[Nat]

I was listening to a podcast today and found it so interesting that I went ahead and transcribed it as best I could. The reason I'm pasting it here is because I think some of the issues we run into on the site may involve issues of the group mind vs the individual.

------------

Transcribed from Philosophy Bites podcast "Philip Pettit on Group Agency"

"I think of group agents as a subspecies of groups in general. There are groups - their members will have a property in common and so-on, but the feature that makes them a group agent is that they mimic an individual agent in how they behave.

If you take a small group like the three of us - suppose we constitute a group perhaps that meet every evening for a drink. Now in order to constitute a group agent, what would have to be the case that there is some goal or purpose that we together want to advance or further, and equally we form shared views about the best means to advance this goal, about the opportunities, about the sub-goals, about the order of importance, about the sub-goals - and of course we're going to form views about changing the goal, about adding to the goal. We've got to as a group, the three of us, organize ourselves in order to advance a purpose according to certain accounts of how best to advance that purpose. We've got to do together what we all do individually to pursue a goal.

Groups obviously vary enormously in size and enormously in structure. The three of us could be a group. We could develop some political purposes, for example, we're speaking on advancing the green agenda in Oxford. So that's the big goal we might agree on. We then would want sub-goals: what are the important aspects of the green agenda? Keeping the streets clean, keeping the waters clean? We've got to decide on all of these, and then with these things we've got to decide what do we do? Together we've got to agree on an agenda of what we do and a means of doing it, which is just what I as an individual would do if I decided on my own to pursue a green agenda.

There's quite a dominant tradition that says, look, it's just a fashion of speaking to talk about a group as an agent - they are just individual human beings who are agents and they coordinate their actions together. And at a certain level that's true, but it misses something very important, which is: When you ask, "What's the relationship between the goal of a group or the goal or beliefs or the the judgments of the group and the judgments of individuals?" People who take the view that groups are just individuals, they will always answer, "Well, the goals which the group holds - they just answer to the views held by the majority of the members."

When you begin to see what's wrong with that, you get an introduction as to why group agency is an interesting topic.

Let me serve it to you abstractly: Suppose the three of us are involved in having to make a set of judgments and decisions because we want to mimic an individual agent. As we agree, we have to put together our judgments on various matters and also our desires, self-preferences. Suppose for example we've got to put together our judgments on matters like the matter of whether P is the case and the matter of whether Q is the case. The issue is whether then the conjunction of P and Q is the case.

So, it comes to the matter of whether we as a group believe P, guys? Well, here beside me I have David and Nigel. Let's suppose that I think P is the case and David things P is the case, so that's the majority. So Nigel doesn't think P is the case, but that's okay - as a group we believe P.

Okay, then it comes to Q. We vote on Q. It turns out I think that Q also is the case as well as P. David this time rejects it - he doesn't think Q is the case. But you, Nigel, accept that Q is the case. Ah, majority ticked off Q - the group believes Q.

We believe P, we believe Q. Does the group believe P and Q? Ah. The majority rejects P and Q, because I'm the only one who believes P and Q. I believe both P and Q, so I believe P and Q. David doesn't, because he rejected Q. Nigel doesn't, because he rejected P. So as a group, if we follow majority voting, we're stuck with saying we believe P, Q, but reject P and Q. That's a real problem. If we behaved like that, we couldn't operate as a group.

There are many many examples of this. Actually, I became alerted myself to that particular paradox through some legal literature - what is called the doctrinal paradox. I gave it the name, "the discursive delimma," arguing there was a wider problem than you have in the legal case. Let's focus on that paradox - the discursive delimma. Here's an example of it:

Suppose the three of us make up the board of some organization, maybe it's a housing association. Someone comes to us with a complaint against the landlord which is that the heater in his room blew up and caused him great psychological damage or harm. He's brought a complaint against the landlord to us, and the three of us have to decide on that complaint. Imagine now that we have decided in the way courts would decide an issue like this - which is to say the landlord will indeed be liable, he'd be blameworthy, if he had a duty of care in this matter - looking after the heater - and indeed the tenant was actually harmed - he really was traumatized by the blowing up of the heater. If both of those are the case, then we think the landlord is liable or culpable and maybe there's some punishment due to him as a result.

The three of us have to make up our mind on these matters. The first issue - was the tenant traumatized? David and I might agree, like P. And Nigel might think no - but as a group we think yes, he was indeed traumatized by the incident.

Second question: Did the landlord have a duty of care for the heaters and looking after them or was it somebody else maybe in this housing association we're imagining? And let's suppose David thinks no, the landlord didn't have a duty of care, but Nigel thinks he did and I think he did. So again, Q - the landlord had a duty of care - we agree to that as a group.

But now it comes to the issue of do we agree that the landlord should be held culpable, blameworthy? But remember I'm the only one who thought P and Q - that the landlord had a duty of care and that this poor tenant was damaged. Each of you rejected one of those, so you're going to say no to that and I'm going to say yes. Now as a group we are being incoherent.

So as a group, we've got to avoid just forming the group view that answers by majoritarian methods to our individual views. If the three of us are going to behave as a group agent - that means we are going to mimic an individual which means as a group we are going to be capable of advancing purposes, sharing purposes, according to shared representation. We can now see in order to achieve that agency, we have to make sure that our representation of how things are and our purposes are actually consistent. What you've seen from this example is that you can't guarantee they'll be consistent if you just rely on majority voting within the group.

Christian List and I were happily able to establish a more general result which is that there is no simple way of starting with individuals and their views and then determining the group views by the majority vote or any other simple sort of aggregation of those individual views into a group view. You just can't do it.

It is tragic in this sense: it's called an impossibility result. It's impossible to ensure that the views of a group are coherent, that they are rational, that they stand together, that they are consistent - it's impossible to establish that kind of rationality, "collective rationality," let's call it - and at the same time establish another condition which you might call "individual responsiveness." In other words, have the group hold views that are responsible case by case to the views of the individuals.

If we're going to get our act together, say on the housing association example, what we have to do is decide NOT to go by majority views but on some issues to adopt a view as a group that the majority of us actually reject. That is called constructing, as it were, a group mind.

There's actually a very long tradition of recognizing that individual human beings can combine into units or groups which themselves can act like agents, having purposes, having representations, having means of changing these purposes and representations, and being pretty rational and coherent about doing all of that. Although interestingly by the standard history, the Romans and the Greeks for example didn't have that notion of a corporate entity. That suddenly happens in the Middle Ages to answer to realities like the guild or the town or the monastary or the monastic order or whatever. Groups which now have a real salience in this world. And as these groups become salient, the lawyers and the philosophers begin to talk precisely of the artificial person, the persona ficta, some understand it as a pretend person but many as an artificial person. Already in the 1300s there's a strong particularly legal group of people who are arguing that these corporate entities are persons in their own right and that they should be treated as persons, they've got standing in law like persons, they've got property, they have contracts, they can have a mind of their own.

Myself looking at those midieval thinkers, you sort of feel, gosh, they are thinking the same sort of things that I think. But then you move down to the 17th century for example and people like Thomas Hobbes, who begins to think about the State and the Commonwealth in precisely the same way, modeling it as he says "on the company of merchants." Hobbes does something that I think misleads other people for a long time - he suggests that what happens when you do get an incorporation of individuals is that they go by majority voting. That's endorsed also by Locke and by Rouseau when they equally talk about the Commonwealth as an incorporated agent.

They are wrong about that, but after Hobbes, there are two developments really. One development is in actual practice and the other development is in legal theory or philosophical theory.

The development in practice is in the 19th century - the rise of the commercial corporation. Remember that in the early 1700s, corporations were severely limited in English law as a result of the South Sea Bubble and the bad experience there. Things began to loosen up again only in the 1820s and from then on over the next 30 or 40 years, and the same thing happens in America. You get an amazing development in which commercial corporations become capable of more and more and more independence. So the corporation can be formed just by registering it, it can operate in any area within the domain of the legislation, it can change sphere of activity without going to parliament or back to the registry, and you get limited liability of course. And of course, corporations are allowed to own other corporations and to control other corporations, so you get the possibility of a whole biomass of corporate entities, which is what we've been experiencing over the last 200 years.

The conceptual development - very interesting things happen. One is that a preeminently german thinker Arthur Gilke? goes back to the medieval sources and develops a whole theory of corporate entities - churches and commercial corporations and political parties and states and so on - about how these corporate entities have got a life of their own, are "real persons" - a phrase used - and that becomes highly influential. Unfortunately as I think, that whole development, this new sort of interest, philosophical, theoretical interest in corporations, in corporate agents - that all gets held up because it becomes associated in the popular mind with fascism, because they talk the language of corporations, of corporate entities a lot and so in the fight against fascism, intellectual and otherwise, you get a cult of what comes to be called "individualism" which is taken mistakenly to commit us to thinking there are no such things as group agents. Frankly, I think we are only recovering in a way from that triumph mid-century of a line that said "there are no group agents, there are only individual agents." Of course there are group agents."