Perron Tollefsen, D. (2003). Participant Reactive Attitudes and Collective Responsibility. Philosophical Explorations, 6(3), 218–234.
In this paper I suggest that we can approach the metaphysical problems surrounding the issue of collective responsibility in a roundabout manner. My approach is reminiscent of that taken by P.F. Strawson in “Freedom and Resentment” (1968). In that paper Strawson attempts to detach the issue of moral responsibility from the metaphysical issue of determinism. Strawson argues that the participant reactive attitudes – attitudes like resentment, gratitude, forgiveness and so on – to which we are subject by virtue of our participation in certain interpersonal relationships, provide the justification for holding individuals morally responsible. These reactive attitudes constitute moral responsibility. That is, to hold another responsible is just to be prone to have the appropriate reactive atti tude toward them. Even if the thesis of determinism is true, this will not and should not overturn our practice of holding others responsible. My aim in this paper is to reveal how these reactive attitudes and the features of agents to which they are sensitive can help us justify the attribution of moral responsibility to col lectives.
1. Strawson’s Expressivism
Strawson’s concern in “Freedom and Resentment” is to offer a version of compatibilism. The truth of determinism should not, according to Strawson, undermine our practice of holding others responsible and of praising, blaming, and punish ing. This is not, as the utilitarian would have it, because the practices of punish ment and reward are efficacious in regulating human behavior in society. Rather, it is because the very notion of moral responsibility is constituted by a set of com plex attitudes; attitudes like resentment, indignation, love, and so on. To hold someone responsible is just to be prone to have these attitudes towards others and to be responsible is just to be the appropriate target of these attitudes.
The reactive attitudes can be classified into three categories: the personal, the vicarious, and the self-reactive. The personal reactive attitudes are those attitudes we feel when either good will or ill will is shown to us (e.g. resentment, jealousy). The vicarious reactive attitudes are those attitudes we feel in response to ill or good will shown to others (e.g. disapprobation and indignation). The self-reac tive attitudes are attitudes directed at ourselves in response to how we treat oth ers and ourselves. Guilt, for instance, is paradigmatically a response to our own actions.
Because the practice of holding responsible is a natural fact of adult human social life, it requires, according to Strawson, no theoretical truth as a basis for its justification. Libertarians appeal for justification to the theoretical claim that in order to hold a person responsible one must be metaphysically free. Utilitarians appeal for justification to the claim that holding people morally responsible reg ulates human behavior. But this “over-intellectualization” of our practices is a mistake, according to Strawson. The propriety of the reactive attitudes is not established by some independent notion of responsibility. The explanatory prior ity is the other way around. As Gary Watson puts it:
It is not that we hold people responsible because they are responsible; rather the idea (our idea) that we are responsible is to be understood by the practice, which itself is not a matter of holding some propositions to be true, but of expressing our concerns and demand about our treatment of one another. Holding responsible is as natural and primitive in human life as friendship and animosity, sympathy and antipathy. It rests on needs and concerns that are not so much to be justified as acknowledged” (1987, 259).
Despite the fact that Strawson denies the need for an external, rational, justifica tion for our reactive attitudes, he does acknowledge that these attitudes have their own internal criteria of application. These criteria can be identified by reflecting on two kinds of cases. In the first case, a person does something that normally would indicate a lack of good will toward another, but in the current circum stances there isn’t a manifestation of such ill will. If a good friend walks by me without saying hello, I am likely to feel a bit of resentment. But if I find out that my friend did not recognize me, I will excuse his behavior and alter my attitudes. It would be inappropriate for me to feel resentment under those circumstances. I will refer to these sorts of cases as involving “local” excuses.9 These “local” excus es inhibit the reactive attitudes.
In the second kind of case, we don’t simply excuse a person but we exempt them. Very young children, the mentally deranged, severely handicapped, and so on are not appropriate targets of our reactive attitudes. Why is this? Following Lawrence Stern (1974) and Watson (1988), we can say that what is missing in these cases is the potential for moral address. Membership in the moral commu nity requires that the subject be a potential participant in interpersonal interac tions. To the extent that very young children and others in “unfortunate forma tive circumstances” lack the capacity for moral address, they are not appropriate targets for our reactive attitudes and, therefore, not appropriate subjects of moral responsibility. We can say that in these cases there are “global”10 considerations that exempt a subject from moral responsibility.
When we exempt another from moral responsibility on a long term basis, as we do with the mentally deranged, we take, what Strawson calls, the objective attitude toward the person. To take the objective attitude toward a person is to treat the person as something to be managed, directed, handled, cured, manipu lated, and so on. One cannot, according to Strawson, have a mature relationship with an individual that one views with an objective attitude. “If your attitude towards someone is wholly objective, then though you may fight him, you can not quarrel with him, and though you may talk to him, even negotiate with him, you cannot reason with him. You can at most pretend to quarrel, or to reason, with him.” (Strawson 1968, 79) To view someone with an objective attitude, then, is to view them as outside the moral community.
2. Reactive Attitudes and Collective Responsibility
Strawson’s focus is on the attitudes we have toward individuals and ourselves. But we also have reactive attitudes toward collectives. Consider, for instance, con temporary attitudes toward tobacco companies. On Dec 15, 1953, Tobacco exec utives met at the Plaza Hotel in Washington, D.C. to discuss what they consid ered to be a crisis: research linking smoking and cancer. In less than two weeks the industry released an advertisement that appeared in newspapers across the United States entitled “A Frank Statement to Cigarette Smokers.” The advertise ment claimed that there was no evidence linking cigarette smoking and cancer and that the industry believed that there was no reason to think that tobacco products were harmful to one’s health. That meeting, according to lawsuits filed in federal and state courts across the United States, began a decades-long cam paign to deceive the public about the health risks of smoking.
This decade-long deception has produced a great deal of anger. Many, of course, are angry with the original executives that initiated the cover up. But the anger is also directed at the tobacco companies themselves. Here is the expres sion of such anger from a young anti-smoking activist fighting her own addiction.
The anger here is palpable. Clearly the target of this woman’s anger is not some CEO but the tobacco industry itself. We do not reserve our reactive attitudes for individuals. We unleash our indignation and resentment on collectives themselves.
Our indignation on behalf of the victims is directed not just at the individuals who committed and participated in the cover-up, but also at the institution which condoned it and, in certain cases, made it possible.
Although many fans love individual players, there also seem to be attitudes that are directed at the teams themselves. Indeed, many fans remain committed to their team despite the fact that the players, management, and even ownership changes.
Faced with the fact that we do have reactive attitudes toward collectives, we can now ask whether these attitudes are justified. Those skeptical of the notion of collective responsibility will argue that these attitudes are never justified. Get ting angry at a Tobacco company is like getting angry with one’s pet or the table on which one stubs a toe.13 I am not justified in resenting the table that met my foot because the table is not a moral agent – a person, and reactive attitudes are only appropriately directed at persons. Likewise, according to the skeptic, we are not justified in having reactive attitudes towards collectives because collectives are not moral persons. [224]
Now, one might respond to this sort of skepticism as French does by devel oping an account of personhood and extending it to groups. I think, however, that Strawson provides us with a much easier response. This sort of skepticism presupposes that the only way our reactive attitudes toward collectives could be rationally justified is by appeal to some metaphysical fact about the status of collectives as persons. But to look for a metaphysical fact to justify our attitudes would be to commit the very mistake that the determinists or utilitarian compat ibilists make when they attempt to justify our reactive attitudes toward individu als. Strawson reminds us that the framework of the reactive attitudes is a given “as a whole, it neither calls for, nor permits, an external ‘rational’ justification” (1968, 23). The only justification we can give for these attitudes is pragmatic and is internal to the structure of the reactive attitudes.
But could organizations and other sorts of collectives be excused all the time? That is, are they, like the mentally deranged, exempt? To ask this question is to ask whether collectives meet the criterion we find implicit in the exempting conditions. That is, are collectives capable of moral address? To answer these questions we need to get clear on the notion of moral address and what the capacity for moral address entails. As Watson says, the reactive attitudes are really a form of “incipient communication.” What does an agent have to have in order to be a participant in this sort of communication?
Prima facie, the agent must be able to understand the moral demands that are placed upon him or her. Further, there must be present in the agent the capacity to guide his or her behavior and attitudes in light of these moral demands. If the capacity to direct one’s actions in light of reasons was absent, moral address would be futile. The reasons one provided would never result in the alteration of action and attitude. The capacity to guide one’s behavior and attitudes in light of moral demands, in turn, requires the capacity for deliberation. Agents must be able to attain some conformity between their values and their actions. And this conformi ty cannot come about in any old manner. Conformity secured by electric shock does not involve deliberative capacities.16 Conformity must be achieved by an act of reflection. The capacity for moral address, therefore, presupposes what John Doris has called normative competence – “a complex capacity enabling the posses sor to appreciate the normative considerations, ascertain information relevant to particular normative judgments, and engage in effective deliberation” (2002, 136).
If the dialogue gets off the ground, if the agent engages in the prac tice of giving and taking reasons, responds to criticism, and so on, then we have very good reason to believe that the individual has that capacity. Anyone who has tried to reason with a 10 month old child will have good evidence that the child lacks normative competency. [226]
Do collectives, like corporations, have this capacity? One way to answer this question is to reflect on judgments of normative competency at the individual level. How do we determine if an individual has the capacity for normative com petence? Well, we figure out if an agent has normative competency by starting the conversation. If the dialogue gets off the ground, if the agent engages in the prac tice of giving and taking reasons, responds to criticism, and so on, then we have very good reason to believe that the individual has that capacity. Anyone who has tried to reason with a 10 month old child will have good evidence that the child lacks normative competency. The conversation is always one way. If we provide the child with reasons at all, it is not in order to provide him with additional fod der for the deliberative mill. We want him to change his behavior but we don’t think the change will occur via complex deliberative processes that assess and evaluate reasons. Our interactions with very young children resemble Strawson’s objective stance. Our aim in addressing the child is to shape his or her behavior. The conversation, if there is one at all, is a pretense.
In order to answer the question of whether collectives have the capacity for moral address, then, we can simply ask whether we engage in dialogue with them. I think it is obvious that we do. We bring them to court, we file complaints, we address them in public forums, in the boardroom, before Congress. We, as Watson describes it, “address them with a complaint and a demand.” Although we might talk to individual spokespeople, we intend in many cases to address the corporation itself. We communicate that it has failed to meet the moral demand, the demand to be treated with a modicum of good will and respect. Here is one example of many such dialogues:
This distinctively third-person approach is likely to be met with skepticism. If we do engage in discussion with collectives, the skeptic will say, it is only because the collective is made up of individuals with the capacity for moral address. Thus, the fact that we engage in dialogue with a collective is not evidence that the col lective itself has the capacity. Individuals do the deliberating and talking, not groups, according to this objection.
I think this skepticism fails to understand the complex nature of social institu tions, the ways in which authority structures and roles transform individual actions into collective actions, and the distinct properties that arise at the collective level. First, although we engage in dialogue with institutions through their members, these members are speaking, in many cases, on behalf of the company. Given their role and the structure of the institution, their words are the words of the compa ny. Second, collectives themselves deliberate. They engage in the same deliberative processes that individuals engage in. They propose solutions, consider means to ends, consider possible alternatives, provide critical reflection on possibilities, assess options, draw conclusions and so on. These tasks are distributed in the con text of group deliberation. One person might offer a solution, another evaluate it, another person proposes an amendment to the original proposal, and so on. Group deliberation is distributed cognition. And group cognition has the potential to be a significantly more efficient process than individual deliberation. Thus, groups could actually have greater normative competency than individuals.
No doubt some aspects of the normative competency of the group can be understood reductively. Normative competency involves the appreciation of norms. This appreciation is best understood in terms of the capacities of individ ual group members. But group members will appreciate norms qua group mem bers. Outside of one’s group these norms may not have any force for the individ ual qua individual. Further, in some cases these norms apply only at the collec tive level. They govern collective actions that, in some instances, cannot be understood reductively. Finally, some groups have the capacity to engage in joint deliberation. These deliberative processes are not merely individuals engaging in deliberation along side others. Distributed cognition is a property of groups. Col lective norms and distributed cognition suggest that we cannot understand the moral competency of the group simply in terms of the moral competency of the individual members. Collective normative competency is an emergent property.
But group members will appreciate norms qua group mem bers. Outside of one’s group these norms may not have any force for the individ ual qua individual. Further, in some cases these norms apply only at the collec tive level. They govern collective actions that, in some instances, cannot be understood reductively. Finally, some groups have the capacity to engage in joint deliberation. These deliberative processes are not merely individuals engaging in deliberation along side others. Distributed cognition is a property of groups. Collective norms and distributed cognition suggest that we cannot understand the moral competency of the group simply in terms of the moral competency of the individual members. Collective normative competency is an emergent property.
If certain collectives have normative competency then they are the appropri ate targets of our reactive attitudes. If they are the appropriate targets of our reac tive attitudes, and as Strawson has argued, the reactive attitudes are constitutive of moral responsibility, then our practice of holding groups morally responsible has been vindicated. To hold a corporation or other collective morally responsi ble is simply to be prone to have the appropriate reactive attitudes toward them and for a collective to be responsible is for them to be subject to the appropriate attitudes.
We can extend the Strawsonian line even further. Strawson argues that if we were to relinquish our commitment to the reactive attitudes at the individual level, our lives would be very much impoverished. For instance, he points to the human isolation that would occur if we were to always view others and ourselves from the objective standpoint. We are justified on pragmatic grounds in holding on to the reactive attitudes even if determinism is true because there is no set of goods that would compensate for the isolation we would feel if we were to always see others as mere objects moved by forces. Although we could, psychologically speaking, give up the reactive attitudes, we shouldn’t.
Should we give up the reactive attitudes toward groups? I think not. Follow ing Strawson, I think it is clear that our lives would be greatly impoverished by relinquishing the reactive attitudes toward collectives. This is particularly clear when we focus on the collectives of which we are members and from which our own identities are created. Consider one’s family, team, or political party. Can we imagine treating these groups as mere objects to be manipulated, controlled, and managed? What would it be like to watch one’s favorite football team only from the objective stance?
The question is whether we should respond to collective actions and collective properties in an unemotional way. I think it is clear that if we failed to respond to collective properties with appropriate sentiments it would reveal a serious moral deficit on our part. This is even clearer in cases where the collective property or action is morally abhorrent. Consider the case of war crimes. There seem to be cases where there are evil groups without evil individuals. Not to respond to the evil of this kind with the appropriate sentiments would be to exhibit impoverished moral sensibilities.
3. Objections
A second, but related, objection is that in order for collectives to be the appro priate targets of our moral sentiments they must themselves be capable of having reactive attitudes.24 To exhibit ill will or good will, according to this objection, pre suppose the existence of certain sentiments. We exhibit ill will, for instance, when we are angry. But how could a collective itself have reactive attitudes? These atti tudes are emotions and emotions are feelings. How could the collective, itself, feel?
I do not have space here to give this worry a thorough response, but let me make an attempt to diminish its force. First, it is not at all clear that in order to exhibit ill will or good will one must be capable of moral sentiments. Couldn’t there be a crea ture that had evil aims without it also having feelings like hatred and resentment? But even if one must have reactive attitudes in order to be an appropriate target of the attitudes of others, there may be a legitimate sense in which collectives can have reac tive attitudes. Consider Margaret Gilbert’s account of collective remorse:
“Group G feels remorse over an act A if and only if the members of G are jointly com mitted to feeling remorse as a body over act A” (Gilbert, 2000:135).
Group remorse, according to this analysis, is a function of the joint commitment to form a unified subject that expresses remorse. Individuals qua members of the group will do what they can in terms of their actions and utterances to form such a subject
Now one might argue that this is not an account of collective remorse because there is no subject that “feels” remorse. But as Gilbert points out, it is not at all clear that phenomenology is constitutive of the emotions.
“Consider the case of an individual human being. When I say to you ‘I feel great remorse’ must I be saying something false unless there are pangs or the like in the background? On the face of it, I need not be saying something false. Note that some apparently equivalent expressions do not use the term ‘feel’ at all: ‘I am full of remorse’; ‘I am truly remorseful’” (2000, 135).
It might be argued, then, that the essential nature of emotion is to be characterized not in terms of phenomenology, but in terms of judgments and other intentional states like desires. The fact that collectives lack phenomenology, then, does not mean that they lack the capacity for reactive attitudes.
Even if we admit that phenomenology is essential to emotion, one could understand collective emotions as those emotions that are expressed through the members qua group members.26 Collective reactive attitudes would be differenti ated on this approach from individual reactive attitudes in terms of the norms gov erning the relevant reactions. Certain emotional responses might be licensed for an individual only because of her group membership. They would also be differenti ated in terms of their motivational upshot. The “pangs” of remorse you feel qua group member may lead you to do actions that you would not do qua individual.
Although this approach does not have the collective itself feeling the emotion, it does identify a way in which emotions can be collective. Perhaps this is all we need in order for collectives to be appropriate targets of our reactive attitudes. Finally, one might argue that my extension of the Strawsonian view to collec tives is really a reductio of the Strawsonian view. The argument can be formulat ed in the following way: If Strawson is correct then collectives, themselves, can be held morally responsible. But it is absurd to hold collectives, themselves, morally responsible. Therefore, Strawson is not correct.