Introduction

The background (though most emphatically not the topic) of this discussion is the liberal/communitarian debate. Many believe that debate has now run its course, but it has left an indelible mark on the way that perennial questions about the relations between individual and community are framed. In this chapter I attempt to articulate the idea of one kind of community, pertinent to social and political questions, which is present in many areas of actual human life. In section 1 I discuss the general idea of community, then offer and explore a specific conception of community as collective agency. In section 2 I suggest that membership of a collective agency raises, but does not of itself settle, important questions about loyalty, allegiance and dissociation. In section 3 I suggest that the existence of collective agencies casts doubt on the adequacy of the doctrine of the distinctness of persons.

1 Community and collective agency

The concept of community is a protean one. At its broadest it applies simply to a number of individuals who share something in common. But what they share in common, and indeed how the idea of sharing is to be understood, are matters for further elaboration. For example, they may constitute a community by virtue of sharing the same physical location: in that sense, the squire and the peasant may belong to the same community though in other important respects they stand in relations of separation and even opposition. By contrast, people talk of the gay community or a linguistic community, where the individuals who compose that community may be spatially separated and unknown to one another. Presumably, what underlies the idea of community in the non-spatial sense is some notion of common or shared interests. That in its turn would have to be distinguished from a community involving not merely shared interests but, as it is often put, shared meanings and understandings. Charles Taylor suggests:

The idea of community can also vary depending on whether we think of the individual as belonging to one overarching community or to a series of communities. It has been a matter of contention whether communitarianism insinuates the idea that there is just one relevant community in which an individual is located.² I do not attempt to settle that question. Some communitarians certainly acknowledge the fact of multiple and conflicting communities (as they must). Sandel, for example, says ‘There is no such thing as “the society as a whole” . . . Each of us moves in an indefinite number of communities’.³ But it is another matter whether that explicit acknowledgement is accompanied, either in Sandel’s or in others’ case, by the acknowledged fact’s playing an appropriately prominent role in subsequent thinking.⁴

Once the existence of multiple communities is acknowledged, questions arise about the priority among them. Amitai Etzioni has argued for layered loyalties ‘divided between commitment to one’s immediate community and to the more encompassing community, and according priority to the overarching one on key select matters’.⁵ But it may be less than clear which community counts as the overarching one. Neera Badhwar says that she ‘will follow communitarian practice in using “society”, “nation”, “state”, and “political community” interchangeably’.⁶ In a discussion of state authority which is pertinent for considerations of community, Joseph Raz says ‘Throughout the discussion I refer interchangeably to the state, which is the political organization of a society, its government, the agent through which it acts, and the law, the vehicle through which much of its power is exercised’.⁷ But since these different terms refer to distinct institutions, that raises problems about priority. What if a government has acted illegally, for example? What if the actions of the state are inimical to the interests of the nation? What if the state represents some sectional interest rather than the interests of the whole society? In these circumstances it will be a matter of deep contention what is required for according priority to the overarching community, because it will be contentious which community is the overarching one.

‘Community’, then, can be used for a variety of different purposes, to pick out different phenomena in our life as social creatures: shared location, shared interests, shared meanings and understandings, and so on.⁸ In the midst of these varying conceptions of community, there is no point in being essentialist or prescriptive: in what follows I attempt to isolate and characterise one form of community which is highly salient in the social and political lives of individuals, and to indicate what follows from its existence for some of the issues which were at stake in the liberal/communitarian debate. The form of community in question is a collective agency: what its members share in common is participation in collective action.

Sometimes the actions of individual human beings are best seen as part of some collective action. For example, I may join a number of other people in collectively pushing a broken-down car. The most appropriate and informative description of what I am doing will make reference to the fact that I am acting with others in this way. In this kind of case, there is little conceptual distance between individual and collective action. Each of us individually is attempting to do that very same thing which all of us collectively succeed in doing. Often, moreover, the collection of people involved will be an ad hoc one which dissolves after the task in hand has been achieved. But it is a significant contingent fact about the world we inhabit that there are collective agencies of a more persistent and distinctive character.

Consider two examples of more persistent and distinctive collective agencies (rather special examples, as it will turn out). I may be not merely kicking a ball around a field but playing in a football team. Or I may be not just playing a clarinet but participating in an orchestral performance. Collective agencies like football teams and orchestras typically exist over a period of time and engage in a whole series of related actions – in other words, they persist in a way that, typically, a collection of car-pushers does not. Connected to that persistence is the further fact that the collective agency can survive a change in its constituent membership. Particular individuals come and go but the team or the orchestra goes on. Moreover, these collective agencies are distinctive in that what they do is distinct from what their individual constituents do: it is only the collective agent, the team, which wins a match and is awarded points for doing so; it is only the orchestra which produces an orchestral performance. (Indeed, one reason why these examples are special is that these collective agencies do things which it would be conceptually impossible for individuals to do.)

I shall refer to collective agencies which exhibit these properties of persistence and distinctiveness as CAs.⁹ They have an ineliminable presence in our social world, in that we cannot say all that we need to say about that world without referring to them. We may insist that a team’s or an orchestra’s playing is just a matter of a number of individuals acting in various ways, and in a sense this is true. But it is a matter of their doing things as members of that entity, and something important is left out of any description of their activities which does not make that clear. There is, then, a certain kind of irreducibility and priority here: our best descriptions of the social world will contain irreducibly collective terms, and there will be a portion of individuals’ behaviour where an adequate description will require prior reference to the collective agency in which they are acting. Whether any kind of ontological or moral priority attach to CAs are further questions.¹⁰ Our social world contains many instances of CAs as described here. Committees, neighbour associations, trade unions, churches, electorates, governments, classes, business corporations, for example, all exhibit the characteristics of persistence and distinctiveness of action. They do things which individuals do not, they possess resources which individuals do not, and their existence is recognised in law.

Collective agencies, communities of individuals sharing collective action, will often be co-extensive with communities of individuals sharing some of the different characteristics mentioned earlier, such as shared location, shared interests, shared meanings and understandings. But for any given CA it will be an open question whether it possesses all or any of these other characteristics. Thus, a CA may consist of individuals located in the same place or it may not: a team does, but a trade union does not. Similarly, a CA may consist of individuals sharing a common interest or it may not: The National Association for the Advancement of Colored People does, but a court does not.¹¹ Most importantly, and least easily seen, a CA may consist of shared meanings and understandings in the way described earlier or it may not.

The point is not easily seen because shared meanings and understandings are certainly necessary for the collective actions of some CAs, such as teams and orchestras, to take place at all. A team’s winning is an ‘institutional fact’. There are rules which specify what counts as winning, and without those rules winning is not possible at all. The fact that Team X won, unlike the fact that the sun is 93 million miles from the earth, depends on a complicated set of attitudes taken up by the agents involved.¹² (That is a further respect in which the cases of teams and orchestras are special.) But not all collective agency will have this character: there is also a phenomenon which might be called hidden collective agency.

This may also be introduced by example. A number of individuals may form a clique. (Perhaps they all went to the same school or belong to the same leisure interest club.) They interact in ways which have an excluding effect on others: they make allusions which they, but not others, immediately recognise; they anticipate each other’s reactions as others cannot; they share a history and a set of attitudes which others do not. The consequence is that non-members of the clique cannot engage in social exchange in the same way, and feel a general sense of exclusion. Now this phenomenon exhibits the following features of the original examples of collective agency: the entity is a continuing one rather than an ad hoc one lasting only briefly; we may assume that it can persist while some (or over a period perhaps even all) of its constituents change; and it does something distinctive which its individual members do not. (Indeed, perhaps the individuals even cannot do the same, since no individual could have a general excluding effect in this way.) But the phenomenon precisely does not exhibit the feature that its activities come into being as a result of the attitude which its constituents have towards what they are doing. On the contrary, they may simply be unaware of what they are doing collectively, though each is perfectly aware of what they are doing individually.¹³ Even here the point may easily be missed, since shared understandings abound in cliques. However, what is not necessarily present is a shared understanding among its members that they constitute a clique! The example itself may be of no great moment, but its structural features are reproduced in more important contexts. For example, an indigenous population may unwittingly act towards strangers in its midst as members of a clique do, but with results which are politically much more serious.

The notion of a hidden CA actually covers several different possibilities. We may be unaware that a CA exists at all, or unaware of its exact nature; we may be unaware that it has acted on some particular occasion or unaware of the actual significance of what it has done. And ‘we’ here may be either the constituents of the collective agency or observers. In any event, a CA does not necessarily involve shared meanings and understandings in the way intended in some conceptions of community. Though it is necessary for a number of individuals to act together for a CA to be in operation, they may or may not have any shared conception of what they are collectively doing (because individually they may not have any conception at all of what they are collectively doing).

Notice that a CA is not necessarily an overall community. True, a whole village or a whole culture may act in some way significantly different from its constituents taken severally, so that we wish to characterise it as a CA; but at the same time there will many CAs which are very local and partial communities. This has consequences for the issues discussed in section 2.

2 Community, identification and dissociation

One of the central matters of contention in the liberal/communitarian debate was whether ‘the self is prior to the ends which are affirmed by it’.¹⁴ The negative communitarian answer held that the self is an embedded self: it approaches the selection of ends with a particular social identity which predetermines its mode of selecting them, so that, for example, a shared communal end is ‘not a relationship [people] choose (as in a voluntary association) but an attachment they discover’¹⁵ and ‘agency consists less in summoning the will than in seeking self-understanding’.¹⁶

Now any tolerably adequate description of me as an individual will include reference to various social roles which I inhabit – teacher, parent, voter, and so on – and these descriptions will therefore constitute part of my identity. It is then tempting to infer that ‘what is good for me has to be the good for one who inhabits these roles’.¹⁷ Or, as Ross Poole has recently expressed it: ‘An identity defines a perspective on the world and our place in it . . . It calls upon us – or those who have the appropriate identity – to act in one way rather than another.’¹⁸ But the inference is too hasty. There is an important distinction to be observed between identity and identification. As a creature capable of self-consciousness, deliberation and action, it is always open to me to reflect on my identity, to consider whether I wish to continue in the roles I occupy, and (sometimes) to act to divest myself of one or more of them. I can, in other words, choose to identify with or dissociate from a given role. So, for example, if I am a victim of racial or domestic violence, what is good for me is to cease having those descriptions applicable to me. Arguably, something similar is true if I am a member of the Ku Klux Klan. In that way, what is good for me may not be what is good for the inhabitant of a given role. On the contrary, what is good for me is to divest myself of the role. Of course, divesting oneself of a role is not always an option. I cannot cease being someone’s parent or offspring (though even there I may choose not to act in ways associated with the role) and perhaps in practice I cannot cease being a citizen of the state I live in, for example.

Similarly, dissociation specifically from a CA¹⁹ is not always an option. Here, as in the case of individual action, there are the possibilities of compulsive or coerced action, where some of the normal features of control and decision are lacking. (I may act in an army as a conscript, for example). But membership of a CA is peculiarly susceptible to the possibility of dissociation, for reflection on such membership is, precisely, reflection on what one is participating in doing; and, anxieties about determinism aside, what one does is a matter where choices and decision are in principle involved. Moreover, even in cases of coercion a shadow of the options of identification and dissociation persists, in the form of the attitude with which someone participates. If I have been coerced, for example, into taking part in some collective practices which humiliate others, I can still do so reluctantly, affirming to myself that this is something I do not wish to be doing, rather than willingly and with relish.

The need for choice and decision and the possibility of dissociation, rather than solely discovery, are all the more apparent given that CAs are typically partial rather than overall communities. Collective agencies engage in courses of action which sometimes conflict. There may be deep-rooted conflicts between classes or nations or ethnic groups, there may be more tractable conflicts between neighbourhood associations and residents of a particular street. And then sometimes an individual finds that they belong to a number of different CAs which are locked in conflict. You are, say, a parent, an employee, a manager, a member of the board of school governors; and the CAs associated with these descriptions are pulling in different directions. The conflicts between CAs are then reproduced within an individual, who will experience the pull of acting in different directions and will have to make decisions about priorities.

It is not clear to me how wise it is to take a stand on the blanket question whether or not the self is prior to its ends. What can be said with more confidence, however, is that the self is importantly distinct from its ends in the context of particular CAs. Where an individual is participating in collective action with others, a space must always be left for critical reflection, options of identification with or dissociation from the CA and (where this is a live possibility) even actual detachment from a CA. None of this will be settled by mere membership of a CA. But then since CAs are sometimes co-extensive with communities defined in other ways, exactly the same options must remain open in those contexts. To that extent, an individual’s embeddedness fails to settle questions of ends without the addition of critical reflection. (Whether the critical reflection proceeds by reference to abstract principles or to the values of some other community to which someone belongs will be a further question.)

3 Community and the distinctness of persons

A perennial concern in relations between individual and community is the question whether the one type of entity has priority over the other (though it is an important philosophical error to suppose that there is only one kind of priority and that therefore priority must always attach entirely either to the one or to the other). Here, too, current thinking has been influenced by the liberal/communitarian debate, in particular by the appeal to the distinctness of persons frequently made by liberals. Rawls, for example, has argued that ‘the plurality of distinct persons with separate systems of ends is an essential feature of human societies.’²⁰ What we cannot then do, according to Rawls, is use the same kind of reasoning when arriving at social decisions as that used by one individual with one set of ends: the ‘reasoning which balances the gains and losses of different persons as if they were one person is excluded’.²¹

Nozick makes a similar point. Individually, we sometimes choose to undergo some pain or sacrifice for a greater benefit or to avoid a greater harm: in other words, we accept some cost for the sake of the greater overall good. Why should we not also argue that some people must bear some costs so that others may gain, for the sake of the greater overall social good? Nozick’s reply is that

According to the doctrine, it is particularly important to bundle together the desires of a single individual. By contrast, no special importance attaches to a bundle which represents the desires of different individuals for the same end. That explains why the doctrine is invoked to criticise classical utilitarianism, which is taken to allot special importance to the latter kind of bundle, in the interest of maximising overall desire-satisfaction, regardless of whose desires they happen to be.

There is both an implausibility and an incompleteness in the doctrine of the distinctness of persons.²³ The implausibility arises from neglecting the complexity of individuals’ desires. They can reflect on them, accord some higher priority than others, and also acquire meta-desires (as when I desire to smoke but desire not to desire to smoke, or desire that people’s desires should be less conventional). Consequently, individuals themselves may attach importance to the fact that a given end is desired by a number of other people, and they may themselves attach more importance to some desire jointly held by a number of people than to the bundle of their own individual desires. Thus, I might desire a cessation of some incidence of racial oppression, regard this desire as having much higher priority than any other desires I have, recognise that it is a desire held by others, and believe that it is important that there is a high level of desire for this cessation. This combination of beliefs and desires among a number of individuals can itself generate a sense of community – in terms of shared desires – across individuals: we identify with one another as desirers of the same end. Where individuals themselves sum desires across individuals in this way, rather than only seeing them as desires which each of them has individually, it is not clear that we can be so confident that no importance should be attached to such a process of summation across individuals.

Consider an objection. It might be said that in the case in hand the desire is for a state of the world, cessation of racial oppression, rather than for a state of an individual. But, it might be objected, it is only the latter kind of desire which the doctrine is meant to cover. This objection is weak, because all the essentials of the claim could be re-run with individuals’ desire for states of themselves. The earlier example of the National Association for the Advancement of Colored People would illustrate the point. It may matter to me what happens to me as a coloured person, and it may therefore matter to me what happens to people who are like me in that respect. And that may matter more to me than any other questions about my own well-being. Moreover, since many desires are for general states of the world rather than for the individual, it would be a considerable restriction on the scope of the doctrine if it were thought not to be applicable to such general desires. For example, suppose I desire to own a watch. The realisation of the desired state of affairs would involve others (those who make, sell and transport watches). Although the desiring is a state of an individual, the realisation of what is desired would involve other individuals, and it is not clear why the latter fact should be thought any less important than the place of residence of the desire, as it were.²⁴

The incompleteness of the doctrine of the distinctness of persons lies in its failure to allow for the existence of necessarily collective ends, such as winning a team game, performing a symphony or electing a government. It is not at all surprising that we encounter collective desires for such ends, since it is only collectivities of individuals which can actually bring them about. And where CAs have their own characteristic good, such as winning a game, performing a symphony, furthering the interests of the nation or the culture, they or their constituents can indeed undergo sacrifices for the good of that entity: the team sells some of its collective assets, or a member foregoes their wages, in order to buy a player whose presence will enhance the team’s results. Hence, pace Nozick, it is not true that there is no social entity which can undergo a sacrifice for its own good.²⁵

At this point a kind of premature moral panic may occur. It may be felt that, once we reject the distinctness of persons and allow a place for the possibility of a collectivity or an individual being sacrificed for the collective good, we are on the slippery slope to allowing the eclipse of individuals and the incursion of totalitarian collectivism. Such a thought clearly exercises Nozick. The panic is premature for two reasons. First, we are at this stage exploring the appropriate characterisation of the world we live in which will allow us then to draw defensible moral conclusions. To say that the local football team and the Ku Klux Klan each has a characteristic good is not to say that either good should be promoted: that is a further question. If the distinctness of persons erroneously characterises that world, it has to be rejected, and a more suitable characterisation found which will support any set of moral convictions we wish to retain. Second, even if all the claims in section 1 about the irreducibility of CAs are correct, we need to bear in mind that CAs are themselves composed of individuals and nothing else. In the context of a CA individuals have not been eclipsed, but we have to take seriously the idea that in this context they are indissolubly linked. There are, that is, circumstances where we collectively desire something, and that fact is not further dissolvable into circumstances where I do and you do and she and he do. We want to win and we can only want that as a team.

The existence of CAs, therefore, can be invoked in challenging liberal claims that there are only distinct individuals and that desires cannot be summed in any way except as belonging to individuals, just as, in the previous section, it provided the context for challenging the claim implicit in some communitarian thinking that socially embedded people discover rather than selecting their ends.²⁶

Conclusion

In one way the nature of the contemporary world is congenial to a stress on the idea of community, and in another way not. On the one hand, globalisation is a cliché and the interconnections between large numbers of people on a worldwide scale are ever more apparent. The idea of human beings as isolated units seems in that respect less defensible than ever and the expression ‘global village’ more appropriate than ever. On the other hand, the contemporary world signally lacks a feature possessed by at least some literal villages at some times and places, namely that of providing an all-embracing community in which an individual’s life gained its significance from their place in a closed social network. In that connection, it has been a familiar criticism of communitarian theory that it might have been appropriate for well integrated societies, where there was a workable notion of an all-embracing community, but that this presupposes a world which no longer exists (or perhaps never was). As I have indicated in this chapter, there are many conceptions of community more circumscribed than this all-embracing one. The conception I have concentrated on, where what people share in common is participation in collective action, has many instances in actual life, including some (such as multinational corporations) which are peculiar to modern conditions.

We need a fuller account of individuals’ relations to communities of this kind than I have been able to provide here, an account of the forms which identification with them can take and the circumstances in which dissociation is justified, as well as an account of how the actions of collectivities are to be compared and contrasted with the actions of individuals. One aspect of these further matters, alluded to earlier, seems to me particularly important. Just as an individual is not necessarily in the best position to understand the existence or the nature of all of their own actions without further reflection, so collectivities of individuals may be unaware of the existence or the nature of their collective actions. People can co-ordinate their actions in subtle and complex ways and collectively produce results of which they are quite unaware. In that respect, there may be more communities around than are dreamt of in our political philosophy.