# Keyboard Shortcuts?

×
• Next step
• Previous step
• Skip this slide
• Previous slide
• mShow slide thumbnails
• nShow notes
• hShow handout latex source
• NShow talk notes latex source

Click here and press the right key for the next slide (or swipe left)

also ...

Press the left key to go backwards (or swipe right)

Press n to toggle whether notes are shown (or add '?notes' to the url before the #)

Press m or double tap to slide thumbnails (menu)

Press ? at any time to show the keyboard shortcuts

\title {Joint Action \\ Lecture 02}

\maketitle

# Joint Action

\def \ititle {Lecture 02}
\def \isubtitle {Joint Action}
\begin{center}
{\Large
\textbf{\ititle}: \isubtitle
}

\iemail %
\end{center}

\section{From Individual to Joint Action}

\section{From Individual to Joint Action}
Provides an alternative introduction to the question around which this module is organised.
Let me inform (or remind) you how philosophers standardly think about action ...
‘What events in the life of a person reveal agency; what are his [sic] deeds and his doings in contrast to mere happenings in his history; what is the mark that distinguishes his actions?
\citep[p~43]{Davidson:1971fz}

Davidson, 1971 p. 43

So what is the mark that distinguishes her actions? The passage that I’ve quoted doesn’t fully answer the question. (You need to know about basic actions too.) But here you can see part of the answer. Davidson holds that the mark is intention. That is not to say we do only things that we intend to---after all, alerting the prowler is something Davidson’s man without intending to do so. Rather, the idea is this. If you intend something, to turn the light on perhaps, and this intention is appropriately related to an event, then that event is an action. (This doesn’t yet tell us which events are actions.)
Brilliant theory, so simple. But what’s missing here? We are focussed entirely on an individual acting alone. There’s a parallel question about joint action: \textbf{What are our doings in contrast to things we merely happen to do in parallel?}
But why are we asking this seemingly quite technical question?

Question

What distinguishes genuine joint actions from parallel but merely individual actions?

Requirement

Any account of shared agency must draw a line between joint actions and parallel but merely individual actions.

Aim

Which forms of shared agency underpin our social nature?

This is the organising question for our project (the project to be investigated in this series of lectures). Of course there will be lots of further questions, but I like to have something simple to frame our thinking and this question serves that purpose.
My hope is that by answering this seemingly straightforward question, we will be in a position to answer the hard question about which forms of shared agency underpin our social nature.

They each intend that
they, the sisters, cycle to school together.

Our task, which we started at the end of last lecture, is to find objections to the Simple View.

\section{Walking Together in the Mafia Sense}

\section{Walking Together in the Mafia Sense}
Does Bratman’s ‘mafia case’ provide a reason to reject the Simple View?
Bratman offers a counterexample to something related to the Simple View \citep[see][]{Bratman:1992mi,bratman:2014_book}. Suppose that you and I each intend that we, you and I, go to New York together. But your plan is to point a gun at me and bundle me into the boot (or trunk) of your car. Then you intend that we go to New York together, but in a way that doesn't depend on my intentions. As you see things, I'm going to New York with you whether I like it or not. Does this provide the basis for an objection to the Simple View?

The Simple View

Two or more agents perform an intentional joint action
exactly when there is an act-type, φ, such that
each agent intends that
they, these agents, φ together
and their intentions are appropriately related to their actions.

Here’s the simple view again. My aim now is to present the most convincing objection to it that I can.

Bratman’s ‘mafia case’

Michael Bratman offers a counterexample to something related to the Simple View. Suppose that you and I each intend that we, you and I, go to New York together. But your plan is to point a gun at me and bundle me into the boot (or trunk) of your car. Then you intend that we go to New York together, but in a way that doesn't depend on my intentions. As you see things, I'm going to New York with you whether I like it or not. This doesn't seem like the basis for shared agency. After all, your plan involves me being abducted.
But it is still a case in which we each intend that we go to New York together and we do. So, apparently, the conditions of the Simple View are met (or almost met) and yet there is no shared agency.

1. I intend that we, you and I, go to NYC together.

2. You intend that we, you and I, go to NYC together.

3. You intend that we, you and I, go to NYC together by way of you forcing me into the back of my car.

We’re considering that Bratman’s ‘mafia case’ provides a counterexample to the Simple View. But does it really?
The mafia case fails as a counterexample to the Simple View because if you go through with your plan, my actions won’t be appropriately related to my intention.
And, on the other hand, if you don’t go through with your plan, that it is at best unclear that your having had that plan matters for whether we have shared agency.
I suggest that what is wrong in the Mafia Case is not that the agent’s need further intentions, but just that if their intentions don’t connect to their actions in the right way then there won’t be intentional joint action.
But the mafia case fails as a counterexample to the Simple View because if you go through with your plan, my actions won’t be appropriately related to my intention.
And, on the other hand, if you don’t go through with your plan, that it is at best unclear that your having had that plan matters for whether we exercise shared agency.
Bratman uses the Mafia case to motivate adding further intentions to those specified by the Simple View. But I suggest that an alternative response to the Mafia case is no less adequate and simpler: what is wrong in the Mafia Case is not that the agents need further intentions, but just that, if they act as they intend, their intentions won’t all be appropriately related to their actions.
So Bratman’s ‘mafia case’ is not a counterexample to the Simple View.
I note that Bratman is clearly aiming to identify intentions whose fulfilment requires shared agency. But I don’t think this is necessary. It seems to me that what matters is that the Simple View as a whole distingiushes shared agency from parallel but merely individual agency, not that it does so by way of fulfilment conditions of intentions.
Rather than continuing to discuss whether the Mafia case really motivates rejecting the Simple View, let me consider other ways to generate what seem to be more plausible candidates for counterexamples to the Simple View ...

## Walking Together in the Tarantino Sense

\section{Walking Together in the Tarantino Sense}

\section{Walking Together in the Tarantino Sense}
The Simple View, having survived the objection that it involves circularity and Bratman’s ‘mafia’ objection, now faces a yet more challenging objection. Apparently the Simple View cannot distinguish between all the contrast cases that an account of shared agency must distinguish. (Contrast cases are pairs of cases where one involves shared agency and the other does not and which are otherwise as similar as possible).

The Simple View

Two or more agents perform an intentional joint action
exactly when there is an act-type, φ, such that
each agent intends that
they, these agents, φ together
and their intentions are appropriately related to their actions.

Here’s the simple view again. My aim now is to present the most convincing objection to it that I can.

Walking together in the Tarantino sense

Here is my attempt to improve on Bratman’s counterexample. Contrast friends walking together in the way friends ordinarily walk, which is a paradigm example of joint action, with two gangsters who walk together like this ...
... Gangster 1 pulls a gun on Gangster 2 and says: “let’s walk” But Gangster 2 does the same thing to Gangster 1 simultaneously.
We might call this ‘walking together in the Tarrantino sense’.
The conditions of the Simple View are met both in ordinary walking together and in walking together in the Tarantino sense. [*Discuss ‘appropriately related’]. So according to the Simple View, both are intentional joint actions.

1. I intend that we, you and I, walk together.

... by means of my forcing you at gun point.

2. You intend that we, you and I, walk together.

... by means of you forcing me at gun point.

The interdependence of the guns means that our actions can be appropriately related to our intentions.
Now I wanted to say that walking together in the Tarantino sense is not an intentional joint action unless the central event of of Reservoir Dogs is also a case of joint action. And I think it’s pretty clear that that isn’t a joint action. But I was surprised to find that at least two people responded, independently of each other, to this suggestion by saying that walking together in the Tarantino sense really is a joint action.
My opponent reasoned that each is acting intentionally, and that coercion is no bar to shared agency.

the threat of collapse: trading intuitions

Just here we come to a tricky issue. There is a danger that we will just end up trying to say something systematic about one or another set of intuitions, where nothing deep underpins these intuitions.
I think this is a real threat; you’ll see that most philosophers are not careful about their starting point in theorising about shared agency. They merely give examples or a couple of contrast cases and off they go. Adopting this undisciplined approach risks achieving nothing more than organising one’s own intuitions. (It’s fine to organise intuitions on weekends and evenings, but it shouldn’t be your day job.)
That’s why I want to go slowly here --- maybe this is very frustrating and you want to get into the really exciting, weird and crazy stuff about plural subjects, shared emotions or aggregate animals. But before we can do this seriously we need some sort of foundation that will ensure we aren’t merely organising intuitions.

another contrast case: blocking the aisle

Imagine two sisters who, getting off an aeroplane, tacitly agree to exact revenge on the unruly mob of drunken hens behind them by standing so as to block the aisle together. This is a joint action. Meanwhile on another flight, two strangers happen to be so configured that they are collectively blocking the aisle. The first passenger correctly anticipates that the other passenger, who is a complete stranger, will not be moving from her current position for some time. This creates an opportunity for the first passenger: she intends that they, she and the stranger, block the aisle. And, as it happens, the second passenger’s thoughts mirror the first’s.

1. The sisters perform a joint action; the strangers’ actions are parallel but merely individual.

2. In both cases, the conditions of the Simple View are met.

The feature under consideration as distinctive of joint action is present: each passenger is acting on her intention that they, the two passengers, block the aisle.

therefore:

3. The Simple View does not correctly answer the question, What distinguishes genuine joint actions from parallel but merely individual actions?

Explain the case to your partner. Is it a genuine counterexample to the Simple View?

Is it a genuine counterexample?

Recall our earlier contrast cases ...

Question

What distinguishes genuine joint actions from parallel but merely individual actions?

The Simple View

I’ve been arguing that the Simple View is either outright wrong or else radically incomplete as an account of shared agency.
Apparently, it is possible for two or more agents to each intend that they do one thing together and to act on these intentions without them thereby exercising shared agency a strong-ish sense.
So the Simple View fails to provide a satisfying answer to the question, What distinguishes genuine joint actions from parallel but merely individual actions?
Let me pause to say why this matters and how it fits into the big picture ...
Philosophers have offered a tremendous variety of incompatible, wildly complicated and conceptually innovative theories about shared agency. The Simple View is an obstacle to discussing these theories. If the Simple View is correct, none of the complexity philosophers have offered is needed.
The first problem I encounter in thinking about shared agency is that philosophers seem to take for granted without argument that the Simple View can be excluded. In fact it is surprisingly difficult to show that the Simple View is wrong. The usual argument against it is that it is circular, but we saw that this argument depends on the mistaken assumption that all cases of acting together involve joint action.
A better objection to the Simple View involves counterexamples. But we saw that the standard counterexample, Bratman’s mafia cases, does not work. However refining that counterexample does appear to present a problem for the Simple View.
Note that I don’t claim that the objection to the Simple View is decisive; in fact one of my aims in these lectures is to show that it is possible to save the Simple View. Nevertheless I do think that the objections to it are serious enough that we must now explore what proper philosophers have to say about shared agency.
That’s why your first seminar task is to read and write about Searle’s early article. This along with some less-readable but perhaps deeper efforts by Raimo Tuomela are often regarded as having initiated contemporary discussions of joint action.
***BRATMAN's DIAGNOSIS - have to intend to do it by way of the other’s intentions. This is what is wrong in blocking the aisle shared agency means connecting with each other as agents, not merely as bodies
Bratman’s brilliant idea for avoiding this sort of problem is to suggest that we don’t just each intend the action but rather we each intend to act by way of the other's intentions.
We can put this by saying that our intentions must interlock: mine specify yours and yours mind.
Now this appeal to interlocking intentions enables Bratman to avoid counterexamples like the Tarantino walkers; if I intend that we walk by way of your intention that we walk, I suppose can't rationally also point a gun at you and coerce you to walk.

‘each agent does not just intend that the group perform the […] joint action.

‘Rather, each agent intends as well that the group perform this joint action in accordance with subplans (of the intentions in favor of the joint action) that mesh’

(Bratman 1992: 332)

each agent does not just intend that the group perform the […] joint action. Rather, each agent intends as well that the group perform this joint action in accordance with subplans (of the intentions in favor of the joint action) that mesh' \citep[p.\ 332]{Bratman:1992mi}.
Our plans are \emph{interconnected} just if facts about your plans feature in mine and conversely.
‘shared intentional [i.e.\ collective] agency consists, at bottom, in interconnected planning agency of the participants’ \citep{Bratman:2011fk}.
In making this idea more precise, Bratman proposes sufficient conditions for us to have a shared intention that we J ... ... the idea is then that an intentional joint action is an action that is appropriately related to a shared intention.

We have a shared intention that we J if

‘1. (a) I intend that we J and (b) you intend that we J

‘2. I intend that we J in accordance with and because of la, lb, and meshing subplans of la and lb; you intend [likewise] …

‘3. 1 and 2 are common knowledge between us’

(Bratman 1993: View 4)

\begin{minipage}{\columnwidth}
\emph{Bratman’s claim}. For you and I to have a collective/shared intention that we J it is sufficient that:
\begin{enumerate}[label=({\arabic*}),itemsep=0pt,topsep=0pt]
\item (a) I intend that we J and (b) you intend that we J;
\item I intend that we J in accordance with and because of la, lb, and meshing subplans of la and lb; you intend that we J in accordance with and because of la, lb, and meshing subplans of la and lb;
\item 1 and 2 are common knowledge between us' \citep[View 4]{Bratman:1993je}
\end{enumerate}
\end{minipage}
Note that these conditions are sufficient but not plausibly necessary. If sharing a smile is a joint action, and if human infants in their first year of life are incapable of knowing things about other's knowledge of their intentions about the other's intentions, then to suppose that these conditions were necessary would be to imply that you can't share a smile with an infant.
We cannot therefore regard Bratman’s insight as yet giving us an answer to the question, What distinguishes genuine joint actions from parallel but merely individual actions?
In developing the insight, Bratman produces a complex view. So let’s step right back and consider his own starting point (which isn’t quite to answer our question, although it is certainly important to him.)

## Bratman on Shared Intention

\section{Bratman on Shared Intention}

\section{Bratman on Shared Intention}
The leading, best developed account of shared intention is due to Michael Bratman. What are the main features of his account?
So I wanted to start by showing you his picture.
Bratman (2009, p. 150) starts with our question, What distinguishes ...?
But he immediately goes on to say, ‘Step back: what do we want from an answer to this question?’

‘three main concerns: conceptual, metaphysical, and normative.

‘We seek an articulated conceptual framework that adequately supports our theorizing about modest sociality;

‘... to understand what in the world constitutes such modest sociality;

‘and ... an understanding of the kinds of normativity—the kinds of ‘oughts’—that are central to modest sociality.

‘And throughout we are interested in the relations—conceptual, metaphysical, normative—between individual agency and modest sociality.’

Modest sociality:
‘small scale shared intentional agency in the absence of asymmetric authority relations’

Braman 2009, p. 150

\citep[p.~150]{Bratman:2009lv}
Why is this a reasonable strategy? One consideration is that, developmentally and evolutionarily, modest sociality almost certainly came first. So it is reasonable to hope that if we can understand this, we can build up to more complex cases. Another, related thought is Fred Dretske’s maxim that you don’t know how something works until you know how to build it. It is a reasonable guess that in building agents capable of exercising shared agency, a good first target is to get them started with modest sociality.
Note that this is only a starting point: we are also going to consider large scale joint actions (like electing a president) and very small scale joint actions (like sharing a smile or moving an egg).

the continuity thesis

‘once God created individual planning agents and ... they have relevant knowledge of each other’s minds, nothing fundamentally new--conceptually, metaphysically, or normatively--needs to be added for there to be modest sociality.’

Bratman (2015, p. 8)

\citep[p.~8]{bratman:2014_book}

What is shared intention?

Functional characterisation:

shared intention serves to (a) coordinate activities,
(b) coordinate planning, and
(c) structure bargaining

Constraint:

Inferential integration... and normative integration (e.g. agglomeration)

Substantial account:

To illustrate: if we share an intention that we cook dinner, this shared intention will (iii) structure bargaining insofar as we may need to decide what to cook or how to cook it on the assumption that we are cooking it together; the shared intention will also require us to (ii) coordinate our planning by each bringing complementary ingredients and tools, and to (i) coordinate our activities by preparing the ingredients in the right order.
The functional characterisation is really important: if we accept it, then it tells a lot about what shared intentions could and could not be. In particular, it either rules out Searle’s account or at least shows that the account is not clearly an account of shared intention because it does not explain how the attitude Searle characterises, the ‘we-intention’ could coordinate activities, coordinate planning and structure bargaining.
inferential integration ... [illustrate with planning example where you have to plan both individual and joint action]
normative integration (e.g. agglommeration)
[illustrate agglomeration for individual case first, then joint]
These points are extremely simple but also extremely powerful. They are powerful because they create problems for many approaches to shared agency. Consider Searle’s view again. He thinks that shared intentions are not intentions but a new, sui generis kind of attitude (which is why he uses the term ‘we-intentions’). If you think this, you have to explain how come the new attitudes are inferentially and normatively integrated with ordinary intentions. (I’m not saying this can’t be done, just that doing it is challenging, and certainly not something that Searle has attempted as far as I know.)
We’ll shortly see how the substantial account is built step by step. But maybe it’s helpful to mention the strategy
creature construction is an idea from Grice ...
the construction ...

step 1

‘Our shared intention to paint together involves your intention that we paint and my intention that we paint.’

Bratman (2015, p. 12)

\citep[p.~12]{bratman:2014_book}

(Compare the Simple View)

This is roughly what the simple view said This might seem completely innocuous, but it is interestingly controversial.

the ‘mafia case’* motivates ... and painting the house different colours motivates ...

step 2

We each intend that we paint by way of the intentions that we paint* and by meshing* subplans of these intentions.

Given what I said earlier, I don’t think the mafia case actually motivates step 2. But I did provide other cases (the Tarantino walkers and blocking the asile) which do seem to motivate step 2.
star: complication ‘and that the route from these intentions to our joint activity satisfies the connection condition’ \citep[p.~52]{bratman:2014_book}.
On the connection condition: It is ‘the condition that specifies the nature of [the] explanatory relation’ between shared intention and joint action \citep[p.~78]{bratman:2014_book}.. ‘the basic idea is that what is central to the connection condition is that each is responsive to the intentions and actions of the other in ways that track the intended end of the joint action--where all this is out in the open.’ \citep[p.~79]{bratman:2014_book}.
We intend to paint the house, but I blue and you red. Earlier work: I trick you ... In the book: ‘we have a problem. In a case of shared intention we will normally try to resolve that problem by making adjustments in one or both of these sub-plans, perhaps by way of bargaining, in the direction of co-possibility. So we want our construction to account for is standard social­ norm-responsive functioning of the shared intention.’ \citep[p.~53]{Bratman:2012fk}
meshing subplans are required
star: meshing
‘The sub-plans of the participants \emph{mesh} when it is possible that all of these sub-plans taken to­ gether be successfully executed.’ \citep[p.~53]{bratman:2014_book}
So much for step 2; now we come to the last major step (I’m skipping some details.)

why??

step 3

‘there is common knowledge among the participants of the conditions cited in this construction’

Bratman (2015, p. 58)

\citep[p.~58]{bratman:2014_book}
Why impose the common knowledge condition? Before discussing this [I might have to skip discussion of this, but there is a useful quote on the handout], let me provide a summary of where we are with Bratman’s account.
Why require common knowledge in the construction of shared intention? ‘in shared intention the fact of the shared intention will normally be out in the open: there will be public access to the fact of shared intention. Such public access to the shared intention will normally be involved in further thought that is characteristic of shared intention, as when we plan together how to carry out our shared intention. Since such shared planning about how to carry out our shared intention is part of the normal functioning of that shared intention, we need an element in our construction of shared intention whose functioning supports some such thinking of each about our shared intention.’ \citep[p.~57]{bratman:2014_book}

What is shared intention?

Functional characterisation:

shared intention serves to (a) coordinate activities, (b) coordinate planning and (c) structure bargaining

Constraint:

Inferential integration... and normative integration (e.g. agglomeration)

Substantial account:

We have a shared intention that we J if

‘1. (a) I intend that we J and (b) you intend that we J

‘2. I intend that we J in accordance with and because of la, lb, and meshing subplans of la and lb; you intend [likewise] …

‘3. 1 and 2 are common knowledge between us’

(Bratman 1993: View 4)

These are the conditions that we have been discussing.
Note that these conditions are offered as sufficient but not necessary. (Bratman originally claimed that they were necessary and sufficient, but nothing in the construction rules out alternative realisations of the functional characterisation of shared intention.)
Are sufficient conditions sufficient for achieving Bratman’s aims? Bratman’s pitch is this. Recall the continuity thesis (‘once God created individual planning agents and ... they have relevant knowledge of each other’s minds, nothing fundamentally new--conceptually, metaphysically, or normatively--needs to be added for there to be modest sociality.’ p.8) Bratman reasons that if we can give sufficient conditions for shared agency that are consistent with the continuity thesis, then our default assumption should be that shared agency does not require concepual, metaphysical or normative innovation.
So if we accept Bratman’s sufficient conditions, then we should also accept the continuity thesis. (There might issues about whether merely sufficient conditions are enough to fulfil his aim of providing a framework for theorising about shared agency; more on this when we come to consider joint action and development.)

lecture: too much discussion?

## Two Objections to Bratman

\section{Two Objections to Bratman}

\section{Two Objections to Bratman}
Searle and Velleman have offered objections to Bratman’s account of shared intention. (These are also objections to the Simple View.) Are Bratman’s replies successful?

‘Our shared intention to paint together involves your intention that we paint and my intention that we paint.’

Bratman (2015, p. 12)

‘the team intention ... is in part expressed by "We are executing a pass play." But ... no individual member of the team has this as the entire content of his intention, for no one can execute a pass play by himself.’

Searle (1990, pp.~92--3)

\citep[pp.~92--3]{Searle:1990em}

the own-action condition:

‘it is always true that the subject of an intention is the intended agent of the intended activity’

Bratman (2015, p. 13)

\citep[p.~13]{bratman:2014_book} [Note that Bratman *denies* this claim.]
Bratman rejects the own-action condition. He notes that there are cases in which it seems I can intend things which involve others’ actions (e.g. I can intend that my son tidy his room). In his 2015 book (chapter 3, section 1), Bratman considers two arguments for it which do not seem to succeed. I follow Bratman in thinking that there is no reason to accept the own-action condition.
Is the ‘own action’ condition a genuine requirement on intending.

Is it a genuine requirement?

So are we done with the objections? Not quite!
A further problem arises from Velleman’s observation about intentions ...

the settle condition:

‘intentions . . . are the attitudes that resolve deliberative questions, thereby settling issues’

Velleman (1997, p. 32)

\citep[p.~32]{Velleman:1997oo}
If we accept the settle condition, then there is a challenge to Bratman: I can only rationally and knowingly intend that we paint if I can settle whether we paint; and likewise for you. But it seems that, in many ordinary cases, I can’t unilaterally settle whether we paint. So, it seems, I can’t intend that we paint without relevant irrationality or ignorance. \citep[pp.~64--5]{bratman:2014_book}
How can we meet this challenge?
Is the settle condition a genuine requirement on intending.

Does Bratmans’s view
violate the settle condition?

A solution?:

(a) if we both do as we intend, we will paint

(b) our intentions that we paint are interdependent*

Strictly speaking, what is required is persistence interdependence ...
Our intentions have \emph{persistence interdependence} just if (a) each of us ‘will continue so to intend if, but only if the other continues so to intend’ and (b) ‘there is this interdepen­dence because each will know whether or not the other continues so to intend, and each will adjust to this knowledge in a way that involves responsiveness to norms of individual plan-theoretic rationality.’ \citep[p.~65]{bratman:2014_book}

(The persistence of my intention is interdependent with the persistence of yours, and this is because ...)

conclusion

Bratman’s account is excellent and it is difficult (but perhaps not impossible) to provide convincing objections to it. If it enables us to answer our question about distinguishing parallel from joint, we can then go on to explore its consequences for normative aspects of joint action.
But things are perhaps not so simple. counterexample is coming.