Intro

This will be something of a long and theoretical post, but I'm interested in others' opinions on this - and a quick google search of Robert Nozick and Fediverse turned up literally nothing, so I'm thinking that this is a new connection. The recent news about Beehaw defederating from other instances, and the wider discussions about how federation might or might not work reminded me of Robert Nozick's Anarchy, State, and Utopia (1974), which I imagine anyone who's formally studied philosophy will have come across. The main point of this book is to make the case for the libertarian minimal state, with the overall thesis in the preface being:

Our main conclusions about the state are that a minimal state, limited to the narrow functions of protection against force, theft, fraud, enforcement of contracts, and so on, is justified; that any more extensive state will violate persons' rights not to be forced to do certain things, and is unjustified; and that the minimal state is inspiring as well as right. (p. ix)

The book, while influential and important, is I think deeply flawed, and there's some general agreement about this in the philosophy departments I've been involved with. (Same with many of Nozick’s general opinions.) Unfortunately, the parts of the book that tend to be taught are the first two, and in particular the Wilt Chamberlain argument (pp. 160-2) in which he argues that unequal distribution of wealth and goods is fine as long as the unequal distributions were caused by a history of mutual freely consenting exchanges.

I say 'unfortunately' because– while the first two sections and the Chamberlain argument are definitely important and influential– Part III, Utopia, is the strongest. I'm not a libertarian, but it's a novel, well-structured, and interesting argument for the minimal state, based in part on possible-world semantics, and I think it looks a lot like what the Fediverse is going for, which is why I'm interested in the crossover.

The Possible-World Model

Nozick begins by defining Utopia and identifying its main issues:

The totality of conditions we would wish to impose on societies which are (preeminently) to qualify as utopias, taken jointly, are inconsistent... The best of all possible worlds for me will not be that for you. The world, of all those I can imagine, which I would most prefer to live in, will not be precisely the one you would choose. Utopia, though, must be, in some restricted sense, the best for all of us; the best world imaginable, for each of us. In what sense can this be? (pp. 297-8)

He then sets up this perhaps rather convoluted idea, based on the concept of imagining possible worlds. The core idea is this: that in any possible world you can imagine, it must include that all other rational agents in that world will also be able to imagine other possible worlds, and that (if they prefer) they can then move to those possible worlds.

The question then moves to: is it possible for this to be stable? Because Nozick is interested in whether utopia as traditionally explored by utopian theorists and authors (and note that to an extent he’s subtextually talking to socialist utopians throughout) is possible, the key question is whether worlds will keep being created over and over, with people moving over and over, or whether there'll ever be a world where everyone in that world chooses to stay. And stable worlds must then:

[satisfy] one very desirable description... namely, none of the inhabitants of the world can imagine an alternative world they would rather live in, which (they believe) would continue to exist if all of its rational inhabitants had the same rights of imagining and emigrating. (p. 299, his emphasis)

This is, given that people are able to move to worlds they imagine (which Nozick calls 'associations' - as opposed to 'east-berlins' in which inhabitants are unable to move to other worlds).

He puts this also in set theory terms (quoted just below), and then points out an equivalency of members of S choosing to form an association of their own, vs. members of S refusing entry to those members of A who are not also members of S.

if A is a set of persons in a stable association then there is no proper subset S of A [note from me: 'proper subset' means it's a part of the whole but not equal to it. So {1, 3} is a proper subset of {1, 3, 5}, but {1, 3, 5}, although a subset of itself, is not a proper subset of itself] such that each member of S is better off in an association consisting only of members of S, than he is in A. For, if there were such a subset S, its members would secede from A, establishing their own association. (p. 300)

There is then a fairly lengthy section expanding on this, caveating it, and also doing some more in-depth logic/set theory, which I'll skip over as it's not as relevant (and this is already getting long). It's pages 301-6 if anyone's interested in reading, though. Page 307 onwards is where Nozick begins analysing how this model laid out above could be seen in the real world.

The Real World

Obviously, the above possible-worlds model is very idealised, and there are several limitations in the real world. Nozick lays out the following four:

1. In the model, we can imagine infinite possible people to associate with (although we cannot have an infinite number of people in an association); in the real world there are firstly not infinite people and secondly we can't create them. So even if I can imagine the perfect association for me, it might not exist; same with a community I might want to join.
2. In the model, the only ways associations interfere with each other is by drawing away its members - in the real world, communities impinge on each other in all kinds of ways.
3. Information costs - it takes effort to find out about other communities in the real world; in the model it's instantaneous and easy.
4. In the real world, some communities don't let their members know about, or move to, other communities.

It’s worth noting here that Nozick was writing in 1974, before the advent of the internet (and to a lesser extent, globalisation in general), so point 3 is less of an issue here. Particularly regarding moving and travel costs, which are vastly, vastly, reduced online. In fact, I think these issues are all reduced on the internet, which is relevant when it comes to the potential for implementation. I say more about this at the end of this post, and it’s one of the main things I’m interested in hearing opinions about.

Nozick, now, is interested in the implementation (or influence) of the possible-world model in the real world, and his key point ends up being this:

The idea that there is one best composite answer to all of these questions, one best society for everyone to live in, seems to me to be an incredible one. (And the idea that, if there is one, we now know enough to describe it is even more incredible.) (p. 311, his emphasis)

The ‘questions’ he refers to are questions of values, of activities, of interests. Security or adventure? Luxury or austerity? Private property? Religion? The fact, Nozick thinks, that utopian authors attempt to imagine a utopian society demonstrates a blindness to the heterogeneity of human nature. Which is demonstrated by the fact that they all have their own visions of utopia, and the fact that the inhabitants of their visions all lead different lives.

The conclusion Nozick draws is that there is no sense in having one type of community in a utopia - rather, that “Utopia is a framework for utopias” (p. 312, my emphasis because it’s the most important point here). We should be aiming for a kind of “meta-utopia”, and this is where the real-world limitations flagged above come into play. The meta-utopia is necessary precisely because of these real-world limitations. What does this look like?

[T]he environment in which utopian experiments may be tried out; the environment in which people are free to do their own thing; the environment which must, to a great extent, be realized first if more particular utopian visions are to be realised stably. (p.312)

Nozick thinks this conclusion can arise from a few forms of arguments. One is, that people are different, and so thinking there’s any ‘best’ world for everyone is foolish. That’s what’s happening when he states that there’s no composite answer to the questions of how best to live/structure society. But what if there is a society that’s the best society for everyone? Nozick reckons that that still leads to this meta-utopia. His reasoning for this involves what he calls “design devices” and “filter devices”.

Design devices basically attempt to structure an ideal society from the ground up, with a bunch of people discussing what the best society is, constructing a model for this society, and then implementing it.

Nozick basically thinks that this is a non-starter. I think this analogy puts his ideas well:

It is helpful to imagine cavemen sitting together to think up what, for all time, will be the best possible society and then setting out to institute it. Do none of the reasons that make you smile at this apply to us? (pp. 313-4)

EDIT: I want to note that I mean that this analogy puts his ideas well. I share the scepticism of design devices, while simultaneously thinking that many societies denoted as 'primitive' tap into important and valuable aspects of human communities which 'modern' societies dismiss.

The complexity of the human condition is also a large part of his reasoning here why design devices don’t work. So, the alternative is filter devices, which “involve a process which eliminates (filters out) many from a large set of alternatives” (p. 314). This is desirable for a few reasons:

1. It requires less knowledge than design devices. Filtering processes don’t need to know precisely what an end-product looks like; they can just have some ideas about what they don’t want and begin with that.
2. The filtering process naturally improves with time. When you have a filter for new candidates, then those candidates are, on average, of better quality (however that’s defined in this particular community), so the filtering process now has better material to work with.
3. New material creates novel ideas, which would not be accessible with a design process (Nozick doesn’t outright state this, but I think it’s clear that he thinks it).

Moreover, one single filtering process will be insufficient. Nozick describes it thus:

[P]eople try out living in various communities, and they leave or slightly modify the ones they don’t like… Some communities will be abandoned, others will struggle along, others will split, others will flourish, gain members, and be duplicated elsewhere. Each community must win and hold the voluntary adherence of its members. No pattern is imposed on everyone, and the result will be one pattern if and only if everyone voluntarily chooses to live in accordance with that pattern of community. (p. 316, his emphasis)

Some advantages he lists to this: given that the filtering process is largely constituted by people leaving communities they don’t like, this will cause communities which people want to live in; mechanical processes are limited “given our inability to formulate explicitly principles which adequately handle, in advance, all of the complex, multifarious situations which arise” (p. 317) - this one is very, very similar to many recent discussions I’ve seen about moderation and the ‘don’t be an arsehole’ clause.

So what does all this lead to? Basically, that the utopian framework should be one that is informational. Whichever framework provides the best means for finding out about various communities, is the one that should be adopted– first, because that is what best facilitates the filtering process, and second, because it best mitigates the real-world issues laid out at the beginning of this section.

Conclusion

Ok, so, that’s the bulk of what I wanted to put down. The rest of Utopia focuses very much on the physical world - it’s well worth a read. (NB I’m using ‘physical world’ rather than ‘real world’ or ‘actual world’ (which Nozick uses) because the internet is part of the real world. As opposed to the ideal world, which one example of is the possible-worlds model laid out above.)

I was initially going to offer my own thoughts about how this connects to the Fediverse and the internet in general, but just the sharing of Nozick’s framework has gotten long enough that I think I’ll leave it there. Part of my motivation for sharing this is that, although I’m very much not a libertarian, this is imo one of the strongest defences of the minimal state; in addition to this, I’m much more sympathetic for a kind of meta-libertarianism when it comes to spaces like the internet, especially if they function to facilitate filtering processes.

That said, I still have worries about the way this can be, and is, implemented. There’s been a lot of discussion on Beehaw defederating from lemmy.world and sh.itjust.works, and although I personally don’t see a problem with it, I can understand why people are annoyed. I wonder if this is a consequence of people thinking they’re existing in the meta-utopia, when in fact they’re existing in an instance of it. I also don’t want to label the Fediverse as the meta-utopia, although I do get the sense that that’s what they’re aiming to become.

Nozick has a relevant section on the Beehaw thing, actually, and I’m just going to quote it because I’ve just about reached my energy limit for explaining/elucidating philosophy lol. Feel free to skip it, it’s a side-note to this post and not a main point.

A person will swallow the imperfections of a package P (which may be a protective arrangement, a consumer good, a community) that is desirable on the whole rather than purchase a different package (a completely different package, or P with some changes), when no more desirable attainable different package is worth to him its greater costs over P, including the costs of inducing enough others to participate in making the alternative package. One assumes that the cost calculation for nations is such as to permit internal opting out. But this is not the whole story for two reasons. First, it may be feasible in individual communities also to arrange internal opting out at little administrative cost… yet this needn’t always be done. Second, nations differ from other packages in that the individual himself isn’t to bear the administrative costs of opting out of some otherwise compulsory provision. The other people must pay for finely designing their compulsory arrangements so that they don’t apply to those who wish to opt out. (p. 321-2)

Another reason why I’m interested in opening up this discussion, is that I’ve experienced almost no discussion on this section of Nozick’s work in my experience of academic philosophy. The other two sections– and particularly Wilt Chamberlain– are talked to death, but Utopia has relatively little engagement. On one hand, I get this - a large part of philosophical education is understanding the history of ideas, and Utopia is comparatively uninfluential. You need to know Wilt Chamberlain if you’re entering academic philosophy; you don’t need to know all this. On the other hand, it’s a shame, because I think it’s the strongest part of Nozick’s work.

I also think that it’s somewhat more relevant to the internet than it is to the physical world. Not because of the legitimacy of its ideals, but purely because of the relative ease of implementation. The four issues mentioned above are really reduced in online spaces.

1. We still don’t have infinite people, but the variety of people we can interact with is potentially wider. Potentially. The issues with lack of– or exodus of– minorities, which I’ve seen discussions about on Tildes while searching past posts, is an important one here. I’m not necessarily referring specifically to Tildes here - I’m too new to the site to really have a good sense of the community. But just like in the real-world, I can’t conjure up people and create my own version of Tildes which includes all the people here currently and also all the other people I’d like to see.
2. Communities on the internet obviously interfere with each other, just like physical-world communities. This isn’t that reduced, perhaps only in terms of stakes. Physical-world community interference can cause wars, financial boycotts, etc. Maybe doxxing or the like is analogous? Regardless, it’s reduced although not eliminated in severity, as far as I can see.
3. The difference in information costs should be immediately obvious. It’s much easier for someone to try out Tildes, than it is for someone to try out France.
4. Relatedly, internet communities don’t have the same power as physical communities to limit information, although there are definitely still issues here, especially with an increasingly-corporatised internet. On the other hand, the internet itself does work in breaking down these barriers in the physical world, at least in terms of information (not in ease of moving countries). At least, there’s usually no real financial costs to hopping platforms.

Guess I’m leaving it here? Maybe all I’ve accomplished is sharing some cool philosophy– at least, I think it’s cool.

The page numbers all reference Anarchy, State, and Utopia - I don’t know if I’m allowed to link PDFs here, but suffice to say it’s the first one that shows up.

This Stanford Encyclopaedia of Philosophy page also includes some useful context, and a bit of discussion on the Utopia section - although, again, relatively brief. Nevertheless a great source.