Second-level Libertarianism

So if you haven't yet read Patri's Cato Unbound essay, you should go do that now. And, if you didn't happen to catch it live, you should also check out the podcast of his Seasteading talk at Cato. Of course, regular DR readers are already familiar with Patri's basic thesis: governments are inefficient local monopolies that impose (a) high barriers to entry and (b) exceptionally strong customer lock-in. That's not an especially controversial claim. What is, perhaps, somewhat more controversial Patri's diagnosis: Policy Libertarianism is useless in the face of overwhelming institutional obstacles, so if we want a truly free society, we should instead focus on changing its fundamental structure.*

But it strikes me that a lot of PL types (and certainly a lot of the commenters at Cato) rather significantly misunderstand Patri's vision. Will Wilkinson offers the clearest -- though hardly the only -- such example:

That said, one of the merits of Friedman’s "dynamic geography" is that it is not really a "libertarian" project at all...I think there’s good reason to expect competing sea-top jurisdictions to settle on a scheme of governance more libertarian than what the world’s current nation states have to offer. But I also think there’s little reason to expect a seastead to embody the system of most libertarians’ dreams unless a lot of libertarians coordinate and settle there.

I don't want to get into any sort of I'm-more-libertarian-than-you shouting match here. Still, I think it's wrong to dismiss dynamic geography as not libertarian. I submit that it's best to understand Patri's project as what I'm going to call second-level libertarianism.** I'll try to explain what that means below.

Many (perhaps most) PLs are concerned with bringing about what we might call first-level libertarianism. That is, they want to live in a society with a minimal (or possibly even nonexistent) state, one with little-to-no taxation, an unfettered market, purely voluntary associations, and no restrictions on self-regarding or consensual behavior. Appropriately enough, PLs concentrate on ways of making our society more like that. And, as Patri allows in his talk, this can be a positive development. Lowering tax rates and increasing growth (even if only by a little) can have large positive-sum utility when aggregated over lots of people.

Moreover, the PLs have a point when they complain that Patri may be giving them too little credit for the work that they are doing. To the extent that seasteading is practical, it will be because enough people in the United States are willing to let seasteaders go their own way. To put the point another way, the best way to keep the U.S. Navy from parking a carrier group outside Floating Libertopia is to convince lots of Americans of the value of allowing small groups to pursue alternative political arrangements. Wilkinson and Reason's Brian Doherty are right to suggest that seasteading would be even less likely to succeed without Reason, Cato and the rest of the PL crowd. For that matter, it's not totally unreasonable to suggest that absent the work of Patri's grandfather, there's a real possibility that the Seasteading Institute's venture capital would have gone to cover Peter Thiel's 90% marginal tax rate.

But Patri's critique is fair in another way. See, where the SLs and the PLs part company, I think, is that Patri isn't really aiming at a first-level libertarian society. Will is right in that respect. But Patri is aiming at what I'm calling second-level libertarianism. The analogy I have in mind is similar to the distinction made between act- and rule-utilitarianism.***

Basically, the idea is that, like with utilitarianism, libertarianism can be applied in two different ways. It can be applied to particular governments -- that is, we can say that a particular state is more or less libertarian depending on the extent to which it conforms with whatever principles we take libertarianism to consist of. But both utilitarianism and libertarianism can also be used in a second-level fashion, as a way of generating the specific rules (in the case of utilitarianism) or the specific societies (in the case of libertarianism) that govern our ordinary, day-to-day lives.

What the particular rules are don't really matter here; the point is just that, on the second-level view, we don't necessarily appeal directly to utility at the level of action. We might, if it works out that direct appeals to utility are the best rule-of-thumb available. But if they aren't, then we might well end up being guided by rules that don't look particularly utilitarian.****

I think Patri is doing much the same thing. Indeed, he says as much during the talk, though it's a point that seems to have escaped most of the audience. Dynamic geography -- and by extension, the seasteading project -- isn't about creating a first-level libertarian society. It is, rather, aimed at creating a libertarian framework for creating new governments. If we take seriously Patri's argument that government is just another industry, then we should aim at a world in which the industry of government is a competitive market. We ought, in other words, to apply the principles of libertarianism to the generation of government itself.

To put the point another way, libertarianism -- much like utilitarianism -- can be applied at either of two levels. It can be used as a set of guidelines for a particular government. But it can also be used as a set of guidelines for generating the set of existent governments. PLs are aiming primarily at the former. SLs like Patri are aiming primarily at the latter. The idea, of course, is that a libertarian framework will -- as Will suggests -- be likely to generate particular libertarian societies. But even if it doesn't, I think that there is a good case to be made for the argument that a world in which governments are generated via a libertarian framework is, in some very real sense, a libertarian world even if none of the actual societies generated from that system looks like Libertopia.

* I'd like to point out how thoroughly -- and quickly -- Jacob's PL/SL distinction has become standard parlance; Cato's Doug Bandow, for example, uses the term throughout his response to Patri.

** For those philosophers out there, yes, I am borrowing the idea of two-level libertarianism from R.M. Hare. Obviously the analogy is far from perfect. That's why the actual post uses act- and rule-utilitarianism rather than Hare's critical and intuitive levels. I've stuck with stealing Hare's terms, though, because rule-libertarianism sounds like such a glaring oxymoron.

*** I am, by the way, particularly delighted by this analogy since (a) many of the Catallarchs are on board with Patri's vision of dynamic geography and (b) most of them also hate rule-utilitarianism. It's a nice analogy in another way, too: as with "utilitarianism" people use the word "libertarian" as if it has but one meaning, when in fact, it encompasses a huge array of possible meanings. Still, I think it's useful to talk about both in a general fashion, as long as we're aware that we are really talking about a set of theories with a strong family resemblance and not any particular concrete theory.

**** Again, for the philosophers out there, yes, I am arguing that second-level libertarianism might well turn out to be self-effacing.

Share this

So if I understand the

So if I understand the argument correctly, second-level libertarianism is about generating libertarian constitutions, and first-level libertarianism is about generating libertarian policies within constitutions that aren't necessarily libertarian?

It would strike me, then, that second-level libertarianism implies first-order libertarianism. So it seems odd to say that Patri isn't interested in first-level libertarianism... I'd say it's quite the opposite: He's so interested in first-level libertarianism that he's willing to impose libertarianism at a higher level in order to guarantee that first-level libertarianism will attain.

But this distinction doesn't say anything about whether dynamic geography is essentially libertarian or not. At first I read your post as implying that states generated through dynamic geographical projects would necessarily be done so out of a libertarian process, and I found this confusing. But if you're not trying to imply this, then how is dynamic geography libertarian at heart?

Not quite

So if I understand the argument correctly, second-level libertarianism is about generating libertarian constitutions, and first-level libertarianism is about generating libertarian policies within constitutions that aren't necessarily libertarian?

That's not quite what I meant to be arguing. It's not that dynamic geography automatically generates a libertarian constitution. It's that dynamic geography puts in place a structure that presupposes exit rather than a structure that presupposes voice.

At bottom, as I see it, libertarianism is about providing people with the freedom to arrange their lives in whatever way they see fit. One way of doing that is by minimizing the role of the state in the society in which I already live. That's what first-order libertarians are up to. But another way of giving me real autonomy is by setting up lots and lots and lots of different states and allowing me to choose between them -- or, failing that, to easily go start my own. That, I think, is second-order libertarianism.

Essentially, the strategy is to say, look, I don't care if almost all of you want to live in a social democracy. Those of us who don't will go and do something else. And if it turns out that other people want still other forms of government, well, they are welcome to those, too.

So, no, I don't think that Patri's second-order libertarianism implies anything like first-order libertarianism. Indeed, Patri's whole argument -- at least the one he advances in public most often -- is a purely pragmatic one. He thinks that a first-order libertarian society is the best type, but is at least in principle open to the possibility that it won't be. Indeed, even if it is not, a second-order libertarian world -- even if it's just a world full of competing non-libertarian governments -- is still better than what we have now.

In the world of competition, some things win (see: VHS, Blu-Ray) and some things don't (so long, Beta and HD DVD). Patri thinks that Libertopia will be the Blu-Ray of government competition. I'm inclined to think that it'll be something more like a Rawlsian political system that listens more carefully to Hayek wrt making the worst off person better off. The best way to solve the debate, though, is to try both and see which one wins out.

Whatever the answer we get, we'll be better off for having had the competition. After all, anyone really think it would have been better if we'd just abolished competition in hi-definition movie formats and simply picked one winner in advance?

Both/and

Not acquainted with the first level/second level distinction.

Here's my take-away from Patri's op-ed: Governments pose a big, fat libertarian problem by being oligopolies. If we could provide people with more options for how to be governed -- even if many of those options were not very libertarian in themselves -- we could promote choice, and thereby enhance autonomy. Even the existence of a fundamentalist Christian sea-stead would enrich me by 1) providing me with one more option than I currently have and 2) providing my fundamentalist Christian neighbor with one more option to leave me the hell alone. Thus, the fact that any given sea-stead is not a model of libertarianism isn't really a criticism of sea-steading as a road to libertarianism.

At the risk of sounding like a policy libertarian, this seems like a no-brainer to me, at least in theory. Doubtless there are legions of practical challenges.

I sense Will Wilkinson's real beef is with Patri's dismissal of policy advocacy. I find merit in advocacy, but I also find merit in Patri's conclusion that advocacy alone is unlikely to win electoral majorities. Politics entails paying off various constituency groups. Thus, while we can help elect politicians with greater respect for autonomy and defeat politicians with less regard for autonomy, I don't expect to ever elect a real libertarian candidate.

That said, I find little point in pitting advocacy against practice. They each play a role, and they each appeal to different people. Why not have both? More options for promoting libertarianism rather than less -- what could be more libertarian than that?

That said, I find little

That said, I find little point in pitting advocacy against practice. They each play a role, and they each appeal to different people. Why not have both? More options for promoting libertarianism rather than less -- what could be more libertarian than that?

Because we have limited resources (time, money) which we want to use efficiently - what could be more libertarian than that? :)

Fair enough.

Fair enough. I think the voice/exit explanation is much clearer.

Though one of the questions that came up at Patri's seasteading presentation at GMU yesterday was to what extent dynamic geography really would shift the locus of policy protest from voice to exit. Are the fixed costs of moving really that policy-related?

Voice/Exit

Though one of the questions that came up at Patri's seasteading presentation at GMU yesterday was to what extent dynamic geography really would shift the locus of policy protest from voice to exit. Are the fixed costs of moving really that policy-related?

That's an interesting point. As a practical matter, it seems likely that voice will still predominate. After all, we rarely leave voluntary groups (churches or clubs or so on) without a pretty lengthy period of first trying to convince others to do things our way. A lot of particular seasteads are likely to be governed by something like democracy on a day-to-day basis.

Importantly, though, those democracies will be (a) much smaller and (b) predicated on the understanding that people can leave easily enough should the rules of the majority prove too unpopular with the minority. The ease of exit should act as a constraint on what kinds of things people are able to do via voice.

"The threat is greater than

"The threat is greater than the execution" - we don't need to exercise exit to have it make our voices louder.

That said, there are substantial fixed costs to moving besides that of just the physical ones.

Good comments

Hadn't thought of it that way, but it makes sense.

I want second-level libertarianism both because I think it is the best (perhaps only) way to get first-level libertarianism, and because I think it is a good thing in and of itself to give more people their desired societies. But, as I think you are saying, the latter is much more important. That is, I think it is much more likely that I am wrong about my specific desired society than about the virtues of a market for societies. Maybe libertarian societies will lose out in the market - in which case, maybe they deserve to.

As you say - "Whatever the answer we get, we'll be better off for having had the competition."

Anyway, great post.

Language

One way to lower barriers between States, and increase competition between them, is a common language. I suggest that those who wish to weaken the power of the state settle on a particular language as the language of freedom and standardize on it. Start speaking it in your native countries, and teach it to your children.

Since English is already the language of commerce, and for personal reasons, I suggest English.

If one is willing to sever your ties with Terra Firma in the pursuit of a libertarian utopia then surely severing your ties to your native language isn't such a big deal.

The primary merit of seasteading..

...is that it is an attempt to produce private goods. If the profitable production of such private goods can be demonstrated then individuals have all the incentive they need to produce them.

Of course this says nothing about the feasibility of seasteading.

Movement libertarianism is futile because it seeks to produce public goods through collective politics. it's really not difficult to see why these public goods are always under-produced in collective politics - the incentives for the individual are not there.

I think it's great for libertarians to pursue projects which may enhance liberty through the production of private goods, like seasteading, but I also think it's counter-productive to ultimately promote such projects as libertarian. They should be promoted as profitable to the individual, because that's the only way they'll work.