Shared Intention Author(s): Michael E. Bratman Source: Ethics, Vol. 104, No. 1 (Oct., 1993), pp.

97-113 Published by: The University of Chicago Press Stable URL: http://www.jstor.org/stable/2381695 Accessed: 04/08/2010 12:40
Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at http://www.jstor.org/page/info/about/policies/terms.jsp. JSTOR's Terms and Conditions of Use provides, in part, that unless you have obtained prior permission, you may not download an entire issue of a journal or multiple copies of articles, and you may use content in the JSTOR archive only for your personal, non-commercial use. Please contact the publisher regarding any further use of this work. Publisher contact information may be obtained at http://www.jstor.org/action/showPublisher?publisherCode=ucpress. Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printed page of such transmission. JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact support@jstor.org.

The University of Chicago Press is collaborating with JSTOR to digitize, preserve and extend access to Ethics.

http://www.jstor.org

Shared Intention* MichaelE. Bratman

In Choice:TheEssential Element in Human Action Alan Donagan argued forthe importanceof "will"to our shared understanding of intelligent action.1By "will"Donagan meant a complex of capacitiesforforming, changing,retaining, and sometimesabandoning our choices and intentions. (Choice is, forDonagan, a "determinate variety of intending.")2 Our capacity to intend is to be distinguishedboth fromour capacity to believe and fromour capacityto be moved by desires.And Donagan thoughtthat intentionsinvolve what,followingAustin,he called "'as it were' plans."3 I am broadly in agreementwiththese main themesin Donagan's book, and I will pretty much take them for granted in what follows.4 I will suppose thatintentionis a distinctive not to be reduced attitude, to ordinary desiresand beliefs;thatintentions are centralto our shared of ourselves as intelligent understanding agents; and that "the study
* Thanks to Margaret Gilbertand Raimo Tuomela, thoughtful commentatorson presentationsof earlier, shorterversions of this article. Thanks also to Philip Clark, Rachel Cohon, Fred Dretske, David Hilbert, Henry Richardson, and Debra Satz for theiruseful philosophical advice. Barbara Herman and David Velleman provided rich and probingcommentswhen thisarticlewas presentedat the September 1992 Memorial Conferencein Honor of Alan Donagan, held at the University of Chicago. Some of the issues theyraised are discussed further in my"Shared Intentionand Mutual Obligation" (presented at the Pacific Division American Philosophical Association,San Francisco, March 1993). Work on this articlewas supported in part by the Center for the Study of Language and Information, made possible in part throughan award fromthe System Development Foundation. 1. Alan Donagan, Choice:TheEssential Element in HumanAction (London: Routledge & Kegan Paul, 1987). 2. Ibid., p. 97. 3. Ibid., p. 96. 4. I developed ideas that are in some respectssimilarto Donagan's themes in my Intention, Plans, and Practical Reason (Cambridge, Mass.: Harvard UniversityPress, 1987). Of course, there are various differences in our views. Donagan discusses one of these-concerning the consistency demands to whichintentions are subject-in Choice, pp. 98-105. My detailed treatmentof choice differs in certain ways fromDonagan's (see Intention, Plans and PracticalReason, chap. 10). And there are other differences as well. But these differences are not relevanthere. Ethics104 (October 1993): 97-113 of Chicago.All rights reserved. 0014-1704/94/0401-8731$01.00 X 1993 byThe University

97

98

Ethics

October 1993

of intention"is in part the "studyof planning."5My hope is thatthese common elements in our views about intentioncan serve as a basis for reflection on the phenomenon of shared intention. I

That we do sometimeshave intentionsthatare in an importantsense shared seems clear. We commonly reportor express such shared intentions by speaking of what we intendor of what we are going to do or are doing. Speaking foryou and myselfI mightsay thatwe intendto to sing a duet together;and I mightsay that paint the house together, we are going to New York together.In each case I reportor express a shared intention. of structured social groups: Sometimeswe speak of the intentions the Philosophy Department, for example, intends to strengthenits undergraduateprogram. But some shared intentionsare not embedstructures. These will be my main concern ded in such institutional here: I will focus on cases of shared intentionthat involve only a structures and pair of agents and do not depend on such institutional authorityrelations. Supposing, for example, that you and I have a shared intentionto paint the house together,I want to know in what that shared intentionconsists.6 On the one hand, it is clearlynot enough for a shared intention to paint the house togetherthateach intendsto paint the house. Such do not even insurethateach knowsof the other's coincidentintentions intentionor that each is appropriatelycommittedto thejoint activity itself.On the other hand, a shared intentionis not an attitudein the of some fusionof the two mindof some superagentconsisting literally agents. There is no single mind which is the fusion of your mind and mine. Now, one way in whichyou and I may arriveat a shared intention is to make an appropriate, explicit promise to each other. But such promises do not ensure a shared intention,for one or both parties the promise. Nor are may be insincereand have no intentionto fulfill explicit promises necessary for shared intentions.Consider Hume's "tho' theyhave example of two people in a row boat who row together

5. Donagan, Choice,p. 95. 6. There is a recentliterature in artificial intelligencethatfocuseson similarissues. See, e.g., Philip R. Cohen and Hector J. Levesque, "Teamwork," Nous 25 (1991): 487-512; Barbara J. Grosz and Candace L. Sidner, "Plans for Discourse"; and Jerry Hobbs, "Artificial Intelligence and Collective Intentionality: Comments on Searle and on Grosz and Sidner." The lattertwo essaysare in Intentions in Communication, ed. Philip R. Cohen, JerryMorgan, and Martha E. Pollack (Cambridge, Mass.: MIT Press, 1990), pp. 417-44 and pp. 445-59, respectively.

Bratman

SharedIntention

never given promises to each other."7Such rowers may well have shared intentionto row the boat together. To understandshared intention,then, we should not appeal t an attitudein the mind of some superagent; nor should we assum that shared intentionsare always grounded in prior promises. M conjecture is thatwe should, instead,understandshared intention, i the basic case, as a state of affairsconsistingprimarily of appropriat attitudesof each individual participantand theirinterrelations.8 How do we determinein what thiscomplex of attitudesconsist Begin witha related query: What do shared intentionsdo, whatjob do theyhave in ourilives?I thinkwe can identify three main answer to this query. First,our shared intentionto paint togetherwill help coordinat my activitieswithyours (and yours with mine) in ways that trackth goal of our paintingthe house. Someone willscrape before,not afte the new paint is applied by someone. Second, our shared intentio willcoordinateour actionsin partby ensuringthatmyplanningabou my role in the house-paintingis coordinated withyour relevantplan ning, and vice versa. If I plan to get the paint but not the brushes will likely check whether you plan to get the brushes. Third, ou shared intentionwill tend to provide a background framework tha structures relevantbargaining.Though we share the intention to pain together we mighthave conflicting preferences about who scrapes an who paints, or about what color paint to use. Such conflicts call fo bargaining in some form-not bargaining about whether to pain togetherbut, rather,bargaining about how we are to paint togethe Our shared intention,then, performsat least three interrelat jobs: it helps coordinate our intentionalactions; it helps coordinat our planning; and it can structurerelevant bargaining. And it doe all thisin ways thattrackthe goal of our paintingthe house togethe Thus does our shared intentionhelp to organize and to unifyou intentionalagency in ways to some extent analogous to the ways in whichthe intentions of an individualorganize and unify her individu agency over time. An account of what shared intentionis should ex plain how it does all this. So what we want to know is this: Are there attitudesof each o the individual agents-attitudes that have appropriate contentsand

7. David Hume, A Treatise ofHumanNature,ed. L. A. Selby-Bigge(Oxford: Oxfor University Press), p. 490. See David Lewis's remarksabout thisexample in his Conven tion:A Philosophical Study (Cambridge,Mass.: Harvard University Press, 1969), p. 44, e.g 8. Let me explain why I say only "primarily." I claim below thatshared intention involve "common knowledge." I do not tryhere to say what common knowledge is But it may be thatit involvessome external situationin the environment of the agent that functionsas what Lewis calls a "basisfor common knowledge" (p. 56).

100

Ethics

October 1993

are interrelatedin appropriate ways-such that the complex consistingof such attitudeswould, if functioning properly,do thejobs of shared intention? Can we describe an appropriate complex from whose proper functioning would emerge the coordinated action and for bargaining,characteristi planning, and the relevant framework of shared intention?If so, we would have reason to identify shared intentionwiththis complex. II

Such an approach to shared intentionwill need to draw on an understandingof the intentionsof individuals,withspecial attentionto the roles of such intentionsin coordination. Here I briefly sketchan approach to such mattersthat I have developed elsewhere.9 Suppose I intend now to practice the tenor part tomorrowat noon. If all goes well my activity between now and then will include all necessary preliminary steps-for example, gettingthe music if I don't already have it-and it won't include activity incompatiblewith my practicingthen-for example, screamingtoo much at an athletic eventthe nightbefore.And when tomorrow noon arrivesI willbe in a a movie.This normally positionto practice;I willnot be, say,attending happens, if it does happen, because of my intention.My intentionto practicemy part tomorrowcoordinates my activity between now and then in a way that supports my practicingat noon. How does my intentionplay this coordinatingrole? In part, by to practice shaping my planning between now and later. My intention is an elementof a partialplan. As timegoes by I need to fillin thisplan frommeans-endincoherence.So appropriately;otherwiseitwillsuffer my intentionposes relatively specificproblemsof means and preliminary steps for my planning. I am faced, for example, witha problem about how to get a copy of the tenor part by noon. In contrast,my plan poses no special problem about how to get a copy of The Iliad, even if I would much like one. Further,my intentionconstrainsmy plans in ways necessary to ensure that my plans remain internally consistentand consistentwith my beliefs: for example, it precludes at noon. In these waysmyintentionhelps going to a movie tomorrow insure that my activities between now and tomorroware coordinated witheach other in ways that support my practicingthen. For all this to work my intentionwill need to have a further property.Prior intentionsare revocable. If thingschange in relevant will waysit may behoove me to change my plan. Still,priorintentions If we were constantly need to have a certainstability.'0 reconsidering
9. Primarilyin my Intention, Plans, and PracticalReason. See also my "What Is Intention?"in Cohen, Morgan, and Pollack, eds., pp. 15-31. 10. See my discussion in Intention, Plans, and PracticalReason, and in "Planning and the Stability of Intention,"Minds and Machines2 (1992): 1-16.

Bratman

SharedIntention

10

our prior plans they would be of littleuse. The nonreconsiderati be the default. of one's prior intentionswill typically Intentions,then, are normallystable elements of partial plan These plans are subject to demands for coherence and consistenc further demands which help structure planning. Such planning is no the onlymechanismthatcoordinatesan individual'spurposiveactivi coord over time. A tigerhuntingher prey may exhibitwonderfully withoutbeing capable of such planning. But for crea nated activity tures like us-as Donagan says, "creatures . . . of will"-planning i an importantcoordinatingmechanism. III

I need now to discuss two more preliminary issues. First:my strateg of attitud is to see our shared intentiontoJ as consistingprimarily At least some of these attitud of each of us and theirinterrelations. concernourjoining actionofj-ing; afterall, our share willspecifically in the pursuitof ou intentiontoJ supports coordinationspecifically J-ing.But much talk ofjoint action already builds in the veryidea o forexam shared intention.For us to tryto solve a problem together, ple, we need an appropriate shared intention.We would riskcritic if our analysis of shared intentionitselfappealed t able circularity So w thatinvolvedthe veryidea of shared intention."2 joint-act-types thatare, as I willsay willwant to limitour analysans tojoint-act-types neutral with respect to shared intention.For example, we will wan to use a notion of painting the house togetherthat does not itse I assume that w require that the agents have a shared intention.13 that ar will have available appropriate conceptions of joint activity neutral withrespect to shared intention;or anyway,my discussion i limitedto such cases. A second problem: the attitudes of the individualparticipants th of a shared intentionwill include intentionsof thos are constitutive But what I intendto do is to performactions of myown participants. I cannot intendto performthejoint actionJ. So how will the concep tion of thejoint action get into the intentionsof the individuals? Distinguish two strategies.First,we can appeal to my intentio to play my part in ourJ-ing, where this entails that ourJ-ing, whi not somethingI strictly speakingintend,is somethingI want.14 Second

11. The quote fromDonagan is fromChoice,p. 137. 12. Donagan discusses an analogous problem for individualintentionalaction i Choice,pp. 87-88. 13. Think of a case in which we paint it during the same time period but we ar each ignorant of the other's activity. 14. An appeal to myintention to play mypartin ourJ-ingis similarto the approac of Raimo Tuomela and Kaarlo Miller to what theycall "we-intention" (see "We-Inte tions,"Philosophical Studies53 [1988]: 367-89, esp. pp. 375-76).

102

Ethics

October 1993

to, we can tryto exploit the factthat we speak not only of intentions but also of intentions that-for example, myintention thatScottclean up his room. Accordingly, we can speak of my intentionthat weJ. 15 The idea here is not to introduce Consider the second strategy. some fundamentally new and distinctive attitude.The attitudewe are appealing to is intention-an attitudealready needed in an account of individual intelligentagency. But we are allowing this attitudeto include in its content thejoint activity-our J-ing.6 Such appeals to myintentionthatweJ willseem reasonablynaturalgivenan emphasis on the roles of intentionsin plans. This is because my conception of our J-ingcan functionin my plans in ways similarto my conception of my own A-ing: in each case I face problemsof means and preliminary steps; and in each case I need to constrainthe rest of my plans in the light of demands for consistency.And susceptibility to these demands for coherence and consistencyis a characteristicsign of intention. It mightbe objected that talk of an intentionthat we J conflicts withthe plausible idea thatone must see what one intendsas to some extentwithinone's influenceor control.That is why I can intend to raise my arm but not that the sun shine tomorrow.But, in fact,this need be no objection to the second strategy;for that strategycan of my build an appropriate influenceconditioninto itsunderstanding intendingthat weJ. It can say, roughly,thatfor me to intendthatwe J I need to see your playing your role in our J-ing as in some way affectedby me. So the second strategy coheres with the planning conception of intention and can acknowledge a plausible influence condition. In what follows I will pursue this second strategy:my account of our shared intentionto J will appeal to your and my intentionthat we J. I will not try to settle the question of exactly what version of the influence condition we should accept, for none of my main points strategy depends on this issue. Nor will I tryto argue that the first is fruitful. must fail. My claim here is only that the second strategy IV

I want to say what it is for us to intendsomethingprimarily in terms of (a) intentionsand other attitudesof each and (b) the relations of

15. A strategy similarto one once urged on me by Philip Cohen. In "Objects of in press) Bruce Vermazen defendsappeals to intentions Intention,"(Philosophical Studies, that are not intentionsto act. 16. This contrasts withJohn Searle's conceptionof "we-intending" in his "Collective Intentions and Actions," in Cohen, Morgan, and Pollack, eds., pp. 401-15. A weintention,for Searle, is a distinctive attitudeof an individual-an irreducibleaddition to the kindsof attitudes of whichwe are capable. On the tack I am taking,myintention thatweJ and my intentionto play my part in ourJ-ing are both intentions-they are both instancesof the same attitude;but theyare intentions in theircontents. thatdiffer

Bratman

SharedIntention

103

these attitudesto each other.'7 This account should explain how it is that shared intentions support the goal-directed coordination of in part by way of coordinated planning and relevant shared activity, bargaining. Limiting myselfto joint-act-types that are neutral with I proceed by consideringa seriesof views. respectto shared intention,
VIEW 1: We intendtoJ if and only if I intendthatwe J and you intend that we J.

View 1 does ensure thatthe participantsin a shared intentionto But View 1 is nevertheJ each are, in a way,committedto theirJ-ing. less too weak. Afterall, each of us can intendthatwe J withouteven knowing of the other's intentionthat we J.'8 Yet at least that much cognitivelinkage is involvedin shared intention.Indeed, it seems reasonable to suppose that in shared intentionthe factthateach has the relevantattitudesis itselfout in the open, is public. This suggeststhat we turn to:
VIEW

2: We intend toJ if and only if 1. I intend that we J and you intendthat we J, and 3.19 1 is common knowledge20 between us.21

Now consider an example: you and I each intend that we go to New York together;and thisis commonknowledge.However, I intend that we go togetheras a result of my kidnapping you, throwingyou in mycar, and forcing you tojoin me. The expressionof myintention,
17. Note that my targetis our shared intention.My directtargetis not what Tuomela calls a "we-intention";for a we-intentionis an intentionof an individual that concernsa group's activity (see Raimo Tuomela, "We Will Do It: An Analysisof GroupIntentions,"Philosophy and Phenomenological Research51 [1991]: 249-77). Nor is my targetwhat John Searle calls a "collectiveintention"in his "Collective Intentionsand Actions."A collectiveintention, as Searle understands it,is an intention of an individual concerning a collective'sactivity. Indeed, both Tuomela and Searle want to allow that therecan be a we-intention/collective intentioneven if thereis in factonly one individual-one who falselybelieves others are involved (see Searle, "Collective Intentions and Actions,"pp. 406-7; and Tuomela, "We Will Do It," p. 254). In contrast, it takes at least two not only to tango but even for there to be a shared intentionto tango. 18. This is true even if, to intend that we J, I must believe that your relevant activity depends on mine. 19. This numberingwill help keep mattersclearer as we proceed. on the idea of common knowledge.See, e.g., Lewis. 20. There is a large literature I use here an unanalyzed notion of common knowledge. 21. View 2 is in the spiritof Raimo Tuomela's analysisof "intentional joint goal" (see his "What Are Goals and JointGoals?" Theory and Decision28 [1990]: 1-20, esp. p. 10). View 2 is also close to what Margaret Gilbert calls a "strongshared personal goal analysis" of the psychologicalbackground of what she calls "acting together"(see "Walking Together: A Paradigmatic Social Phenomenon," Midwest Studies15 [1990]: 1-14, esp. p. 3). Gilbertrejects such an analysis: she argues thatit does not guarantee appropriate obligations and entitlements. My reasons for rejectingView 2 are quite I turn to Gilbert'sconcerns later. different.

104

Ethics

October 1993

we mightsay,is the Mafiasense of "we'regoing to New York together In intendingto coerce you in this way I intendto bypass your inte tional agency. And that seems to rule out a shared intentionto go t New York: myintentionwill surelynot support coordinatedplannin about how we are going to get to New York. Granted,if I succeed i will in a way be unified: we will indeed g what I intend,our activity togetherto New York. But since the way our activity is tied togeth thisis not the kindof unifiedagenc bypassesyourrelevantintentions, of shared intention.22 characteristic This suggests that in shared intentionI not only intend that w J; I also intend that we J in part because of your relevantintentio I intendthatour performanceof thejoint activity be in partexplaine I intendthatyo by your intentionthatwe performthejoint activity; participateas an intentional agent in ajoint activity that,as I know,yo too intend.However, once we bringintothe contentof an intention o mine the efficacy of your intention,it is a short step to including a well the efficacy of myown intention.In a case of shared intention see each of the participants, intention includingme, as participating, of your intentioni agents. If this obliges me to include the efficacy the contentof myrelevantintention, thenitseems plausibleto suppos that it also obliges me to include the efficacy of my own intentio Afterall, I see each of us as participantsin the shared intentionan the shared activity. Why would what I intend include a requireme that your intention that we J be effective, and yet not include a analogous requirementconcerningmy own intentionthatweJ? These considerations, taken together,argue for:
VIEW

3: We intend toJ if and only if 1. (a) I intend that we J and (b) you intendthatwe J 2. I intend that we J because of la and lb; you intendth we J because of la and lb 3. 1 and 2 are common knowledge between us.

In shared intentionthe constitutive intentionsof the individualsar for each agent has an intentionin favorof the effica interlocking, of an intentionof the other.And the intentions of each involvea kin for each has an intentionconcerningthe efficacy of reflexivity, of a intentionof her own. of indivi Now, Donagan has argued thatthe choice characteristic ual intentionalaction is a choice that one act in a way explained b that verychoice: "The choices that explain actions are explanatori

22. This example, and the one to follow afterView 3, are also discussed in m "Shared Cooperative Activity," Philosophical Review 101 (1992): 327-41. See esp. pp 332-33 (where I have more to say in defense of conditionsto be added below in View 3 and 4) and pp. 334-35 (where I have more to say about the kindof coercion involv in the Mafia example).

Bratman

SharedIntention

105

self-referential."23 The idea that shared intentioninvolves reflexive intentionsof the individualsis in a way similarin spiritto this claim of Donagan's. Nevertheless,my claim about shared intentionis compatible withthe rejectionof the need forself-referentiality in the case of individual intentionalaction. In a case of shared intentioneach agent sees herself as one of a pair of participants.Given that she intendsthatthe relevantintentionof the otherbe effective, and given that she recognizes that she and the other each have an intentionin favorof thejoint activity, there is pressure on her also to intendthat her intention be effective.But this pressure arises from the social contextof the shared intentionand need not be present in the case of individual,nonshared intentionalactivity. So there is room for the conjecture that it is only when we get to shared intentionthat each agent is obliged to include in what she intendsa referenceto the role of her-own intentions. To returnto the main thread,note that View 3 does not require that you and I either have or aim at having a shared conception of how we are to J. Suppose you and I each intend that we paint the house togetherin part because of each of our intentions.However, I intend that we paint it red all over, and you intend that we paint it blue all over. All thisis commonknowledge;and neitherofus is willing to compromise.24 On View 3 we have a shared intentionto paint the house. But this seems wrong, for neither of us is committedto the interpersonalcoordination of our relevantsubplans. Granted, for me to intend that we paint the house, despite my knowledgeof our differences, I need to thinkthereis some real possithatwe willnevertheless bility paintit.But perhaps I thinkthisbecause I thinkI can trickyou about the color of the paint in your can. We mightthen satisfy1-3 of View 3; and yet we would stillnot have a shared intention.For our intentionto be shared neither of us can intendthatthe other'srelevantsubplans be subverted.A shared intention should function to unifyour intentional agency at least to this extent; otherwise it would not support appropriately coordinated planning. So we need to go beyond View 3. But we also need to be careful not to go too far. First,it would be too strong to require that the subplans of our intentionsin la and lb completelymatch, for there

23. Donagan, Choice, p. 88. Others who have defendedsimilarviewsabout the selfreferential causes of intentionalaction include GilbertHarman, "PracticalReasoning," ReviewofMetaphysics 29 (1976): 431-63, and Changein View(Cambridge, Mass.: MIT Press, 1986); JohnSearle, Intentionality (Cambridge: Cambridge University Press, 1983); and David Velleman, PracticalReflection (Princeton,N.J.: Princeton University Press, 1989). For a trenchantcritique of such views, see Alfred Mele, "Are IntentionsSelfReferential?" Philosophical Studies52 (1987): 309-29. 24. Rachel Cohon helped me get this example into shape.

106

Ethics

1993 October

can be featuresof your subplan thatI do not even knowor care about, and vice versa. Perhaps your subplan includes paintingin overalls or buying the brushes at a certain store. While I need to know you will show up withthe brushes,I may well neitherknow nor care how you are dressed or where you get the brushes. So our subplans may well not completelymatch. Still,it seems that we will each want them in the end to mesh:our individual subplans concerning our J-ing mesh just in case there is some way we couldJ thatwould not violate either of our subplans but would, rather,involvethe successfulexecution of those subplans. If I intendthatwe paint solelywithred paint and you intendthat we paint solelywithblue, our subplans do not mesh. But ifyou intendto get the paint at Greg's Hardware, and I simplydo not know or care about where you get the paint,then our subplans,while theydo not completelymatch,may stillmesh. And it is meshing subplans that are our concern in shared intention. There is a second way in which we must be carefulnot to go too far. For you and I to have a shared intentiontoJ we need not already have arrived at subplans that mesh. Much of our relevant planning may occur afterwe have arrived at our shared intention.All that is plausiblyrequired is thatwe each intendthatweJ by way of meshing subplans. This leads us to: 4: We intend toJ if and only if 1. (a) I intend that we J and (b) you intendthat we J 2. I intend that we J in accordance withand because of la, lb, and meshing subplans of la and Ib; you intendthat we J in accordance withand because of la, lb, and meshingsubplans of la and lb. 3. 1 and 2 are common knowledge between us.
VIEW

On View 4, then, I need neitherknow nor seek to know of all your subplans for us to have a shared intention;nor need we already have arrivedat complete,meshingsubplans. What is required is thatI intend that we J by way of meshing subplans. I can so intend even though there are as yet no specific,meshing subplans such that I intendthat weJ by way of them. You and I may not yethave filledin each of our subplans, or we may have filled them in in ways which do not yet preferencesconcerningsubplans and mesh. We may have conflicting be involved in negotiationsabout how to fillin our plans even while we have already startedtoJ. on thislast point. Our shared intentioncan It is worthreflecting serve as a relativelyfixed background against which relevant bargaining can take place. Suppose you and I jointlyintendto paint the but we have yetto agree on the colorsor on the division house together preferenceswe may engage in various of roles. Given our conflicting in such bargaining may, of course, formsof bargaining. Difficulties

Bratman

SharedIntention

107

lead either of us to reconsiderthe intentionthat we paint together. But so long as we continue so to intend,our bargainingwill concern but how. Our bargainingwillbe framed not whetherto paint together by our shared intention. Recognitionof such potentialbargaining raises a question. Supconditions 1-3 of View 4 withrespect to pose that you and I satisfy our going to New York togetherbut that there are large differences betweenus in relevantbargainingpower. Perhaps itis a veryimportant matterforyou but only a welcome break fromworkforme. Suppose I plan to use this differenceto bargain hard for meshing subplans thatare verymuch to myliking.Perhaps I plan to put a lot of pressure Accordingto View 4 we could stillhave on you to pay forboth tickets. a shared intention.Is that an acceptable result? I believe that it is; though, of course, too much stubbornness might result in the dissolution of our shared intention.Granted, at in bargainingpower some point the exploitationof large differences of going to New York becomes coercive. When it does our activity cooperative (ifthatis whatwe manage to do) willnot be a fully together But it may stillbe one that is jointly intentional;and we may activity. There stillmay be appropriate stillhave a shared intentionso to act.25 kindsof coordinationin our planning and action. even when A virtueof View 4 is thatit allows forshared intention We can intendto reasons for participating. the agents have different sing the duet togethereven though myreason is the love of the music and yours is, instead, the chance to impress the audience. View 4 does have a drawback: it does not yetprovidefora shared intentionto play a competitivegame together.You and I mighthave a shared intentionto play chess togetherand yetneitherof us intend that our subplans mesh all the way down. Afterall, I intendto tryto scuttleyour plans for checkmatingme. I thinksuch cases will force in View 4; but I will not tryto get this straight modest modifications whether,cases of competitive here. Instead, I want to explore further games to one side, View 4 provides for appropriate explanations of the coordinated planning and action,and associated bargaining,charof shared intention. acteristic V as I on three basic points. First,shared intention, Begin by reflecting understandit, is not an attitudein any mind. It is not an attitudein the mind of some fused agent, forthere is no such mind; and it is not Rather, an attitudein the mindor mindsof eitheror both participants. in attitudes (none of which thatconsistsprimarily it is a stateof affairs
thatare not insured 25. For suggestionsof otherconditionson cooperativeactivity see my "Shared Cooperative Activity." by the successfulexecution of a shared intention,

108

Ethics

October 1993

are themselvesthe shared intention)of the participants and interrelations between those attitudes. Second, to say in what shared intentionconsistsI have sought to combine two main elements: (1) a general treatment of the intentions of individualsand (2) an account of the special contentsof the intentions of the individual participantsin a shared intention.Intentions of individualsare normallystable elements in larger,partial plans of those individuals.These plans are subject to demands for means-end coherence and consistency. Because of these demands,intentions tend to pose problemsforfurther practicalreasoningand to constrainsolutions to those problems. Given these features of the intentionsof individuals,and given the special contentsidentified in View 4, I want to explain how that in which a shared intention consists supports coordinated planning and action,and appropriate bargaining,in pursuit of thejoint activity. Shared intentionconsists primarilyof a web of attitudesof the individual participants.These attitudesof the individualsare subject to various rationalpressures.In particular, the intentions of the participants are subject to demands for consistencyand coherence. The specificimpact of these demands will depend, of course, on the contentsof these intentions.And in shared intentionthe relevantintentions of the individualparticipantshave the special contentswe have been discussing. So-and this is the third point-what we want to show is thatintentions of individualswiththese special contentsshould lead to planning, bargaining,and action of those individualswhich, taken together,constituteappropriately coordinated planning and unifiedshared activity. The unifiedaction and coordinated planning characteristic of shared intention is to be explained primarily byappeal to the functioningof the attitudes which are constituentsof the shared intention. Let us see how steps in the directionof View 4 contribute to such an explanation. Begin with View 2. Condition 1 of View 2 requires that each intendsthat weJ. So the demand for means-end coherence of the plans of each insures rational pressure on each participantto pursue means to thejoint J-ing.It also follows,given the demand for consistencyof each agent's plans, that there is rational pressure on each to eschew courses of action believed by her to be incompatible withthejoint J-ing. So far so good. But what we learn fromthe Mafia case is thatthis does not insure that there is rational pressure on each participantto aim at coordinationwiththe other'ssuccessfulexecution of herintention. Yet the pursuitof coordinationwiththe other'ssuccessfulexecution of her relevant intentionis essential to the kind of coordinated of shared intention. planning characteristic

Bratman

SharedIntention

109

This bringsus to View 3. The conditionsof View 3 insurerational pressure on each participantto seek means not only to thejointJ-ing but also to the joint J-ing by way of the other's intention. Now, I frequently formmy intentionsin the lightof my expectationsabout your intentionsand actions, including expectations about how my intentionswill influenceyours. Since my expectationsabout how my intentions will influenceyoursmay depend on my expectationsabout how you expect my intentionsto be influenced by yours, this can get quite complex. But in this,as Schelling says, "spiral of reciprocal expectations,"we stilleach see the other's intentionsmerelyas data affected by our forour deliberations, albeit as data thatare potentially In contrast, the conditionscited in own decisions.26 agents who satisfy View 3 do not see each other's relevantintentionmerelyas a datum, foreach intendsthatthejoint activity go in part by way of the efficacy of the other's intention. Each is rationallycommitted to pursuing by means, and eschewingobstacles,to the complex goal of theirJ-ing way of the other agent's relevantintention.Each aims at the efficacy of the intentionof the other. in thisway, interlock In requiringthatthe participants' intentions View 3 gives up on the idea, implicitin View 2, thatthe cruciallinkage between the attitudesof those who share an intentionis merelycognitive.27 Appropriate common knowledge,or the like,is not a sufficien link for shared intention.Each agent needs also to embrace as her of the other's relevantintention. own end the efficacy However, the conditionsof View 3 stilldo not insure that each agent aims at there being meshing subplans. The conditionsof View 3 do insure that each agent seeks a consistentindividualplan in support of a joint J-ing in which each agent's intentionthat theyJ is efficacious. But these conditionsdo not insure thateach agent intends that the subplans of both, taken together,be jointly consistent:that is the lesson of the painting case. But shared intentionshould bring withit rational pressure in the directionof subplans of both particiBy requiringthatthe jointlyconsistent. pants thatare, taken together, participantsintend that theyJ by way of meshing subplans, View 4 insures such rational pressure. Finally,View 4 makes it clear why shared intentionswill sometimes frame relevant bargaining. On View 4 each agent aims at a performanceof thejoint J-ing that goes by way of each participant's relevantintention and itsmeshingsubplans. So even ifthe participants
ofConflict (Cambridge, Mass.: Harvard Univer26. Thomas Schelling,The Strategy sityPress, 1960), p. 87. 27. Gilbertin "WalkingTogether,"and Searle in "CollectiveIntentions"also reject reasons. related ideas, though for different

110

Ethics

October 1993

have differing preferencesabout how theyare toJ,neitherparticipant will be in a position to pursue such preferencesin ways that bypass the other's intentions/subplans. This makes it likelythatin such cases the demand on each agent thather plans be means-end coherentwill lead to rational pressure in the directionof bargainingthatis framed by the shared intention. Suppose, then,thatthe intentions of individualparticipants have the contentsand interrelations cited in View 4; and suppose thatthese intentions-like intentions generally-are subjectto demandsforconsistency and means-end coherence. These rational pressureson these intentionsof those individualswill issue in pressure in the direction of coordinated planning and action, and appropriate bargaining,directed at thejoint action ofJ-ing.And that is what I wanted to show. VI Margaret Gilbert has argued that in an importantsense of "acting together"each participanthas associated nonconditionalobligations to act and nonconditionalentitlements to rebuke the otherforfailures to act.28On View 4, if you and I have a shared intentionto J then you ought to performyour role if you continue to intend that we J. But View 4 by itselfseems to offerno guarantee thatby virtueof our having a shared intentionyou have a nonconditional obligation to perform.Does this suggest that somethingis missingin View 4? Recall that intentionsare subject to a demand for stability. One reason for this is that the reconsiderationof an intention already formedcan itselfhave significant costs; a second is thatan agent who too easilyreconsidersher priorintentions willbe a less reliablepartner in social coordination. This latter,social pressure toward stability is particularlyrelevant to the stabilityof intentionsconstitutiveof a shared intention.So our approach to shared intentioncan account forrationalpressureon a participating agent not too easilyto abandon her relevantintentions. Note furtherthat if each agent's relevant intentionsare fairly stable it will normallybe reasonable for each to relyon the other to stickwiththejoint project. The stability of the constituent intentions therebysupports each in planning on the contributions of the other, just as we would want in coordinated planning. When I too easily abandon my intention that we take a walk togetherI am, then, being unreasonable. But it does not followthat in abandoning myintentionI am violatinga nonconditionalobligation to you, a nonconditionalobligationgroundedin our shared intention. To be sure, shared intentionsare frequentlyaccompanied by such
28. For example, pp. 5-6 of "WalkingTogether." This summarizesaspects of her much longer discussion in On Social Facts (London: Routledge, 1989).

Bratman

SharedIntention

111

obligations. In arriving at a shared intention we frequentlymake promisesor reach agreementswhichgeneratecorresponding nonconditionalobligations.Further,once we begin executinga shared intention implicitpromises frequently arise-promises thatgenerate nonconditional obligations. Still, such a promise or agreement does not seem to be, strictly speaking, necessaryfora shared intention. Imagine two singerswho each highlyvalue theirduet-singing but neverthelesshave a clear understanding between them thatneitheris makingany bindingpromise to or agreementwiththe otherconcerning their singing. Each publiclystates that she reserves the rightto change her mind. These two could stillshare an intentionto sing a duet together.29 They could stillengage in coordinatedplanningaimed at theirsingingthe duet and in which each relies on the participation of the other. Granted,the normal case of shared intentionwillnot be like this. In a normal case there will likelybe some promise or agreement; and that will further contributeto the confidenceof each that she can plan on the participationof the other. Nevertheless,such a promise or agreement does not seem essential to shared intention. And when there is no such promise or agreement, or some other obligation-generating process, the shared intentionmay not impose a nonconditionalobligation to stickwiththejoint action. Consider two different responses to this. First,one mighttryto insist that the mere satisfaction of the conditionsof View 4, in the absence of some furtherobligation-generating agreement,does not ensure shared intention.30So our singers do not in fact have a shared intention. At this point perhaps the dispute is merelyverbal and we should simplyspeak of shared intentionin a weaker and in a strongersense. The weaker sense is captured, pretty much, by View 4. The stronger sense involves yet a further condition,that there be a bindingagreement.3"I have argued that shared intentionin the supposed weaker sense supports coordinated planning and action, and relevant bar29. Lewis makes a similarpoint (p. 34). 30. This is roughlyin the spiritof some of Gilbert'sremarksas commentatoron an earlier and shorterversion of this article at the Central Division of the American Philosophical Association (APA), Louisville, Ky. (April 1992). (Gilbertput the point in termsof a special notion of 'joint commitment," indicatingthat "it may be reasonable enough to think of [joint commitment]as an 'implicitagreement'.") This was also Raimo Tuomela's tack in his replies as commentatoron an earlier shorterversion of this article (presented at the meetings of the Society for Philosophy and Psychology, Montreal,June 1992). 31. In her comments at the Central Division of the APA, Gilbert suggested (as Paul Weirichbroughtout in the discussion period) thatsuch a bindingagreement,and the resultingobligationsand entitlements, would itselfbe sufficient fora shared intention. But thatseems to me wrong,since bindingagreementsdo not guarantee intentions on the part of the individual agents to act accordingly.That is why I understandthe

112

Ethics

October 1993

gaining, aimed at the joint activityand that it is typicallybut not necessarilyaccompanied by relevantnonconditionalobligations.That seems to me a reason to see the phenomenon captured by View 4 as at the heart of the matter.At the least we have seen that there is an importantkind of shared intentionthat does not essentiallyinvolve such obligations. Such shared intention is primarilya psychological-rather than primarily a normative-phenomenon. The step to nonconditionalobligationsand entitlements is a step beyondthismore basic phenomenon. Consider a second response to my defense of View 4. One might urge that a shared intentionin the sense of View 4 could only come about by way of a process of a sort that generates corresponding nonconditionalobligations.Perhaps the process is not, strictly speaking, one of agreement or the exchange of promises; it mayjust be a more general kindof mutual assurance. But thisprocesswillnevertheless be sufficient to support correspondingobligations. My reply to this is twofold.First,the main claim-that shared intentionmustalways come about by way of an obligation-generating process-does not seem to me veryplausible: the case of the cautious singerswho disavow obligationseems a fairly clear counterexample.32 But, second, even if I were wrong about this, this need not be an objection to View 4. We could stillallow thatView 4 says what shared intention is, while notingthatthe creationof a shared intention brings withit certainnormativeconsequences. We could stillagree withView 4 thatshared intention consistsprimarily of a web of individualpsychological states and their interrelations. It would just turn out that the creation of this psychologicalweb has normativeconsequences. VII

This approach to shared intentionis broadlyindividualistic in spirit.33 Granted, much recent work in the philosophy of mind has argued thatour ordinarywaysof specifying the contentsof the attitudes draw on featuresoutside of the individualwhose attitudesare in question.
strongersense of shared intention,if such there be, to include the conditionscited in View 4 as well as a furthercondition specifyingan appropriate normative relation between the participants. 32. I believe that certain cases of coerced shared intentionwould also provide counterexamples to this overly general claim. Other potential counterexamplesmay come fromcases of shared intentionin which the common knowledge is grounded in the background knowledge of the participantsand is not the resultof assurances each gives the other (a point David Velleman helped me see-though he did this while tryingto convince me that such cases posed problems for View 4). I discuss these mattersfurtherin my "Shared Intentionand Mutual Obligation." 33. Assuming that the common knowledge condition can be understood along individualistic lines.

Bratman

SharedIntention

113

Such external featuresmay include the causal context of the use of names or natural kind terms,34 as well as relevantlinguisticpractices of the community in which the individualis located.35 The individualism of myapproach to shared intention can grantthese insights about what determinesthe contentof an individual'sattitudes. The claim is not that we can specifythese contentsin ways that do not appeal to elements outside the individual whose attitudesare in question. The claim, rather,is that shared intentionconsists primarily of attitudes of individualsand theirinterrelations. The coordinated planning and action, and framework for bargaining,characteristic of shared intention emerge from the proper functioningof these attitudesof the individual participants.

34. Hilary Putnam, "The Meaning of 'Meaning'," in his Mind,Languageand Reality (Cambridge: Cambridge University Press, 1975), vol. 2. 35. Tyler Burge, "Individualism and the Mental," Midwest Studiesin Philosophy 4 (1979): 73-121.