Joint Action Research Paper

1. Some Initial Considerations On Joint Action

1.1 Joint Intentions Are Integral To Joint Actions

Contemporary philosophy of action has tended to focus not on joint action but on the action of an individual person. A notion central to the philosophy of action is that of intention. It is commonly argued that what distinguishes an action (such as a man raising his arm) from a mere bodily movement (such as his arm rising) is an intention. When a man raises his arm, his arm rises in part by virtue of his intention to raise it. The sociologist Max Weber made a similar point when he characterized action as meaningful behavior.

Joint actions, similarly, appear to presuppose intentions, in this case joint intentions. Thus, for example, people cannot be conversing unless there is a joint intention to converse. The parties to a conversation act in light of a joint intention to converse, so as to implement that intention. A joint intention may be referred to by such statements as ‘We intend to go shopping, ’where our intending to go shopping is not, or is not simply, a matter of each of the people in question personally intending to go shopping. Precisely what is at issue is not obvious. Much of the growing body of philosophical work relevant to joint action is concerned with the nature of joint intention, or, in less formal terms, what it is for us to intend.

1.2 Joint Action Must Be Cooperative Only In A Broad Sense

Joint action is not necessarily a matter of amicable cooperation. We may be quarrelling with one another or engaged in a fight. Quarrelling and fighting are species of joint action. Though it need not be amicable, there is clearly a broad sense in which joint action is always cooperative or collaborative. I cannot quarrel with you unless you quarrel with me; I cannot fight with you unless you fight with me. If you refuse to take me seriously we cannot quarrel. If you lay down your arms, we are no longer fighting.

1.3 Some Consequences Of Involvement In Joint Action

Informal observation of the way people behave when doing things together makes it clear that the parties often rebuke one another for behavior contrary to the relevant joint intention. Such rebukes are commonly understood to be in order. For example, Jill and John are walking together, John begins to lag behind, and Jill rebukes John for doing so. John understands that, in the context of their walking together, Jill has the standing to rebuke him.

Less punitive responses are also commonplace, and routinely understood to be in order. For example, Jill asks John why he is lagging behind. John may not like being questioned like this, but he understands that Jill has a legitimate interest in the matter.

Underlying these responses is the understanding that, all else being equal, participation in a joint intention entitles each of the parties to relevant actions from the others. If one party does not produce the relevant actions, therefore, the others have the standing to question and to rebuke that party, as they see fit.

A further point is this. Suppose John and Jill jointly intend to go for a five mile walk. At some point along the way John decides that this will be too tiring for him. He may then announce that he is turning back, and he may indeed turn back. If he does these things, then, failing some special background circumstances, he will be acting contrary to understandings that have been set up between him and Jill. It would be different if he said or did something that amounted to asking for and gaining Jill’s permission for his turning back.

In general terms, no one party to a shared intention is in a position to cancel or change that intention unilaterally. Though I am in a position unilaterally to change my own mind, I am not in a position unilaterally to change our mind, that is, to change our joint intention.

1.4 Inferences From Statements About What ‘We Intend’

A final observation on joint action is this. From ‘We intend to walk for five miles’ and ‘In order for us to walk for five miles, I must walk for five miles,’ it seems that John can infer that he has reason to walk for five miles. In other words, our intention can give me a reason for acting. It is at least not obvious that there need be any interposition in this reasoning of a premise about what John himself intends.

2. Two Central Questions

A satisfying account of joint action will incorporate an explanation of the observations just given. Two relevant questions deserve particular emphasis.

2.1 What Is The Relation Of Joint Intentions To Personal Intentions?

Are the joint intentions that underlie joint actions basically concatenations of personal intentions of the participants, perhaps personal intentions in favor of the joint action? This may be the conjecture that most immediately springs to mind. As we shall see, however, serious problems attend this idea.

2.2 Of What Kind Are The Entitlements Associated With Joint Intention?

The parties to a joint intention understand their participation to entitle each of them to actions in conformity with the relevant joint intention. How is this understanding to be explicated? In particular, what kind of entitlement is at issue? One might assume that it is moral entitlement founded on moral principle. On the other hand, it may seem that joint intention is too basic and primitive a phenomenon to require moral understanding in all participants. It is certainly not obvious that it does require such understanding.

3. Some Problems With Accounts Of Joint Intention In Terms Of Meshing Personal Intentions

A carefully constructed account of joint intention that puts personal intentions at center stage is that developed by Bratman (Bratman 1999 contains a number of relevant articles). This may serve as an example of this type of account for the present purposes.

A somewhat simplified version of Bratman’s account runs as follows. We jointly intend to go for a walk if and only if (a) (i) I intend that we go for a walk, (ii) you intend that we go for a walk. (b) I intend that we go for a walk in accordance with and because of (a) (i) and (b) (ii); you intend likewise. (c) (a) and (b) are common knowledge between us.

Bratman does not define the ‘common knowledge’ that he requires in condition (c). What is intended is, roughly, that (in the example) the fact that conditions (a) and (b) are fulfilled is completely out in the open as far as the two people involved are concerned. (For the first published discussion of ‘common knowledge’ see Lewis 1969.)

There are several problems with this account of joint intention. Two will be detailed here. Both are common to accounts that put personal intentions of the participants at the foundation of joint intention and hence joint action.

One problem has to do with the entitlements associated with joint intention. It does not appear that anyone is entitled to relevant actions from anyone else just by virtue of the satisfaction of the conditions laid down in the account. My intending that we go for a walk, and intending that this occur by virtue of my so intending and your so intending, does not seem to entitle you to any particular action from me. Nor does the fact that my various intentions are entirely out in the open between us seem sufficient to bring any such entitlement into existence.

If you know that I intend that we go for a walk, and have reason to think that I won’t change my mind, then there is a sense in which you are ‘entitled to expect’ that I am going to make the appropriate contribution to our walking together. But this just means that you have a good basis for predicting that I am going to act thus. It does not mean that you are entitled to my acting thus in such a way that questioning or rebuking me is in order.

In related discussion, Bratman has invoked a complex moral principle to argue for the presence of the relevant entitlements in many situations in which his conditions on joint intention hold. The principle in question is the ‘principle of fidelity’ proposed in Scanlon (1990). This principle runs as follows (Scanlon 1990, p. 208): If (a) A voluntarily and intentionally leads B to expect that A will do x (unless B consents to A not doing x); (b) A knows that B wants to be assured of this; (c) A acts with the aim of providing this assurance, and has good reason to believe that he or she has done so; (d) B knows that A has the beliefs and intentions just described; (e) A intends for B to know this, and knows that B does know it; and B knows that A has this knowledge and intent; then, in the absence of some special justification, A must do x unless B consents to x not being done.

Scanlon’s principle of fidelity is not applicable in all situations where there are joint intentions on Bratman’s account of them. So appeal to this principle will not rescue Bratman’s account of joint intention insofar as joint intention by its nature entitles its participants to conforming actions.

Another problematic aspect of Bratman’s type of account is that it does not explain the need for the concurrence of all parties in order to dissolve the joint intention. One with a personal intention is in a position to change it unilaterally. It follows that an individual who is party to a joint intention in Bratman’s sense is in a position unilaterally to do away with one of the foundational personal intentions at any time. There is no need for the permission of the other party or parties. Thus a joint intention in this sense appears to be something relatively flimsy, flimsier than joint intentions as these are ordinarily understood.

4. The Irreducibility Of ‘We Intend To Do Such-and-Such’

John Searle has claimed that at the core of joint intention and action lie irreducible thoughts of the form ‘We intend to do such-and-such’ in the mind of each participant (see, for instance, Searle 1995; see also Sellars 1963). These thoughts are irreducible in the sense that they cannot be reduced or broken down into thoughts about what ‘I intend’ and what ‘you intend.’ Searle’s central concern is to describe the inner or subjective side of joint intention with respect to each participant. He evidently believes that there are requirements for the correctness of thoughts of the form ‘We intend to do such-and-such,’ though he does not focus on this issue.

What he writes at one point suggests that, in his view, one person’s thought of the form ‘We intend to do such-and-such’ is correct as long as the other person or persons in question also think that thought with respect to the same people. This proposal does not seem to be right, if ‘We intend to do such-and-such’ is understood in accordance with everyday understandings about joint intention.

Suppose that Ben is currently thinking, with respect to himself and Elaine, ‘We intend to get married.’ Elaine is in a similar position. According to the proposal under discussion, it follows that each one is correct in thinking ‘We intend to get married.’ Yet according to everyday understandings it would not be plausible to claim that, solely by virtue of the facts described, Ben and Elaine jointly intend to get married. If one learned that they had never discussed the matter one would judge that their thoughts were not properly grounded.

5. A Role For Joint Commitment

5.1 Joint Commitment

When is one right in thinking an irreducible thought of the form ‘We intend to do such-and-such,’ according to everyday understandings? Margaret Gilbert has proposed that, in a sense she explains, the parties need to have established a joint commitment to intend as a body to do such-and-such (see Gilbert 1999).

The relevant notion of commitment may be introduced by reference to a personal decision. If Jane decides to go shopping this afternoon, she is then subject to a personal commitment to go shopping. She is not subject to such a commitment if she merely inclines to go shopping or hopes to do so.

According to Gilbert, people can be subject both to personal and to joint commitments. A joint commitment in Gilbert’s sense is not an aggregate or conjunction of corresponding personal commitments of each party. Rather, it is a commitment of all the parties, or of the whole. By its nature a joint commitment requires the participation of each party for its creation and its dissolution.

Gilbert has argued that the idea of a joint commitment is a fundamental concept that underpins not only joint intentions and goals but also joint beliefs, and social rules and principles (see Gilbert 1989, 1996, 1999). Those engaged in joint action act in light of the joint commitment underlying the relevant joint intention or goal.

There are both basic and derived forms of joint commitment. In order to create a joint commitment of the basic type, each party must express to the rest his or her readiness to be committed with them through a joint commitment whose content is specified. When this has been done in conditions of common knowledge, the relevant joint commitment is in place. In general, when a joint commitment is formed the parties to it are committed simultaneously. They are committed through the joint commitment. Similarly, no one can be rid of their commitment through the joint commitment unless the others are. Once it is rescinded for one it is rescinded for all.

Without prior authorization, no one can unilaterally rescind or dissolve a joint commitment: all must rescind it together. Otherwise it stands as a constraint on the parties. A joint commitment can of course be violated by one party acting alone, but this is precisely a matter of violation as opposed to dissolution. Unlike those who rescind or dissolve it, the violator acts against a standing joint commitment.

Gilbert construes the idea of being jointly committed to intend as a body somewhat as follows. The parties are jointly committed to constitute together, as far as is possible, a single body that intends to do such-and-such. Though they are plural, or many, they are jointly committed to make themselves as much like a singular subject of intention as they can. A joint commitment to intend as a body is sufficient to create, in Gilbert’s terminology, a ‘plural subject’ of intention. In other words, in her terms, we intend to do such-and-such if and only if we constitute a plural subject of intention.

How do people express to one another their readiness to be jointly committed to intend as a body to do such-and-such? There are many such ways. Among these are such humdrum exchanges as ‘Shall we go for a walk this afternoon?’ with a following ‘Sure, let’s.’

Joint commitments are not necessarily set up through verbal exchanges. Two men who jump into a rowing boat in hot pursuit of a fugitive may signal by their attempts to coordinate the movements of the oars that, in effect, they stand ready to be jointly committed to intend as a body to row the boat. As each moves the oars in light of their ensuing commitment they engage in the joint action of rowing the boat.

In principle it may take some time for a particular joint commitment to be established. Each party’s readiness to be party to the joint commitment may be made clear only gradually and over the course of many interactions with the other people in question.

5.2 Derived Joint Commitments

The foregoing examples of the formation of a joint commitment involve joint commitments of the basic type. A joint intention in Gilbert’s sense may also involve a derived commitment, as in the following example.

Don asks his wife Sally ‘Where do we plan to go on vacation this summer?’ He presumes that though he is unaware of any such plan, there may already be such a plan, and Sally will know what it is. This presumption is based on the existence of a background commitment of the basic type. Don and Sally are jointly committed to accept as a body that Sally may make vacation plans for the two of them.

With this background commitment in place, Sally is in a position unilaterally to create joint vacation plans. In other words, by virtue of the background joint commitment and Sally’s planning, Don and Sally may take on a derived joint commitment to plan or intend as a body to go to Spain this summer, for instance.

5.3 Large Populations

Members of large populations, such as the many inhabitants of a large island, are unlikely to know one another personally, or even to know of one another as individuals. If there are to be population-wide joint commitments in such populations, this will be through expressions of readiness to be jointly committed with all others with a certain common feature, such as ‘those living on this island.’

By virtue of background joint commitments of the kind discussed in Sect. 5.2, the members of a large population can have a particular joint intention in Gilbert’s sense though many of the members do not know of the intention in question. This accords with everyday understandings according to which it may be said that, for instance, a particular country intends a particular action, without it being assumed that everyone in that country knows of the intention, or, indeed, of the ensuing action.

Those who form intentions for the whole population may be authorized to ensure that these intentions are implemented. Their implementation need not involve all members of the population in order that the action in question be considered an action of the population as a whole.

5.4 Obligations And Entitlements Of Joint Commitment

Gilbert has argued at some length that the concept of a joint commitment underlies a central concept of obligation, a concept of obligation for which there is a corresponding concept of entitlement or right (see Gilbert 1996, 1999). A brief indication of her position might note that if one violates a commitment that is joint, one is evidently answerable to the other parties to the commitment. They can observe that one has violated ‘our’ commitment as opposed to a commitment of one’s own. They have a clear basis for questioning the violator and for administering rebukes on their own behalf. A natural way to characterize the situation in familiar terms is to say that the parties to a joint commitment are all entitled to conformity from the other parties. These others have corresponding obligations to conform: they are obligated to all parties to the commitment, their obligation being to conform to the commitment.

This and related arguments are arguments from the nature and structure of joint commitment. It may thus be best not to think of the relevant entitlements and obligations as moral entitlements and obligations. All that one needs to understand, in order to understand that they are there, is the nature of a joint commitment. One need not appeal to the value of anything such as, in the case of Scanlon’s principle of fidelity, the value of what Scanlon refers to as ‘assurance.’

Going back now to Gilbert’s account of joint intention and action, it offers an explanation of how those who are party to a joint intention have associated entitlements to conformity from one another. These are entitlements of joint commitment. There are derivative entitlements to ask questions, rebuke one another, and so on, should deviation from the joint intention be mooted. Gilbert can allow that in some situations where there are joint intentions it is morally permissible not to act in accordance with them. Such situations may include those where one party coerced another into participating in the relevant joint commitment, and those where the joint commitment is to intend something immoral. All things considered, then, it may be morally permissible not to fulfil the relevant obligations of joint commitment. This does not entail that those obligations no longer exist. They exist as long as the sustaining joint commitment exists (see Gilbert 1996).

5.5 Joint Intentions As The Intentions Of Plural Subjects

Gilbert’s plural subject account of joint intention and action does not appeal to personal intentions in favor of some joint action. She holds, in effect, that one who says ‘We intend to do such-and-such,’ referring to a joint intention, does not say something reducible to statements about what ‘I’ intend and what ‘you’ intend. To that extent she and Searle are in agreement.

The plural subject account of joint intention provides an explanation of how from an irreducible ‘We intend to row to the other side’ and ‘My rowing more slowly than I am doing will best facilitate our rowing to the other side,’ I can infer ‘I should row more slowly.’ My participation in the relevant joint commitment gives me reason to do what I can to implement the joint intention. It is not necessary, given this account, that I first form a personal intention that we row to the other side.

6. Applications Of The Plural Subject Theory Of Joint Intention And Related Phenomena

Concurring with the judgement of sociologists, Gilbert has argued that those who participate in joint intentions and actions constitute a social group, though the group may be a small and transient one (see Gilbert 1989). She argues, more generally, for the claim that plural subjects in her sense constitute an important, central kind of social group. Thus it can be argued that obligations of a special kind are part and parcel of group membership.

This line of argument makes it possible to bring Gilbert’s plural subject theory of joint action and related phenomena to bear on a classic question of political philosophy, the question of political obligation. Is there an obligation to obey the laws of one’s own country in particular, and, if so, what is the source of this obligation? Evidently, one possible source is participation in a joint commitment. Thus one might participate, for instance, in a joint commitment to accept as a body the rules prescribed by a certain person as the rules of the jointly committed population as a whole (see Gilbert 1996, 1999).

Gilbert’s plural subject theory is also relevant to the topic of collective guilt in the sense of collective moral responsibility. According to Gilbert, an act is constituted as a joint act by virtue of an underlying joint commitment. That joint commitment unifies the people in question into a group. It may then sometimes be reasonable to ascribe moral responsibility for a joint act to a group as such, quite apart from the question of the moral responsibility of particular group members. This perspective on joint action also allows for the intelligibility of a common phenomenon: the experience of pride or guilt over the actions of one’s group, as opposed to one’s own actions (see Gilbert 1996).

7. The Nature And Role Of Agreements

Another philosopher who has written at length on joint intention and action is Raimo Tuomela. In Tuomela (1984), he put what he called ‘we-intentions’ at the center of an account of joint action. Each party to a joint action must have a relevant we-intention. Tuomela’s complex definition of a we-intention runs roughly as follows: a member Ai of a collective G weintends to do X, where X is some act of G, if and only if (a) Ai intends to do his own part of X, given that every member of G will do X his own part of X; (b) Ai believes that every member of G will do his own part of X; (c) there is a mutual belief in G to the effect that (a) and (b).

Tuomela (1995) introduces the ‘underlying assumption that we-intentions are formed in a process involving the making of an explicit or implicit agreement (or at least believed agreement)’ (p. 127). Agreements determine what joint action is afoot; we-intentions ensue. If all goes as the parties we-intend, joint actions follow.

Clearly, agreements are a paradigmatic context for the production of joint intentions and actions. However, as the example of the rowers suggests, agreements as these are ordinarily understood and are not essential. If one says that there is an implicit agreement in such cases, the question arises as to what precisely this claim amounts to, since an implicit agreement does not seem to be an agreement proper. One might do well to start with an account of paradigmatic agreements. There has been less concentrated philosophical focus on such agreements than might have been expected, given the undoubted importance of agreements within the scheme of everyday social concepts. A standard claim, perhaps echoing statements in contract law, has been that an agreement is an exchange of promises. Gilbert has argued at length against this view, on one interpretation of it, and in favor of the proposal that typical everyday agreements are, in effect, joint decisions, where a joint decision is constituted by a joint commitment to endorse as a body the decision in question (see in particular Gilbert 1996, Chap. 13).

Suppose that Phyllis and Joe jointly commit to endorse as a body the decision to get married next year. Presumably it is reasonable then to regard them as jointly committed to intend as a body to get married next year. Given Gilbert’s account of agreements and joint intentions, then, agreements give rise to joint intentions, providing a basis for joint action. Agreements are not necessary for joint action, however, insofar as a joint commitment to intend as a body can be brought into being in the absence of a joint commitment to endorse as a body a certain decision. This accords with the way the relationship of agreements to joint actions is commonly understood.

8. Individualism, Holism, And Joint Action

Annette Baier has objected to the ‘individualistic bias’ she finds in recent philosophical work on joint intention and action (see Baier 1997). She objects in particular to the assumption that multiperson actions are constructed out of single persons, their abilities, and intentions. Baier proposes that doing things with others is more basic than doing things alone. To say this, however, is not yet to explain what doing things with others amounts to.

Clearly, the range of existing philosophical accounts of joint intention and action is wide. They include accounts in terms of personal intentions, with appeal to moral principles such as Scanlon’s principle of fidelity to account for the associated entitlements and obligations. (This appeal is to be found in both Bratman’s and Tuomela’s work.) They also include Gilbert’s account, which derives the associated entitlements and obligations from a joint commitment. Gilbert’s account is a nonindividualistic, or holistic, account, at least insofar as a joint commitment in her sense is not simply a conjunction of personal commitments.

Accounts that make personal intentions central accord with the approach to social phenomena of Max Weber who put at the center of his sociological concepts the concept of an action that is oriented to the actions of others (called by Weber a ‘social action’). They harmonize well, also, with the somewhat more recent idea of characterizing social processes in terms of the mathematical theory of games. Gilbert’s plural subject account of social phenomena accords rather with the approaches of Emile Durkheim and Georg Simmel, among sociologists, and to Jean-Jacques Rousseau among political philosophers. (Gilbert 1989 proposes an interpretation of Durkheim that suggests a commonality between Gilbert’s approach and that of Durkheim.) These thinkers all emphasized in one way or another that people engaged in joint action were unified in a significant sense.

Bibliography:

1. Baier A C 1997 The Commons of the Mind. Open Court, Chicago
2. Bratman M E 1999 Faces of Intention: Selected Essays on Intention and Agency. Cambridge University Press, Cambridge, UK
3. Durkheim E 1982 The Rules of Sociological Method, 1^st American edn. Free Press, New York
4. Gilbert M 1989 On Social Facts. Routledge, London
5. Gilbert M 1996 Living Together: Rationality, Sociality, and Obligation. Rowman and Littlefield, Lanham, MD
6. Gilbert M 1999 Sociality and Responsibility: New Essays in Plural Subject Theory. Rowman and Littlefield, Lanham, MD
7. Lewis D K 1969 Convention: A Philosophical Study. Harvard University Press, Cambridge, MA
8. Rousseau J-J 1983 On The Social Contract. Hackett, Indianapolis, IN
9. Scanlon T 1990 Promises and practices. Philosophy and Public Affairs 19: 199–226
10. Searle J R 1995 The Construction of Social Reality. Free Press, New York
11. Sellars W 1963 Imperatives, intentions, and the logic of ‘ought.’ In: Castaneda H-N, Nakhnikian G (eds.) Morality and the Language of Conduct. Wayne State University Press, Detroit, MI
12. Simmel G 1971 Georg Simmel: On Individuality and Social Forms. University of Chicago Press, Chicago
13. Tuomela R 1984 A Theory of Social Action. Dordrecht, Boston
14. Tuomela R 1995 The Importance of Us. Stanford University Press, Stanford, CA
15. Weber M 1922 Economy and Society. University of California Press, Berkeley, CA, Vol. 1