This maybe a little long for a blog post, but it is a version of the paper I will be presenting at the collective intentionality conference and I would welcome any comments on it before I present it on August 26th
Some believe that there is a certain kind of direct normative constraint which occurs in collective action; Roth aptly names it ‘Practical intersubjectivity’. In this paper I will firstly set out the contradiction between the existence of practical intersubjectivty and the notion of individual agents as essentially autonomous. I will suggest that we can understand the force of collective intentions over individuals by virtue of the part their having such a force plays in allowing the collective to be a coherent agent. In doing so I will be attempting to apply Bratman’s argument, about the force of intentions for individuals, to collectives.
The Puzzle of Intersubjectivity:
Imagine the ‘Leicester Tigers’ (a Rugby club) are travelling to a match and their bus breaks down. Wishing to get it started again they decide to push it up a hill. Commonplace social events such as this are not simply constituted by agents deciding to collaborate in their actions, rather we can say, more strongly, that they also involve agents setting out to jointly perform a collective action.
The Leicester Tigers performing their collective action seems to give the individuals who are its members special reasons to participate in this action. That is, it seems to give those individuals reasons to act ways that further the goal of the action; in this case to put their backs into pushing the bus. These reasons seem to continue to hold, and compel each individual to actively participate as the collective action takes place.
In general then; membership of a social group with a collective intention to X, or that is doing X intentionally, appears to provide individuals with reasons to play their part in the group’s performance of X. If these reasons were purely instrumental then they would be interesting but not overly philosophically problematic. However, they appear to arise and constrain in a way that suggests that something more intriguing is going on. As well detailed in Gilbert’s many writings on the topic1, the intentional attitudes of collectives seem to have direct normative power over their individual members.
As noted, Roth refers to this phenomena as ‘practical intersubjectivity’2. It is ‘practical’ because it concerns practical rationality, that is, it is about what we ought to do in the face of standing intentional attitudes. It is ‘intersubjective’ because the standing intentional attitudes in question are not each subjects own intentions, but rather are intentions that are held collectively, that is between the subjects.
So what is so puzzling about intersubjectivity for individual autonomy? Well, first we might ask, what is unproblematic about the normative constraint that occurs in the process of individual action. If I intend to perform the action X, or if I am intentionally X-ing, then this directly gives me reason to act in such a way that I will facilitate this action.3 My standing intentional attitudes thus govern what I can appropriately do. Roughly we might say that, there is no challenge to my autonomy in this because it is my own attitudes that are governing me. When we move to the normative constraint that occurs in the process of collectively action (i.e. intersubjective normative constraint) it seems as if participating individuals are being given reasons by standing intentional attitudes that are not their own, but rather they are intentions that belong to the group. The challenge to each individuals autonomy, then, is that it appears to involve individuals being governed by attitudes that are not their own.
Individual identity, intention and irrational action
Imagine the following: Kim’s car breaks down at the bottom of the hill. Being a strong woman, Kim forms the intention to push the car to the top of the hill and starts to do so. Now suppose that, without changing her mind as to her goal, Kim abruptly stops pushing her car and instead starts lying down in the road. In such a situation we would clearly say that she is doing something wrong.
Why can’t Kim turn around and say that she does not value acting in line with her own intentions? What would be wrong with such a person not acting in line with their intentions? Why she be normatively condemnable? Let’s call one who asks these questions a sceptic.
Bratman’s answer to the sceptic, that is what he sees as the wrongness of acting against ones own intentional attitudes, is that they have failed to take note of the connection between the necessity of the normative force of intentions and what it take to be an autonomous agent. According to Bratman, unless Kim sees her own intentions as having force over her, and acts accordingly, she will not be an autonomous agent.
At first this might seem odd. How can something that constrains someone add to their autonomous agency? Well, imagine someone for whom all future options were open. We might think that this is the most autonomous agent imaginable. However, they can only be an autonomous agent if they are indeed an agent. Bratman argues that being an agent conceptually requires that one’s practical options are linked to one’s prior mental attitudes. This is because if we do not have such a link then we do not have a coherent whole but merely disparate parts.
Here we must have a stronger sense of ‘link’ than mere causal connection, because ones mental attitudes could be causally linked, i.e. they could follow in a causal chain, without this necessarily meaning that, in terms of their content, they were linked in such a way that we could conceive of them as creating a coherent agent across time.4 For Bratman then, we have reason to abide by our personal intentional attitudes, and thus these attitudes rationally constrain us, because we have, “a distinctive, noninstrumental practical reason to avoid a correspondingly incoherent psychological complex.”5
The argument can be presented as follows:
- Being an agent requires having a coherent psychological complex;
- Having a coherent psychological complex requires that ones mental attitudes are linked in a structured way;
- Intentional attitudes provide this structure when the rules that attend them are seen as normatively compelling;
- Being an agent requires seeing the rules attending intentional attitudes as being normatively compelling.
- For any particular piece of practical reasoning there is a normative force to following these rules because failure to do so will result in failure to understand oneself as a potential agent of ones actions in that particular case.
The force of Bratman’s argument can be seen by imagining the phenomenology of realising that one is transversing the rules of practical reason, and thus acting against the constraint of ones intentional attitudes. As Bratman says; “When I recognise inconsistency in my own intentions, I see that in this specific case there is no clear answer to the question; “Where do I stand?””6
Collective identity, intention and irrational action
Now, let us move from the individual to the collective.7 Might it be that Bratman’s planning theory of agency can incorporate the notion of us seeing ourselves as socially coherent as well as individually coherent agents? If so then we could say the following; collective intentions gain their normative force from the role they play in structuring the identity of social groups across time.
Taking the form of the argument above and applying it to collectives we get the following:
- Being a collective agent requires having a coherent psychological complex;
- Having a coherent psychological complex requires that the collectives mental attitudes8 are linked in a structured way;
- Intentional attitudes provide this structure when the rules that attend them are seen as normatively compelling;
- Being a collective agent requires seeing the rules attending intentional attitudes as being normatively compelling.
- For any particular piece of practical reasoning there is a normative force to following these rules because failure to do so will result in failure to understand the collective as a potential agent of a collective action in that particular case.
If this argument is sound, as it seems to me to be, then it establishes that, just as individual intentional attitudes have force over the individual, collective intentional attitudes have force over the collective. However we still have a gap between the collective and its members as it does not establish why the individual group member is subject to the normative constraint of the collective intentions.
What makes not knowing where one stands as an individual agent so problematic is that one thereby undermines one’s ability to understand oneself as, and in fact to be, a coherent united agent. As Bratman puts it, not knowing where one stands means that one fails to see oneself as an agent rather than merely an aggregate of conflicting needs, desires and considerations. Such a point is fine for individual agency however its transferability to collective agency is unclear. This is because failure to understand where we stand does is not transparently devastating to ones individual agency in this way.
If one is able to retreat from the perspective of the ‘we’ back to the perspective of the ‘I’ then what is to stop the individual being a sceptic and simply accepting that the collective is a conflicting stew of collective needs, desires and considerations rather than a unified agent?
A possible answer to the sceptic is that they are not acknowledging the importance of being social agents to our identities as individual agents. If, in a strong sense, we see ourselves as socially located agents then in order to understand ourselves as being coherently motivated by collective intentions we must abide by the norms that attend intentions. For brevity I’ll call this individual-identity-incorporates-collective-identity solution the ‘I/C identity solution’ in the brief exploration of it that follows.
So, it is clear that if one recognises an inconsistency between one’s own intentions and those of the a collective – that one see one’s self as part of – then this means that (in this specific case) there is no clear answer to the question “where do we stand”. However, we need a further move to say why this is problematic. The I/C identity solution to this is that it is problematic because, as a basic fact of the kind of agents we are, we do see ourselves as able to do things as collectives.
Take the Rugby players in the example I gave before. Even if we imagine that each of them is reluctant to play their part in pushing the bus up the hill, it seems a fair assumption that they do think that such an action is at least conceptually possible. They think of themselves as part of a collective that has the could potentially act.
The fact that the rugby players see themselves as agents who can potentially not only act individually but also collectively, will be reflected in many of their individual goals. For example, If they did not see themselves as part of a collective that could act then it would be hard to make sense of the fact that many of them might have as their principal goal for the year together winning the Heineken Cup. Such a goal is a collective one and seems to require that the individual players can conceptualise themselves as able to act as a coherent collective agent.
So running with the I/C identity solution we might add something like the following to the 1-5 above:
- Failure to understand the collective as a potential agent of its actions means that one cannot understand oneself as potentially part of a collective that can act;
- Part of ones conception of ones self as an (individual) agent is that one has the potential to be part of such a collective (i.e. to be a collective agent);
- For any particular piece of practical reasoning (regarding collective actions) the normative force of the collective intentional attitudes for the collective caries over to the individual because this is required for the individual to see themselves as potentially part of a collective that can act.
Further fleshing out of an argument such as this could, it seems to me provide a useful answer to those who baulk at the idea that individuals can be directly normatively constrained by the process of being involved in collective actions. In saying that it is our connection to our social identities that gives collective intentional attitudes normative force over participating individuals, we can further understand how an agent can be capable of both individual and collective action.
1Gilbert, (1992), On Social facts, Princeton University Press. (2000), Sociality and Responsibility: New Essays in Plural Subject Theory, Rowman & Littlefield. (2003), “The Structure of the Social Atom: Joint Commitment as the Foundation of Human Behaviour”, in Socializing Metaphysics: The Nature of Social Reality, Ed. Schmitt, F., Rowman & Littlefield. (1996), Living Together: Rationality, Sociality and Obligation, Rowman & Littlefield
2Roth, A. S. (2003), “Practical Intersubjectivity”, in Socializing Metaphysics: The Nature of Social Reality, Ed. Schmitt, F., Rowman & Littlefield
3Though we probably want to say that it provides a ‘framework’ reason rather than a normal reason to avoid charges of unwarranted “bootstrapping” see Bratman, M. (1999) Faces of Intention: Selected Essays on Intention and Agency, Cambridge University Press, p. 24 – 27
4Bratman acknowledges that we might conceivably argue that other mental attitudes, such as memories, could serve as bridges creating a coherent agent across time. However, memories do not have the power to structure our future choices. I can remember that I wanted to go to the beach yesterday. This does not tell me what I must do today. Intentional attitudes on the other hand carry with them practical authority. If I continue to hold an intention to go to the beach then this not only tells me what I must do now (e.g. find my towel) but also structures the choices I can make into the future (e.g. getting out of the house on time to catch the bus) . Intentional attitudes are special for Bratman because they are “authoritative policies” that are “embedded in structures of planning agency.” Op. Cit., Bratman, p.430
5Op. cit., Bratman, p.421
6Op. cit., Bratman p.431
7Note that Bratman would not want to make the following argument, seeing himself as part of the individualist camp on the subject of collective action. As he wants to resist the idea that collective action directly puts normative constraints on the individual in the sense I gave at the start of this paper he can do so. If, however, we accept the force of the arguments for collective normative constraint then attempting to extend Bratman’s theory seems legitimate.
8One may of course doubt that a collective can have ‘mental’ attitudes as it seems to lack a mind. While this may be the case it does seem that collectives can have attitudes that are at least like mental attitudes. We can have collective beliefs and even perhaps collective emotions.