View source for Thinking About Groups

<!-- you can have any number of categories here -->
[[Category:John Holbo]]
[[Category:Jacob Levy]]
[[Category:Rationalism, Pluralism, and Freedom]]
<!-- 1 URL must be followed by >= 0 Other URL and Old URL and 1 End URL.-->
{{URL | url = http://crookedtimber.org/2017/08/20/thinking-about-groups/}}
<!-- {{Other URL | url = }} -->
<!-- {{Old URL | url = }} -->
{{End URL}}
{{DES | des = A review of [[Rationalism, Pluralism, and Freedom]] by [[Jacob Levy]].  "... can both the government and your neighbor be an abusive jerk? Obviously. There isn’t some a priori theory that tells us whether we are more at risk, relatively, of being busybodied by our neighbor or excessively nose-counted by the state. Or killed by a soldier of the state, or lynched by a voluntary local association of neighbors wearing pillowcases on their heads." | show=}}
<!-- insert wiki page text here -->
<!-- DPL has problems with categories that have a single quote in them.  Use these explicit workarounds. -->
<!-- otherwise, we would use {{Links}} and {{Quotes}}  -->
{{List|title=Thinking About Groups|links=true}}
{{Quotations|title=Thinking About Groups|quotes=true}}

{{Text |
In the hopes of writing something that isn’t rendered obsolete by Donald Trump in 48 hours, I’m going to say a few (thousand) words about how I got a lot out of Jacob Levy’s good new book, Rationalism, Pluralism, Freedom. At its core is a dilemma – an antinomy: two models of the optimal form and function of groups within a liberal order. Neither model can be quite it. It seems we need to split the difference or synthesize. But there is no coherent or necessarily stable way. (Well, that’s life.) There, I gave away the ending.

Groups? Yes, you know the sort: families, political parties, ethnic groups, clans, churches, professional organizations, civic organizations, unions, corporations, neighborhood groups, bowling leagues. The lot. There is a lot to distinguish in this lot, but let’s not. Let’s lump. Anything that might be a locus of loyalty or ‘identity’, above the individual level, below the state level. There are citizens; there is the state; there are intermediate groups betwixt and between. (Use ‘civil society’ if you are happy with that, but maybe that term has baggage we might need to rummage through. Jacob has things to say about how the term is tricky.)

Continuing on cartoonishly, for (possibly illusory) clarity: most of us – I’ll wager – subscribe to the view that life in a modern liberal democratic society is like this: there are citizens and there is the state. Citizens enjoy a basket of liberties and rights, over and against each other and the state. A lot of critics say ‘liberals’ think too much in terms of autonomous individuals – little social atoms, bouncing freely off each other – but even those critics would mostly be slow to replace what we’ve got with another system in which some other circle was deemed more elementary than this mutually orbiting citizen-state binary: the family, the estate, the deme, the guild, the tribe, whatever. If you seriously want to scrap the US Constitution in favor of some kind of neo-feudal system, for example … well, that’s pretty radical. I’m just going to assume I’m addressing an audience that is basically ok with making individual citizens primary units over and against the state. (All pretty vague? Yes, quite. But we haven’t got all day. On we go.)

Now, if you have these two basic units, the state and the individual, it makes it kind of tricky what the normative status is of intermediate groups, eh? What is all that in-between stuff good for or bad for? What sorts of ‘mediating groups’ need to exist – because they’re great! possibly vital for the health of citizens and/or the state itself! What sorts of stuff should not be permitted, because it’s toxic – either to the state or to some individuals. And what sorts of stuff should be merely tolerated, even though its a bit dicey, but pragmatically what are you going to do? (Trying to rule out certain groups would cause more trouble – do more harm – than good. Even so, those groups may do more harm than good, to individuals and the state. A lot of dumb, bad groups aren’t banned, and that’s how it should be.) At this point some people might say: I care about the health/power/status of my group way more than I care about either the stability of the state or respecting the rights and liberties of my fellow citizens. If what it takes to keep my group dominant is depriving my fellow citizens of rights and liberties, I’m willing to do that. But if you say that, you really are not on the same page. You aren’t committed to liberal democracy. In this post I’m only considering what attitude you should have towards groups, in the abstract, if you have some normative commitment to making sure individuals can exercise their rights and enjoy their liberty, equally, in a stable liberal-democratic state.

This is a timely issue: identity politics and groupthink and partisanship and tribalism. (I am going to assume you’ve noticed these are hot-button issues, and I’ll omit reviewing recent events for your benefit. Some horrible thing is about to happen to make my report obsolete, if I tried.) The worry is that a lot of this groupishness is bad, inimical to the health of a free society. Everyone sees bad groups – bad identity formations – they don’t like. But everyone is also attached to groups they like.

How do we think about that? I think most people think about it pretty roughly. Now, let’s turn to Jacob’s book, which can help. As I said, he outlines two views about the proper/permissible form and function of groups, each of which looks too extreme in its purity (he says so and I agree.)

[UPDATE: it occurs to me I’m mixing my own thoughts with Jacob’s here. Tell the truth, I’m a bit mixed up myself about what I learned from him, what I thought before, and what I thought in response to him that maybe wasn’t explicit in his text. It’s a mystery.]

On the one hand, we have the ‘pure liberal’ theory.

Basically, the idea is that what any individual has the right to do, individually, many individuals have the right to do, associatively. Call this unfettered freedom of associations (note the ’s’). It’s a function of unfettered individual freedom of association. (This is a libertarian move, and you have no doubt met with it on the internet.) This makes sense abstractly. It has an attractive principled simplicity and clarity. It also provides plausible answers in a wide range of contexts. Consider, for example, groups that many people won’t find attractive or healthy-looking – from s&m dungeons to monastic orders! The  ‘pure liberal’ theory is clear about what you say: if you folks want to get together and bind yourselves to all that, tight as all that – you go be your bad selves! Don’t force anyone else to play along and join in. And there needs to be something like a safe word so you can get out if you change your mind. You still enjoy a right of exit from any group, as a citizen of the larger state in which this group exists.

That sounds good to me. How about you? I’m not interested in forbidding recreational bondage play or the Benedict Option.

Selling yourself into slavery is a bit dicier. Can you do that? I think the ‘pure liberal’ theory would have to say: yes. (If you get a good price for yourself and then you can give the money to your kids to make better lives for themselves? and you are happy with that exchange? – fine! But you can’t sell your kids into slavery.) I’m pretty much not on board with selling yourself into slavery. I’m off the bus.

The looming practical problem with the ‘pure liberal’ theory is not so much the selling-yourself-into-slavery bit, however. To me the more salient, sure-to-arise-in-real-practice problem with infinite freedom of associations is going to be something more like, for example, Jim Crow (just not state-sponsored). If all the white families in town form a voluntary association that collectively commits to disassociate from that one black family – no one will hire/sell to/be in the same room with them – that’s going to get pretty ugly for the one family. If the association likewise ostracizes any white who refuses to get on board with this? … well, you do the math. No reason why this viciously-illiberal, apartheid-like social order won’t be stable and self-sustaining, yet everyone is playing according to ‘pure’ liberal rules. The moral of the story: private, voluntary groups can – and predictably will, if unchecked – cast long, sinister shadows over large areas of the public sphere. Some citizens will not, in practice, enjoy a right of exit out from under the oppressive shadow of some group, even though all the group is doing is exercising aggregate freedom of association of its membership, doing things at the group level that would be fine at the individual level.

To sum up: if it turns out that ‘freedom’ is, in fact, another word for nothing left to lose, maybe you are using ‘freedom’ wrong? So the pure theory is wrong? (Don’t object that social Jim Crow – black-white racial prejudice – is not necessarily going to be such a problem. If not that, then something else like it. There’s going to be something.)

So that’s pure theory 1. On the other side, we have pure theory 2, what Jacob calls ‘congruence’ theories (which, before his book, didn’t really enjoy a collective label.) The idea here is that liberalism has to go all the way down – and through-and-through. You can’t have significant pockets of illiberal order existing within the matrix of liberal order, because such pockets are a toxic threat to that order, and to individuals living in it, inherently. For a representative expression of this view, see Susan Moller Okin’s classic, Justice, Gender and the Family. (Jacob cites it as a good example; it is.) The family needs to cease to be patriarchal. Same goes for churches and communities that enforce illiberal norms within their group borders. (Okin focuses on gender, but obviously that’s not the only possible issue. But it’s a good one. I think all the general types of problems groups can raise can be exemplified in terms of gender.) For Okin and other ‘congruence’ theorists it’s not enough to say that these groups are voluntary, or that citizens enjoy formal right of exit upon attaining the age of majority. I’m not going to run through the arguments for why it’s not enough. I take it the gist is clear and familiar. If you are a female, born into a patriarchal family, within a patriarchal ethnic community, raised in a patriarchal religious sect in which women do not have a significant voice in church leadership, it is not deemed sufficient that, at the age of 18 you can leave the church, turn your back on your family and community, up stakes and start again somewhere else. The requirement that you go to such extremes, to achieve exit, just so you can enjoy basic liberal standards of equality and opportunity and so forth, is too burdensome.

As Jacob notes, ‘congruence’ is what you arrive at on Charles Taylor’s ‘long march’. It’s what you get with ‘modular man’, in Gellner’s sense. You could say what we are insisting on, here, is ‘modularity’ all the way down. Not because we are the Borg. (The popular conservative idea that this view is motivated by an evil obsessive-compulsion to institute homogeneity-for-its-own-sake is mostly just bad psychology, with a bit of sophistry on top.) The motive is obvious: it’s not fair if equal citizens can’t exercise their rights and liberties fully without taking nearly heroic steps to get clear of groups that make it hard for them to do that. That’s a totally obvious and plausible moral reason to favor this approach.

So what’s the problem if not that congruence theorists are all evil Borg-types? The practical problems with congruence not likely to be so extreme, in practice, as those with the pure view are. (I think it’s almost certain that the pure view would produce extreme problems, if instituted, whereas it’s merely a likely possibility that the congruence view would produce extreme problems, in practice. That’s my back-of-the-envelope estimate. YMMV.) But the problems here are real, and may be extreme. You can’t get thee to a nunnery – or even into your favorite s&m dungeon – if that is your druthers. Or even just attend an all-girl’s school.

There is a flattening out of a lot of the texture of cultural and social and institutional and religious life. This can seem pointless, at worst horrifying (like rows and rows of identical buildings.) People can’t choose to belong to all sorts of distinctive groups because every group is required, not to be exactly the same as every other, but similar in essential points. So, in the name of freedom, you have restricted choice. Maybe – maybe – some dumb Harrison Bergeron stuff is going to be mandated as well. That’s less clearly required in principle by ‘congruence’ but worth thinking about. It’s pretty clear what it means to say that freedom of associations is in principle unlimited. It’s not so clear what it implies, in principle, to mandate the replication of liberal order at all levels and scales of groupish life.

So we have our two pure views: the pure liberal view; the congruence view. Take your pick. Which do you like: maybe Jim Crow, or maybe Harrison Bergeron?

Obviously we don’t like the bad extremes into which either side may fall. So how do you split the difference? It’s not so easy to see how to do so in a clear, principled (or just plain sensible-seeming) way.

Take the ‘pure’ side. The obvious danger comes when groups are big enough to generate a kind of critical mass problem. So you forestall that by imposing limits – kind of like zoning regulations. Or anti-trust, anti-monopoly rules. A rough first cut might be this: the degree of incongruity of your group with liberal norms has to be inversely proportional to its size/power. The more you can throw your weight around, the more careful you have to be not to trap your own members, so they can’t get out; and the more careful you have to be not to crush smaller neighbors, because they can’t get out from under you, into the light of the public sphere. The main problem with this fix is that 1) it’s a lot less clear than the categorical freedom of associations rule; 2) it’s philosophically orthogonal to it, because it’s being hybridized with the congruence view. Crudely: libertarians (like Nozick) don’t like mandated ‘patterned distributions’. You let people run free and the chips fall where they may. If you aren’t fine with Jim Crow, if that’s how the chips fall, you have to adopt at least a negative ‘patterned distribution’ proscription. Some patterns are forbidden. But then: you have to stop raging against ‘patterned distributions’ because that’s you now. (Maybe you are ok with that.)

We’ll get to the hybrid fix for the congruence side down below. But it will be clearest it we get something else out of the way first.

One thing that makes it hard to synthesize the two views, theoretically, is that each side – the pure view and the congruence view – is preoccupied with forestalling a particular class of practical, downside risks. (This is a very important point for Jacob, and he does a really good job of discussing it.) It’s reasonable to fault each side with downplaying the risk the other side sees, and excessively playing up the risk it worries about. But it’s not right to say that, theoretically, the risk has to be equal in every case. So it’s hard to theorize the right degree of correction in a general way.

Let me just fill in the details, so you see what I’m talking about. ‘Pure’ liberals are attuned to the danger posed by ‘the Man of Systems’ (which is paradoxical, because, to a man – and most of them are – pure liberals are men of systems. But that’s an irony for another day.) I actually don’t like the term because I think it implausibly exaggerates the incidence of planners who are inhumanely Borg-like in their normative outlook. It’s far more common just to be over-optimistic about what can be achieved – or just plain corrupt. Everyone knows that there are always cost-overruns, and the thing isn’t built on time, and there are construction flaws, due to planning failure and/or corruption. That applies to state planning as much as to building anything else – a new house. That’s really what we are talking about here. What are the problems going to be if, predictably, those in power are always over-straining their native capacity to plan and execute plans in a virtuous manner?

As a tonic correction to the limits of central planning and rationalization, you can insert your Burkean ‘little platoons’ (even though Burke was no pure liberal). You can slot in standard Tocquevillean insights (or just your copy of Tocqueville. I think some folks keep an unread copy of Democracy In America just to lob, self-righteously, at the heads of ‘progressives’, who allegedly haven’t read Tocqueville, even though they totally probably did read some in college.) Insert also your Montesquieu and your Constant, as well as Ferguson and Smith. Basically, pile up all cogent worries about how, possibly, the central power will have a tendency to stupidity and/or downright rottenness. On the day that risk is realized, you will be glad if intermediate groups exist that are of a proper size and character to counter central power. Even if the day isn’t here yet, it’s good for those in central power to know that if they start misbehaving, the periphery will get pissed and start acting as counterweight. Checks and balances.

Those attuned to the risk that central planners may go wrong are bothered that certain schools of liberal theory quite literally stipulate away such risk, assuming we’ve instituted justice just to see what perfection looks like. (I’m looking at you, Rawls!) Assuming the best, government-wise, is not the best way to prepare for the worst. At the same time, it’s worth pointing out that two very different argument streams are crossing here: 1) a principled argument that, since individuals should enjoy free association, that ought to scale up to groups. Why not? 2) a more pragmatic argument that the state should contain within itself powerful groups as a kind of insurance policy against the state going insane and bad. 1) and 2) can lean on each other, but they are not mutually entailing, by any means, and they are obviously very different thoughts.

Now we turn to face the other direction, the congruence group. The thinkers who are so sensitive to the dangers posed by central levers of power in the hands of The Man of Systems can be total idiots – willfully and/or self-servingly blind – when it comes to thinking about more local dangers. In his book Jacob talks about the ‘busybody’, as opposed to the ‘Man of Systems’. I think his foot slips a bit, in choosing this mild-sounding tag. It does catch a lot. But not all. But Jacob doesn’t play down the life-and-death risks. (Armed KKK militias aren’t ‘busybodies’ by half.) Local tyranny – from your dad, the boss, the intolerant folks in your neighborhood, the ‘little platoon’ burning a cross on your lawn – can be way worse for your health than what some dumb or even corrupt king or state planner gets up to, off at the distant center. J.S. Mill was sensitive to this, and most Americans are, too, without having to read Mill. When we tell ourselves history stories in which freedom triumphs over tyranny, some of the proudest moments take this form. If you are an African-American, under Jim Crow, you are glad when central government reaches out its universalizing, systematizing, visible hand, and gives you some civil rights for a change. Libertarians who can’t see any threats to liberty except from the government are just never going to win over people who know enough US history to know that, sometimes, government frees you from local, perhaps private tyranny.

So basically people can be dumb in both directions. They can be over-optimistic about the good government can do, systematically, for everyone – who are all supposed to be equal. They can incautiously play down the risks of government being dumb or corrupt, if local groups can’t counter-balance them. They can be over-optimistic about good in ‘little platoons’. They can fail to see the threat to liberty in what Mill called ‘the tyranny of society’, as opposed to the magistrate.

When you put it like that, it’s obvious: can both the government and your neighbor be an abusive jerk? Obviously. There isn’t some a priori theory that tells us whether we are more at risk, relatively, of being busybodied by our neighbor or excessively nose-counted by the state. Or killed by a soldier of the state, or lynched by a voluntary local association of neighbors wearing pillowcases on their heads. You just have to be not-dumb about perceiving risks of both sorts. But this obviously makes it hard to have an abstract theory, in which is inscribed some stable preconception of the relative salience of those risks.

In other words – and to sum up – the ‘pure’ theory says: worry about this more! The ‘congruence’ theory says: worry about that more! The only way to synthesize these is to say: girls, girls, you’re both hideous! But it’s just not obvious they are equally hideous.

And that’s Jacob Levy, folks! It’s a good book. I tried to make it sounds kind of duh-obvious. Like we knew this already. Well, that’s good books for you. After you are done you are kind of like: why didn’t I write that? (Except for the Montesquieu bits. It’s all I can do to spell that guy’s name. I get it wrong half the time.)

Now, let me chip in my own 2 cents on top. I said I would say something about the abstract problems with kludging the congruence side, by injecting it with some of the wisdom of the other side. I get the sense that Jacob would probably agree with what I’m about to say, but I’m not totally sure. Maybe I’m actually channeling him. Not sure.

A sociological observation: conservatives like to complain that liberals are victimologists and special snowflakes. The reason for this is that conservatives are themselves a bunch of big fat, moist, special snowflakes who like to think of themselves as endlessly victimized by victimologists. If liberals ever stopped feeling so damn sorry for themselves, conservatives would have to cancel their own eternal pity party. But seriously, the more charitable way to put it is: people of all walks of liberal democratic life like to feel like they are good, decent people. They couldn’t feel that way if their groups – the groups they identify with – were systematically, on a daily basis, infringing good, decent people’s rights and liberties. (Who could sleep at night thinking they are a total bastard for belonging to a group that is undermining liberal democracy and depriving innocent people of liberty?) So people, as a condition of feeling decent about themselves, need to feel like they are underdogs, in the groupishness department. It’s a form of moral self-justification or insurance: knowing your group is on the outs, on the defensive, is the little guy, the victim. So long as that is true, you don’t need to worry too much that the tribalism you are perpetrating might be a bad thing, not a healthy thing.

It’s like everyone is working with the following maxim: incongruency is ok so long as it counter-balances incongruency elsewhere in the system (so really the fact that you are being locally illiberal makes the overall system more liberal)

Putting it another way: groupishness is permissible iff it’s publicly anodyne (we can ignore those mild cases) or favors underdogs because it functions to level the public playing field by introducing an unleveling that is equal and opposite to some other unleveling. (To mix dog and leveling metaphors.) Obviously then it’s impermissible to have groupishness that favors overdogs, because that functions to unlevel the playing field. Overdogs really are morally obliged to make themselves like social atoms, because when they run in packs they are just intolerable.

This means, of course, that everyone is always figuring out how their group is actually an underdog, and trying to scope out angles from which they look underdoggy.

Is this a bad approach? Well, I think it’s kind of morally plausible. I’m attracted to congruence as an ideal, but I like my groups, too, and I feel that they are good for liberal democracy. When testing groupishness for permissibility, test it for underdogginess. But on reflection, that’s a pretty rough test. And it motivates not just victimology (which is sort of sour) but confabulation (which is not the way to think clearly.) It’s often confused or dishonest or otherwise inadequate.

This is not to deny that there are victims, and groups that have way better titles to victimhood than other groups. In saying a lot of people are motivated to think confusedly about how they are underdogs, I’m not saying there’s no such thing as thinking clearly, or that it’s just in the eye of the beholder.

I’ve tried to avoid naming group names in this post. I’ve talked very abstractly about ‘groups’. I think that’s actually a good thing, even though abstraction has its limits. If I had named names, we would have gotten in bitter arguments about the particulars in comments. (Because that’s what comments are for! Let the games begin!)

In what way does my abstract framework help in the least? Well, probably not much. Who pays attention to abstractions when they’ve got the least skin in the game? So when any actual case comes up, people will just be that way. But here’s a test, if you can just remember to run it when the time comes. When some group you identify with is being accused of bad groupishness because of something-something, and when you are inclined to make an underdoggish whadabboutish defense – whaddabout that other group that is against us, and is bigger? – think whether you would find your group’s behavior permissible on the (counterfactual!) assumption that it isn’t an underdog. You are free to go on believing your group actually is the underdog. I’m not asking you to doubt that. You could very well be right. But still perform the exercise. Imagine your group is dominant, or at least comfortably secure, not disadvantaged hence in need of actively pushing back when pushed. Would that thing your group did that bothers people still be ok in those non-dire straits? This should clarify, in some small way, whether the group behavior you defend, that someone else finds oppressive or bullying, is justifiable in your own eyes ONLY because you conceive of at a defensive response to larger oppression and bullying. That is, you buy into the ideal of congruence, and you only excuse local incongruence in the service of larger congruence. You just happen to believe that everything your groups do serves that larger congruence, even though others (wrongly) don’t see it that way.

This post is long enough. I just hammered it out to get my abstract thoughts all out. Gosh, I wonder whether they are true! No doubt you will tell me they are false. Oh, what a relief it will be to be properly and severely corrected by your superior wisdom!
}}