Human beings act together in characteristic ways. Forms of shared activity matter to us a great deal, both intrinsically – think of friendship and love, singing duets, and the joys of conversation – and instrumentally – think of how we frequently manage to work together to achieve complex goals. My focus will be on activities of small, adult groups in the absence of asymmetric authority relations within those groups. My approach begins with an underlying model of individual planning agency, and then seeks a conceptual and metaphysical bridge from such individual planning agency to modest forms of sociality.

Shared Intention

Suppose you and I are painting a house together. What makes this a shared intentional activity? We could imagine a contrast case in which we each intentionally go through the same motions as we do when we paint the house together, and yet there is no shared intentional activity. Perhaps we are each set only on our individual painting project and respond to each other only with an eye to avoiding collisions. Echoing Wittgenstein's question about the difference, in the individual case, between my arm's rising and my raising it, we can ask: what is the difference between such a contrast case and shared intentional activity? In the case of individual human action we can see the difference as involving an explanatory role of relevant intentions of the individual agent.

TWO APPROACHES TO INSTRUMENTAL RATIONALITY

Suppose I intend end E, believe that a necessary means to E is M, and believe that M requires that I intend M. My attitudes concerning E and M engage a basic requirement of practical rationality, a requirement that, barring a change in my cited beliefs, I either intend M or give up intending E. Call this the Instrumental Rationality requirement – for short, the IR requirement.

Suppose now that I believe that E, and I also believe that E will only occur if M. My beliefs engage a basic demand of theoretical rationality, a demand that, roughly, either there be a change in at least one of these two beliefs or I believe M. Call this the Belief-Closure requirement – for short, the BC requirement. BC, note, is not a consistency demand on my beliefs: failure to add the further belief that M need not involve inconsistency in the way that adding a belief that not-M would. Nevertheless, something like BC seems a basic rationality constraint on belief.

Both IR and BC express constraints on the coherence of the agent's relevant attitudes; and these constraints are aspects of the normal rational functioning, in the psychic economy of believing-and-intending agents, of the cited attitudes. The intentions and beliefs of such agents will tend to be responsive to these constraints. But the requirements differ in important ways.

PLANNING AND CORE ELEMENTS OF AUTONOMY

Humans seem sometimes to be autonomous, self-governed agents: Their actions seem at times to be not merely the upshot of antecedent causes but, rather, under the direction of the agent herself in ways that qualify as a form of governance by that agent. What sense can we make of this apparent phenomenon of governance by the agent herself?

Well, we can take as given for present purposes that human agents have complex psychological economies and that we frequently can explain what they do by appeal to the functioning of these psychological economies. She raised her arm because she wanted to warn her friend; she worked on the chapter because of her plan to finish her book; she helped the stranger because she knew this was the right thing to do; he left the room because he did not want to show his anger. These are all common, everyday instances of explaining action by appeal to psychological functioning. In doing this, we appeal to attitudes of the agent: beliefs, intentions, desires, and so on. The agent herself is part of the story; it is, after all, her attitudes that we cite. These explanations do not, however, simply refer to the agent; they appeal to attitudes that are elements in her psychic economy. The attitudes they cite may include attitudes that are themselves about the agent and her attitudes – desires about desires, perhaps. But what does the explanatory work is, in the end, the functioning of (perhaps in some cases higher-order) attitudes. These explanations are, I will say, nonhomuncular.

THE AUTONOMY-HIERARCHY THESIS

In autonomous action the agent herself directs and governs the action. But what is it for the agent herself to direct and to govern? One theme in a series of articles by Harry G. Frankfurt is that we can make progress in answering this question by appeal to higher-order conative attitudes. Frankfurt's original version of this idea is that in acting of one's own free will, one is not acting simply because one desires so to act. Rather, it is also true that this desire motivates one's action because one desires that this desire motivate one's action. This latter desire about the motivational role of one's desire is a second-order desire. It is, in particular, what Frankfurt calls a second-order “volition.” And, according to Frankfurt's original proposal, acting of one's own free will involves in this way such second-order, and sometimes yet higher order, volitions.

Frankfurt's hierarchical proposal has met with a number of challenges and has been subject to clarification and emendation. I myself have elsewhere tried to map out some details of this debate. My concern here, however, is with the very idea that there is a close connection between autonomous agency and motivational hierarchy.

Of course, much depends on what kind of close connection one has in mind. Some might argue that all cases of human autonomous agency essentially involve motivational hierarchy. But I will focus on a somewhat weaker claim.

In Choice: The Essential Element in Human Action Alan Donagan argued for the importance of “will” to our shared understanding of intelligent action. By “will” Donagan meant a complex of capacities for forming, changing, retaining, and sometimes abandoning our choices and intentions. (Choice is, for Donagan, a “determinate variety of intending.”) Our capacity to intend is to be distinguished both from our capacity to believe and from our capacity to be moved by desires. And Donagan thought that intentions involve what, following Austin, he called “ ‘as it were’ plans.”

I am broadly in agreement with these main themes in Donagan's book, and I will pretty much take them for granted in what follows. I will suppose that intention is a distinctive attitude, not to be reduced to ordinary desires and beliefs; that intentions are central to our shared understanding of ourselves as intelligent agents; and that “the study of intention” is in part the “study of planning.” My hope is that these common elements in our views about intention can serve as a basis for reflection on the phenomenon of shared intention.

That we do sometimes have intentions that are in an important sense shared seems clear. We commonly report or express such shared intentions by speaking of what we intend or of what we are going to do or are doing. Speaking for you and myself I might say that we intend to paint the house together, to sing a duet together; and I might say that we are going to New York together.

We are planning agents. Our purposive activity is typically embedded in multiple, interwoven quilts of partial, future-directed plans of action. We settle in advance on such plans of action, fill them in, adjust them, and follow through with them as time goes by. We thereby support complex forms of organization in our own, temporally extended lives and in our interactions with others; and we do this in ways that are sensitive to the limits on our cognitive resources. These facts are, I believe, an important key to an adequate philosophical treatment of (1) the very idea of intention, (2) basic features of our agency, (3) important forms of shared agency, and (4) important forms of responsible agency.

I discussed planning agency, and its significance to (1) and (2), in my 1987 book, Intention, Plans, and Practical Reason. Since then I have tried to elaborate and to deepen this approach, to extend it to (3) and to (4), and to explore its relations to the work of others. The present volume of previously published work includes eleven essays that were the result of this effort, as well as a pair of earlier critical studies that also seemed useful to include.

At the heart of my theory is a model of our planning agency and the attempt to use it to understand intention. I call this the planning theory of intention. The main idea is to see intentions as elements of stable, partial plans of action concerning present and future conduct.

REASONING: THEORETICAL AND PRACTICAL

In theoretical reasoning our concern is with what to believe, in practical reasoning with what to intend and to do. In the former we are trying to find out how the world is; in the latter we are concerned with how to (try to) change it. These can seem to be two different – though no doubt importantly related – enterprises. J. David Velleman argues otherwise. His view is “that practical reasoning is a kind of theoretical reasoning, and that practical conclusions, or intentions, are the corresponding theoretical conclusions, or beliefs” (p.15). These twin identifications – of practical reasoning with a kind of theoretical reasoning, and of intention with a kind of belief – are at the heart of this wide-ranging, imaginative, and important study of autonomous agency, rationality, and value. Let us call this pair of identifications “cognitivism about practical reason.” Velleman's cognitivism about practical reason lies at the foundations of his book. I want to know if it is defensible.

SPONTANEOUS SELF-KNOWLEDGE

Velleman's route to his main conclusions begins with reflection on the special kind of knowledge and understanding we normally seem to have of our own intentional conduct. When I act intentionally I usually know what I am doing and I usually know at least a rudimentary explanation of why. Further, I seem to know this in a distinctive way. As Velleman puts it, there is a kind of “spontaneity” in my knowledge: I do not need to step back and observe what I do and infer what my motives are in the way you would need to for you to have such knowledge about me.

Much of our behavior is organized. It is organized over time within the life of the agent, and it is organized interpersonally. This morning I began gathering tools to fix the bicycles in the garage. I did this because I was planning to go on a bike trip tomorrow with my son and knew I needed first to fix the bikes. Just before I gathered the tools, I was on the telephone ordering tickets for a trip to Philadelphia next week. I did that because I was planning to meet my friend in Philadelphia then. If all goes well, each of these actions will be part of a distinct coordinated, organized sequence of actions. Each sequence will involve coordinated actions, both of mine and of others. These sequences will also need to be coordinated with each other. Such coordination of action – between different actions of the same agent and between the actions of different agents – is central to our lives.

How do we accomplish this organization? It seems plausible to suppose that part of the answer will appeal to commonsense ideas about planning. It is part of our commonsense conception of ourselves that we are planning agents (Bratman 1983, 1987). We achieve coordination – both intrapersonal and social – in part by making decisions concerning what to do in the further future. Given such decisions, we try to shape our actions in the nearer future in ways that fit with and support what it is we have decided to do in the later future. We do that, in large part, by planning.

In several recent papers I have sketched a general approach to phenomena of shared agency that do not involve relations of authority (Bratman, 1992; 1993). I focused, in particular, on what I called shared intention, shared intentional activity, and shared cooperative activity. The basic idea was that at the heart of these phenomena is shared intention – a shared intention, for example, to paint the house together. Shared intentional activity, in the basic case, is activity suitably explainable by a shared intention and associated forms of mutual responsiveness. Shared cooperative activity requires, further, the absence of certain kinds of coercion, and commitments to mutual support in the pursuit of the joint activity.

What are shared intentions? My strategy here was two-pronged. I tried to specify roles distinctive of shared intention: roles such that it would be plausible to identify shared intention with what plays those roles. I argued, in particular, that our shared intention to J plays three interrelated roles: It supports coordination of our intentional activities in the pursuit of J, it supports associated coordination of our planning, and it structures relevant bargaining (Bratman, 1993, p. 99 [this volume, p. 112]). I then argued that a certain kind of public, interlocking web of intentions of each of us would play those roles. This supported my conjecture that shared intention could be identified with that web of intentions of the individuals.

Shared intentions are intentions of the group. But I argued that what they consist in is a public, interlocking web of the intentions of the individuals.

The Sources of Normativity derives from the 1992 Tanner Lectures and Seminar in Cambridge, England. It consists of a brief introduction by Onora O'Neill, a “Prologue” and four lectures by Christine Korsgaard, separate discussions of these lectures by G. A. Cohen, Raymond Geuss, Thomas Nagel, and Bernard Williams, and a concluding “Reply” by Korsgaard. It is a major work of the first importance.

We have beliefs and intentions, and we perform actions. In each case we make ought judgments that purport to say what to believe, what to intend, what to do. In making claims on what we believe, intend, or do such judgments are normative, and we can ask what the source, the justification, of this normativity is. Morality, in particular, makes normative claims – sometimes quite demanding – on (at least) what we intend and do. We can ask what the grounds are for the normativity of such moral demands.

Korsgaard's main question – which she labels “the normative question” (10) – is this last question about the “claims morality makes on us.” (10). But her answer involves a general approach to grounds for normative claims on what we intend and do. Her answer – which she sees as Kantian, and notes (99n) has similarities to views of Harry Frankfurt – is that the basic ground of practical normativity is “the reflective structure” of our minds (100). We – normal, mature human beings – have the capacity to be reflectively aware of the desires and inclinations that tend to move us to intention and action.