, information exchange, or other relational content. Given a type of tie, the second determinant of a network is the structure of those ties as they relate to one another. Our treatment of brokerage thus far assumes an unnamed tie content and a structure: a broker who stands between two others where, as previously discussed, the alters may or may not have ties to one another. There are other tie and social network constructs, whether centrality (Freeman 1977, 1979) or structural equivalence (Lorrain and White 1971; Burt 1978), that we might consider, but I will restrict myself to network structures that directly implicate open or closed brokerage structures.

Brokerage Process

The second explanatory variable, brokerage process, concerns the way that brokers, whether individuals or firms, engage their network. I will focus on three variants of brokering: conduit, where the third party relays knowledge or information from one alter to the second alter without attempting to change the relationship between the alters; the tertius iungens (i.e., third who joins), defined as a strategic behavioral orientation toward connecting people in their social network, by either introducing disconnected individuals or facilitating new coordination between connected individuals (Obstfeld 2005); and the tertius gaudens (i.e., third who enjoys or benefits), where the broker exploits unfamiliarity, competition, or conflict between parties that the broker leverages actively or through purposeful inaction (Obstfeld, Borgatti, and Davis 2014). I will use this typology to develop a more detailed theory of how strategic actors mobilize support for initiatives.3 The simplest rendering of this view is that the tertius iungens brokerage orientation presents the core mechanism by which strategic actors induce cooperation in order to get new things done. The more complex argument is that while tertius iungens-induced cooperation is the dominant theme in mobilizing action, social skill is a function of complex combinations of these three brokerage behaviors. We don’t invite everyone we know to a dinner party but, rather, selectively invite those who, as a group, will make for the best social gathering (iungens), while choosing to exclude those we judge an inappropriate fit for the occasion (gaudens).


Knowledge and knowledge articulation constitute the second pair of explanatory variables. Consider that what any actor conceives as a goal, or a means for achieving it, is shaped by what he or she knows. Alongside the connecting of actors in a social network context, the organizational and entrepreneurship literature has recognized the importance of the combination and recombination of knowledge (Kogut and Zander 1992; Grant 1996; Spender 1996; Nahapiet and Ghoshal 1998; Hargadon 2003).

Individual stocks of knowledge are critical as a starting point for the recombining of knowledge central to innovation. Cohen and Levinthal (1990), for example, argue that a firm’s capacity to recognize and assimilate information is a function of a firm’s level of prior related knowledge. Their theorization about this capacity to locate and integrate new knowledge based on prior knowledge draws on research on knowledge acquisition at the individual level. In short, an individual’s prior knowledge acquired through experience and education provides a critical resource, as well as a basis with which to evaluate and assimilate new knowledge critical to the pursuit of innovation.

Knowledge Articulation

The effectiveness of actors is also determined by their ability to communicate what they know to others whom they wish to enlist—a skill I refer to as knowledge articulation. One aspect of this communication skill involves the surfacing of tacit technical or social knowledge. Tacit knowledge refers to more unconscious, automatic, taken-for-granted understandings that are more difficult to surface or communicate (e.g., Collins 1985; Nonaka and Takeuchi 1995; Spender 1996; Gourlay 2004). In addition, in any social context where knowledge is in use, an actor must achieve intersubjective understanding of the matter at hand as a precondition to influence and enlistment. Polanyi (1958) used the term “articulation” to describe the communication of tacit knowledge but never defined it. Others have employed the articulation concept to reference the processes by which knowledge is made more explicit (Winter 1987; Benner 1994; Spinosa, Flores, and Dreyfus 1997; Zollo and Winter 2002; Dougherty 2004; Dougherty 2006). Building on this work, I define knowledge articulation as the social process by which knowledge is made more explicit, usable, or relevant to the situation at hand (Obstfeld 2001; Weick, Sutcliffe, and Obstfeld 2005). This articulation of knowledge is particularly important when people attempt to communicate some aspect of their tacit knowing for the first time, or move knowledge across a boundary in the pursuit of joint innovative action. In its various forms, articulation involves advocacy through talk (Mische and White 1998; Gibson 1999) and the active pursuit of intersubjective or shared understanding, which distinguishes it from a more general category of communication. This active orientation toward shared understanding is imperative, given the change in thinking, behavior, and even social order necessary to mobilize support around innovation.

Methodological Approach to Studying How New Things Get Done

The elusiveness and idiosyncrasy of getting new things done makes the social scientist’s task of locating and theorizing about it unusually difficult. To systematically study the social processes through which actors mobilize networks and knowledge, there are at least two empirical approaches. The first method, fieldwork or ethnography, involves direct observation and yields rich, in-depth insights about how new things begin. The value of fieldwork lies in its insight into the qualitative look and feel of previously misunderstood or overlooked phenomena. However, the risks and costs associated with direct observation are great, given the challenges of anticipating where to watch and how to watch with sufficient depth and rigor to come away with something useful to say. How do we recognize the sought-after newness when it occurs? What if an innovation is already underway when we begin our observation? If so, are we then still in a position to say something about how it got started? The central problem here is that the new doesn’t officially “begin” at a single point in time. Rather, it is always preceded by some other antecedent conditions and events. There is rarely an Adam-and-Eve moment in the progression of technological or social change.

A second approach is to select an already-large organization or movement of interest and work backward historically (e.g., McAdam 1990; Tilly 1978). Such an approach presents the scholar with the obvious advantage of selecting in advance the larger organization or movement outcome that the study of incipient conditions will predictably yield. The historical approach, however, poses other methodological predicaments, such as the danger of sampling only on successes or failing to capture the true origins or mechanisms of social phenomena begun some time ago and likely far away. The historical approach may yield synoptic clarity, but often at the expense of unpacking the precise social mechanisms central to the specific mobilization process in question. There is also important insight to be gained from combining insights derived from closer-in field observation and survey work with examples from historical analysis, and from sampling briefer cases of different kinds of beginnings. Such a blended approach yields a broadened perspective on emergence, an illumination of the twists and turns that lead (or fail to lead) to bigger things, salient founding figures with strong intentions and visions of future states, and perhaps most importantly, specific mechanisms and practices that make for getting new things done.

This book presents a parsimonious, broadly applicable theoretical framework drawn from these multiple sources to account for how new things get done. To portray how networks and knowledge are mobilized, I will draw on my own field observations, quantitative survey research (sometimes in the same setting), and other research and examples from a broad range of social phenomena and academic literatures. While I began this project with extensive ethnography of a single organization, I subsequently employed supplementary methods and analysis to derive a theory with much broader application to the range of organizing.

Field Observations Illustrating How New Things Get Done

My essential argument is that the microsocial mobilization of shared action has a set of dynamic, interlocking characteristics common to organizational innovation, entrepreneurship, collective action, and transformation of institutional fields, among others. To illustrate such a dynamic social process, I begin with a brief vignette drawn from my extensive field observation and survey study of an automotive design process (Obstfeld 2005, 2012) involving the efforts of several strategic actors, primarily middle managers, collaborating to reengineer the prototype parts-purchasing (PPP) process at a major U.S. automotive manufacturer, which I refer to as AllCar. The initial instigators of the mobilization effort were Brian, an AllCar project manager, and Dan, the program manager for the build of an entirely new vehicle, the G5.4 Brian and Dan were both long-term employees socialized into a homegrown AllCar “activist” tradition that arose from the organization’s several decades’ effort to innovate and bootstrap in the face of substantial resource constraints. The scrappy, risk-taking heirs to this self-described “cowboy” culture formed a loose network who shared an understanding and experience of how to surreptitiously leverage their network to assemble a critical mass of support before going above ground with a more formal advocacy for a given innovation. Dan once said of Brian,

Brian is a mole. He’s a gopher. He’s an underground player. . . . He’s partnering. He’s out there [burrowing] around. . . . Brian is underground with Johnson [a high-ranking manager in engineering operations]. He’s underground with Carl. He’s underground with Brad of [the] GreatCar [division].

Dan, whom I had observed for hundreds of hours, displayed the same underground collaborative behaviors. This loose cluster of cowboys had achieved notable success. In one case, for example, they were instrumental in pushing the organization to develop and employ the world’s first three-dimensional automotive design process for a production automobile.

When I first spoke with Brian, he was preparing to orchestrate a cross-division initiative to redesign the company’s antiquated, inefficient PPP process. Dan, in league with Brian, had also roughed out a somewhat synchronized scheme to leverage his senior management position and network in response to similar issues identified by the AllCar group that his G5 division was also encountering. Dan’s plan was to redesign the G5’s prototyping procedure, while simultaneously creating and staffing a new G5 PPP unit, under his direction, to operate it.

The two efforts met very different fates. Brian’s more ambitious AllCar group, after an underground planning phase, was initially successful in mobilizing broader and higher levels of support. Dan’s narrower, within-division plans, while encountering periodic resistance, ultimately succeeded as he assembled several constellations of lateral and upper-level support while skirting procedural roadblocks and resistant stakeholders. Critical to Dan’s (and the AllCar group’s) efforts was his ability to fashion a clear vision and logic through which he recruited several high-ranking executives, as well as the actual team members for the new PPP unit, before it was approved. Dan’s compelling “pitch” for forming a new G5 PPP team, which he refined in numerous meetings and backroom conversations, led to a repeated reference to the importance of “getting the right parts to the right car at the right time.” The success of AllCar’s and Dan’s efforts hinged on intimate knowledge of the AllCar community and its subgroups, the capacity to mobilize networks across corporate departments and professional ranks—a mobilization that was made possible by the protagonists’ ability to translate their ideas into compelling, well-timed appeals that often successfully enlisted support, shaped that support into collaborative efforts, and defused opposition. “Exogenous” factors also had a continual role, ultimately derailing the AllCar effort and periodically disrupting and altering, but not sinking, Dan’s initiative. Though Dan’s implementation constituted a local success (whose subsequent merit was yet to be tested), the same network dynamics were responsible for several previous AllCar innovations, some of which resulted in broad impact across the organization and in some cases the industry.

I use this illustration to bring to life, in brief form, the nature of mobilization, which I will elaborate in far greater empirical and theoretical detail throughout this book, but also to point out the family resemblance of this innovation-focused effort to mobilization efforts in other domains that also involve local action motivated by a desired future state pursued by actors skilled in enlisting and connecting others in projects of growing scope and impact. For example, similar dynamics are found in most early-stage entrepreneurial efforts. As an alternative to a research tradition that has tended to focus either on new-firm creation as an event or on heroic individual entrepreneurs, Ruef recently proposed an “emerging organization” perspective that begins “with the intuition that startup efforts . . . involve collective action that is oriented toward the founding of a new organization.” According to Ruef, “entrepreneurs, in this conception, are defined by their intention to form a social group” (2010, 7). This microsocial view of entrepreneurship is captured in Ruef’s (2010) illustration of group formation, in which an entrepreneur, Luis Hernandez, persuades two partners to start a wholesale clothing business. Much like Dan at AllCar, Luis’s initial formation of this ownership core, along with his subsequent enlistment of support, constitutes the same initial nexus of collaboration found in many start-ups. In this case, the necessary but frequently overlooked feature of start-ups, even to some extent by Ruef, is the connecting activity that entrepreneurs like Hernandez orchestrate between entrepreneurs, investors, and other sources of support (e.g., employees and advisors) that provides resources necessary to the firm’s growth.

In the social movement literature, McAdam, Tarrow, and Tilly argue that brokerage is one of the core mechanisms in collective action, where brokerage involves “the linking of two or more unconnected social sites by a unit that mediates their relations with each other and/or another site” (2001, 142). That common social processes are invoked here is no accident, though what remains to be offered is a more precise microsocial account of these social processes with application to multiple domains. Fligstein and McAdam (2012) offer a compatible grasp of the microsocial origins of institutional fields. It is such an account that I hope to present here.

Structure of This Book

The structure of this book is as follows. In Chapter 1, I introduce a more detailed account of brokerage process, which I describe as composed of three basic brokerage orientations toward action. I then explain how strategic actors employ these different action orientations, to provide a more detailed account of how mobilization is accomplished. In Chapter 2, I explore knowledge and knowledge articulation as complementary resources and skills. Collaborative action takes place within the field of social networks, but the discursive practices associated with knowledge articulation are a critical means by which coordination is accomplished. I describe the interaction between networks and knowledge articulation with ethnographic data. In Chapter 3, I provide a more in-depth explanation of the mobilization outcome—creative projects—and its contrast to organizational routines. While “getting new things done” is evocative, a more rigorous account of nonroutine creative outcomes is necessary. In Chapters 4 and 5, I provide in-depth ethnographic data to illustrate how strategic actors interweave brokerage and knowledge articulation in the pursuit of routine-based innovation (Chapter 4) and nonroutine innovation (Chapter 5) in an automotive engineering and design environment.

Having framed this basic model for the mobilization of action, I take a deeper look in Chapter 6 at the theory of using social skill to mobilize innovative action, drawing on and extending Mead’s symbolic interactionist perspective and examining insights from fieldwork regarding perspective taking and use of multiple voices (riffing). In Chapter 7, I explore how the BKAP model can be employed to explain creative outcomes in artistic movements, entrepreneurship, collective action, and several other issues in organization theory and strategy. I have also alluded here to unique challenges posed by digital tools and contexts. It is now not unusual for a layperson to assume that any reference to social networks primarily alludes to Facebook and LinkedIn. I explore how digital contexts shape and reshape mobilization processes and brokerage using the Arab Spring as an illustration. On a more practical note, I then reflect on opportunities that social skill as conceptualized here may have for addressing pressing issues associated with education, social inequality, and social mobility. I conclude by arguing, in the spirit of de Tocqueville, for the “science of association” as the master social science.

My emphasis on process and structure is reminiscent of the stance taken by Weick (1969) almost fifty years ago in his book The Social Psychology of Organizing. Weick’s book title involved a small alteration to the title of another well-established book, The Social Psychology of Organizations, by the famous social psychologists Katz and Kahn, also at the University of Michigan. Weick’s understated but clever title change suggested a shift from viewing organizations as structures to organizing as a process. In a similar approach, I attempt to consolidate a shift already implicit in some social network literature, from social networks as a structure to the brokerage processes that constitute network behavior. The objective, of course, is not to abandon the structural view, but to build a complementary, and ultimately more complete, theoretical perspective. In the past, a focus on “social networking” was often suspect as a province of hackneyed truisms and idiosyncratic ruminations developed in the absence of data or theory. Nevertheless, I argue that this process-based perspective presents valuable new insights for understanding both mobilization as it has unfolded in historical contexts, and the accelerated collaborative environments that characterize the world we live in today.


1. Taleb (2007), for example, points out that while “black swan” events are comparatively rare occurrences, they have a disproportionate ability to reshape social systems.

2. Gould and Fernandez’s (1989) foundational work unpacks the brokerage phenomenon through five fundamentally distinct brokerage types that are distinguished by the different memberships that the broker and each of the two alters might have (i.e., coordinator, itinerant broker, gatekeeper, representative, liaison). Recognizing that different parties to a brokerage arrangement hold different interests, Gould and Fernandez use subgroup memberships, whether shared or distinct, to distinguish between these brokerage categories. While it is beyond the scope of this volume to fully engage this important work, consider some of the strengths and limitations of the Gould and Fernandez brokerage scheme. First, the approach, by tracing subgroup memberships of all three parties to an open triad, provides simple, objective quantifiable distinctions to further distinguish between five variations on the open triad structure. Second, by formalizing these five variations on the open triad, the authors elucidate important implications of differing membership and alignments in these coordinative contexts. Third, the Gould and Fernandez model captures a complexity in the brokerage role that involves both the individual and the subgroups from which the different parties originate, speaking simultaneously to the microfoundations and interorganizational contexts of many brokerage cases.

There are, however, several limitations with the fidelity of the Gould and Fernandez model to brokerage phenomena that speak to the alternative empirical and theoretical approach provided in this book. First, the Gould and Fernandez model is based entirely on structural distinctions specifically with respect to the open triad (i.e., uncompleted two-paths); while their approach substantively advances brokerage analysis by allowing for the evaluation of different structurally-defined brokerage opportunities, it is agnostic to the specific brokerage processes involved (Spiro, Acton, and Butts 2013). Second, Gould and Fernandez’s cross-sectional design does not allow for variation in the social process within those structures or their empirical consequences as brokerage process evolves over time. Third, Gould and Fernandez assume clearly defined membership roles when brokerage roles are, almost by definition, ambiguous, ill-defined, multiple, hidden, and evolving. Finally, Gould and Fernandez assume that brokerage, by definition, involves a non-tie between alters. Despite the importance of the non-tie condition, I argue that brokerage can also take place in the presence of the alter-alter tie, a condition that the predominant definitions of brokerage rule out.

3. Tertius iungens (YUNG-gains) is based on the Latin verb “iungo” which means to join, unite, or connect. In early Latin, it means literally to yoke, harness, or mate and serves as the root of such modern words as junction, conjugal, and yoga. In one context it is used in the phrase “to throw a bridge over a river.” In later Latin, it seems to be used in a more metaphorical sense, “to unite” or “to form” (as in a friendship). Cicero used the phrase “iungere amicitiam cum aliquot,” that is, “to form a friendship or alliance with another.” For more on the origins of tertius gaudens, see Burt (1992). Both of these orientations will be discussed in great depth in the next chapter.

4. All persons’ names, organizational names, and acronyms in this book are pseudonyms.