The problem with trying to punish an institution that is judged to be “delinquent”—whether a “rogue state,” the United Nations, BP, or the United States Army—might be understood as one of responding to an entity that (to invoke Edward, First Baron Thurlow's eighteenth-century account of the corporation) “has no soul to be damned and no body to be kicked.”Footnote 1 Perhaps this seems a fairly obvious point. After all, even if one can draw some carefully qualified analogies between individual human actors and institutions (as I will attempt to do in the first part of this article), the two types of entity are different in important ways. One might thereby conclude that the corporeal—and, depending on one's beliefs, even the spiritual—nature of individual human beings renders them vulnerable to forms of punitive harm to which institutions, in the sense of formal organizations, are simply impervious. Alternatively, one might counter that such an observation has little relevance when we are talking about “delinquent” institutions in international relations. We do not, one might argue, need to be able to anthropomorphize formal organizations in order to be able to punish them. Indeed, we frequently justify actions toward states, multinational corporations, and intergovernmental organizations—including economic sanctions, boycotts, imposed reparations, fines, “naming and shaming,” and even dismantlement—in terms of punishment, and these actions can arguably serve quite successful deterrent, retributive, and even rehabilitative functions.
In this article, I want to take a path somewhere between these two responses to the idea of punishing institutions. The distinction between individual human actors and corporations stressed in Baron Thurlow's statement is, indeed, a fairly straightforward one. Nevertheless, it is also a distinction the implications of which are largely ignored when we make calls to punish institutions. This distinction does not preclude the possibility of punishment at the level of the institution, but it does point to significant conceptual and practical complexities in how we can coherently respond to formal organizations that are seen to evade their moral responsibilities in international relations.
I have argued elsewhere that formal organizations can be moral agents.Footnote 2 They can therefore be bearers of duties.Footnote 3 They can also be objects of blame when these duties are not discharged. While this article begins by surveying the terrain of this argument—an argument that I feel fairly confident with, despite its arguably radical bentFootnote 4—it then turns to a proposition that, for me, elicits both hesitation and worry: that we can punish institutions for abrogating their moral responsibilities. Even more daunting than determining which institutions in international relations can be assigned moral responsibilities and burdened with blame (and in what circumstances) is the task of responding to the delinquent institution once it has been blamed for a particular act or omission. Especially problematic is judging how (and if) an institution can be punished in a way that does not punish its constituents as individuals. The central aims of this article are, first, to outline a conceptual framework for blaming an organization at the corporate level for harm or wrongdoing; and, second, to highlight the grave difficulties in turning to punishment as an appropriate response to such institutional delinquency.
This article is divided into four sections. In the first section I outline what I call a “model of institutional moral agency.” This model not only attempts to define criteria that a collectivity, or group, must meet in order to qualify as a moral agent but also endeavors to define those circumstances in which an institutional moral agent can be expected to discharge specific duties. The purpose of this section is to establish why it makes sense, conceptually, to talk about blame and punishment vis-à-vis certain collectivities, even if the attempt to introduce punishment in practice is accompanied by a plethora of difficulties. In the second section I draw on this model of institutional moral agency in order to clarify what I mean by a “delinquent” institution. The third section contains a preliminary discussion of problems involved in attempting to punish such institutions. Specifically, I address three types of concern: “guilt by association,” “misdirected harm,” and “overspill.” In the fourth section I illustrate these problems by turning to the danger of harming innocent individuals while ostensibly punishing delinquent states through organized violence. To conclude, I offer some points for future development on two themes: the relative culpability of individual members of delinquent states that have democratic versus nondemocratic decision-making structures; and the logic of nondistributive, as opposed to distributive, forms of punishment in response to institutional delinquency.Footnote 5
A MODEL OF INSTITUTIONAL MORAL AGENCY
In international relations, assertions of moral responsibility are commonplace. These assertions take two forms: claims of prospective moral responsibility, in which ex ante judgments are made regarding tasks that an agent ought to perform given certain conditions; and (necessarily related) statements of retrospective responsibility, involving ex post facto assessments of a particular event or set of circumstances for which an agent's acts or omissions were such that the agent is deemed deserving of moral praise or blame. For example, when confronted with the possibility of crises in forms as diverse as financial collapse, famine, genocide, terrorist attacks, and environmental destruction, we speak passionately of duties to avert or mitigate disaster: “Something must be done!” In the wake of such crises, we demand to know who acted wrongly, whether maliciously, recklessly, or negligently. We ask who should have acted but did not, or who should have responded more quickly, more efficiently, or more robustly. We point fingers and apportion blame. We call for the guilty to be held to account. We also, frequently, insist that someone—or something—be punished.
The important detail that is often overlooked, however, is that any meaningful assertion of moral responsibility requires that those who are called on to uphold duties, and those who are held to account for failing to discharge them, must be moral agents—entities that, by definition, possess capacities to contemplate, recognize the significance of, and ultimately execute different courses of action in the first place. Divorcing difficult questions of moral agency from those assertions of moral responsibility regularly voiced in international relations variously results in incoherent policy-making, the effective evasion of duties despite nominal calls to action, the creation of “scapegoats,” and, perhaps most relevant to the current discussion, the tendency for punishment to be tenuously justified and carelessly directed. In short, in order to be able to speak coherently about moral responsibility in international relations, we need a clear understanding of moral agency. Moreover, moral agency cannot be the exclusive realm of individual human beings unless some actions and outcomes are to be left beyond comprehension and critical evaluation. Of course, this proposed move from individual human beings to groups as the bearers of moral burdens must be carefully examined.
Is it possible for some groups to be moral agents? According to what criteria would they qualify as such, and under what conditions can duties be assigned to them, or blame be attributed? As we understand specifically moral agency first and foremost in the context of individual human beings, this seems a logical starting point in confronting these questions. Although philosophers rely on quite different standards in identifying those specific features of individual human beings that define them as moral agents, they agree, in general terms, that to qualify as a moral agent the individual must possess capacities both for understanding and reflecting upon moral requirements and for acting in such a way as to conform to these requirements. Moreover, and importantly, for the individual then to exercise moral agency, and be able to discharge specific duties, he or she must also enjoy the freedom to do so. Exercising moral agency requires that one not be constrained by structures, or other agents, in a way that prevents one from acting.Footnote 6
The question of when—if ever—a group is a moral agent has been largely neglected by theorists of international relations. Writing in the area of business ethics, Peter French challenges what he identifies as an “anthropocentric bias” in our moral reasoning and aims to illustrate that the corporation can be a “moral person.”Footnote 7 Drawing on parallels between the capacities of individual human agents and those of institutions, the moral philosopher Onora O'Neill argues that institutions can also be agents for whom ethical reasoning is both accessible and action-guiding.Footnote 8 Adding to and elaborating upon French's account of those groups that are “full-fledged members of the moral community,”Footnote 9 and inspired by O'Neill's “thin theory of institutional agency,”Footnote 10 I propose that a group qualifies as a moral agent if it possesses five characteristics. First, a group must have an identity that is more than the sum of identities of its constitutive parts; it must have what might be called a “corporate” identity. Second, to be a moral agent a group requires a decision-making structure. This not only enables the group to deliberate but also provides a means of arriving at a goal or policy that is more than simply the aggregate of individual aims and intentions. A third requisite feature is an executive function linked to this decision-making structure, which allows it to translate decisions into actions and ensures that policies can be implemented. Decision-making procedures and structures for carrying out the resulting resolutions come together to ensure that a group has a capacity for purposive action. Fourth, the group must have an identity over time. Finally, to be a moral agent a group must have a conception of itself as a unit—that is, it cannot be merely externally defined.Footnote 11
I refer to groups that meet these criteria as “institutional moral agents.” This terminology is significant. The label “institution” can, of course, mean different things, and I use it here specifically in the sense of a formal organization, or what might be called a “structured institution.”Footnote 12 Yet even while I am focusing on institutions as formal organizations, another connotation of the term draws attention to the norms, rules, procedures, practices, and cultures that frame and channel the decisions and actions of individual human actors within these organizations. The label institutional moral agent thereby highlights why certain types of groups reach decisions and act in ways that cannot be described simply in terms of the sum of decisions and actions of their individual members. Such collectivities as mobs, crowds, markets, nations (unless they have their own political structures and decision-making bodies), and the often-invoked “international community” do not qualify as institutional moral agents. Examples of groups that do qualify include Hamas, Amnesty International, Microsoft, the Catholic Church, Aberystwyth University, and the World Bank. Moreover, the state, which will be central to the subsequent analysis of punishment, also qualifies as an institutional moral agent.Footnote 13 As institutional moral agents, these bodies can reasonably be expected to bear duties, and can be blamed for the acts or omissions that derogate from these duties.
Three points of clarification should be offered here to avoid misunderstanding about the proposal that institutions can be moral agents. The first responds to the criticism that I am ascribing to institutions features of the individual human actor that one could not realistically expect an institution to possess. I am not suggesting that institutions are the same as individuals. Indeed, one of the starting points of this article is that they differ in significant ways that have a direct bearing on discussions of punishment. I am simply suggesting that institutions share with individual human beings certain capacities that allow both to be considered moral agents.
The second point is that the moral agency of institutions in no way precludes or undermines the moral agency of those individual human actors, and subgroups, that comprise them. Moral agency is understood here to exist simultaneously at different levels. This means that the assignment of duty or the apportioning of blame at the level of the state, for example, does not allow those multifarious agents within it (such as individual citizens, a particular government or administration, the state leader, or its military organizations) to evade either moral expectations or censure for those discrete actions that are ascribable separately to them. Rather, moral agents at all levels can be responsible for concurrent, complementary, or even coordinated acts and omissions.
Finally, the model of institutional moral agency offered here says nothing about the substance of the moral demands to which formal organizations might be expected to conform. There is no attempt being made to adjudicate among different sources of obligation, among different codes of conduct, or among different accounts of morality in international relations. This must, I think, be a separate project, and one upon which the current endeavor need not rest. As has been argued, claims are frequently made that institutions in international relations are morally responsible for certain failures. They are variously held to account, blamed, and made targets of attempts at punishment (when, for example, a state that fails to adhere to international human rights norms of racial equality is ostracized within the international community, a clothing manufacturer is boycotted for using child labor, an oil company is fined for negligence that leads to environmental catastrophe, or, indeed, when war is waged against a state that is deemed to have supported terrorist operations). The value of this model of institutional moral agency is to address the coherence of apportioning blame to these types of corporate bodies and, by extension, to evaluate the logic of particular practical responses to alleged institutional delinquency. Without making assertions about the source, nature, or force of perceived duties in international relations, one can nevertheless ask who—or what—can respond to what we understand to be ethical imperatives, and who—or what—can be blamed for their breach in the context of specific acts or omissions. One can then go on to ask whether there is agreement between these objects of condemnation and the bodies that we, in fact, harm in our attempts to punish perceived abrogations of responsibility and gross injustices.
Having offered this final caveat, it is not the case that we are left without any moorings in discussions of duty. On questions of obligation in international relations, it is possible to point to principles that represent areas of near-universal agreement. Following Mervyn Frost, one might refer to these principles as “settled norms,” defined not by their universal observance but, more soberly, by the perceived need to provide special justification for any attempt to either deny or override them.Footnote 14 The principle of noncombatant immunity, according to which noncombatants are illegitimate intended targets of organized violence, is a good example of such a norm.Footnote 15 Furthermore, in cases in which states sign up to codes of conduct, such as the 1948 Genocide Convention or the 1984 Convention against Torture, certain acts (such as the abuse of prisoners at Abu Ghraib) and omissions (such as the failure to intervene in the Rwandan genocide) would seem to provide uncontentious examples of derelictions of duty.Footnote 16 Of course, identifying the appropriate objects of blame in cases where there is widespread agreement that a particular moral responsibility has been abrogated, and deciding how best to respond to this culpable body, remain complex issues. I will begin to confront these issues in the following sections.
DEFINING DELINQUENCY
Just as formal organizations possess the capacities to address injustices and respond to crises in ways that individuals on their own cannot, and might thereby be assigned responsibilities that could not be borne by individual human actors, formal organizations can also be blamed for wrongdoing, misjudgment, neglect, or harm that is not attributable on the same scale to particular individuals within the organization. In other words, when a relevant act or omission that has led to moral failure is best described at the corporate level, the institution as a whole can be blamed. Yet, to return to the account of individual human agency introduced at the start of the previous section, this statement must be qualified somewhat, and in a way that is deeply significant for addressing the proposed category of “delinquent institution” with respect to actual institutions in real-world contexts. Perhaps the best place to begin confronting the question of when an institutional moral agent can reasonably be labeled “delinquent” is with cases in which such actors fail to respond to what are understood to be moral imperatives and yet cannot coherently be blamed for their apparent transgressions. From there it is possible to move on to cases in which institutional moral agents that do not discharge particular duties can reasonably be held to account.
When Duties Cannot Be Discharged: “Weak” and “Constrained” Institutions
Importantly, fulfilling the criteria for institutional moral agency is not enough to determine that a particular body can be blamed for failing to respond to a moral imperative in a specific set of circumstances. Even those entities that we can call moral agents cannot be expected to discharge duties for which they are unable to perform requisite actions. As with individual moral agents, institutional moral agents might be prevented from responding to particular ethical demands for two reasons. Both reasons are extremely important to how I will go on to define institutional delinquency.
First, the organization in question might possess capacities for deliberation and action that are limited in some respects. A group can qualify as a moral agent if such capacities are apparent, even if they are also restricted. Indeed, each of the criteria posed above for determining which groups qualify as moral agents can be met in degrees. This can be illustrated if one applies the model of institutional moral agency to states. Although so-called quasi-states, or states that lack positive sovereignty, tend to have, for example, weak decision-making structures,Footnote 17 they can, nevertheless, satisfy the criteria for moral agency and be held accountable for some actions at the corporate level of the state.Footnote 18 In other words, a group may possess capacities for deliberation and action that allow it to qualify as a moral agent even when it faces internal impediments that render these capacities limited or unreliable in a way that prevents the group from discharging some duties that it would otherwise be understood to bear. I will label such an entity a “weak institution.”
Second, an actor might face external limits to discharging particular duties. Perfect freedom from other actors and influences is neither achievable nor necessary for exercising moral agency.Footnote 19 However, to be considered vulnerable to the apportioning of blame if particular duties are abrogated, institutional moral agents must be able to pursue their own objectives relatively free from external impediments. Some obstacles faced by institutions in international relations are disenabling: for example, states are constrained by the financial demands (and imposed policies) of foreign creditors; and certain intergovernmental organizations are designed to perform particular functions, and are delegated responsibilities accordingly, but are not provided with the resources necessary to fulfil them. That such external factors can undermine an institution's ability to exercise moral agency in particular cases, and thereby render unreasonable expectations that certain duties be discharged, follows from the understanding of individual moral agency offered above. I will label a formal organization thus restricted a “constrained institution.”
Both categories of institutional moral agent are crucial to the discussion of holding collectivities to account for moral transgressions in international relations. While collectivities that meet the criteria for institutional moral agency are able to discharge specific responsibilities, and can be blamed for their acts and omissions in certain circumstances, an actor—whether an individual or an institution—cannot coherently be blamed, held to account, or punished for abrogating a duty that he, she, or it was unable to discharge in the first place.
Delinquent Institutions
The considerations of, first, whether an institution possesses the sophisticated, integrated capacities for deliberation and action that would allow it to qualify as a moral agent and, second, whether, as an institutional moral agent, it is either weak (due to internal limitations) or constrained (as a result of external circumstance) in ways that would prevent it from discharging duties in the context of specific actions are fundamental to discussions of moral responsibility in world politics. These steps do not lead to an account of whether certain groups are somehow inherently moral or immoral, and on their own cannot determine whether these groups deserve praise or condemnation. Asking whether a group is a moral agent—and then examining the degree to which it is able to act in specific circumstances—are necessarily prior questions to determining whether it is vulnerable to assignments of duty and ascriptions of blame in the context of specific actions. An analysis of an organization's internal capacities and external conditions must come before any attempt to evaluate the degree to which it can be said to have breached its obligations and thereby be vulnerable to condemnation—and possibly punitive action.
In cases in which an institutional moral agent enjoys both the capacities and enabling conditions to discharge specific duties but still fails to do so, blaming that institution for the consequences of its failure is a coherent response. I will refer to institutions in these circumstances as “delinquent institutions.” Delinquent institutions in this sense might be understood to include Union Carbide in Bhopal, India; the UN in the context of Rwanda, Srebrenica, and Darfur; France in Rwanda; the U.S. military with respect to both civilian deaths and prisoner abuse in Iraq; and BP in relation to the Gulf of Mexico oil spill.Footnote 20 In each case, careful attention must be paid to establishing whether the institution in question is the appropriate object of blame, taking into account the moral guidelines to which it is expected to conform, its particular capacity for purposive action, the degree to which it is actually able to act given external conditions, and the level at which the relevant act or omission is most accurately described. Assuming that there is no lack of delinquent institutions, thus defined, in international relations, I want to turn here to the separate question of how we can respond to them.
THE PROBLEMS WITH PUNISHING DELINQUENT INSTITUTIONS
Punishment is not the only possible response to institutional delinquency. Other types of response include forgiveness, reconciliation, reform, apology, and compensation. Each of these also deserves attention. Making assertions of responsibility in international relations need not be understood to lead either logically or inevitably to calls for punishment.Footnote 21 Punishment is, however, a prominent justification for action—and response to alleged delinquency—in international relations. Moreover, the argument that such bodies as states are moral agents, and thus can be blamed for wrongdoing in a way not reducible to their individual human members, invariably prompts questions of what appropriate forms of punishment might be. Yet, is it possible to punish institutions at the corporate level, even when that is where blame is most accurately apportioned? Or does such a proposition take the understanding of institutional moral agency a step beyond what the realities of formal organizations allow?
By punishment I mean quite simply the deliberate infliction of some burden (in the form of suffering or deprivation) on an offender in response to an offense. Of course, there are different types of justification for punishment, and the underlying logic that one accepts will inform how one then evaluates any particular case of punitive action. Theories of punishment are generally divided between those labeled consequentialist and those referred to as retributivist. Consequentialist positions justify punishment according to its contingent contribution to achieving some good and thereby look ahead to what the punishment will achieve. Retributivist positions are, by contrast, backward-looking and justify punishment with reference to what the agent has done (or failed to do); the guilty deserve to be punished, and the punishment is only justified if it is they who are punished.Footnote 22 Acknowledging that many theories of punishment rely on some alliance between consequentialist and retributivist logics,Footnote 23 the concerns I set out below assume that just punishment would necessarily preclude the innocent being punished (or the guilty being punished disproportionately), even if this were seen to benefit some independent good, such as deterrence, for example. In other words, the concerns I explore below rely on the view that a morally sound justification of punishment requires, at the very least, robust retributivist constraints.Footnote 24
Even if a compelling case can be made for punishment in response to an institution's moral infractions, it is difficult to conceptualize how to actually punish such a body, or make it suffer at the corporate level. While an institutional moral agent is characterized, in part, by an identity that is greater than the sum of identities of its constituent parts, it is, nevertheless, made up of individual human actors. Punishing the entity that these individuals together help to form without punishing them each as individuals is a challenge. A colorful illustration of this conundrum can be found in an unlikely place. In A Christmas Carol, Charles Dickens provides an imperfect depiction of what it would look like to punish an institution (and, in doing so, leaves us with a problematic, but thought-provoking, response to Baron Thurlow's question of how one can punish a corporate entity when it has no “soul to be damned”). Beginning a journey of self-reflection, Dickens's protagonist, Ebenezer Scrooge, is forced to witness an unsettling scene of suffering spirits and damned souls who are being punished for failing to lead ethical lives. Upon hearing “incoherent sounds of lamentation and regret; wailings inexpressibly sorrowful and self-accusatory,” Scrooge, to his horror, observed that
the air was filled with phantoms, wandering hither and thither in restless haste, and moaning as they went. Every one of them wore chains … some few (they might be guilty governments) were linked together.Footnote 25
In this portrayal, the purported punishment of guilty governments entails no more than the synchronized suffering of individual human actors. In other words, punishment might be described as distributive. Whether we consider a fictional purgatory (and the ghosts of governments) or real-life sanctions imposed on concrete regimes, Dickens's image points to a crucial problem. Can we effectively punish an institution while remaining faithful to the understanding of responsibility as nondistributive that the model of institutional moral agency supports?
In what follows, I will touch on three concerns that arise in the attempt to punish institutions. I do not mean to suggest that these pose necessarily insurmountable obstacles to the viability of institutional punishment. They do, however, represent problems that must be raised and addressed in any preliminary exploration of the subject. The first, which I will address under the heading of “guilt by association,” deals with the danger of punishing all of the members of a group for the misdeeds, or failures to act, of certain constituents of that group. The second concern is over the possibility of “misdirected harm,” which would include the punishment of individuals as individuals in response to institutional delinquency. The third and final apprehension is about the harmful side effects of punishing an institution on those individual human actors (and other institutional actors) who are adversely affected as a result of the institution's incapacitation. Drawing on an article by John C. Coffee, the title of which invokes Baron Thurlow, I will refer to this final concern as the “overspill problem.”Footnote 26
Guilt by Association
One serious concern with the proposal to punish institutions is that such a move might be based on the unfair assumption of guilt by association.Footnote 27 This concern is rooted in an objection to blaming, and punishing, the many for the misdeeds of the few. This is a valid objection, but one that I want to suggest flows from a conception of moral responsibility that departs radically from the notion of institutional moral agency I have proposed.
The alternative understanding of group responsibility that generates this apprehension is referred to by the frequently invoked and much maligned phrase “collective responsibility,” and is no stranger to discussions of guilt and apology or of retribution and reprisal in international politics. According to this conception, a degree of solidarity within a group, or a shared aspect of identity, allows those who are not party to a specific action to be morally praised or blamed for the action of an agent or agents within the same group.Footnote 28 Responsibility in this sense has been addressed in the specific context of exploring the degree to which individual citizens must bear and retain guilt for the past deeds of their governments. It has garnered particular attention with regard to the issue of whether the German people are collectively responsible for the Holocaust.Footnote 29 By this account, membership within the group in question, and therefore the distribution of responsibility among its members, may even be understood to extend transgenerationally—and, in another variation on this position, to extend universally where the relevant group is conceived of as humanity as a whole.Footnote 30
Aversion to this notion of responsibility—and any call to punishment that might arise from it—is not difficult to explain. It runs counter to the understanding that one cannot be blamed for the wrongdoing of another. This conception of collective responsibility diverges significantly from the understanding of responsibility associated with the model of institutional moral agency outlined above. An implication of this model cannot be that all members of a group are responsible for the actions of discrete members within that group. Institutional responsibility, as I have defined it, is simply not distributive in this way. The group itself is the moral agent. If the group itself is also the proposed object of punishment (in response to misconduct or neglect that cannot be described in a way reducible to its constituents), then the fear that punitive action is being directed against individuals who are being unfairly held to account is unwarranted.
This particular concern is based on the conceptual difficulties of talking about blaming and punishing a formal organization. These difficulties can be overcome by rejecting the equation of institutional blame with guilt by association and reserving the apportioning of responsibility to institutions for instances in which the relevant action is one that can genuinely be described at the corporate level. The subsequent two concerns, however, relate to the practical difficulties of actually punishing an institution, even after we have established that it is the institution itself that deserves blame. The following problem appears to be particularly intractable.
Misdirected Harm
In cases in which we are responding to acts or omissions that cannot be reduced to the agents that constitute the group, punishing the institution, as an institution, makes sense, at least in theory. One problem that remains, however, is that it is not apparent how one goes about doing this. Here we are back to Baron Thurlow's observation that an institution “has no soul to be damned and no body to be kicked.” So, how does one punish it? Perhaps more to the point of my immediate concern here is the question of how to punish the institution at the corporate level (at which blame has been apportioned) rather than inflicting punishment on its individual human constituents (who do have bodies to be kicked and souls to be damned). In other words, if my previous concern might be referred to as the danger of misdirected responsibility, then this concern is one of misdirected harm.
Overspill
The final concern also relates to the danger of harming individuals in the attempt to punish institutions. However, in this case such harm is the indirect consequence of effectively punishing the institution. If we can assume that it is possible to get past the previous sources of worry by, first, appropriately identifying the relevant institutional agent as an object of blame in the context of a specific action, and, second, arriving at a means of punishment that targets this agent at the corporate level, this third problem remains. Even punishment that is directed at the corporate level of the institution risks indirectly harming individuals—both those who are members of the institution and those who benefit from the services it provides. As Coffee quips, “when the corporation catches a cold, someone else sneezes.”Footnote 31 He goes on to observe that overspill from a penalty levied against a corporation occurs at different levels.Footnote 32 Individuals benefit from the functions performed and, indeed, the duties discharged by an institution. (Exactly this reasoning was put forward in arguments against punishing banks in the recent global financial crisis.Footnote 33) Although one might argue that the long-term deterrent or rehabilitative effect of punishing an institution could result in the institution in question better fulfilling its functions and discharging its duties—to the ultimate benefit of the individuals that it serves—in the short term, these individuals might suffer.Footnote 34
In the subsequent section I will aim to illustrate each concern in the context of attempts to apportion punishment through military means. Specifically, I will touch on examples taken from the recent wars in Afghanistan and Iraq. I am not accepting either state as a “delinquent institution” but, rather, using these cases of alleged delinquency in order to interrogate the logic of justifying military action in response. Particularly useful in the current context are the discrepancies that these cases entail between the purported objects and the actual victims of punitive harm.
INNOCENTS AND THE PROBLEM OF PUNISHING THE STATE
Talk of punishment is nothing new to discussions of the ethics of war. Indeed, a just cause for engaging in organized violence was traditionally seen to include punishment of wrongdoing. Although accepted justifications for war have narrowed to self-defense in most contemporary articulations of the just war tradition, punishment as a reason for war has not disappeared—and, arguably, has experienced a reemergence.Footnote 35 This is apparent in both academic and political commentary on the so-called war on terror. For example, in response to the 2001 attacks on the World Trade Center and the Pentagon, the just war theorist Jean Bethke Elshtain maintained that “a carefully worked out and unprovoked act of terror against non-combatants of one's own country is an injury—an act of war—that demands a response. That response is just punishment.”Footnote 36 President Bush, defending wars against both Afghanistan and Iraq, also spoke passionately of “bringing justice to the enemies.”Footnote 37 Punishment has been unmistakably present as an underlying justification for military action in the war on terror—even though it has not been the only or, indeed, in the case of Iraq, the predominant (explicit) justification for engaging in war. Analyzing the idea of war as punishment with the aid of the conceptions of responsibility and blame developed above serves, then, as more than a mere philosophical puzzle. It is an endeavor that reveals far-reaching and important implications pertaining to how we understand, and respond to, purported cases of institutional delinquency in international politics. These implications will be addressed by revisiting the three categories of concern generated by the idea of punishing institutions.
Narrowing the Enemy: The Specter of Guilt by Association
The perceived danger of being seen to rely on a logic of guilt by association when attempting to punish a delinquent state is apparent in the context of the war on terror. Awareness of this danger is revealed in dedicated attempts to avoid it—through explicit endeavors to narrow the purported objects of punishment. Breaking with the conventional understanding of “the enemy” in war as the collective body against which one is fighting, there have been recent moves, particularly with respect to the war in Iraq, to significantly limit who is described as falling within this category. “Enemy,” it seems, has taken on the connotation of those who are “guilty”—and who are, therefore, deemed to be legitimate targets. For example, President Bush declared at the start of the war that “our enemy in this war is the Iraqi regime, not the people who have suffered under it.”Footnote 38 Expressing a similar sentiment, Prime Minister Tony Blair assured the Iraqi people that “our enemy is not you but your barbarous rulers.”Footnote 39 Drawing this nominal distinction might be seen to be particularly important because of the underlying justification of war as punishment.
Punishment implies a guilty party, yet in many cases there is a dissonance between treating the state as the object of organized violence (and therefore, implicitly, as the object of punishment) and accurately portraying the party whose delinquency has provoked a military response. In the case of Iraq, eliding punishment with other just cause arguments required an overt denial that all members of the Iraqi state were being held collectively responsible for the wrongdoing of a corrupt few. Falling into the trap of assuming guilt by association was seen to be a genuine hazard. Of course, narrowing the category of those labeled “guilty,” and thus justifiably vulnerable to punitive action, is more straightforward than effectively limiting those who ultimately become targets of attack. Indeed, the apparent attempt to avoid the charge of apportioning guilt by association in this case raises the question of whether war can be an appropriate means of punishment when culpability is thought to lie only with certain individuals—or a subgroup—within a state. This question brings us back to the separate problem of misdirected harm.
Individual and Innocent Deaths: The Reality of Misdirected Harm
Misdirected harm, as described above, occurs when the objects of a punitive response are distinct from the entity whose delinquency is invoked to justify the punishment in the first place. Even if one avoids the charge of bringing notions of guilt by association into justifications for punishing the state—by, for example, making the sorts of statements offered by Bush and Blair in the context of the war in Iraq—the danger of misdirecting punishment remains. There are two ways in which this problem is manifest when punishment is directed against the state through military means. The first brings us back to the difficulty of punishing the state as a corporate entity (in a way that does not effectively target its individual constituents) when it is the corporate entity that is being held to account for some wrong. The second introduces the problem of whether one can coherently use war as an instrument of punishment against a state if it is not the state as a whole that is deemed to be delinquent but, rather, a subgroup within the state. I will address each in turn.
The just war tradition has a long-established and complex means of limiting legitimate human targets in war: the principle of noncombatant immunity. Accounts of this principle describe noncombatants as deriving their immunity from their “innocence”—a term variously used to connote their lack of moral guilt or, more commonly in contemporary arguments, the fact that they are “not harming.” Of course, killing in war is never completely delimited and discriminate. Noncombatants are regularly killed. When this is unintended, they are designated “collateral damage.”Footnote 40 Where this line between permissible and prohibited human targets is drawn, why it is morally significant, and which individual and institutional actors are to be blamed when noncombatants become victims of attack (intentionally, disproportionately, or due to insufficient efforts to safeguard them) are difficult and important questions. They will not, however, be addressed here. Arguably, these dilemmas fade into the background if war is portrayed as a means of institutional punishment. This is because killing noncombatants and soldiers alike becomes problematic if we endeavor to respond accurately to wrongdoing at the corporate level of the state. I realize that one might be tempted to argue that soldiers would remain permissible objects of punishment—not as individuals, but as instruments of the state. Nevertheless, if this argument is simply that their particular roles mean that they can be made to suffer on behalf of the state, then this clearly constitutes what I have labeled misdirected harm.Footnote 41 Indeed, even accounts of combatant vulnerability that rest either implicitly or explicitly on notions of the guilty soldier sit uneasily with justifications for war that rely on the notion of institutional (rather than individual) culpability.Footnote 42 Punitive harm is misdirected toward individual human beings when it is the state that is being blamed for an alleged delinquency.
There is at least one conceivable retort to my claim that harm is misdirected in this case. As I argued earlier in this article, institutional responsibility does not preclude concomitant individual responsibility. One might invoke this point to argue that war can legitimately achieve the aim of punishing both the state at the corporate level and those individual citizens who are understood to be either directly or indirectly responsible for the policies of the state. This is an argument that warrants serious consideration. Yet, even if one were to defend this extreme position, it must be conceded that, in some cases, it is particularly difficult to point the finger of blame at individual citizens for contributing—to whatever degree—to the delinquent policies of the state to which they belong. Indeed, such an argument is stretched to the breaking point when membership within a nondemocratic state means that individual citizens play no part in its decision-making process (and, perhaps, risk having any opposition to the ruling government and any attempt to introduce democratic reform brutally oppressed).Footnote 43 It is exactly this scenario that leads to the second possible manifestation of “misdirected harm.”
Misdirected harm also occurs when the state is the proposed object of punishment yet delinquency is so narrowly associated with a specific group within the state that it becomes impossible to talk about delinquency at the corporate level of the state at all. To defend punishing the state in such a case would involve either embracing the idea of guilt by association (and treating the many as collectively responsible for the misdeeds of a few) or eschewing this idea of transferred guilt and succumbing to the equally serious misdemeanor of misdirected harm. This second variation on misdirected harm is exemplified in the recent wars against Afghanistan and Iraq to the extent that attempts were made to justify them in terms of punishment, even while explicitly asserting that the delinquent parties did not include those citizens who were inevitably placed under attack.
I have described attempts by President Bush and Prime Minister Blair to make explicit that they were not holding the Iraqi people collectively responsible for what they presented as the grievous misdeeds of their rulers. Perhaps the sharp distinction between the overwhelming majority of members of this “rogue state” who bore no responsibility for its alleged delinquency and the culpable minority in power served to intensify the perceived need to make clear that civilians would be protected. Prime Minister Blair announced in the context of the attack on Afghanistan that “this military plan has been put together mindful of our determination to do all we can to avoid civilian casualties”; President Bush stated at the start of the Iraq war that “protecting innocent civilians is a central commitment of our war plan.”Footnote 44 Of course, a great many noncombatants were killed. The way in which this represents a case of misdirected harm resulting from the state being punished is subtly different from that addressed above. Whereas in the previous manifestation of misdirected harm the problem was identified as one of effectively punishing individuals as individuals in the attempt to punish the institutions to which they belong, in this case harm is directed outside the corporate entity (whether this be the Taliban or Saddam Hussein's regime) whose delinquency is invoked as a justification for engaging in punitive action. Even somehow successfully punishing the state at the corporate level would, in this case, nevertheless entail misdirected harm as the state as a whole is not being held responsible for the alleged delinquency. Alternatively, apportioning punishment that affects individuals as individuals in such a case (which I have argued is the necessary, if unintended, outcome of trying to punish the state through war) would result in compounded misdirected harm. One of the reasons that the people of Afghanistan and Iraq prove to be such deeply problematic recipients of punitive harm is that they are not the perpetrators (nor, one might add, are they in any way the beneficiaries) of the delinquent deeds with reference to which punishment is being justified.
Indirect Harm: The Problem of Overspill
The problem of overspill brings us back to the indirect harm with which both individuals and groups might be afflicted when an institution is punished and is no longer able to perform certain functions toward either its members or those outside the institution who depend on it. Overspill can be identified in the ongoing problems associated with the occupation of Iraq, for example. In the case of such military action against a state, overspill might take the form of suffering caused by destruction of the state's infrastructure, including disruption to health services, intermittent or absent electricity, poor sanitation, and damage to water supplies and water treatment facilities (each contributing to malnutrition and the spread of infectious diseases) as well as population displacement, widespread unemployment, and the erosion of the rule of law. Moreover, environmental damage might affect those within and without the state. Finally, military action might place burdens on neighboring states in terms of refugee flows and on external institutions with respect to rebuilding the affected region.
Overspill is significantly different from misdirected harm. Misdirected harm represents a disjuncture between the objects of blame and the direct objects of punishment. Overspill does not involve such a disjuncture. Rather, overspill highlights the indirect and unintended harm of those who are neither held morally responsible for an alleged delinquency nor targeted in the ensuing punishment. Overspill is likely to be unavoidable to some degree in all cases of punishment, whether individual or institutional.Footnote 45 It is a problem that should be carefully considered in the context of determining what constitutes legitimate and effective institutional punishment. Yet, unlike the problem of misdirected harm, I do not think that overspill necessarily undermines the logic of a particular form of punishment in response to institutional delinquency. Given the indirect and unintended nature of the harm that characterizes overspill, it is, arguably, conditionally morally acceptable. Overspill might be acceptable if there are concerted efforts made to avoid it, and if the indirect harm is proportionate to the overall good to be achieved by the resort to institutional punishment.
CONCLUSION
Limited overspill might be an acceptable consequence of punitive war. Moreover, while guilt by association is a profoundly problematic assumption upon which to base an argument for punishing the state, it is also an unnecessary one. Institutional (rather than individual or collective) responsibility provides a sound conceptual foundation for punishing the state at the corporate level. However, the problem of misdirected harm is unavoidable if war is the means by which the state as an institution is ostensibly punished. Indeed, this problem renders attempts to justify war in terms of the culpability of the state morally incoherent (unless, perhaps, such a position is accompanied by a compelling argument for the complicity of each individual member of the state). In other words, although any justification of war as a means of attempting to punish the state risks stumbling at all three hurdles that I have identified, it is misdirected harm that proves impossible to clear.
These considerations might usefully be invoked to evaluate other attempts at institutional punishment. By identifying and illustrating these three categories of concern in the context of this problematic case of punitive war (where they emerge most starkly), it has been possible to construct an initial framework for critically exploring other endeavors to punish institutions that are deemed to be delinquent—for example, the levying of fines against reckless multinational corporations and the imposition of economic sanctions on wayward states.Footnote 46 In arriving at this preliminary framework, two themes have arisen that deserve attention in such future attempts to explore different means of punishing institutional actors. I will touch on these briefly in the specific context of the state before reiterating the broader points that we can—and cannot—take from the discussion to this point.
The first theme that warrants further attention is that of democratic versus nondemocratic decision-making and the extent to which this should affect our considerations of how to respond to a delinquent state.Footnote 47 Barry Buzan's provocative analysis of who we may bomb—an analysis tied to the degree to which a people deserves their government—might be instructive here.Footnote 48 To what degree are citizens of a democracy responsible for the foreign policy of their state (in a way that citizens in repressive regimes with no say in such decision-making are not)?Footnote 49 A classically ambiguous statement by President Bush is interesting in this context: “I want each and every American to know for certain that I am responsible for the decisions that I make, and each of you are as well.”Footnote 50 To what extent can we take this literally? To what extent does taking this literally render unnecessary any distinction between holding a government to account for its policies, practices, and delinquencies, and holding the citizens of a state to account for their individual actions? To what extent (and this is a separate question) does taking literally the statement that each and every citizen is responsible for the policies of the state make the idea of punishing the state as an institutional moral agent, through organized violence, more coherent—and more palatable? Each of these questions requires careful consideration of how culpability, both individual and corporate, is affected by the degree to which citizens contribute to the state's decision-making process.
A second theme that follows from this is that of distributive versus nondistributive blame and punishment. If culpability is established at the corporate level of the state, then one might argue that an appropriate response must be one that is nondistributive (in other words, one that is directed at the institution). Only if one is responding to an aggregate of individuals who are individually deemed responsible for a harmful act or omission should blame and punishment be distributive, or directed at the individuals themselves.
In order to think about what this means for broader considerations of institutional punishment, it is perhaps important to end by clarifying the things that I am not arguing. First, I am not suggesting that institutional punishment is necessarily unviable. There are a variety of possible means of punishing formal organizations—including dismantlement, boycott, and “naming and shaming”—and a variety of types of organizations against which some form of punishment might be deemed appropriate. Organized violence against a state is only one (admittedly extreme) example of punishment that one might attempt to justify on the basis of institutional culpability. Other forms of punishment might be more effective in responding to institutional delinquency in a way that avoids, or minimizes, harming its constituents as individuals. Second, I am not proposing that punishment is necessarily an incoherent basis on which to engage in military action. There are a plethora of issues regarding the legality, and indeed prudence, of justifying war in terms of punishment that deserve attention and have not been addressed here. In the context of this discussion of institutional moral agency and responsibility, the focus of concern has been on whether it makes sense to justify military actions toward a state in terms of culpability at the corporate level. My point here is that war waged against the state represents a form of punishment that is necessarily distributive (in that it directly harms those within the group as individuals) and can only represent a coherent response when responsibility for the acts or omissions that have motivated the punitive action is also judged to be distributive (in that blame is clearly reducible to the constituents of the state). To return to Baron Thurlow, one might observe that the state has many bodies to kick and many souls to damn, but engaging in either practice fails to respond coherently, and proportionately, to institutional delinquency. The important and challenging question of what would constitute a viable, effective, and morally acceptable (or nondistributive) means of punishing a delinquent institution must still be answered.
