Apparently, an increasing number of people say that they believe that climate change will cause human extinction.1 To what extent all of these people are genuinely convinced about this I don’t know, although, for reasons explained elsewhere,2 I expect that for most of them it is a desperate attempt to find meaning and make sense of an increasingly meaningless and senseless world more than a genuine conviction. Regardless, the topic of human extinction is an interesting one. How likely is human extinction really? And would human extinction actually be a bad thing? These are the two questions I will focus on – and try to answer – in this article.
Will mankind go extinct?
The answer to the question whether humanity will go extinct is “yes, of course”, but that answer is obvious only because the question is too vague. The real question is (or should be) when humanity most likely will go extinct. So, to give a more useful answer, let’s distinguish some timescales. For the purpose of this discussion, I’ll define the short term as within roughly a century; the middle term as within (roughly!) 10,000 years (but not within a century); and the long term as the period after that. It is important to note that while relevant circumstances in the short and middle term are mainly determined by climate change, this is not the case in the long term.
If mankind survives the short and middle term, we will surely at some point go extinct in the long term. This extinction can have several causes. Even if we master space travel,3 we’ll go extinct when the universe “dies” (in the distant, distant future). If we don’t master space travel, we’ll go extinct when Earth becomes uninhabitable because the Sun becomes too hot (in the slightly less distant future, but still about a billion years away). But much more likely, we’ll go extinct long before that due to evolution (or something else). On the time scale of (probably many) 100,000s of years, we’ll gradually evolve into something else. At some point our descendants will have become a different species. If we’d somehow preserve a specimen of our species, then that specimen will (at that point) no longer be able to interbreed with our distant descendants. That new and different species might still be “human” in some sense, however,4 but it will evolve further, and as some point it will be as different from us as chickens are from tyrannosaurs (which both belong to the same clade of dinosauria). So, even if nothing kills us on the way, just because we’ll gradually evolve into something else, mankind will inevitably go extinct in the long term.
There are plenty of things that can kill us on the way, however. A large meteorite could kill most or even all life on Earth, for example, and while the likelihood of a meteorite impact of sufficient size is vanishingly small on short time scales, it becomes a near certainty if the time scale is long enough. An even more likely cause of mass extinction – including human(-oid) extinction – in the long term doesn’t come from outer space, but from inside the planet, however. Most major mass extinction events in the Earth’s history were caused by immense volcanoes covering whole landmasses with flood basalt and raising atmospheric CO₂ to levels high enough to change the oceans into acid baths and to raise the global temperature to very inhospitable levels. Some microbial life will probably survive that, but we or our descendants won’t. This has happened several times in the past, and there is no reason why it wouldn’t happen again in the future. It almost certainly will.
There are other possible causes of human(-oid) extinction in the long term, but it’s not particularly interesting to try and discuss – or even list – all of those. The important point to realize is just that in the long term, human extinction is a certainty. If we’re “lucky”, we’ll evolve into something else; if we’re “unlucky”, we’ll go extinct before that due to some kind of planetary disaster; but even if we’re “lucky” we’ll eventually go extinct when the Sun or the universe dies. Hence, the “yes, of course” answer above. Now, let’s turn to the more interesting question: How likely is human extinction on the short or middle term?
The “doomist” belief mentioned in the first paragraph of this article is that mankind will go extinct within a century or so from now due to the effects of climate change. There is considerable variation in the expected time scale, however. Some people believe that humanity will be extinct by 2040 or even 2030,5 while others believe that it might take slightly longer than a century. There are, moreover, scenarios of short-term human extinction that do not depend on climate change – such as the idea that artificial intelligence will (accidentally or intentionally) exterminate us – but I’ll ignore such scenarios here to focus on the direct and indirect effects of climate change.
Climate change won’t kill humanity directly on the short term. Even if continued CO₂ emissions together with various feedback effects and passed tipping points lead to a realization of the most hellish climate change scenario that is geophysically possible – namely, to some kind of uninhabitable “wet greenhouse” – then it will take at least several centuries to get there.6 In that case, humanity would indeed go extinct, but it wouldn’t be in the short term. Given physical limitations, climate change just cannot be fast and extreme enough to kill all of us directly. (But it can and will kill many of us.)
The real threat are the indirect effects. Climate change will lead to increasing conflict over increasingly scarce resources such as food and water, and some (perhaps, many) of these conflicts will turn violent. War itself won’t lead to human extinction, but nuclear war might bring us close. Even a full scale nuclear war isn’t likely to lead to human extinction, however. It would lead to a nuclear winter lasting for a few years, killing most plants and animals, and Ice Age-like conditions for a quarter century after that, killing many of the survivors,7 but it wouldn’t kill everything. Some humans and other animals would survive – maybe 1% of the global population, maybe slightly less, maybe more – and those survivors would eventually repopulate the planet. They might actually have better survival changes on the middle term because, although their number would be small, nuclear war would terminate most CO₂ emissions and re-freeze the permafrost, and thereby limit global warming.
Another cause of extinction has been suggested by Guy McPherson. In conditions of societal collapse – which will eventually envelop the planet – nuclear power plants won’t be maintained, and unmaintained nuclear plants will sooner or later melt down or cause another kind of nuclear disaster (unless they were shut down in the right way, but that takes a long time – probably too long). Many modern nuclear reactors are designed to contain a meltdown in the reactor vat, but that won’t work in all cases and there are many older, less safe reactors as well. Hence, McPherson is right that global societal collapse will lead to nuclear disasters. There are about 400 active nuclear reactors spread around the planet (albeit mostly in the northern hemisphere). Not all of those will meltdown, but some – and perhaps even many – of them will. This will cause radioactive clouds and fallout covering much of the planet (but not evenly). People (very!) close to nuclear disasters will die almost immediately due to radiation exposure, but their number will be tiny. For the vast majority of people, the only effects are those caused by nuclear fallout and a small increase in background radiation. These will significantly increase cancer rates, but won’t kill anyone directly. Life expectancy will drop (due to cancer) and there will be many more birth defects and other health problems, but it is very unlikely that this will lead to human extinction.8 Nevertheless, while McPherson’s scenario is unlikely to cause human extinction, it is a scenario that deserves serious consideration anyway because its basic premise is almost certainly true: in conditions of societal collapse nuclear power plants cannot and will not be maintained and that will almost inevitably lead to nuclear disaster. Hence, if global societal collapse is inevitable – and I think it is9 – then global nuclear disaster is inevitable.
If neither nuclear war nor global nuclear disaster are likely to completely exterminate mankind, does that mean that human extinction on the short term is unlikely? I’m not sure, but I think it is (very) unlikely indeed. Except, of course, if there is some other indirect effect of climate change that could do us in. I have no idea what that could be, however, so my provisional answer to the question whether short-term human extinction is likely is “no”. In the contrary, I’m quite convinced that mankind will not go extinct within a century (or even two or three centuries) from now.
I’m less confident about the middle term, however. Or less confident than I used to be, at least. In optimistic but still at least somewhat plausible scenarios we’ll manage to limit global warming to a few degrees. That will still make parts of the planet uninhabitable and kill a significant part of ocean life, but it leaves enough space and enough resources for at least several billions of humans. I don’t really believe in such scenarios, however, for reasons explained so often in this blog already that I’ll be very brief here: we won’t significantly limit CO₂ emissions (or at least not until it is too late and/or circumstances limit them for us), and because of that climate change will not be limited to just a few degrees. Most likely, global warming will be in the 6°C to 10°C range, and that will leave only a small part of the planet inhabitable and will almost completely kill the oceans.
The 10°C scenario explained in Stages of the Anthropocene leaves enough space for roughly (!) 100 million people, which seems more than enough to secure longer-term survival chances, but there are a few problems. (In case of a 6°C scenario, the number would be a bit higher; up to a billion, perhaps.) One problem is that the conditions in such a world are very harsh, and if they would become only a little bit harsher due to some kind of natural disaster, for example, that may very well be enough to drive humans and many other surviving animals to extinction. Another problem is the instability in the first one or two thousand years of the middle term – that is, what I called Stage 2 of the Anthropocene before.10 The problem is that in much of this period the parts of the globe that are inhabitable now will have become too hot or too dry (or both) to sustain human life, while the parts of the planet that are likely to become inhabitable in the more distant future (i.e. in Stage 3.) will not be inhabitable yet. The main limiting factor is probably soil formation. Some parts of the planet that are now too cold to be productive (tundra and other arctic landscapes mainly) will eventually become available for agriculture, but only after fertile soils have been formed there, and that takes many centuries. (And in case of land that is presently covered by ice caps much longer than that.)
Hence, there will be a long period – perhaps a millennium or more – in which previously inhabitable areas have become mostly uninhabitable, while new areas have not become available yet. This will lead to a bottleneck in human population. The number of humans that can survive on the planet during that period is not 100 million (or even a billion), but only a fraction of that, and the lower that number, the greater the chance of extinction.
The reason I used to be “optimistic” about humanity’s survival chances during that bottleneck period is that it seemed that humanity experienced and survived such a bottleneck before. Supposedly, the supereruption of the Toba volcano in Sumatra approximately 73,500 years ago,11 lead to a deterioration of global conditions that reduced humanity to less than 10,000 people. However, recent evidence suggests that, although the Toba eruption had severe effects on the global planet, its effects were more unevenly spread than previously thought and did not cause a genetic bottleneck among humans living in Africa (i.e. the ancestors of modern humans).12 In other words, humanity never experienced a real bottleneck. That doesn’t mean that we won’t survive the (likely) bottleneck in the coming millennium, but it gives us no reason to be optimistic or confident that we will.
And there is another problem – thus far I have focused on the effects of CO₂, but there are other pollution-related dangers in the middle term. There are growing “dead zones” in the oceans, areas that suffer from lack of oxygen caused by the run-off of soil minerals and (especially!) agricultural fertilizers. According to Itsuki Handoh and Timothy Lenton the main culprit is phosphorus, which creates a positive feedback of phosphate concentration, changes in biological productivity, and anoxia (lack of oxygen) in the oceans.13 In the geological past there have been several occasions in which phosphorus run-offs (due to natural erosion) set off a feedback loop that turned almost all of the oceans anoxic, which lead to an extermination of most marine life, but also to burps of hydrogen sulfide that poisoned the atmosphere, killing most life on land as well. Furthermore, ocean anoxia also leads to an increase of gases that eat away the ozone layer, further helping the extinction of land animals (such as humans). Probably, the amount of phosphorus we have dumped into the oceans isn’t enough yet to set off runaway ocean anoxia, but increasing temperatures also increase anoxia so there is reason for concern, and Tim Lenton and Andrew Watson even suggest that ocean anoxia may be more dangerous – that is, more likely to cause human extinction – than CO₂.14 So, even if climate change doesn’t kill us in the middle term, agricultural fertilizers might.
Is human extinction bad?
Perhaps counter-intuitively, the second question – whether human extinction would be bad – is a much more difficult question than the first. Possibly the best known argument related to human extinction is Ninian Smart’s benevolent world-exploder argument (not his name). That argument is a response to Karl Popper’s suggestion to replace the utilitarian principle to maximize happiness with a principle to minimize suffering.15 This idea is now generally called “negative utilitarianism”. While classical utilitarianism aims for the greatest happiness for the greatest number of people, negative utilitarianism aims for “the least amount of suffering for anybody”.16 Smart’s objection to negative utilitarianism is the following: “Suppose that a ruler controls a weapon capable of instantly and painlessly destroying the human race”. Then, given that this would end all human suffering, according to negative utilitarianism, that ruler would be morally obliged to use that weapon. And because “we should assuredly regard such an action as wicked,” negative utilitarianism is wrong.17
The crux of Smart’s argument is, of course, in his premise that “instantly and painlessly destroying the human race” is “wicked”. Smart doesn’t argue that human extinction is bad, but just assumes that it is bad, and thus that causing human extinction is bad as well. He doesn’t defend these assumptions, but appears to take them for granted. That, I think, is significant – like Smart, most people just assume that human extinction would be a bad thing. And in many cases, this assumption is consciously or unconsciously rooted in a religious belief in mankind as the pinnacle of creation.
One of the most powerful arguments against that belief can be found in Robert Nozick’s “The Holocaust”.18 According to Nozick, after the Holocaust, “mankind has fallen” and “humanity has lost its claim to continue”.19 He imagines alien observers from another galaxy looking at human history:
It would not seem unfitting to them . . . if that story came to an end, if the species they see with that history ended, destroying itself in nuclear warfare or otherwise failing to be able to continue. These observers would see the individual tragedies involved, but they would not see . . . any further tragedy in the ending of the species. That species, the one that has committed that, has lost its worthy status.20
Aside from the general point that humanity may not be the greatest thing ever and doesn’t automatically deserve to continue, Nozick also makes another important point in this passage: there is a difference between the “individual tragedies” of people dying and the disappearance of the species homo sapiens. Obviously, the extinction of mankind would involve the death of very many individuals, but the badness of the extinction of mankind is not (just) the sum of the badness of the deaths of all those individuals. An answer to the question whether human extinction is bad is (or should be) primarily an answer to the question whether it is bad in addition to the individual deaths involved.
If we don’t just assume that humanity is the greatest thing ever and (therefore) deserves to continue, then how do we decide whether human extinction would be a bad thing? It seems that we’d need some kind of objective criteria to decide badness. That’s not exactly right, however – intersubjectivity is enough. That is, if there are criteria of badness that everyone agrees about, then we can use those as our standard. I think that there are at least two such criteria: everyone agrees that undeserved suffering is bad (and no one believes that they deserve to suffer themselves), and everyone agrees that dying is bad (except that sometimes suffering is even worse than death). I won’t give a more extensive argument for my claim that these two criteria of badness are universal (but I hope to do so elsewhere some day), because I think that it is fairly obvious that they are (and I also think that objections are the result of misunderstanding or confusion rather than genuine refutations).
It may seem that the badness of death implies that human extinction is bad, but that would be a mistake for two reasons. Firstly, extinction doesn’t necessarily result from excess deaths, but may also be the consequence of insufficient births. (And in such a case, death could not even be what makes extinction bad.) And secondly, as explained above, we need to distinguish the individual deaths from the disappearance of the human species. The badness of death is the badness of the first and is unrelated to the second, so if the second is what we’re interested in, then the badness of death is irrelevant to the question whether the disappearance of mankind would be bad.
It is quite likely that the process of human extinction would involve a lot of suffering, but it is undeniable that the result of that process – i.e. the non-existence of members of the human species – would preclude any future human suffering. Hence, human extinction would certainly lead to less (human) suffering in the long run. One could, of course, argue that there are things that offset suffering: happiness, beauty, artistic and scientific achievements, and so forth.21 Anti-natalists such as David Benatar deny that there are such offsets and thus that it is always better not to be born,22 and there are (at least) two reasons to be skeptical about such a supposed offset. Firstly, throughout history, it has always been a small majority of humans that enjoyed (most of) the fruits of human achievements – the vast majority of people just struggled through life, often suffered, and only occasionally experienced something approaching happiness. That achievements for the few offset the misery and suffering of the many would require a very good argument, and I don’t see any.
Secondly, while there is universal agreement about the badness of suffering, there is very little agreement about the things that would make suffering worthwhile – that is, about beauty, truth, and other achievements. Perhaps, this second ground for skepticism is easily countered, however. Everyone agrees that there is beauty, that there is pleasure, and that there are human achievements. That we disagree about which things count as such doesn’t really matter. And if there are human achievements (etc.) that matter – regardless of what those achievements (etc.) are – then the continued existence of mankind matters, because everything we value and everything that matters has value only to us and matters only to us. Even the threat of human existence already undermines the value of everything that matters to us, as Samuel Scheffler argued in Death and the Afterlife.23
So, let’s assume that human achievements, beauty, pleasure, happiness, and so forth have value. The question, then, is whether that value outweighs suffering? Perhaps, it would in conditions of decreasing suffering. If a civilization would make a serious effort to decrease suffering and to make happiness, beauty, pleasure, and the fruits of various human achievements available to more and more people, then that may very well outweigh the badness of suffering. But we’re not living in such a society. To some extent, we have tried. Elsewhere I argued that human civilization is (or can be seen as) a project to decrease suffering,24 but that project – if it ever was really successful – is failing now and climate change will terminate it forever.25
The next century will destroy almost everything that matters. Many artistic and scientific achievements won’t survive the global societal collapse that is now almost certain.26 Neither will more purely immaterial achievements like ideas and ways of doing things. Cities, libraries, and landscapes will be lost to flood, drought, or other “natural” disasters. In less than a century from now, most of the things that matters to us will (almost?) inevitably be lost. And there isn’t much chance to rebuild any of it. The planet of our (great-) grandchildren will be too chaotic, too hostile, and too resource-strapped to enable the rise of a new civilization. There will be much suffering and very little beauty, very little happiness, and very few achievements to offset that.
If we survive long enough, in many tens of thousands of years the planet may have recovered, but even then there is little chance of real recovery. The easily accessible fossil fuels that kick-started the industrial revolution and that enabled the fast technological and economic development of the past centuries have all been exhausted by us and thus won’t be available to our distant ancestors, and without those, they will be doomed to forever live in pre-modern conditions. A civilization that isn’t based on fossil fuels may be possible in principle, but it would need fossil fuels in an intermediary stage to get there. Of course, pre-modern conditions are not necessarily much worse than modern conditions (certainly, 19th century industrialism wasn’t an improvement over serfdom),27 except that such conditions offer very little opportunity for the aforementioned democratization of happiness, beauty, and the fruits of human achievement. Without the cheap energy needed for economic and technological growth, any supposed offsets of suffering will only be available to a very small elite.
Hence, while there may be things that offset the badness of suffering, the future will offer few of those “things” and when it offers them, it will offer them few people. Furthermore, climate change will lead to more disasters and much more hostile conditions in the parts of our planet that remains inhabitable. The mathematics are easy: more suffering and less potential offsets of that suffering equal less badness of human extinction. Indeed, the conclusion seems to be that human extinction would be the best of all possible outcomes.
This should not be understood as an argument for some kind of collective euthanasia, however. Euthanasia is – strictly speaking – a conscious choice for death by an individual to end that individual’s suffering, but mankind as a whole cannot make choices. Furthermore, whether (continued) suffering is worse than death is subjective, and thus not something one can decide for others. Consequently the extermination of mankind would be more like murder than like euthanasia, and therefore, that human extinction is not bad – and probably even preferable to any real alternative – doesn’t automatically imply that the extermination of mankind is not bad either. The badness (or lack thereof) of human extermination might require additional arguments. “Might”, because a utilitarian or other kind of consequentialist would argue that all that matters is the result and thus that if extinction is the preferred outcome, then the extermination of mankind would be morally justified or even required as well (as Ninian Smart suggested). I won’t discuss additional arguments for extermination, however, for two reasons. Firstly, the question I aimed to answer was about extinction, not about extermination. And secondly, the extermination question is moot as there is no feasible way to completely wipe out mankind anyway.28
My provisional answer to the first question – whether and when humanity will go extinct – is that human extinction is very unlikely in the near future, but that there may be a climate-change related population bottleneck in the coming millennium, which could contribute to human extinction, but not cause it on its own. So, while there is a significant change of human extinction in the next millennium or so, there is also a good chance (perhaps, even a better chance) of survival, and after that the conditions improve. (Except if we set off runaway ocean anoxia – then only some microbial life will survive.) On the long run, human extinction is a certainty, however, but that shouldn’t be a surprise as nothing lasts forever.
My answer to the second question – whether human extinction would be bad – is “no”. Given that the future only promises an increase in misery and suffering with virtually no chance of real recovery from that (even in the distant future), and given that almost everything of value that may offset this suffering will be lost and cannot be regained, human extinction is almost certainly preferable (from a moral point of view) to any other plausible scenario.29
All things considered, it would be better if we would go extinct. But unfortunately we won’t. (At least not soon enough.) Instead, we’re doomed to experience prolonged suffering. Unfortunately, extinction won’t end the misery that the rich and powerful have brought upon us and upon our children. But perhaps, if we’re lucky, some big meteorite will end it all.
If you found this article and/or other articles in this blog useful or valuable, please consider making a small financial contribution to support this blog, 𝐹=𝑚𝑎, and its author. You can find 𝐹=𝑚𝑎’s Patreon page here.
- According to the 2017 report Climate Change in the American Mind “four in ten Americans (39%) think the odds that global warming will cause humans to become extinct are 50% or higher” and another study from 2015 reports that “in four Western nations: the US, UK, Canada and Australia . . . a majority (54%) rated the risk of our way of life ending within the next 100 years at 50% or greater, and a quarter (24%) rated the risk of humans being wiped out at 50% or greater”. Sources: A. Leiserowitz et al. (2017), Climate change in the American mind: May 2017. (New Haven: Yale Program on Climate Change Communication); and M.J. Randle & Eckersley, R. (2015), “Public Perceptions of Future Threats to Humanity and Different Societal Responses: a Cross-National Study”, Futures 72: 4-16.cc.
- See the section “The Nature of Doomism” in Facing the Anthropocene – Attitudes towards Climate Change.
- I don’t believe we ever will, however. Space is too hostile and the physical limitations are too great.
- In the same sense that Neanderthals were a different species, but human nevertheless.
- This is sometimes called “near-term extinction”.
- But such scenarios are extremely unlikely. We won’t be able to burn enough fossil fuels to realize anything approaching a wet greenhouse. See Stages of the Anthropocene.
- See the section on nuclear weapons in Crisis and Inertia (3) – Technological Threats and Crises, as well as the papers referred to there, for further details.
- Probably even the increase in cancer won’t have a significant effect because other factors – such as war, epidemics, drought, and hunger – already decrease life expectancy so much that the demographic effect of cancer is negligible.
- See On the Fragility of Civilization and A Theory of Disaster-driven Societal Collapse and How to Prevent It.
- See: Stages of the Anthropocene.
- Michael Rampino & Stephen Self (1992). “Volcanic Winter and Accelerated Glaciation Following the Toba Super-Eruption”, Nature 359: 50-52.
- Chad Yost et al. (2018). “Subdecadal Phytolith and Charcoal Records from Lake Malawi, East Africa Imply Minimal Effects on Human Evolution from the ~74 ka Toba Supereruption”, Journal of Human Evolution 116: 75-94.
- Itsuki Handoh & Timothy Lenton (2003). “Periodic Mid-Cretaceous Oceanic Anoxic Events Linked by Oscillations of the Phosphorus and Oxygen Biochemical Cycles”, Global Biogeochemical Cycles 17.4, 1092.
- Tim Lenton & Andrew Watson (2011). Revolutions that Made the Earth (Oxford University Press).
- Karl Popper (1947). “The Spell of Plato”, in: The Open Society and its Enemies, Vol. 1 (London: Routledge). See especially note 6 to chapter 5 and note 2 to chapter 9.
- Karl Popper, The Open Society and its Enemies, 241n2.
- R. Ninian Smart (1958). “Negative Utilitarianism”, Mind 67.268: 542-43.
- Robert Nozick (1989). “The Holocaust,” in: The Examined Life: Philosophical Meditations (New York: Simon and Schuster): 236-42.
- Idem, p. 238.
- Idem, pp. 238-9.
- Unfortunately, there is a lot of ugliness, stupidity, and ignorance as well, and probably much more than beauty and positive achievements.
- David Benatar (2006). Better Never to Have Been: The Harm of Coming into Existence (Oxford University Press).
- Samuel Scheffler (2013). Death and the Afterlife (Oxford: Oxford University Press).
- See Armageddon and Utopia and Crisis and Inertia (5) – Derailing a Speeding Train.
- I’m increasingly having doubts that humanity is even capable of significantly decreasing suffering. Perhaps, civilization is a scam – it deceived us (or me, at least) in believing in the possibility of some kind of better society, but really humans aren’t capable of that. We might be too bigoted, too narrow-minded, too selfish, too stupid, too ignorant. Maybe, we’re not the “greatest thing” in all of “creation” but the worst, and the Holocaust merely underlined that fact. Maybe we deserve to go extinct.
- See The 2020s and Beyond and A Theory of Disaster-driven Societal Collapse and How to Prevent It.
- And it will prevent those future people from destroying the planet again, of course.
- Well, there is – climate change (and/or other forms of pollution) just might exterminate mankind, although the odds of survival still are pretty good. The propositions that human extinction is not bad and that climate change might cause human extinction do not together imply that climate change isn’t bad, however, because the effects of climate change are part of the argument why human extinction isn’t bad. Arguing that climate change isn’t bad would be a bit like poisoning someone, causing that person such excruciating pain that he wishes death would come faster, and then claiming that because death would not be bad for that person, the poison that is (slowly and painfully) killing him in the first place isn’t bad either.
- And I haven’t even taken the positive effects of human extinction on other species into account. If their suffering matters, then that would be an additional argument for the extinction of the species that causes much of that suffering: us.