In the last decade, we have made great progress in recognizing patterns in the use of state-sponsored mass murder and other life integrity violations. As a result of this body of work, policymakers now have better tools with which to predict massive human rights abuses, and fewer excuses to hide behind when confronted with potential or ongoing atrocities. Yet much more needs to be done. On the policymaking side, mass killings continue unabated, with few international actors willing to address them head-on. On the academic side, we have spent so much time and intellectual capital on the structural factors that allow, encourage, exacerbate, or inhibit atrocities that we have often neglected the role of the perpetrators themselves. The two books reviewed here take on this deficit in the literature. These important new books convincingly argue that in order to understand and address the most egregious human rights violations, we must begin with those responsible for devising and implementing these murderous policies.

Rejecting the structural approach—that factors such as sociopolitical structures and/or upheaval best explain or predict atrocities—each starts from the premise that large-scale human rights abuses on the order of mass killings are policy choices made by a small group of elites. Leaders make rational but horrific calculations about whether to employ such atrocities. These are but one set in an arsenal of tactics that leaders may choose to employ when faced with threats, policy problems, or agendas to implement, and are used when they are deemed “useful” or optimal. From this assumption, each author sets out to understand why such atrocities occur, and begins the process of rethinking how to address them.

In Final Solutions, Benjamin A. Valentino examines the “strategic logic” of mass killing. He argues that powerful actors employ mass killing when they come to believe that this particularly abhorrent policy option best resolves a political problem that they face. In most cases, mass killing is used by a small group of decision makers to solidify power, eliminate threats (real or imagined), or advance a political agenda, but only after they have come to believe that alternative options are neither practical nor feasible.

Valentino lays out the strategic logic of mass killing at length and proceeds to examine in separate chapters three different types of cases—communist, ethnic, and counterguerrilla mass killings—each with its own unique and deadly logic. In each chapter, relevant cases of mass killings are subjected to thorough historical process tracing in order to highlight the role of the elite decision-making calculus. In each chapter, the author also briefly discusses cases in which mass killings did not occur. In the chapters on communist and ethnic mass killings, he finds that radical, exclusionary ideologies that call for leaders to “fundamentally reorganize society at the expense of certain groups” all too often yield mass killing (p.153). In the chapter on counterguerrilla mass killings, he concludes that counterinsurgency leads to mass killings because the tactic “may appear to offer the last chance for victory at an acceptable cost” to regimes faced with guerrilla insurgents (p. 233).

Valentino notes that the implications of his findings are twofold: First, successful intervention to prevent mass killing is possible; second, intervention to prevent mass killing can be done better than it has been done recently by the international community. He suggests becoming more proactive by monitoring regimes that he identifies as most likely to see mass killing as a viable policy option—those trying to implement radical social changes that dispossess large numbers of people, and those facing guerrilla insurgencies. Ironically, this suggests that some of the findings Valentino had rejected earlier regarding structural factors, in particular the focus on the presence of major sociopolitical upheaval, may not be so easily dismissed after all. Next, he argues that since a small group of elites is responsible for planning and implementing mass killings, interventions to prevent or halt mass killings should focus on removing them from power. Indeed, this conclusion has found support in recent quantitative research on intervention in genocides and politicides, and may be an important point of departure in rethinking how to prevent or react to instances of mass killing.

While an important, well-written piece of scholarship, Final Solutions does have some flaws worth noting. The cases of mass killing examined are well researched and expertly process-traced, but the cases where no mass killings occurred are addressed briefly and in far less analytical detail. An exception is the chapter on ethnic mass killings, where murderous regimes are more carefully compared to most similar cases, their immediate predecessors, which were often dramatically less lethal. Unfortunately, the other two case chapters are not nearly as rigorously comparative. Given that one of Valentino's primary criticisms of the literature is that it has “focused narrowly on a few cases in which mass killing occurred and neglected the many other cases [that] did not lead to mass killing” (p. 7), a more thorough comparative analysis would have strengthened his already convincing argument.

Valentino is rightly eager to make the case for the contribution of the “strategic logic” framework, but he appears dismissive of alternative approaches. While much of his early discussion of the structural approach is on target, he is too quick to dismiss the relevance of this important approach. Yet by the end of the book, it becomes clear that ultimately, structural factors help determine which regimes are at greater risk for mass killing, while perpetrator decisions ultimately determine who among that subgroup will actually choose the murderous policy. Also, in responding to arguments that mass killings on the scale of genocide can only occur if substantial members of the population support or engage in the killing, Valentino argues that evidence from Rwanda suggests that “less than 9% of the male Hutu population over the age of 13” directly participated in the killings (p. 37). Yet when almost 1 out of every 10 males actively engage in the slaughter of their neighbors (not to mention the large percentage of bystanders), one can hardly argue that leaders act without the support of or murderous action by substantial elements of society. Such evidence calls out for an analysis of both the principals and the agents of mass killing.

These few problems associated with Valentino's otherwise excellent book are addressed at least in part in another important work, Neil J. Mitchell's Agents of Atrocity. In this remarkably readable book, Mitchell applies a principal-agent framework to the policy area of massive violence targeted at civilians during civil wars. In doing so, he gives an appropriate nod to the importance of structural factors as context for these murderous choices, while still emphasizing the role of perpetrator decisions and policy implementation.

Mitchell argues that leaders use violence against civilians or prisoners during civil wars because of either “the self-interested pursuit of power” or “the intolerant logic of a divisive belief system” (p. 3)—arguments that echo Valentino's arguments. However, Mitchell reminds us that there is a third motivation for atrocities that is missed if one concentrates only on the principals and ignores the agents. While principals devise the policy, agents do the killing. Thus, we must also consider “the selfish gratification” of those who actually do the killing, as well as the ability or willingness of principals to restrain these agents (p. 5).

Mitchell engages in a more balanced comparative analysis of three cases of civil wars, all severe threats to the regime's existence, but with divergent outcomes. Three carefully documented, historically rich case chapters examine the English and Russian civil wars and the Israeli-Palestinian conflict. In examining the Russian civil war, the author demonstrates how the exclusionary ideology of the principals, combined with loosed agents, yielded the worst of the three examined sets of atrocities. In the Israeli-Palestinian case, he shows how the Machiavellian pursuit of power and security can lead to serious but less extensive atrocities, and demonstrates how principals can use agents to carry such atrocities out. In the case of the English civil war, Mitchell shows how ideology of the principal and the careful selection and control of agents can limit atrocities, even during a civil war, but only if the principal is committed to restraint.

Agents of Atrocity suggests the potential impact of both outside actors and of “tolerators” or “positive” leadership within affected countries. Mitchell's prescriptions involve ensuring that principals can better monitor agents, and using the already-existing international network of human rights nongovernmental organizations to hold both principals and agents accountable for their actions. While perhaps a bit idealistic, his suggestions, coupled with stronger international action suggested by Valentino and others, might just “reduce the odds of bastards being bastards” (p. 189).