Abstract
Epinets is a demanding and brilliant book. It demands and deserves from its audience a very close read. Its theoretical logic builds “line upon line, precept upon precept,” and so this is not a book to be blithely perused. It also demands much of itself. Moldoveanu and Baum not only engage in an act of intellectual brokerage between epistemic game theory (and related fields) and network analysis to introduce what they refer to as epinets (networks of agents’ beliefs); they also seek—staying with Burt’s (2005) theory—to “seed and catalyze closure” (p. 162) among diverse researchers committed to the epistemic turn in social science that they propose.
Scholars from several fields should find much value in their work. These include, first and foremost, network researchers looking for fresh ideas and new methods but also organization theorists more broadly defined, as well as game theorists, strategy researchers, sociologists of knowledge and of religion, and even students of military intelligence. One of the most interesting discussions we had about Epinets took place with a German intelligence expert whose attention was riveted by the book’s core claim: that what you think others think (and what you think they think you think) matters decisively for strategic behavior. Like a shrewd spy who inserts herself in the learning loop of her country’s enemies’ spies, Moldoveanu and Baum’s ideal social actor is an embedded (though not constrained) actor who has much “level 2” knowledge—she knows what others know—and “level 3” knowledge—she can accurately predict what others think she thinks.
Valuable ideas and leads mark each chapter. Weick (1996) famously asked organization theorists to “drop their tools” if they wish to gain new insights, but many have been reluctant to do so because of the lack of new tools. Moldoveanu and Baum make a case in chapter 1 (p. 3) for a new “tool kit for modeling, measuring, and manipulating” epinets using a series of intriguing vignettes, and in chapter 2 they discuss their tools in detail. They illustrate their epistemic description language (EDL) in a way that is rigorous and accessible, yet not pedantic. We see their tool kit—intuitions, concepts, language, and methods—not as forcing a full-scale abandonment of existing tools but as complementing existing approaches and, in some cases, forcing researchers to soberly rethink their use of those approaches.
For instance, in chapter 2 the authors critique the use of Bonacich’s (1987) centrality measure as a means of capturing status, arguing that it may miss important semantic elements. Standard network measures are often insensitive to the types of connectivity that produce the most important differences in status: for example, two individuals may be the subject of roughly equal amounts of positive chatter among respected peers, but one of the two may be more closely affiliated with those who are known for being known—having what Moldoveanu and Baum call “renown.” There is a difference between being generally respected in Hollywood and being a prominent member of the Ocean’s Eleven cast: Matt Damon (pre-Bourne) perhaps gained an added boost in status over an otherwise near-peer by associating with Brad Pitt, an actor the authors discuss as renowned.
Importantly, the set of tools Moldoveanu and Baum provide will prompt their readers to engage in their own critiques of existing network methods. Consider again the case of two individuals with equal status in the network-analytic sense, say, two senior executives, A and B, who both receive much deference from their highest-potential subordinates. Eigenvector-based measures of centrality will miss important differences that are brought into relief by an epistemic approach. What do A and B really think about the deference accruing to each of them? Does A quietly concede that the deference he receives is mere herd behavior (others are “supposed” to show respect for A), while B knows that the respect she receives is authentic? If so, B’s “status footprint” has a width and depth (with associated implications for B’s behavior) that A’s lacks. Higher-order data of the kind Moldoveanu and Baum highlight can uncover and explain important differences that are obscured by typical approaches. Perhaps most deeply, Moldoveanu and Baum ask us to rethink what actually constitutes a meaningful social tie: beneath the surface of easily observed ties—based on who collaborates with whom, who seeks advice from whom, and so forth—exists a network of beliefs about what others believe that merits close attention in new research.
Chapters 3 to 6 offer engaging material that builds on the foundation laid in the first two chapters. Chapter 3, on the epistemic antecedents of coordination and co-mobilization, is marked by artful storytelling—an account of a dyad composed of Alice the economist and Bob the programmer is a particularly effective reminder of how differences in interpretive frameworks can hinder coordination—and an insightful analysis of an epinet of faculty members. Here, readers will gain a clear sense of how to think about different “levels” of knowledge in an accessible setting. Chapter 4 addresses an important question: “What must agents know or believe about what those whom they trust know or believe in order for their relationship to be one of trust?” (p. 94). This chapter offers one of the best theoretical discussions of trust we have seen. The distinction between trust in competence (I am confident that A will know Z if Z is true) and trust in integrity (I trust A to assert Z if Z matters) is valuable, but it would also be interesting to know if trust in integrity can ever mean trusting A to lie about Z. Can epinets help us understand a situation such as what occurs when an antebellum sheriff interrogates a farmer, asking him if he is hiding former slaves in his attic on their way to freedom? Given Moldoveanu and Baum’s interest not just in rational trust but in moral trust as well, addressing this Augustinian question strikes us as an intriguing next step.
Chapter 5 examines changes in epinets and the strategic actions agents take to foster such changes. Moldoveanu and Baum use the compelling example of a CEO ouster to illustrate these dynamics. Their approach captures complex information-diffusion processes on a fine-grained level. The rest of the chapter will be of interest to students of politics and power in organizations. The authors point out various moves agents can undertake to shape epinets in their interest, such as divisive moves that reduce the commonality of knowledge and create information asymmetries that favor agents and distortive moves that create doubt or misinformation that can result in frictions, again furthering the agent’s benefit. The authors describe these moves effectively.
There are a few areas of the book that could be better executed. We would like to have seen more engagement with the epistemology literatures, a clearer sense of how believing and knowing differ, and more speculation about mechanisms underlying observed patterns. For instance, why did the clique of faculty members in chapter 3 do better than the overall network in terms of level 2 knowledge coherence? Perhaps the equal betweenness-centrality and eigenvector scores of the members of this clique (their structural equivalence) fostered an uncommon degree of competition (and mutual monitoring) broadly in keeping with Gould’s (2003) discussion of social symmetry. But these are relatively minor quibbles. Epinets is a remarkably smart and creative book that is likely to shape the intellectual landscape far into the future. We strongly encourage others to read and especially to use it.