An Alternative Reproductive Imperative Argument

The Challenge

The was some buzz about an essay by Torbjörn Tännsjö which was commissioned by Vox then rejected. I heard about via LGM, and went into quite a bit of discussion in comments. Tännsjö was trying to present an extraction of an argument from his recent book in about 1000 words for a general audience. He is a pretty hard core utilitarian with some focus on the Repugnant Conclusion.

Easy to see that this could go pear shaped.

The essay is a disaster I think. But, what I read of his other work is not so disastrous. This is a hard form to write and his is a hard topic to fit into this form. Ok

Having lots of important, time critical stuff to do, I naturally am going to give it a shot. Yay me!


  1. It must be about 1000 words
  2. It must argue that it is plausible that we have a moral duty to produce a lot of children…close to resource limitations all other things being equal.
  3. It must go by way of the Repugnant Conclusion.


The Reproductive Imperative

While there are many belief systems which impose a duty to have children and even lots of children or as many children as you can produce, many of these justify that duty by a command (“Be fruitful and multiply!”) or by subgroup survival preferences. And there are clearly strong personal, partially biologically supported reasons that many, perhaps most, people want kids. But, is there any positive duty to have kids or a lot of kids? Is the choice whether or not to have children fundamentally a direct (though perhaps overridable) moral imperative? I think it’s fair to say that in advanced industrial societies there’s a strong current of thought that says that having children is a free choice, though having a lot of children is problematic.

There is a somewhat surprising argument that we all have a positive duty to produce as many children as reasonably possible. Indeed, the argument holds that it’s quite likely we should have more children than fewer but happier children. The surprising part is the argument is quite difficult to resist and follows from straightforward moral intuitions.

A Fantasy Scenario

Imagine a world somewhat like ours. It has 10 billion people living on it but it has solved the sustainability issues of such a population and, furthermore, the standard of living of 99% of the population is equivalent to living in a US household with an income of $300k/year. This is a rich rich world full of happy people.

Now, suppose a disaster was about to strike this world which would wipe out the whole population. You have two possible solutions: one (A) will kill off half the population, but leave the rest untouched. The other (B) will maintain the whole population, but half would see their income reduced to about $50k a year.

I trust that most people think that saving 5 billion people is the clearly right thing to do. That is, it’s hard to see how to ethically prefer A to B. Indeed, if you thought that A was the only solution, it would be a profound relief that someone came up with B.

It’s not hard to see why! Those 5 billion people almost certainly prefer to live, even under reduced circumstances. Their lives are filled with good things and thus are a good thing we should preserve.

This all seems very straightforward. But what if our world has only 5 billion people on it (with $300k incomes) and we know we could add another 5 billion (though with only $50k incomes)? How would we add them? Probably by having more babies, but we could imagine that we had the technology to produce babies without anyone having to take on the risk of pregnancy. So, we have worlds A (5 billion, all rich) and B (10 Billion, half rich, half ok) again. If we preferred B to A before, shouldn’t we do the same now? After all, they are exactly the same worlds. The only difference in the scenarios is how we get to them.

My personal gut reaction is that I feel that in the “saving” scenario, I have a strong moral obligation to get to B. Lots of sacrifices would be justified and if I picked A over B the I would be a monster.

But with the multiplying option, I feel a lot more indifferent. I don’t think B should be blocked, but I don’t feel that I’m a monster for not picking it. But, it’s very unclear, from a moral perspective, that I should distinguish the two cases. In the first, I’ve saved a lot of good, but in the second I’ve produced a lot of good! If I value good enough to save it, why doesn’t that push me to produce it?

This preference for worlds with more and more people (even if their individual happiness is lower than in worlds with smaller populations) is known as the Repugnant Conclusion, and it’s one of the trickier bits of moral theory.

Now we see, roughly, that we can get from the Repugnant Conclusion to the Reproductive Imperative: The “easiest” way for fertile people to produce significantly more good in the world is to have another baby (assuming that we aren’t at carrying capacity, or it won’t destroy the parents’ lives). It’s pretty hard to do more good than creating another person (just as it’s hard to do more bad than killing one).

Note that we don’t need to produce the “best” babies or babies in the best circumstances. It’s hard to see that the best off person in the world is better off than two other people (at a reasonable level of welfare). People grow up, fall in love, have families, enjoy food, watch movies, hang out with friends, and do a myriad of other things that imbue a life with value. In general, it’s very hard to make a person twice as happy as they were, but it’s pretty easy to produce two people who are pretty happy. Thus, if we have obligations to produce “better” worlds, we have some obligation to have as many children as we reasonably can.


My gut reaction to this conclusion is extremely negative. After all, the burden of duties to have children have typically fallen heavily on women, including feelings of guilt. It’s easy to see that adding a secular argument pro-having-children could have some nasty effects in current societies. Similarly, I suspect that slowing down population growth might be helpful given some of our more extreme resource bound problems. It’s not a given, of course, because we can up our consumption levels pretty quickly even without population growth. But, it’s an intuition that many share.

But these don’t attack the argument directly. They show that he argument may not be applicable to our society, but that provides an argument that we should be pursuing changes to our society so as to make increased population feasible.

Resisting the argument on its own terms is difficult. We might try to make a difference between “saving” existing people and “producing” future ones (who currently don’t exist). We might claim that while we may have strong obligations toward existing people, we can have no obligations toward non existing people since they aren’t there to owe anything to. This would justify a difference in preference between saving and producing.

But there are costs to this solution. In particular, it seems to ignore something essential about the psychology of preferring B to A in the saving case: We cherish not the bare lives shorn of the living which make them worth living, but the living itself. And that cherishing is similar to the cherishing we have for doing good for others, for bringing more good into existence. This suggests that the repugnance we feel might come from a systematic misunderstanding we have about the case.


I think this is a better essay with close enough content. It’s not a rewrite of Tännsjö’s obviously, but it seems to be a near neighbour.

Part of the problem with Tännsjö’s essay is that it simultaneously does too much and too little and in a very disjointed way. I just re read it and I felt a content sense of whiplash. Now for an audience that is already inclined to be pro-maximal reproduction, perhaps the initial shock would be lower so they’d find the rest of the ride a bit smoother. But if you already agree that we should maximise baby production, why is the rest of the article needed? The target audience has to be one who isn’t on board from the start. But then you really need to address the fact that there will be a lot of issues that aren’t purely argument related. That is, the dialectic is complex. Throwing a wacky conclusion as if it were immediate and obvious, then dealing (poorly) with two objections that are not the immediate reactions people would have, just makes for confusion.

I tried to address this in several ways. First, I put the authorial voice in the same situation as the reluctant reader. It helps that I am such a reader, but that doesn’t matter per se. I want the reluctant reader to come with me through the argument, not reject it reflexively. Second, I separate out the Repugnant Conclusion from the Reproductive Imperative and start with what I think is a very strong presentation of the Repugnant Conclusion. This lets me push the idea that even if we can ultimately reject the RI, the RC makes it difficult. And our rejection of the RI might be different than we thought it was. I certainly now think a bit differently about the RI than I was used to. Perhaps I’ll get back to my earlier perspective eventually, but it’s not as easy to support as I’d like. (More on this in a subsequent post.)

On the plus side, I think we have a good basis in this argument for a better utilitarian handling of disability. Some of Tännsjö’s papers touch on this and I think a direct argument against eugenics is well worth having. More on this later as well.

The shortness of the form is very challenging (mine was 1179 words and ends abruptly). These are not easy waters.