The two books under review differ sharply on the fruitfulness of rational choice theory. Karen Cook, Russell Hardin, and Margaret Levi posit that even trust is best understood as a product of rational, materialist calculations. On their “encapsulated interest” view, trust “exists when one party to [a] relation believes the other party has incentive to act in his or her interest or to take his or her interests to heart” (p. 2). But can incentives truly whisper to the heart? Michael Taylor decries the baleful consequences, both intellectual and practical, that stem from assimilating all reasoned decision making to the numerical weighing of material incentives and ignoring the role of heartfelt personal and moral commitments.

Cook, Hardin, and Levi find it most useful to define trust as resting not only on incentives but specifically on those incentives arising from the desire for repeated personal interactions. The major contention of their book is that trust in this sense can play only a limited role in sustaining cooperation in contemporary mass societies. Even allowing for some transitivity of trust across social networks, the individual in a contemporary society must cooperate with multitudes of people for whom a future of prospective dealings with the individual or that person's trusted associates casts no shadow. Various substitutes for trust are thus required if cooperation is to proceed. The authors consider many examples, among them fictive kinship relations employed by Kenyan pastoralists, associations that license professionals such as doctors or lawyers, and state organs of coercion and regulation.

The sheer variety of these topics is a problem. The terrain of alternatives to trust is vast. Even a selective survey of this terrain yields conclusions of such extreme generality as to be almost devoid of content. Doubtless the authors are on safe ground in stating that “distrust and lack of trust are more likely than trust … to motivate us to create the institutions that allow us to cooperate with each other and treat each other as trustworthy” (p. 186). However, this verity offers little guidance into launching an investigation of concrete empirical situations.

As a result, reading the book's many short treatments of the dilemmas of cooperation in various realms is a frustrating experience: They fail to cumulate and are generally too abbreviated to be satisfactory on their own terms. For instance, in illustrating how large-scale social transitions destabilize established structures of cooperation, the authors contend that the U.S. Civil War eliminated “[t]raditional patterns of reciprocity,” and “slaves once entrusted with children and valued property became subjects of distrust and fear” (p. 169). Whatever its (dubious) empirical merits, the idea that slavery enabled reciprocity and trust between owners and owned flatly contradicts the authors' own earlier analysis of how power asymmetries undermine trust.

While this puzzling vision of slavery serves just a minuscule expository role in the text, the failure to deal consistently with the implications of expressed theoretical views is a broader problem. Cook, Hardin, and Levi argue that actors are always interrogating the contours of counterparties' interests, always looking forward, always assessing whether those in whom they place trust will continue to face a set of incentives prompting trustworthy behavior. Trust is accordingly specific and situational. But another core argument seems to belie the point. The authors repeatedly claim that enforcement-backed formal institutions are an important precursor to trust, “encourag[ing] citizens to take the small risks with others that facilitate their learning who is reliable, even trustworthy, and how to distinguish the reliable from those who are not”(p. 159). However, on an incentive-based view, this argument is incoherent: Loaning you a ballpoint pen gives me no information on whether I could expect return of a loan of a million dollars, the incentive to abscond with which is of an entirely different magnitude. The whole small-risks argument makes sense only if there are fundamentally trustworthy and untrustworthy people whose type can be detected on the basis of their past record. Elsewhere, though, highlighting such backward-looking judgments is explicitly criticized.

Even the materialist microfoundations that ground the whole approach are not carried through consistently. The authors accept that humans are psychologically complicated creatures and not incentive-calculating automata. We are given to “cognitive paranoia” (p. 50), which can prompt blanket hostility to those outside our trusted group, or provoke us to sabotage the goals of overvigilant employers. We often ignore information that contradicts our prior convictions. Yet people “driven primarily by values rather than by interests” would be “an odd breed of human” (p. 115). There is no explanation given for why paranoia and bias are less odd than moral values. Indeed, the authors even concede that trustworthy behavior may sometimes be motivated by moral commitment or personal disposition.

Despite its intellectual amorphousness, Cooperation Without Trust offers readers some rewards. There are vigorous and convincing discussions of the limitations of legal regulation and of the conflicts of interest behind the U.S. corporate scandals of the early 2000s. Stimulating discussions on the limitations of overly general views of trust and a related reformulation of “social capital” are also worthy of attention. However, failures of synthesis and consistency render the whole less than the sum of its parts. While those interested in the topic will wish to sample this volume, teachers of the “encapsulated interest” theory would do better to employ Hardin's more instructionally focused treatment (see his Trust, 2006).

Taylor's book is an impassioned attack on “Rational Choice” theory—capitalized to emphasize what he sees as unfortunate capture of the language of rationality by a viewpoint that actually rejects its importance. A truly rational choice, on Taylor's view, is a reasoned one, in which we reflect upon the implications of our foundational commitments and whether they must override desires. And this is often how we do make decisions. When we are deeply connected to people, to our self-identification as moral, or to places—as, say, some Native Americans are to some traditional fishing grounds—we are not willing to sacrifice this connection even to satisfy desires that we would act on in other circumstances. To conceive of the loss of such connections as a negative number is to imply that there is some positive number that could cancel it, such that “anyone is compensable for any loss” (p. 48). Thus, theorists who take “Rational Choice” to be an arithmetic operation in which numerical pros and cons are mechanically totaled are, in practice, advocates of “an ideology—perhaps the ideology—of disconnection, of disintegration” (p. 87), according to which every connection has its price. The ideology takes especially destructive form in cost–benefit analyses, which, as Taylor relates, have been used to justify horrible depredations against aboriginal populations and the natural environment.

One should not get the impression that Taylor is offering a solely normative argument. He draws on substantial empirical support as well to ground his understanding of human beings as regularly disdaining incentives in favor of the commitment to their moral selves, to their habitats, to fair reciprocity, and to communities. His view helps explain why experimental game theory finds that people cooperate much more often than payoff matrices would predict. He also gives a convincing explanation of research that demonstrates how positively employees react to being treated with trust rather than suspicion. (Dealing with some of the same research, Cook, Hardin, and Levi have a much harder time fitting it into their framework.)

With its clear and lively writing, intelligent argumentation, and forceful point of view, Rationality and the Ideology of Disconnection should be an excellent conversation starter in a seminar setting. Controversy may center on whether Taylor is justified in directing so much fire at rational choice theory. Critiques of disconnection, of a system with, as Marx famously put it in his 1848 Communist Manifesto, “no other nexus between people than naked self-interest, than callous ‘cash payment’” well predate the articulation of neoclassical economics. Taylor's vivid portrait of the effects of markets on community and the environment resonates with this tradition, which of course has made critical use of incentive-based analyses. The author himself suggests that rational choice theory is more legitimation than motivation for the sort of actions he decries, and that other intellectual justifications have been offered in the past.

In any event, the idea that there are no irremediable losses is not a necessary implication of portraying choice as driven by incentives—one can either employ ranked rather than numeric payoffs, or assign a negative payoff so large that it cannot be compensated by anything possible in a particular game. Nor are the interpersonal comparisons of utility involved in cost–benefit analysis sanctioned by rational choice theory. The author is aware of these points, but his responses are not entirely satisfactory.

Taylor's positive intellectual program of offering alternative microfoundations is fully developed, but his positive political program is less so. His rejection of cost–benefit analysis, as incompatible with the importance of regarding certain things as invaluable, risks degenerating into an assertion that there can be politics without hard choices. Deliberation may make people “more likely to consider what ought to be done for the common good” (p. 79) but the oft-posed question of how to deal with dissent over the common good is not addressed. And what is there to do when what is invaluable to me is incompatible with what is invaluable to you? One does not have to support a soulless technocracy to worry that exalting the moral significance of what we consider unique and irreplaceable could have pernicious effects of its own.

Taylor attacks rational choice theory with the fervor of an apostate. As he notes, he was previously a prominent practitioner of the approach himself. When confronted with evidence that people were far more willing to disregard incentives than he had previously held, he reports, his initial inclination was to argue that a numerically understood rationality still ruled human acts, but its reign might be limited to situations of relatively high stakes. In this book he abandons even this retreat, in light of the evidence he presents that quite massive incentives have often proved impotent against convictions. On the other hand, Taylor makes clear that he accepts that incentives do sometimes guide our acts. Indeed, it would be surprising if they did not, given the many institutions (including the cooperation-sustaining ones described by Cook, Hardin, and Levi) that are engineered precisely to steer human behavior through the manipulation of incentives. For their part, these authors note that materialist considerations are not the only foundation of trust. These mutual concessions suggest that to the extent the battle over rational choice theory is a search for the ultimately “correct” microfoundations of action, it is futile. Conviction and cupidity will both have their day, and we would all benefit from regarding them as mechanisms (see Jon Elster, “A Plea for Mechanisms,” in P. Hedström and R. Swedberg eds., Social Mechanisms: An Analytical Approach to Social Theory, 1998, pp. 45–73), rather than final answers about human nature.