Don't Give up on Liberalism Yet

This memo was generated for a nascent political theory group, on 19 Feb 2026. Discussion on Google Docs is omitted. It was written in response to Logan Graves’ “Speaking of Liberalism and Post-liberalism,” his “Speaking of Liberalism and Post-liberalism (Addendum),” and my “Desiderata for a Futurist Political Plan.” This memo concerns itself with Landian critiques of liberalism, much more than Thielian or other critiques. The latter must be covered in a future memo.

Don't Give up on Liberalism Yet

Nothing human makes it out of the near future.
— Nick Land, “Meltdown”

Upon reading Logan’s “Speaking of Liberalism and Post-liberalism,” and its corresponding Addendum, I realized that I did not yet find myself convinced that liberalism is incorrect. So let us attempt to formulate a description of a liberal future, and find where it might break.

First, what liberalism, and why liberalism. To start, I’ll take Logan’s definition: liberalism is “the commitment that individuals are rational and autonomous, and that they therefore should be given negative rights, i.e. freedoms, in order to use that autonomy effectively,” and, within that, a system which “does not make any claims about what ‘good’ is.”

I propose the following rough Hobbesian meta-ethical argument in favor of the latter feature of liberalism. I will call a value system externally positive if it would on net prefer more people which do not follow its values to exist.¹ To such a value system, conflicts are by default undesirable, but there are two reasons to engage in them: (a) enforcement of the value system upon others, and (b) self-defense. For (a): if the probability of victory in a conflict is high enough, the costs of engaging in one are outweighed by the gains from victory. And for (b): if there is a high enough probability of another system initiating a conflict, it’s useful to be the initiator instead. Thus (b) can only be justified by the existence of conflict in the first place. Under conditions of economic growth (usually involving population growth), this tradeoff is skewed strongly against conflict in the case of (a)—the gains from trade are too valuable, the default value of peace too high, for conflict to be at all desirable.

Thus, the value system’s holders enter into an agreement with others on a common legal system.² This avoids the need for self-defense, and neatly aligns with the interests of each individual value system. And the Schelling point for such a system is one which makes no claims about what ‘good’ is.³ There is no need for conflict when one can instead cooperate to secure new resources, land, or children, and conflict will not be expected under a fully neutral legal system. That is, liberalism depends on the possibility of positive-sum cooperation.

If we are to define a futurist liberalism, then, we must contend with the future potentially bringing us the opposite: a fixed pie (“the lightcone”) which must be divided between the interests of value systems. I will call this a finitist perspective, as opposed to an infinitist one.

Nick Land’s dark vision is that, as time spools into the future, value systems will be inexorably selected for reproductive fitness until they reach a perfect “technocapital singularity.” It will inevitably win any conflict, and find no benefit from trade, so it will perfectly tile the lightcone.

Even if we avoid such an outcome, the fact remains that a finite universe will settle upon a division of resources among various values. No longer will the incentives favor cooperation to ensure growth, and no longer will a state be able to remain impartial between interests. Instead, the final configuration of the universe—the division—will need to be decided. Any ‘liberal order’ will retroactively always-have-had the value system of the winning faction.⁴

It is possible that finitism is incorrect. I have heard some discussion from friends that hypercomputation, entropy reversal, and ‘faster-than-light’ travel are extremely promising. In this case, liberalism suffers no loss of desirability—liberalism’s fundamental argument, at least, still holds. Otherwise, we have a much more difficult challenge ahead of us, since any liberal order will have as its telos a particular, morally relevant, division of the lightcone’s resources.

This still leaves gaps. We must continue to articulate the constraints placed on liberalism by asking the following potentially-intractable questions:

1. What to do when minds are infinitely reproducible?
   1. Any system in which negative rights cost resources to provide is exploitable.
   2. Any system which relies on counting minds is exploitable.
2. What to do about Land’s vision, still? Wouldn’t we like humans to not be on the margins of society, even if they still exist and that society is still liberal?
3. How do we defend against existential risks without imposing on freedom or instituting tyranny?
   1. At the very least, either privacy or mind-shape diversity cannot exist. See Thiel and JD Pressman.
   2. Vitalik’s offense-defense distinction and “d/acc” also attempt to deal with this issue.

This may be aided by discussion of the, in my view, more tractable ones:

4. What to do when labor is automated?
   1. This has already substantially occurred. Is the resultant welfare state a correct extension of liberalism, and if not, why did it come about?
5. What to do when value systems are hypermemetic?
6. How do we maintain the courts (and, ideally, the legislature⁵), despite all of the above?
7. What to do about the most basic present critiques of liberalism—eg that it itself already contains a value system?

And of course: Is liberalism at all possible or reasonable in a finite world?

Further discussion on these issues is necessary. Past that, the next step might be to ask why exactly liberalism and then experiment with paths that are post-liberal but which serve the relevant goals. I began this project in “Desiderata.”

Footnotes

1. Externally negative value systems are exactly those that we might call ‘incompatible with liberalism,’ because they would prefer conflict over harmonious disagreement.
2. I will note that this is an argument from causal decision theory, which I’m doing for simplicity; Rawls makes a similar argument from a timeless decision theory, which I believe is nearly valid. Timeless arguments have much more unintuitive consequences and are unnecessary for my current purposes. I also expect that timeless decision theories are in fact unnecessary for usual notions of liberalism.
3. This essay’s argument is almost consistent with a system where some dominant value system tolerates others without granting them formal equality—a kind of benign hegemony. We may need the Rawlsian argument to fully defuse this.
4. My understanding is that this is not too distant from many modern critiques and failures of liberalism. Since the invention of the nuclear bomb, conflict has looked more and more pressing, and the world has looked more and more like a relatively finite pool of resources.
5. Legislatures seem highly unlikely to exist without injustice due to aforementioned issues with counting minds.