Antiantinatalism

Does life suck? I think it does, somewhat. Others would say it sucks even worse than that. David Benatar thinks that life sucks so profoundly that bestowing it on anyone, like your child, is a grave crime. The antinatalist philosopher is having a moment: a profile in the New Yorker, an appearance on Sam Harris’ podcast, and a reader emailing to ask me what I think of the case against being born.

I tried to make sense of antinatalism, and I think it’s bad philosophy. But I also think it’s bad economics, and plays on the widely held intuition that an extra person makes the rest of humanity worse off by taking up some space, some resources, some piece of the pie that would have gone to others. I hold that this zero-sum view is ignorant of the reality of the modern economy. Having children is good for the children, good for you, and good for the world.

heine death

Like a pirate cow, the philosophy of antinatalism stands on three legs:

  1. Some conception of negative utilitarianism, i.e. the view that reducing suffering is the principal (or only) thing that matters morally, and that moderate pain weighs more than great joy.
  2. The observation that life contains more bad things than good things.
  3. A divide between existing and non-existing agents, and the assertion that the latter (i.e. unborn babies) have only an absolute preference to avoid the risk of any suffering, and no preference for enjoying anything at all.

I’ll address these in order.

Negative utilitarianism

Negative utilitarianism isn’t strictly necessary for the antinatalist argument, which is lucky for them because it’s a pretty incoherent philosophy. However, antinatalists sometimes sneak it in by implication. For example, the reader asked me if I’ll take 5 minutes of the worst pain imaginable for an hour of the greatest pleasure. Benatar presented Harris with a similar trade-off: would Sam be willing to have children and grandchildren who will lead lives of hardship if that will ensure a bright future for countless descendants after that?

Both questions expected an answer in the negative, and yet my answer to both is that I most certainly would.

Regarding the first question: evolution hasn’t made great pleasure as accessible to us as it has made pain. Fitness advantages from things like a good meal accumulate slowly but a single injury can drop one’s fitness to zero, so the pain of an injury is felt stronger than the joy of pizza. But even pizza, though quite an achievement, is far from the greatest pleasure imaginable.

Humankind has only recently begun exploring the landscape of bliss, compared to our long evolutionary history of pain. If you can’t imagine a pleasure great enough to make the trade-off worthwhile, consider that you may be falling prey to the availability heuristic. Pain is a lot more plentiful and salient, but it’s not a lot more important. The fact that pleasure is rare should only make it more valuable when offsetting pain, and an hour is a lot longer than 5 minutes.

As for the question about children and distant descendants, answering it negatively seems to me rather selfish. Should my goal be to avoid only that suffering which I can be blamed for and have to deal with? As a parent, I’m going to spend countless hours ensuring a better life for my children, while probably doing little to directly impact the far future. I’d be happy to balance this equation out a bit.

Life is net negative

The contention that life contains more suffering than happiness has to deal with the fact that almost everyone who’s alive finds their life worth living. A big part of that is life looking much better in retrospect than in the moment: we remember both pleasures and ordeals fondly, the latter as meaningful, character-building experiences.

Pessimist philosophers are tempted to discount this positive reminiscing as an illusion, but it’s as real as any other experience. All bliss and agony ultimately matter only as they are experienced in our subjective consciousness, whether we’re reflecting on something happening around us in the moment or something we went through years before.

I don’t see a good reason to separate the experiencing self from the remembering self by assigning one of them greater moral weight. A life’s experience is just an integral over momentary experiencing selves, and in each moment I can choose whether to focus on immediate sensations or reflect on the past.

Taking seriously the position that life is not worth living should lead one to a philosophy of extinctionism – the stance that it would be pretty great if all humans died in their sleep tonight. On the podcast, Benatar desperately tried to wriggle out of the implication that antinatalism leads to extinctionism, and I’m not sure why. As my friend remarked: if you’re biting bullets, don’t just nibble on them.

Non-existing agents

To escape extinctionism, Benatar draws a sharp boundary between existing agents who have an interest in living and non-existing agents who have only an interest in avoiding suffering. Sam Harris reasonably took issue with this arbitrary assertion made on behalf of the non-existent. But as with experiencing vs. remembering, I want to attack the idea of the boundary itself between existing and non-existing agents.

I plan to get rich slowly because I want Jacob.2048 to have more money than he otherwise would have. But is Jacob.2048 an “existing agent”? I have some guesses about Jacob.2048’s interests, but the only thing I’m sure about is that the two of us will have different attitudes on many issues, like the making and spending of money, or moral philosophy. I changed my mind on those topics a lot just in recent years, but I’m still thankful for Jacob.2013 for making mostly good choices.

I know more about Jacob.2048 than I do about my unborn great-grandchildren, but less than I know about myself. Jacob.2048 is certainly less “existing” than I am: his preferences are uncertain, as is his actual physical existence – I could get hit by a bus tomorrow and all my sacrifices for future me will be to no one’s benefit. I care for Jacob.2048 the same way I care about my yet-unborn children, by anticipating what their preferences would be when they’ll actually be around to have those preferences. Calling one of them “existing” and the other not seems like a strange distinction to make.

chodorov quote

Imagine a planet just like ours, but with only 100,000 people on it. Let this group comprise the assembled best experts from every field, and let them be fully devoted to cooperating for the benefit of all. How good would their lives be, all by themselves on an uncrowded planet? You may be imagining a paradise, each person enjoying a life of luxury on their own huge estate in the midst of unspoiled nature.

I’m imagining a world without pencils.

It takes a lot more than 100,000 people to make a pencil. The graphite miner, the driver of the logging truck, the lacquer chemist, the pencil company accountant, each one makes a small and highly specialized contribution that makes the billions of pencils available to everyone who wants them. It’s possible that 100,000 experts could make a pencil if they devoted all their resources to it, but then none of them could afford it.

How many people does it take to make a lightbulb? Or antibiotics? Or sour cream and onion Pringles?

Free trade allows workers to specialize, and specialized workers can create a lot more wealth than people who have to meet all their needs themselves. A cobbler who has to grow their own food is a lot worse off than a cobbler who can trade a pair of shoes for pizza. But they still have to tan their own leather, unless a third person joins who can specialize in doing just that. Every person added to the network may feel that their own job is becoming ever less consequential, but in fact, the benefits of integration keep growing as the size of the network grows.

With a billion people integrated into the global economy, we managed to produce solutions to most direct causes of suffering: food, medicine, power and sewage, porn. Add in a couple more billion, and we moved on to directly providing pleasure: video games, craft beers, great soap.

The fruits of the modern economy aren’t distributed equally, but both the number and the percentage of people who have access to the good stuff has kept growing inexorably for a couple of centuries now. The more people there are, the more the wealth gets spread.

Having more people doesn’t just allow for specialization of work, it also means specialization of fun. In ancient times, like in 1932 Philadelphia, your potential dating pool consisted of a single-digit number of people. Today, I can pick my wife out of thousands of OkCupid profiles in New York, which is itself a strongly filtered population. I’m much more specialized in making my wife happy than the best dude she could pick out of a lineup of 10.

Increasing population can be bad in Malthusian scenarios, but today there isn’t a single resource that we’re using anywhere close to its limit. There’s enough potential energy in the sun and the nuclei of atoms, enough water in the seas, enough room, enough food, enough air. The only resource that’s stretched thin is capable people that can make all that stuff happen.

Do-It-For-Denmark-570x285

If you’re reading this, you’re probably an educated person in a developed country, and your children almost certainly will be. You have every reason to believe that your children’s lives will be better than your own, and that by joining the global economy they’ll contribute to making everyone better off in ways we can’t even imagine yet.

And if you’re reading this, you’re much more likely to be the kind of person who will personally be happier for having children, especially if you follow a few simple guidelines.

Now stop procrastinating on the internet and go make babies!

17 thoughts on “Antiantinatalism

    1. People who are capitalists shouldn’t even offer an opinion on antinatalism. You are in favor of exploitation and every other negative trait of this pathetic plague of a species and this post reflects that in its entirety. Antinatlism is not bad economics, but capitalism is the worst economy of them all, and its not surprising that this deplorable species came up with something to exploit each other.

      You don’t even need Benatar to make an argument. Look around you. Capitalism is making that argument EVERY SINGLE DAY. No need to add more slaves to the pipeline to prop up the stocks on the Wall St.

      Liked by 1 person

  1. “…life looking much better in retrospect than in the moment: we remember both pleasures and ordeals fondly, the latter as meaningful, character-building experiences”
    My memory doesn’t work like that at all. Negative memories stick much better, and often recur at full strength, and I really have to dig for good ones, and those are always more distant, as well.

    Liked by 1 person

  2. To me it’s a lot simpler than that. No need to wrap it up in philosophical argument.

    If I have a child, there is: a 100% chance they will have to suffer through the trauma of watching me age / get sick / die; a 100% chance they’ll have to deal with their own age, sickness, fear of death and eventual annihilation; and a 50% chance they’ll get cancer in their lifetime.

    If I were to have a girl, her current most likely cause of death (according to the stats in my country) is dementia or Alzheimer’s.

    And there’s also the likelihood that with my medical history the child would inherit mental health problems.

    Throw in climate change and overpopulation and the choice is clear. No need for Dr Benatar.

    Like

  3. I lol’d at “pirate cow”. Well played.

    Since you brought up “unborn babies” as “non-existing agents” and argued that life is retrospectively net-positive and Jacob.2048’s preferences matter, I have to ask: Do you think there’s a consistent anti-antinatalist defense of abortion? If so, what is it?

    Like

    1. The defense is that everything is a trade-off, and wanting more babies doesn’t override every other consideration. Also, the marginal extra child born to my readers as a result of my essay is probably set to do much better than the average child whose mother wants to abort it. Finally, the average abortion reduces the total number of children born by less than one; I’m sure that many women who had abortions end up having children later when it’s better for both mother and child.

      There’s also an important difference between personal ethics and public policy. As public policy, I see no reasonable consensus other than letting the pregnant mother make the ultimate decision.

      Like

    2. Note the category “unborn babies” as constructed in the post includes unconceived babies like Jacob’s great-grandchildren.

      If we conclude that children are a public good that the state should try to get more of, it leaves the question of who we should try to make produce it. This is a thorny question, and you could argue for a lot of very different answers (such as “the rich, since they can afford it”, and “whoever we can most cheaply pay to do it”). Banning abortion amounts to choosing the answer, “whoever fails at birth control”, which IMO just doesn’t have a whole lot going for it in our framework.

      Liked by 2 people

  4. “If we conclude that children are a public good that the state should try to get more of, it leaves the question of who we should try to make produce it. This is a thorny question…”

    Or the state could subsidize childbirth... One way to do this is through a minimum guaranteed income where children also have a minimum income (perhaps different then that of adults) whose legal gaurdian(s) get to decide how it's spent. (Issues of fiduciary responsibility in this situation glossed over here).
    
    Otherwise, the government could just try <s>creepy</s> propaganda.
    
    On second thought maybe I don't know what you mean by trying to "make" people reproduce. I trust you don't mean forcible methods but maybe targeted social pressure. Getting rich people to shame other rich people into making more babies perhaps?
    

    Like

  5. Is your stance on anti-natalism is dependent on economic, social and technolocal context? If so, please clarify; your position looks uncomfortably close to the repugnant conclusion, a la Robin Hanson.

    To lay out a less moralistic standpoint (which may eintirely coincide with your viewpoint):
    Dependent on context, there is a desirable target population. This is lower bounded by sustainability of civilization (like 100k people to make a pencil); better robotics/automation/AI may decrease the lower bound. There is a Malthusian upper bound: Beyond carrying capacity, reproduction is a zero-sum game.

    This gives you a curve: living-standards as a function of population (considering technology etc as parameters). Obviously living standards are bad at low populations, even with full automation, because having a diverse set of small niche communities to integrate into is important for people; you don’t want to be the only weird geek caring about topic xyz in the whole world. Likewise cultural stuff.

    This curve, however, is maybe not fully saturating but very sublinear at high populations, even with infinite carrying capacity: Once every reasonable cultural niche is filled with more people than a single person can ever meet, and produces more cultural artifacts than a single person can consume in a lifetime, the gain in living standards from adding more people is pretty limited.

    For example, one could guess that expanding population from 5 billion to 1 trillion will not make a big positive impact on personal living standards, even if we had the carrying capacity. Likewise from 1 billion to 10 billion.

    So, in between peak-living-conditions and Malthusian boundary, you have a hard philisophical question: How much do you value extra people if their existence makes everyone worse off? However you answer it, you will get some peak-goodness between peak-living-conditions and Malthusian boundary.

    Now, is your natalist position that we must go on and colonize all the universe, snuff out all stars to power matroshka brains to simulate a maximal number of people, under minimal living conditions? And today, being far from carrying capacity, we should go on and multiply until we hit subsistence levels, a la Robin Hanson’s EMs, until each person is as materially poor as possible while still keeping him alive?

    Or is your position rather that you refuse to voice an opinion on the big stuff, but we are far away from peak-goodness anyway (maybe even before peak-living-conditions), so for now go on and multiply and punt the question to the future? (Do you really think that there are many cultural niches, or technologies that are feasible with 10 billion people, but not with 1 billion? Having 1 billion people on earth, you could have significantly better material living standards for everyone, at the very least in the sense that everyone could enjoy more untouched nature, less pollution, almost-western living standards, etc)

    Or is your position that reproduction makes people happy; and, while reproduction beyond replacement levels is a finite resource [*], one of my favorite quotes:

    “Resources exist to be consumed. And consumed they will be, if not by this generation then by some future. By what right does this forgotten future seek to deny us our birthright? None I say! Let us take what is ours, chew and eat our fill.”
    [Sid Meier’s Alpha Centauri, in-game attributed to CEO Nwabudike Morgan, “The Ethics of Greed”]

    [*] even if you assume that we turn all the universe into computronium and run until heat death, and regardless of whether you assume immortality or not. It is a trivial and obviously true observation that exponential growth cannot continue indefinitely.

    Like

    1. To clarify my own position:

      Singling out individuals for their personal decisions in this matter is a jerk thing to do (but criticizing groups for their policies is fair game; quiverful is evil)
      Globally we are beyond peak-goodness, given current ecological and technological realities and, and given my moral intuitions on the repugnant conclusion. On average and on the margins, reproduction is a net bad
      I am enough of a narcist to believe that I personally, and most of your readers, beat this average
      I am enough of an ethno-nationalist to advocate against local far below replacement reproduction rates, policy wise
      To some limited degree I go with the Morgan quote
      I am not above committing moral wrongs for selfish reasons. My personal reproductive decisions are entirely unaffected by any global moral considerations. I invite everyone to follow the same line.
      We must globally get down to slightly-above-reproduction levels at least, in order to not run against (local) Malthusian boundaries. I say local, because I have no illusions that any rich country would make significant sacrafices in living conditions, just to prevent mass starvation in some far away place.
      Long term population reduction below peak-goodness makes no sense, policy wise, because at the very latest 200 years into the future this will all become moot due to technological-singularity-stuff anyway (I feel dirty saying this).

      Like

      1. These are all very good questions. My thought on the matter are a bit disorganized, so appreciate you laying out your own for me to react to. I’ll start from the very big and far and move towards the here and now.

        As a matter of ethics, my intuition is that there’s value in diversity of experience. If you duplicated the Earth and all its inhabitants in a faraway corner of the universe, so they would live out the exact same experiences, I wouldn’t consider that to be doubling the universal utility. “Similarity” is a spectrum: if you duplicated me (e.g., in Parfit’s duplication-teleporter thought experiment) the two Jacobs won’t have double the value, and it’s OK to kill one of us. However if one of us escapes the replicator and start accumulating different experiences, the utility of the extra copy will slowly increase from 0 Jacobs to a higher number, although probably never reaching 1.

        This puts some upper limit on how repugnant I’m willing to get, if more and more people will also become more and more similar to each other. But the repugnant conclusion has never seemed very interesting to me anyway. Should we all die so that one utility monster may live? It’s a question with no practical application, and opinions on it depend more on intuition pumps than on rigorous reasoning. One of the intuitions the conclusion depends on is where exactly do you draw the line for a 0-utility-life. The unhappiest person currently living? The 20th percentile of happiness? The 80th?

        One thing I am sure of is that we are very very very far from the derivative turning negative forever. We went from 1 billion to 7.6 billion in 200 years, and the correlation with increasing quality of life has been so dramatic that it’s hard to claim that no causality at all goes the other way. Given my understanding of economics, I do disagree with you that having almost-western living standards is easier at 1 billion than it is at 10 billion. At 7.6 billion we just about have enough pencils, but not nearly enough AI researchers, Mars settlers, aging curers, French rappers. Also not enough rationality bloggers, which why people are forced to read Putanumonit instead of something better.

        Even in a single galaxy, the Malthusian bound is so far it’s hard to estimate. I believe that even humans as we currently exist, with our puny brains, limited consciousness, and tiny lifespans, have enough possible diversity of thought and experience that a trillion of us won’t exhaust it. And given that I currently see the derivatives for happiness, population and wisdom all sloping upwards, I’m OK with punting the 1 trillion or 10 trillion question to the time when there are actually 1 trillion of us to answer.

        Narrowing down even further: there are some populations in the world that will decrease total utility by having kids, but not as many as people think. Following up on my diversity intuition, I’m happy for more people in more cultures to try more stuff. A country only needs to escape abject poverty to start producing scientists, artists, or even just good workers. An extra Chinese person is good, an extra Dane is good, an extra Colombian is good… Certainly an extra Putanumonit reader! There are places that are pushing against local Malthusian limits, but being in a Malthusian situation probably doesn’t leave one with time to read blogs.

        And finally to the very local level: my personal reproductive choices are affected by global moral considerations. Not a lot, but not none at all. Most decisions I make are 90% selfish (insofar as they’re even conscious), but morality is still a consideration, and it can certainly be a tiebreaker on close calls. If I really thought that extra people are horrible for the world, or that my own children would be genetically predisposed to suffer, I would prefer adopting.

        Like

        1. An extra Chinese person is good, an extra Dane is good, an extra Colombian is good…

          Ok, so let’s try to be clear: Telling someone “your existence is bad for the world” is totally a racist jerk thing to do, and I wouldn’t do it.

          We are talking soft policy: Imagine the Chinese government planning a public advertisement campaign, that may either advocate for (A) People, have more kids! More children are morally and patriotically right! (maybe even tax-breaks, subsidized child-care, etc etc) or (B) Hey, here is some free birth control. Also, don’t have too many kids. Chairman Mao said “one child only”, and the latest 5-year plan demands 1.65 children per family (maybe even subsidized care for elderly people who don’t have a lot of grandkids).

          When I am saying that “an extra $nationality person is a net bad”, then I am saying that option (A) would be bad, and option (B) would be good. I am not morally judging individuals because that would be a jerk thing to do.

          (and of course you want to alleviate poverty for everyone, not just families with many children or elderly people with few descendants; that’s beside the point. Same way, too harsh treatment of people who had more than 1 child in China at the times was a jerk thing to do)

          Given this definition, is your viewpoint still that “An extra Chinese person is good, an extra Dane is good, an extra Colombian is good…”?

          In my view it depends very much on circumstances: For various reasons, more people are a bad thing, on global average. But locally is more interesting: And then, I think that, on the margins, an extra Colombian or an extra Chinese is a bad thing (contingent on current facts on the ground), while an extra Polish might be a good thing (ecologically terrible; good for ethno-nationalist reasons, because apart from the number of people in the world, it would be very sad if Polish culture were to die out due to demographics)

          This closely mirrors actual political consensus, by the way. This view is so mainstream that it is agreed upon by the fucking world bank as well as the Chinese communist party, both today and back before they became capitalist.

          Re utility aggregation / repugnant conclusion / duplication: I completely agree that “diversity” is a key part of it. Instead of looking too deep into it, I’d say that global utility scales extremely sublinear with QALYs. Of course QALYs are a totally insufficient model, but bear with me:

          Imagine humanity has a very risky choice to make. We can either grab 1e20 QALYs for certain (e.g. stay in the solar system as bio-humans, that’s 100 billion times 1 billion years of civilization). Or we can take a gamble: 99.9% of immediate and utter extinction, but 0.1% of grabbing a tiny part of the milky way, for a gain of 1e25 QALYs. If your utility scales linearly with QALYs, then the second option is 100x better (“utility” is defined as the nonlinear function that allows you to make choices over probability distributions by maximizing expectation).

          I guess that utility scales very sublinearly after all, at the high ends. I guess even sub-power-law, more like poly-logarithmic. Or does your moral intuition differ? How does it change if you shift zeros around?

          And then the problem of utility monsters vs repugnant conclusion is nicely solved. Diversity is good because it allows you to evade the sublinearity, up to some point. A little bit of inequality is good because it is another dimension of diversity; it’s more interesting if the top 1e-4 are not middle-class (but it would be stupid to give them more than 1e-2 of resources; think power laws).

          Same way, eradicating poverty forever might not necessarily be good: Maybe society is just more diverse in a good sense, if the lowest 1e-3 were not living in middle class circumstances (I am not sure this is a bullet I’m willing to bite; for now I just stay inconsistent on this, but it is hard to escape this rather repugnant …, ahem, conclusion, from the above framework).

          We went from 1 billion to 7.6 billion in 200 years, and the correlation with increasing quality of life has been so dramatic that it’s hard to claim that no causality at all goes the other way.

          I’ll politely disagree. The causality going from improved technology, productivity and industry towards rising population is obviously very strong (vaccinations, antibiotics, cheap sanitation, artificial fertilizer, industrial farming, hybrid crops, etc).

          I do not see how you could then use the observed correlation as Bayesian evidence towards causality in the opposite direction.

          Obviously the converse argument doesn’t work very well either: Bicausality is still a possibility, but I don’t think you can argue that from the observed correlation.

          Like

          1. Xi may think it’s good for China to have 1.65 kids per family, because they can always import people from the poorer neighboring countries and not worry about building houses / providing benefits for them. But that’s not a global point of view. For example, I think every extra Dane is very good for China – they get the benefits of extra productive people in rich countries (more science / Legos / Carlsberg) without bearing any of the costs.

            Speaking of costs, the impact on the ecology and the climate is probably big enough that it can swing the net impact from positive to negative for a lot of people. As I’ve written, I’m not concerned enough about climate change for it to shift the balance, but that’s certainly a crux we disagree on and one where I could change my mind given sufficient evidence.

            As for government policy, I at least think that having kids should be “revenue neutral” for parents. Throughout history, parents used to bear most of the costs of raising children while reaping most of the benefits, e.g. an extra hand on the farm. Today parents still bear most of the costs, but get a much smaller chunk of the benefits (you pay taxes to the government, not to your mom). In rich countries with lof fertility rates old childless people are absolutely free-riding on parents, and everyone would be better off if the allocation of resources to costs was fairer.

            I do not see how you could then use the observed correlation as Bayesian evidence towards causality in the opposite direction.

            All else being equal, any correlation is Bayesian evidence of causality in both directions. If the correlation was negative (as it is in Malthusian cases) you’ll surely agree that it’s evidence against positive causality, won’t you?

            Also, rich countries tend to have a lot fewer kids, which points to the causality from material well being -> fertility being negative in a lot of cases. This is even stronger evidence for population -> well being having positive causality.

            Let’s focus on the micro. In a toy model, 1 out of every 100,000 people decides to become a vaccine scientist / AI researcher / rationality blogger. Let’s assume that the skill for each one of those occupations is distributed on a normal bell curve. In a world of a million people, the best one (out of 10) would be 1.2 SDs more skilled that average. With a billion people that becomes 3.7 SDs, and with 100 billion the best person is one in a million, or 4.8 SDs better than average. I think that a lot of the best things we have are the result of the work of few extraordinary people, and it’s worth having more people around so that those become (statistically) even more extraordinary.

            Like

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s