Immortality and reproduction are incompatible in the long term future

I’ve heard visions of long-term positive futures of humanity incorporate either:

1. people living forever (until they no longer want tomorrow);

2. people having as many children as they want to.

The first motto is from Bryan Johnson’s website; the second is from the largest pronatalist conference in the US. My thinking is that the pronatalists need to step up their slogan game. The Don’t Die movement’s motto hits hard.

In this post, I want to point out that having both at once for all people forever is not possible, if the laws of physics stay the same.

This is extremely irrelevant to the short-term future, where if we manage to solve all problems, we can still solve death and let every couple have many kids if they want to; but at some point, after a few hundred years, the laws of physics will catch up.

This point has been alluded to many times before, but I did not find a link to a concise write-up that I can link to in arguments about positive futures, so here it is.

---

If humanity is bound by the speed of light, we have the following facts:

1. The total resources available to humanity grow less than t³ with time, because the volume of the universe under our control grows as ct³ with time.¹

2. Women having > 1 kids on average rows the population exponentially with time.²

Hence, in the long term, if the population grows, the average amount of resources available for any given person decays exponentially with time.³ This is remarkably stable to whatever is meant by a “person” or “resources.”

We can do with both for a couple of generations; but whatever your vision of the really long-term future is, it has to pick one. In particular, if there is an AI singularity, and we decide to lock in the value system of humanity to make it go well, the value system should better account for this problem, otherwise it will become incoherent.

I believe the fertility crisis to be a non-concern for this argument; as eternal youth / longevity advances raise TFR, for the simple reason that the average number of children born to a woman in her lifetime will increase if her reproductive lifespan is longer. Mothers who raised kids and enjoyed the experience would likely do it again once the kids are grown up, if they are healthy enough.

---

Uploading people and running all human experiences does not work. If what we care about in people is a set of experiences, this spends resources, and we want something similar to “descendants” to experience a similar set of experiences, the same resource constraints apply.

I think what might work for a while is: bite the bullet, and post-singularity, reduce the resource required for a person while maintaining the exact same experiences. The amount of resources (volume, entropy, whatever) required for a person is lower bounded by some constant. There are about 120 halvings between the average human height and the Planck length. We could go on for many generations.

The only thing that works without falling into a Malthusian catastrophe is either (1) convince a sizable fraction of people to not want immortality; or (2) let TFR fall very close to zero. I see a plausible combination of the two: people who have kids might be less attached to immortality than those who don’t.

1. There are entropy-based bounds that push this factor to t² or perhaps lower, but it’s not important for the sake of this argument.

2. Thanks Dynomight for fixing my math!

3. The same argument also works now; so you might wonder: what is currently enabling people’s quality of life to grow somewhat exponentially? The answer is “our ability to extract the resources available to us and convert them into improvements in quality of life”. For various reasons, I expect this growth to become slower in the next hundred years or so.