Planetary euthanasia and future suffering
[Content warning: discussion of suffering (without graphic examples) & astronomical timescales]
[Epistemic status: more of a discussion or research prompt than any kind of recommendation. The most significant uncertainties are the timescale of abolition of suffering relative that of extrasolar space colonisation and the likelihood that Earth-originated sentience is the only sentience that will have ever existed in the reachable part of the universe. Nevertheless, I’m still confident that focusing on “planetary euthanasia” wouldn’t help effective suffering reduction and could even lead to severe backlash.
Also, a much more elaborate and examined account of human extinction and the long-term future of suffering can be found in Magnus Vinding’s Anti-Natalism and the Future of Suffering: Why Negative Utilitarians Should Not Aim For Extinction.]
Even if planetary euthanasia (PE) were sociologically feasible, the risk that there might be extraterrestrial sentience within our theoretical reach or that it might evolve in the future (future which might support emergence of life for a hundred trillion years) suggests a much better usage of such civilisational alignment.
Dropping the alignment assumption weakens the case for staying around to prevent (worse) suffering, as now we need to account for the risk of the civilisation’s causing worse suffering compared to PE. Still, even if the risk isn’t small, it must be compared with the risk of attempting unaligned PE and failing, potentially severely compromising the cause against extreme suffering.
This post (tentatively) concludes that rather than focusing on PE, we should prioritise safeguarding the future from extreme suffering, anywhere in the reachable universe.
According to strict suffering-minimising views (SSVs), minimising suffering should be the foremost ethical priority or even the only ultimate priority.
When one first hears about such ethical views, one may imagine that the best possible outcome for these views is rendering the planet lifeless or otherwise completely and forever uninhabitable by sentient life. And some of those who endorse a SSV may themselves agree with such a putative implication of their views.
But before I go into discussing this intuitively repugnant implication, let me try to make it more, well, intuitive. For in fact it’s not only SSVs or even a much broader range of views that put some emphasis on suffering that may seem to imply this: if things on Earth are bad enough, or if there is a bad enough risk, and if rendering the planet empty were an option, many other ethical views would view this option on similar preventive grounds.
So, for example, a classical utilitarian may deem an empty Earth preferable to a populated one if the “total wellbeing” on the planet is “net negative”. Another example is a deontologist who expects such truly unacceptable “wrongs” on the planet that making it uninhabited (by non-violent means, for example: see below) may seem like the right thing to do.
Hopefully, this makes the question from the next section a bit more relating.
The project of ending sentient life on the planet (e.g. by preventing all births) and ensuring that it never evolves again is sociologically extremely implausible: getting every human aligned with such a project, even if it is a smooth transition through non-procreation (with an automated elimination of the biosphere once no one is around, for example), is, well, not a thing that a typical human will find appealing?
Anyway, for the sake of the argument let’s assume a post-human civilisation that does have such an alignment. The question is then: Would planetary euthanasia (PE) be the best use of such alignment and coordination?
I see several reasons why it would not be among the best options or possibly even a good one, at least not on its own. The following subsection outlines each of the reasons.
The first reason why we may need to be hesitant about PE is the potential existence of extraterrestrial life within our light-cone: if there’s extraterrestrial sentience or life that might give rise to sentience, and if this life may become reachable in the future (e.g. by self-replicating spacecrafts), ending all life on Earth (or ending it prematurely) could amount to allowing eons of Darwinian misery and torture on N other self-containing “worlds”.
Some will find such an objection too speculative or even “Pascalian”, as currently we have no indication of life beyond Earth. Moreover, there are reasons to think that life-bearing planets are indeed exceedingly rare, which accords with the lack of “pro-life” observations.
As someone who themself is wary of arguments from infinitesimal probabilities and enormous impacts, I can relate to such scepticism: afterall, we know that countless sentient beings have been suffering severely on this planet, possibly for hundreds of millions of years, and this will go on, unless some caring forces intervene.
Unfortunately, there doesn’t seem to be substantial grounds for completely discounting the possibility of extraterrestrial life. Therefore, if we take suffering seriously, dismissing the possibility may be ethically frivolous. For, again, if we prematurally chose PE, it would buy us peace at the cost of not preventing much worse suffering in expectation. (And let us also not forget that the operating assumption - i.e. the global alignment and coordination - implies that there will be much less suffering on that hypothetical Earth compared to the actual one. Similarly, the assumption dissolves the risk of unaligned space colonisation.)
Life evolving again
The next consideration for sticking around is that there’s some chance that sentient life will evolve again.
This ties back to the previous section in that this life might be extraterrestrial. So even if in the future we become certain that there’s no extraterrestrial life within our theoretical reach, we would still need to account for the risk of it evolving further into the future.
This argument against premature PE may appear even more speculative, as now we’re considering hypothetical extraterrestrial life that is yet to exist (or not). One may even see this as motivated reasoning to avoid the repugnant implication. Yet, while I’m open to counter arguments and considerations, the numbers below seem to support taking the risk seriously.
For we appear to live relatively early in the expected life-supporting lifespan of the universe: the universe is estimated to be about 13.8 billion years (with Earth and life on Earth being ~4.5 and ~3.5 billion years respectively). Whereas the life-supporting period has been estimated to last at least for another 1-100 trillion years (i.e. 1,000 - 100,000 billion years), which is the time our models suggest when star formation will cease (and thus no stars will remain (relatively) shortly afterwards). Moreover, it’s possible that peak habitability of the universe still lies ahead (although I couldn’t find an estimate to compare this to the possibility that the peak habitability is already behind us).
(For a more detailed review of the possibility of extraterrestrial life and the future of suffering more generally, see Vinding’s Anti-Natalism and the Future of Suffering, especially section “Earth Aside”.)
Even if these staggering numbers overestimated the suffering-risk period of the universe by orders of magnitude, they would still suggest that this point in the history of the universe might be just the beginning of the emergence of suffering. The worst possible tortures may still lie ahead. But they might be preventable by an Earth-originated civilisation.
[Speculative!] Unaligned alien civilisations
Much more _im_probable than discovering extraterrestrial wild life is finding evidence of a technologically advanced civilisation that is still reckless about suffering: e.g. some of groups within that civilisation spread life to other planets without safeguards against emergence of suffering.
If there’s anything an Earth-originated civilisation can do to prevent that suffering (without causing worse suffering), then this would count as another argument against premature PE.
But lest imaging extraterrestrial advocacy or “tidying up” after aliens makes one’s head spin too much, let’s leave this topic for another time.
Without the alignment
So far we held the assumption that PE had no opposition. In practice, we would need to achieve a global acceptance if, despite the counter-considerations from the previous sections, we saw PE as the best possible option of all.
One can argue that PE will be eventually globally supported if human descendants take the unacceptability of extreme suffering to its logical implications.
Perhaps so. Again, I just doubt that premature PE is a good solution from a suffering-minimising perspective in the first place.
One might also think that PE doesn’t require global acceptance, as non-cooperative approaches may be feasible: for example, some future technologies may enable relatively few dedicated agents to launch PE unilaterally.
In my view and that of others (see e.g. chapter 10 “We Should Be Cooperative” in Magnus Vinding’s Suffering-Focused Ethics: Defense and Implications), suffering reducers should avoid attempts at unilateral interventions, as, briefly, such attempts risk creating conflicts and harming the cause in the long run. On the other hand, there is much to be gained from compromises in place of conflicts and other forms of cooperation (even just between longtermist groups).
Risks of staying around
Dropping the alignment assumption cuts both ways, as now we cannot simply assume global cooperation on guarding the long-term future from suffering: we now have to account for the risk that the civilisation on the whole will cause more extreme suffering than it prevents.
One way this might happen is through spreading wild-animal suffering to other planets. Examples of this include creating Darwinian wild habitats as part of space colonisation or simply trying directed panspermia to initiate an extraterrestrial evolution.
Another suffering risk is torture optimised for suffering using future technologies. This can come about if there are still sufficient malevolence and not enough protection in the future (e.g. if invincible wellbeing has not yet been achieved). At worst perhaps, in the future there might be a stable dictatorship that uses torture to subdue and punish its enemies.
Even if suffering remained approximately at today’s levels, this would still be an emergency, as countless sentient beings would suffer severely as they are right now. However, I don’t think it’s realistic to expect suffering to remain at this level indefinitely.
Can anything be done to mitigate these risks? As I wrote before, I think there is a strategy that is able to prevent a range of suffering risks, namely creating a future robust against the risks. And I see replacing the biology of suffering with information-sensitive gradients of wellbeing for all sentience as the most promising route to such a future. For, once invincible wellbeing and caring for others are genetically preprogrammed, harming any post-Darwinian would be practically impossible (and, one hopes, simply inconceivable).
Granted, the abolition of suffering, though technically feasible, may itself not appear sociologically likely. Nevertheless I’m cautiously optimistic. A detailed account is better left for a dedicated post, but in short I think that as somatic gene therapies, preimplantation genetic screening, and, eventually, germline genetic editing become more accessible and accepted to treat recognised human genetic diseases, these technologies will be seen as more and more mundane. This will lead to recognition and demand for broad preventive genetic interventions like being more resistant to many forms of cancer and mentally resilent. Gradually this will percolate to unacceptance of unexamined genetic lottery.
Tentatively, animal exploitation will be abolished too. There are several factors that are likely to be crucial for this: growing awareness about the harms of animal farming, animal suffering in particular; , availability of cheap alternatives to animal “products”; reduced subsidies for and possibly tax on those “products”, alternatives to animal experimentation, and, with time, ban on animal farming, animal experimentation, and other forms of animal exploitation.
I said “cautiously optimistic” that suffering will be abolished because avoiding civilisational collapse is not a given (and an advanced civilisation is our only chance to abolish suffering), and because overcoming currently widespread ideological obstacles and working with our deeply-rooted speciesist biases and intuitions may delay abolition until it’s too late (until we start spreading Darwinian misery to other planets, for example). Ideological disagreements may be especially enduring for the problem of suffering of non-human animals in wild nature - a tremendous problem on its own, although technologies like gene drives and immunocontraception suggest it is tractable.
In short, we shouldn’t rely on the abolition of suffering in all sentience to happen by default. Even if one thinks that suffering will be abolished as part of general civilisational development, there is still a risk of severe suffering persisting - due to, for example, one’s model being wrong, our civilisation’s getting trapped in a different trajectory, or, again, civilisational collapse. And this risk, I submit, even if it were tiny, should be actively mitigated, given what is at stake.
For someone with a strict suffering-minimising view (SSV), a natural question to the idea of at least postponing PE may be: Does this imply natalism?
The good news is that it doesn’t. Specifically, aspiring suffering minimisers and other effective altruists can use their time, money, and other resources much more effectively than trying to raise decent human beings (although this may be different for those aspiring EAs whose personal wellbeing requires having children).
One may see this as a bit dodgy answer though, as currently human procreation is the only way to persist the civilisation past one generation.
So a more straightforward answer is that until rejuvenation technology is mature enough, some level of procreation is necessary to continue the civilisation. Still, since humanity is above this level anyway, the question of natalism can be suspended. Meanwhile, at absolute minimum, we should establish strong right to die as well as wellbeing guarantees. And antinatalists (and others) may want to invest into rejuvenation and potentially other technologies that would make natalism redundant.
PE can seem appealing as a solution to unacceptable suffering on the planet. Unfortunately, proper PE is sociologically implausible. More importantly, even if it were globally supported, the possibility of extraterrestrial worlds with sentient life or evolution of such worlds in the reachable universe suggests much better applications of the alignment, such as persisting as a civilisation to abolish and prevent suffering.
With no alignment guarantees, there is a risk that a future civilisation might even cause worse suffering than it prevents. It’s not clear how big the risk is, but even if it’s likely, there appears to be better ways to mitigate it compared to attempting unilateral PE with a risk of severe backlash against suffering reduction as a cause. (One such mitigating strategy mentioned in the text is replacing the biology of suffering and malevolence with genetically preprogrammed invincible wellbeing before we are capable of spreading Darwinian life outside Earth.)
The tentative proposal of the post is that instead of focusing on PE, the civilisation should ensure that at the very least the future is free from extreme suffering.
I write “possible” here, as an ideal outcome for SSVs (and possibly other views) is an empty universe. ↩︎
David Pearce makes a similar argument in his “The Pinprick Argument”. ↩︎
The technical feasibility of PE isn’t discussed in the post, as it’s simply assumed that for a committed civilisation initiating something like extreme global warming (and thus turning the plant into a Venus) wouldn’t be unresolvable. ↩︎
For example, this paper finds “that unless habitability around low mass stars is suppressed, life is most likely to exist near 0.1 solar-mass stars ten trillion years from now”, while this one finds that:
The Solar System formed at the median age for existing giant planets in the Milky Way, and consistent with past estimates, formed after 80% of Earth-like planets. However, if existing gas within virialised dark matter haloes continues to collapse and form stars and planets, the Universe will form over 10 times more planets than currently exist. We show that this would imply at least a 92% chance that we are not the only civilisation the Universe will ever have, independent of arguments involving the Drake Equation.
Even when it comes to maximising or preserving things of purported intrinsic value like “natural” landscapes or bliss, instead of refusing to collaborate on such efforts, let alone opposing them, suffering reducers could even choose to do their best in supporting others in pursuing those values, with one strict condition. The condition would be that creating and sustaining those things of purported intrinsic value happen in place of suffering (as opposed to at the cost of suffering). The idea (admittedly abstract and perhaps futuristic) is that by engaging in such cooperation, suffering reducers try to make sure that matter and energy comprise anything other than things that cause suffering. (Cf. the “indirect” approach to negative utilitarianism suggested by David Pearce.) ↩︎
As philosopher Ole Martin Moen illustrates this:
If we imagined that from now on, animals started emitting a red light every time they suffered, then from space, Earth would no longer be a blue planet, but a red and glowing one.
Complementary or interim approaches could include reducing the amount of wild habitats and using non-lethal methods of fertility regulation for wild animals. The former, however, will likely be too controversial for the time being, and we should try to stay focused on the main approach (genetically preprogrammed well-being, that is) anyway. ↩︎
In no particular order, as various plausible scenarios emphasise different factors. ↩︎
A reason for optimism is that solving the technical challenges of extrasolar space colonisation would require huge technical progress and political will. David Pearce, for example, thinks that while “in the face of daunting technical obstacles, humans will most likely establish small self-sustaining colonies on the Moon and Mars later this century”, “the challenges of colonising alien solar systems and creating pain-ridden ecosystems light-years away are orders of magnitude more formidable than building lunar settlements, or even terraforming Mars.” In contrast, “compared to the challenges of reaching the stars, getting rid of suffering on Earth is technically trivial”. ↩︎
One caveat, though, suggested by my friend, is that there might be a major genetic component that contributes to one’s accepting a SSV. And if persons with SSVs didn’t have biological children, such a component would be lost, resulting in even less prevalence of SSVs in the population. I’m sceptical about this for many reasons, not least since I think in most cases it’s probably most effective to put one’s resources and best effort into promoting SSVs broadly, as opposed to trying hard to convince one’s child to become an effective suffering reducer (while risking alienating them to the issue altogether). Nevertheless, a deeper look into this consideration is required to justify any certainty. ↩︎
Under open or closed individualism, some may argue that continuing consciousness isn’t fundamentally different from starting a new one, and therefore if we could adjust for practical differences, we should be indifferent between creating a mind and continuing an existing one.
While I do find empty individualism the most accurate as conceptualisation of the nature of personal identity (and in some sense open individualism as well), I don’t think we are anywhere near resolving practical differences between starting a human life and continuing an existing one. Things like dying, causing grief to others, and losing one’s positive influence via one’s insights, knowledge, skills, and connections, and, in the case of a new life, the time and other resources it costs to become an agent of change (at best; and at worst one causes worse suffering than one prevents). Not least, there is a sense of imposition of creating a new sentient being. And even if, fundamentally, enduring personal identities are a myth created by memory (i.e. you now is distinct from your namesake even from a moment ago, despite the strong sense of continuity coming from memory), one a practical level it’s still useful and indeed natural to most humans to believe that one wakes up every morning the same person and that one is not them. ↩︎