Drawing on examples of environmental crisis, the Rwandan genocide, systemic oppressions, the Canadian Red Cross’ handling of the Canadian blood supply, and others, Tracy Isaacs’ Moral Responsibility in Collective Contexts skillfully argues that collective intentions and collective actions exist behind both collective moral wrongs and collective moral successes. Building on, and departing from, the work of a number of philosophers of collective agency,Footnote 1 Isaacs articulates a theory of moral responsibility which includes both individual and thoroughly collectivist dimensions, operating at different levels, as she puts it, both ontologically and explanatorily (24). Isaacs works from practical concerns: recognizing ourselves as belonging to groups can make possibilities for change visible; denying collective responsibility can prevent us from seeing our own moral effectiveness, and, in some cases, from seeing our obligations. I thank Isaacs for the ambition and care of this book, and for her willingness to engage in ongoing discussion about it. In this response, I begin with a summary and evaluation of what I found most instructive about the project, before focusing on three considerations for future work on collective action. As someone who believes ethicists are only at an early stage in fleshing out collectivist models of agency, I want to offer some possibilities for pushing back on the centrality of intention, by demonstrating that intuitions underpinning our sense of what should count as collectives can indicate factors other than a collective’s capacities for collective intention, and for pushing forward on the importance of collective feeling, arguing both for its metaphysical possibility and its practical significance.
According to Isaacs, collective wrongdoing does not eclipse individual wrongdoing, nor does individual wrongdoing eclipse collective wrongdoing—there is space for both. The first half of the book focuses on clarifying and defending an account of collective responsibility; the second half specifies individual responsibility in collective contexts. Focusing on the first of these, I think the goal is chiefly to defend the plausibility of an as-straightforward-as-possible account of collective responsibility against arguments for its implausibility. One of the many virtues of Isaacs’ account is how she deftly maps the terrain of the most common and significant objections to theorizing collective agency, while keeping them manageable and clear for her readers. Isaacs corrals the chief arguments against collective agency from various angles, a number of which take the following form: collective agency account X is missing, or inadequately prioritizing, central feature Y of an ethical framework focused on individuals, and therefore collective responsibility cannot be defended.
Isaacs specifies multiple kinds of collectives along a spectrum from most organized to least organized: organizations (highly structured collectives); goal-oriented collectives (collectives which form around the achievement of a joint goal); and aggregates (collections of people who share some characteristic). Aggregates are not the focus of Isaacs’ account. Organizations and goal-oriented collectives both have collective intentional structures, by virtue of which Isaacs argues that they can be collective agents, with collective intentions, capable of collective action. Collective agents are, therefore, subjects with collective intentions and the corresponding possibility of collective actions. A collective can be responsible without it following that the individual members of a collective are also responsible.
Isaacs summarizes early on what she takes to be the most persistent worry about collective responsibility: that claims about collective responsibility will entail claims about the responsibilities of individual members (19). I am reminded of a passage from Joel Feinberg’s seminal article, “Collective Responsibility”. Because the concept of responsibility had previously been restricted to the level of individual agency, Feinberg points out that the notion of “collective liability” is difficult to understand. Furthermore, he worries that this notion might become a serious threat, especially if the relationship between individuals’ intentions and blameworthiness is no longer taken to be the deciding factor in ascriptions of responsibility. He goes as far as likening collective responsibility to a curse of locusts: “When liability is imposed without such control, then it can ‘fall from the sky,’ like a plague, and land senselessly on random strangers.”Footnote 2
Isaacs’ summary of the six individualist objections in chapter two (59-60) is the most concise I have seen in the literature, and one of my favourite parts of the book. She lists these objections as follows:
1) Collective moral responsibility holds some people responsible for the actions of others.
2) Collective moral responsibility punishes innocent individuals.
3) Collective moral responsibility precludes individual moral responsibility in collective action situations.
4) Collective moral responsibility reifies social structures and other abstract entities.
5) Collective moral responsibility requires collective intentions, but collectives cannot have intentions because intentions are mental states, and collectives lack consciousness.
6) There is no interesting sense of collective moral responsibility. (She elsewhere frames the objection as a question: what is the normative force to be gained by moving to the complexity of the collective level? Or, the opposite, what is the normative loss of staying at the individual level? (57))
The work Isaacs does in taking on each of these objections is invaluable, especially from the perspectives of those who rely upon accounts of collective responsibility, but do not intend to provide their own. Individualist resistance to even the possibility of collective responsibility remains strong and divisive, and Isaacs provides a compelling account of why collectivist accounts must be taken seriously.
Isaacs shares with many ethicists a commitment to a view of moral agency where a capacity to act intentionally is required for an agent to be responsible. She argues that collective intentions, which make it possible for collective agents to act, are not simply collections of individual intentions. Organizations and goal-oriented collectives have different kinds of collective intentional structures. Organizations have intentional structures distinct from those of the individuals performing roles within them. As Isaacs explains: “Requiring commitment to a shared goal overstates what is required for collective action to take place in an organization. The policies, procedures, mission, role definition, and structures of authority are what produce intentional action in these settings” (31). Goal-oriented collectives, by contrast, have collective intentional structures as unities with capacities for different actions than could be performed by the individuals within them. Isaacs writes:
Goal-oriented collectives lack the highly structured character of organizations… Nonetheless, goal-oriented collectives do have the capacity for intentional action…We should think about goal-oriented collectives as unities of a special kind. Insofar as the whole is greater than the sum of the parts, we might consider them to be organic unities or organic wholes… Together we can perform actions we cannot perform alone. I cannot go for a walk together alone. I cannot play street hockey alone. I cannot perform orchestral works alone. Only collectives can be the agents of these actions (31, 34).
According to Isaacs, goal-oriented collectives are capable of performing actions beyond what individuals can do alone (e.g., doing the wave at a sports event, playing street hockey, or playing orchestral music). As in cases where groups come up with ideas that no individual had been able to see alone, or where one semester’s class is enthusiastic and the next semester’s indifferent, no individual has the “chemistry” (35) (in the sense of generative personality or identity) to bring about some goal alone.
The case for the intentional structure of goal-oriented collectives is harder to make than that of organizations, and Isaacs canvasses a number of “hybrid views” (33-34) which aim to make sense of intention partly in terms of the full collective, but also partly in terms of individuals’ agency.Footnote 3 Isaacs holds that the intentions of individuals are still important, but that a uniquely collective level of intention is also relevant. As she summarizes her point:
The way the individual intentions combine and the way members of groups understand themselves as members of groups give rise to collective intentions that, in turn, result in collective actions (36).
Instead of thinking of collective intentions as mental states, Isaacs suggests understanding them as states of affairs. In sum, according to Isaacs, while individuals can have intentions with collective content (e.g., to participate in doing the wave at a sports event in a way that helps contribute to a successful wave), individuals cannot intend collective action (e.g., to do the wave as a group) (40). Collective intentions cannot be reduced to the intentions of individuals. As Isaacs argues:
In the case of collective actions, it is collective agents who are blameworthy or praiseworthy. The individuals who contribute to the outcome do not—actually cannot—perform or intend the whole act even if they may share the collective goal and contribute to its achievement (55).
Isaacs’ account opens up numerous possibilities for future research. I would like to introduce three considerations that could inform such work: 1) reasons for thinking that an account of conflicted collectives is desirable; 2) reasons for thinking factors other than collective intentions might be necessary for collective agency; and 3) reasons for reconsidering Isaacs’ dismissal of the possibility of collective feeling.
1) Conflicted Collectives
Isaacs raises the possibility of collective compromise in the example of the Stratford Theatre Festival (30). In the example, the executive director of the festival hates musicals but decides nonetheless to showcase at least two musicals per season, in keeping with the Festival’s mission to offer a mix of productions, accessible to a variety of audiences. The point of the example is to highlight the director as making the organization’s decision, as distinct from what she herself would decide alone. As Isaacs states:
The decision the artistic director makes may legitimately be redescribed as the festival’s decision… If organizational decisions are taken in the form of votes on motions, for example, it may well be that individuals whose roles require them to participate in the decision cast their votes on the basis of personal reasons. Nonetheless, the decision is the organization’s intention, irrespective of individuals’ reasons for voting as they do (30, 29).
Although Isaacs does not discuss it, I read this example as indicating the desirability of articulating one feature of collective intentions, which is consistent with Isaacs’ account, but not discussed by her: namely, the extent to which collective agents can be conflicted. On Isaacs’ view, collective intentions can be the culmination of, for example, a group of individuals voting, even when those individuals do not all agree. Part of what makes a collective a collective is coming to a singular, decisive intention. If there is disagreement among members of the collective, it is possible for the collective to nonetheless decide to work toward a single goal. So conflictedness is left at the individual level; the collective is partly defined by its capacity for unified, collective intention.
Yet, apart from the collective context, individual agents can and regularly do experience conflictedness. Even when deciding to act in one way rather than another, individuals can feel ambivalent, conflicted, or uncertain. Perhaps then, returning to the collective context, if the members of a group are not in full agreement when making a decision or after the decision is made, we might suspect that the collective itself is similarly conflicted.
Providing an account of conflicted collectives could be important for two reasons. First, if Isaacs held that conflicted collectives are impossible, this would be part of what distinguishes collective from individual agents. Individual agents are the kinds of things that can be conflicted, while collective agents are not. If, on the other hand, conflicted collectives could exist, we would need to ask how to understand such a fact solely in terms of intentions. How could it be possible for an intention that is (on Isaacs’ view) a state of affairs, to behave more like a mental state (that is, the state of being conflicted)?
Second, note that in the case of individuals, a person’s conflictedness can be relevant in our ethical assessment of them. That one is conflicted in the case of a decision can affect our judgments of that individual’s responsibility: conflictedness might be seen by some as a moral failing, and a regular disposition towards conflictedness might be seen as a character flaw. On other views, an individual’s conflictedness might be seen as appropriate given the complexity of a particular domain of action. Either way, at the individual level conflictedness in one’s actions is in some cases morally relevant in evaluations of action or character. Not only what one does but also how one does it can be morally significant. Perhaps the same would hold at the collective level—if disagreement exists among the members of a collective, and that disagreement informs the actions the collective pursues, the conflictedness of a collective might be relevant for moral assessments of it. An account of collective conflictedness would be necessary if such assessments are ever to be made at the collective level.
2) Pre-Intentional Collectives
Isaacs’ account draws support partly from intuitions about what constitutes collective agents—that collective agents come to exist when they intend to perform actions beyond what individuals could do alone. But perhaps apart from her account there are examples of collectives where we would be inclined to say it is something other than collective intentions that solidify them. For example, think of a sports team whose members are each vaguely aware that their coach has been abusing children. The team members all experience a general unease around the coach, hesitate when asked questions about how effective he has been, and have tentative discussions with one another, each individual trying to confirm his or her suspicions while being hesitant about saying too much, and so on. If members of a group have no defined goals, and do not necessarily share traits or characteristics, but share this experience of unease, concern, or a nascent will to do something, do they count as a collective agent?
On Isaacs’ view, they do not. What we see is an example of a group that does not as yet have a goal, but whose members share a sense of unease about a situation, that, in the right circumstances, will turn into a goal. Perhaps what they have is a proto-intention, perhaps it is something more like a set of shared experiences. Yet, however we describe what the sports team has, in this case, perhaps we would want to identify it as a collective before it forms any intention. Identifying this group as a collective may be desirable from a will to be able to identify in retrospect collectives that have failed to establish goals, even when their members shared relevant experiences (e.g., if the team fails to develop a collective intention to bring the coach to justice or act to prevent similar harms in the future, even when the team’s members all experience a general unease). Or, more optimistically, perhaps we want to identify in retrospect collectives that have succeeded in establishing goals, because their members have shared relevant experiences. In either case, it might seem that the sports team should count as a collective agent that has either failed or succeeded, though it does not meet Isaacs’ requirement of having an intention.
My point is not to argue that we should in fact identify the sports team as a collective agent in the absence of collective intention. Rather the point is simply that the fact that intuitions can be roused in a variety of directions (including ones in which the existence of collective agents is not contingent on the existence of collective intentions) either weakens Isaacs’ case or leaves room for the possibility of other bases for collective agency. In other words, the intuitions that support the centrality of intentions in Isaacs’ account are not the only intuitions an account of collective agency should consider.
3) Collective Feelings
Throughout the text, Isaacs argues that collective feelings are not possible. She aims to “dispense with” (11) overly confusing or misleading complexities some other accounts of collective agency have maintained. There are good reasons for doing so: much of the resistance to taking collectivist ethics seriously has come from worries about a lack of clarity in the metaphysics of collective agency. The determination to provide a clear account of collective responsibility is especially explicit in Isaacs’ account of collective guilt (chapter three).
Collective guilt, in Isaacs’ sense, is “blameworthy collective moral responsibility” (71). Collective guilt neither is, nor requires, an agent’s believing itself to be guilty, nor an agent’s feeling guilty. As such, Isaacs distances her account of collective guilt as responsibility from guilt as self-reflection or experience. She refers to three different senses of individual guilt to clarify her claim:
The first, being guilty, is an objective moral condition—someone is guilty when she is morally responsible in the blameworthy sense for doing something wrong. The second, believing oneself to be guilty, involves a self-assessment that may or may not be consistent with the objective moral condition. The third, experiencing feelings of guilt, is an affective response to one’s belief that one is guilty (72-73).
These connections may not be accepted by all philosophers (e.g., the claim that feelings of guilt require a belief about an individual’s guilt); but granting them, Isaacs’ aim is to isolate the first condition of guilt from the other two in the collective context, so that in collective contexts what will be most important is collective guilt as the objective moral condition, rather than accounts of complicated collective dynamics of self-assessment or feeling. Collective guilt as an objective moral condition—where a collective is, in fact, guilty for some wrongful action and therefore blameworthy—follows from the account of collective intention and action Isaacs provides in the first two chapters. If collectives can intend, they can act; and if collectives can act in ways distinct from individual members (as chapter two argues they can), they can be guilty. Nothing then needs to be said about whether collectives can understand themselves to be guilty, or experience guilt.
Isaacs is particularly clear about distancing collective guilt in this objective sense from collective feelings of guilt. As she puts it:
With respect to collective guilt feelings, it is not clear that collectives are capable of affective responses at the collective level or that collective feelings play any necessary normative or practical role in collective guilt…Since feelings of guilt are distinct from the objective moral condition of guilt and from the belief in one’s guilt, an adequate account of collective guilt does not require that we account for collective guilt feelings (73-74).
Isaacs’ argument for the impossibility of collective feelings is based in large part on a criticism of Margaret Gilbert’s particular account of collective feelings. A collectivist account of guilt feelings will be hard to establish, Isaacs argues, because feelings should not be distinguished from bodily sensations—“the sensation of them…is what gets the agent moving” (90)—and collectives do not appear to be the kinds of things that can have sensations. Since feelings seem to have sensory components, unless collectives have bodies, it will be difficult to describe them as having feelings. Isaacs rejects Gilbert’s plural subject account of collective guilt for not providing a sufficient account of the collective affective component. As Isaacs writes, “Unless we distinguish feelings from their sensations—a move that Gilbert makes and that I argue ought to be avoided—collectives just do not appear to be the sorts of things that can have feelings” (85). Isaacs re-iterates her project of developing a workable account of collective responsibility that is not overly burdened by complexities that confuse and incite controversy:
Collective concepts face enough opposition without invoking language that misleads in ways that need to be explained away… We would do better to set aside the language and idea of collective guilt feelings, and focus on collective guilt itself as a species of collective moral responsibility (89).
In sum, Isaacs rejects accounts of collective feelings, for two reasons: (1) such accounts mistakenly require that a collective feel something (e.g., guilt) in order to be something (e.g., guilty in an objective moral sense). By contrast, Isaacs argues that a collective agent’s objective moral condition does not depend on it having any feelings (e.g., a collective need not feel guilty in order to be guilty in an objective sense). (2) Margaret Gilbert’s account of collective feelings does not provide a sufficient account of the affective component of collective feelings, instead holding that collective feelings need not involve sensations. By contrast, Isaacs claims that feelings necessarily involve bodily sensations.
Two responses to Isaacs’ account are important here. On the first point, even if Isaacs is right to reject accounts of collective feeling that mistakenly require that in order to be guilty a collective must feel guilty, this of course leaves open the possibility of an alternative account of collective feeling that does not require that agents feel any particular way as a condition of their moral status. Another account could hold that collective feelings exist and are relevant without having anything to do with the objective moral condition of a collective agent. So rejecting an account of collective feelings for its connection of guilt feelings to the fact of an agent’s guilt is does not yet establish the metaphysical impossibility of collective feelings.
On the second point, the key move in Isaacs’ argument for the impossibility of collective feelings as Gilbert argues for them is her claim that feelings necessarily have a sensory component. In general, it should be noted that if Isaacs is right about Gilbert, that does not rule out collective feelings altogether. More specifically, Isaacs’ claim about the impossibility of collective feelings in Gilbert’s sense is based on a particular understanding of feelings that is not the only available account of what a feeling is. The relation of feelings to sensations has been treated extensively. For example, critics of William James’ account of emotions as reducible to bodily feelings have noted the problem of indiscernibility of different emotions whose sensory reactions cannot be told apart, the failure to adequately account for non-standard emotions, and the failure to account for the relation of emotion to processes of rationality.Footnote 4 Accounts that see feelings as necessarily involving sensations can be contrasted with expressivist accounts of feeling that focus on practices of forming, expressing, and interpreting others’ expressions of psychological states, and that in fact argue for the necessity of expression for the formation of feelings (Campbell Reference Campbell1997).Footnote 5 While Isaacs bases her rejection of the possibility of collective feelings on one understanding of what feelings are, if we consider other accounts of what feelings are, the metaphysical possibility of collective feelings opens up. Because there are theories of feeling that reject the claim that all feelings will have sensory components, it is not obvious that we should accept it as a premise in an argument for the impossibility of collective feelings. In fact, other accounts of feeling that do not tie feelings exclusively to capacities or traits of individual persons, but instead to processes of expression, interpretation, and uptake that collective agents may perform, give us reason to think that collective feelings are possible.
Isaacs herself argues that, in fact, the main interest we would have in articulating a view of collective feelings comes from an interest in motivation (90). As she writes: “Guilt feelings might appear to play an important practical role in motivating guilty parties to act in ways appropriate to their guilt, perhaps even motivating them to repair, make amends for, or accept appropriate punishment for their wrongdoing” (84). Collective feelings of guilt could be part of what motivate collectives to make amends; collective feelings of security or cleanhandedness could be part of what motivate collectives to refuse to accept responsibility. Just as Isaacs’ own account of collective responsibility is motivated in part by her sense of the practical significance of collective action, the practical significance of collective feelings may push us to take their metaphysical possibility more seriously.
Isaacs is wise to resist simplistic analogies between personal feelings and collective feelings—this again showcases Isaacs’ project of understanding collective responsibility non-individualistically. Whatever collective feelings might be, they are not likely to be wholly (or perhaps even largely) analogous to individual feelings. But Isaacs has not given conclusive reasons to think that collective feelings are not metaphysically possible, and there are resources available that suggest that they might be. As someone who agrees with Isaacs that a collectivist account of responsibility is needed, I suggest that an account of collective feelings may be both more complicated and more possible than Isaacs suggests.
In conclusion, Isaacs’ text offers a substantial contribution to the growing literature on collective agency and collective responsibility. Especially given how many questions and considerations are made possible by her analysis, I look forward to the ways the book will be taken up by others, and to her own contributions in the future.
