As part of my critique of pro-natalism, I’m looking at the philosophical foundations of the idea. Most of the explicit discussion takes place within the framework of consequentialism (the idea that the best actions or policies are those with the best consequences) and particularly of utilitarianism, broadly defined to say that the best consequences are those which maximise some aggregate function of individual happiness or wellbeing. Other philosophical traditions either avoid the issue (for example Rawls’ theory of justice) or offer little that can illuminate debate over population numbers (for example, virtue ethics).
Looking at the utilitarian analysis immediately creates a puzzle. All the early utilitarians, from Jeremy Bentham to John Stuart Mill supported population limitation. They accepted Malthus’ claim that, in the absence of limitation, population growth would inevitably reduce the great majority of people to subsistence. But unlike Malthus, who used this argument to say that attempts to make the poor better off were futile (he rejected birth control as “vice”) the early utilitarians accepted the desirability of limiting family size.
Most notable among this group was Francis Place. Taking significant legal risks in the repressive climate of the time, Place sought to spread information about contraceptive methods in pamphlets such as Illustrations and Proofs of the Principle of Population (1822).
Importantly, and in contrast to many 20th century Malthusians, Place rejected coercive measures, placing his trust in families to make prudent decisions given access to information and the methods of family planning. Except for his failure (unsurprising at the time) to consider possible conflict within households, Place’s position stands up very well 200 years later.
The classical utilitarians argued for public policies which promoted the welfare of the community to which they applied, on the basis of “each to count for one, and none for more than one”. This applied both to the current population and to the children who would actually be born as a result of their choices, but not to hypothetical additional people who might raise the sum of total utility.
By contrast, contemporary utilitarian philosophy yields bizarre spectacles like “longtermism” which implies that our primary goal should be to produce as many descendants as possible provided that the result is an increase in aggregate utility. This is taking Parfit’s “Repugnant Conclusion”, which I criticised here, to its logical limit.
This is far from the only problem with contemporary utilitarian philosophy. It’s commonly presented as a theory of individual ethics, saying that our actions ought to be those which promote the maximal happiness of everyone affected, giving ourselves the same weight as everyone else. Apart from being impossibly demanding, this prescription seems perfectly designed to produce absurd counterexamples (trolley problems, organ kidnapping etc). As far as I can tell, the original idea of utilitarianism as a public philosophy is sustained only by a handful of philosophers, most notably Bob Goodin [1]
The task of tracing where and when a tradition of thought has gone off the rails is usually complex and ambiguous. But in this case, the answer is amazingly clear-cut. in 1874, one year after the death of JS Mill[3] (the last of the classic utilitarians), Cambridge philosopher Henry Sidgwick published The Methods of Ethics, which became the standard interpretation of utilitarianism, at least in the philosophical literature
In the space of a few pages, Sidgwick introduced all three of the errors I have criticised above,
First, (Book IV 1.1) he states,
By Utilitarianism is here meant the ethical theory, that the conduct which, under any given circumstances, is objectively right, is that which will produce the greatest amount of happiness on the whole; that is, taking into account all whose happiness is affected by the conduct.
Then, in Book IV 1.2
on Utilitarian principles, population ought to be encouraged to increase, is not that at which average happiness is the greatest possible,—as appears to be often assumed by political economists of the school of Malthus—but that at which the product formed by multiplying the number of persons living into the amount of average happiness reaches its maximum.
I had mistakenly inserted a quote here, which I can’t now locate, although I have clear memory of finding it in the Gutenberg edition
Sidgwick asserts these points, and elaborates some of their consequences, but doesn’t really argue for them. Rather, his style of presentation is tight and rigorous, at least by the standards of 19th century philosophy and his tone is authoritative. The result is that even among philosophical critics, most notably GE Moore, his version of utilitarianism became established as the norm.
Superficially, it might seem that modern social welfare theory has taken Sidgwick’s formal approach and given it a mathematical expression. The standard workhorse of this model, the social welfare function, is an aggregate of individual welfare measures, with properties that ensure that the higher the value of the function the better the outcome. But this function is invariably applied in ways that reject Sidgwick’s errors. First, it is used to evaluate public policy, typically in the context of models that do not assume individuals act as disinterest utilitarian ethicists. Second, it is applied to specific fixed populations. Where comparisons are made between populations, they are almost invariably presented in average rather than aggregate terms.
There are plenty of arguments to be made against the utilitarianism of Bentham and Mill. But the starting point for such arguments ought to be a complete rejection of Sidgwick’s errors.
fn1. I got this wrong in my critique of Parfit. Contrary to what I said there, the Sidgwick-style treatment of utilitarianism as a theory of individual ethics does appear to be the most common interpretation, both for advocates and critics.
fn2. Like Bob Goodin (and me, in a sense, since I studied the subject there) an ANU philosopher.
fn3. While Mill developed utilitarianism in important ways, including a break with Bentham’s crude hedonism, and increased emphasis on rules and secondary principles and, nothing in his work suggests the changes introduced by Sidgwick.
{ 53 comments… read them below or add one }
Matt 01.14.26 at 7:46 am
First, it is used to evaluate public policy, typically in the context of models that do not assume individuals act as disinterest utilitarian ethicists.
For what it’s worth, Sidwick is pretty clear that most people don’t act this way, and he also thinks that most of them both can’t, and shouldn’t, act this way. As Bart Shutlz nicely puts it (early in his massive but highly readable biography of Sidgwick), “[Sidgwick] has often been criticized in more abstract philosophical terms for advancing a doubly indicrect approach to happiness, both individual and social, contenancing the possibility of justifying on utilitarian grounds an ‘esoteric morality’ in which the true (utilitarian) principles of ethics were known to an practiced by an elite group of philosophical sophisticates only.” (It’s this sort of thing that leads people like Bernard Williams and Amartya Sen to raise the charger of “government house utilitarianism”.) And Sidgwick is explicit that most people do, and must, use general rules in their moral deicsion-making. (See pp. 214-16 in The Methods of Ethics for a quick statement.) I also think that looking at Sidgwick’s other writings – his Elements of Politics, for example, which was important at the time but very rarely read these days (*), would give a rather different view of his work. (I think the Methods is as much a work of meta-ethics, though that term wasn’t used at the time, as a book of “applied” or “practical” ethics. That is, he’s trying to understand the grounds of or basis for ethics, and its underlying logic, as much as anything. In the papers collected in a volume called “Practical Ethics” he doesn’t, for the most part, proceed by trying to directly apply the approach set out in Methods to particular practical problems, at least if I’m remembering correctly.)
(*) I struggled for a long time to find a copy of the Elements of Politics, and the copy I finally bought is something like 129 years old, though the cover was re-done by the library it used to belong to!)
Chris Bertram 01.14.26 at 9:53 am
The classical utilitarians argued for public policies which promoted the welfare of the community to which they applied, on the basis of “each to count for one, and none for more than one”. This applied both to the current population and to the children who would actually be born as a result of their choices, but not to hypothetical additional people who might raise the sum of total utility.
Ok, well there are two obvious problems right there. First, we need to define “the community to which they applied” and it seems obvious that in evaluating a law in, say, the United States, we can’t just ignore the effect that law will have on others who live elsewhere. So you are pretty much clearly going to have to commit yourself to Sidgwick’s “all whose happiness is affected”.
Second, given the non-identity problem (or rather the biological facts that give rise to it) there is no clear distinction to be drawn between “the children who would actually be born” and “hypothetical additional people” (since everyone who will be born might not have been but for the policy and other choices people make).
John Q 01.14.26 at 10:27 am
Chris @2 AFAICT, Sidgwick doesn’t address the first issue at all, rather he just abstracts away from it by talking about individual ethics. When he actually comes to public policy, in his role as an economist, his analysis of taxation and public finance is entirely in national terms. Mill had more interesting things to say about broadening the community of concern.
The second point is what matters to me, and what I tried to address in responding to Parfit. The best way to look at it is within the context of a family. In making a decision about whether to have a child, potential parents have to consider whether the family without a extra child would be happier than the one they could bring into being by having a child. In considering the second option, they need to give equal weight to the new child and the existing members of the family. But in considering the first, they don’t need to place any weight on the hypothetical child who will not be born, let alone the many hypothetical children they ought to have on Sidgwick’s reasoning.
MisterMr 01.14.26 at 12:25 pm
My two cents about the personal/public ethics.
I did read a lot of old psychoanalisis stuff (Freud, Adler, Jung) and recently I also read some evo-psy, and some stuff about psychology of ethics (like Haidt).
If we speak of ethics from a descriptive psychological standpoint, there is the question of what exactly are we speaking of, and what are the psychological mechanisms that create ethics, and in reality old dudes Freud, Adler and Jung wrote a lot about this stuff and about religion, and a lot of more modern debate is quite close to these initial positions (Freud and Adler were opposites, Jung tried to build a middle way but he had a religious streak that lead to weird mysticism).
Adler believed that humans have a pro-social tendency, that sometimes goes astray (I believe he was influenced by Darwin, though this influence is never cited), whereas Freud believed that all moral existed only as (interiorized) social pressure; Jung came out with the idea of the archetypes and the collective unconscious mostly to describe this (Adler’s) natural morality.
In absolute sense the two positions are not opposites, and we can think that we have some natural pro social instincts (like empathy) but also that a lot of our moral system is interiorized social norms; in fact, I would argue that we can have interiorized social norms because we have the pro-social instinct of interiorizing norms, something other animals probably don’t do or not to the same extent.
Going back to Freud, his theory is basically that we have a basic-instinct “unconscious” that mostly craves pleasure and mostly sex, but this “unconscious” is filtered by a censor agency, the “superego”, that blocks the more antisocial desires, so that some stay repressed and only part of our desires reach the consciousness (“ego”).
The most famous parts of Freud’s theory are the scandalous ones about sex, and this is what he is mostly remembered for, but acually the core of his theory is in the “superego”: without the censorship agency the whole theory makes no sense, and instead if we keep the superego/repression idea and change the underlying instincts (from sex to, say, social dominance, or fear, or whatever) the theory doesn’t change much.
A lot of more modern theories (like Haidt or many evo-psych theories) in practice require the existence of the superego, or something quite similar, even if they don’t explicitate it (making things rather muddy).
So for simplicity I’ll assume that we, as part of our pro-social instincts, interiorize other people’s expectations (from parents or from others, insert Haidt’s Durkheimian tendencies here), and have therefore a “superego” that makes us feel guilty when we do something that we know other would dislike, and virtuous when we do stuff others would like.
Everybody has a “superego”, but the content of this superego for each person change: maybe my dad told me that “boyus don’t cry” so I have a certain idea about what I should try to be, but your dad told you that a real man is tender so your expectatiuons are different. Thgis kind of logic explains why everyone has a morality, why there are similarities but also big differences about these moralities, and also why we all perceive our morality as something objective that comes for outside but other people’s moralities as their preferences (because it is interiorised stuff).
If we apply this logic to utilitarianism, it is apparent that utilitarianism works well when people have different moral systems (different contents of the superego) because it gives a parameter against which to measure the various moralities.
On the other hand it is also obvious that it sucks as a private morality because of all the common problems ascribed to it: it is too demanding, it is impossible to actually calculate utility values (so one has to go to rule utilitarianism that justifies some form of rule based morality), and in particular utilitarianism if applied properly as a personal morality can never give the needed positive internal feedback of being virtuous, which is something that however psychologically we need.
So the problem of utilitarianism as a private moral system IMHO is very real, but to explain it one has to work on a conception of morality where “private moralities” are by necessity quite imperfect (like the concept of the superego), and this kind of jump is the real problem IMHO because this kind of partial answers are not what people search in a moral theory.
engels 01.14.26 at 5:38 pm
Most of the explicit discussion takes place within the framework of consequentialism… Other philosophical traditions either avoid the issue… or offer little that can illuminate debate
I seem to remember something in the Bible about “go forth and multiply…”
Chris Bertram 01.14.26 at 5:47 pm
@John, obviously people who are sympathetic to utilitarianism would prefer it if utilitarianism did not lead to absurd conclusions, but you can’t save utilitarianism simply by ad hoc jiggery pokery (and restriction of scope) that rules out the absurd conclusions by fiat. Sigdwick and Parfit had the courage to follow the arguments where they led, and they led to places that imo ought to lead us to reject utilitarianism.
Charlie W 01.14.26 at 6:06 pm
Not sure that Sidgwick is representative of utilitarian thought: Smart (1973) is more modern and gives the account of act utilitarianism that Williams argues against, in a reply in the same book. He writes:
“This type of disagreement [over the preferability of average v. total happiness] might have practical relevance. It might be important in discussions of the ethics of birth control. This is not to say that the utilitarian who values total, rather than average, happiness may not have potent arguments in favour of birth control. But he will need more arguments to convince himself than will the other type of utilitarian. In most cases the difference between the two types of utilitarianism will not lead to disagreement in practice. For in most cases the most effective way to increase the total happiness is to increase the average happiness, and vice versa.”
One point is that Smart seems to see himself as offering a utilitarian ethics that can accommodate (at least) two views of happiness maximisation. It’s a broad enough framework for that. A second point is that he doesn’t seem much concerned with remote possible worlds (some future world with a truly huge population). I struggle to read him even as setting the stage for that style of utilitarian thinking (i.e. longtermism). We’re a century on from Sidgwick here, note. Something else must have happened, and more recently?
Charlie W 01.14.26 at 6:21 pm
And to comment twice, like a loon: here’s the utilitarian Peter Singer writing (very accessibly) in the Guardian (in 2023):
“We should think about the long-term future and we ought to try to reduce risks of extinction. Where I disagree with some effective altruists is how dominant longtermism should become in the movement. We need some balance between reducing the extinction risks and making the world a better place now. We shouldn’t negate our present problems or our relatively short-term future, not least because we can have much higher confidence that we can help people in these timeframes. Though the lives of people in the future aren’t of any less value, how we can best help people millennia from now is uncertain.” (my emphasis)
Scott Anderson 01.14.26 at 6:24 pm
I expect you are familiar with it, but since you don’t mention it…. G. E. M. Anscombe offers the original (?) smackdown of Sidgwick in her early paper, “Modern Moral Philosophy” (where, among other things, she coins the term “consequentialism” to describe Sidgwick’s particular transformation of earlier utilitarian philosophy). She’s making a slightly different point there than you are about the difference between expected and actual consequences, but it’s related. For her, the question is whether the fact that someone does not intend for a particular outcome of their action to occur could make any difference to their responsibility for its occurrence. (She gives an account of intention in a book of the same time period that helps explain why this is plausible; the way most philosophers talk about intention today might make the claim seem implausible from the start.) I think her paper brings out the point that the relationship between what one intends to do and what one is morally responsible for should not be based in something like a calculation of probabilities assuming a kind of mechanistic causality, but rather on larger ethical practices by which we assign responsibility to agents based on how all of us are supposed to act. This then allows a defense against, say, an evil-motivated outsider from arbitrarily attaching causal outcomes to one’s actions for which one nonetheless must bear the moral responsibility to prevent them.
MisterMr 01.14.26 at 7:46 pm
So this has nothing to do with my previous comment, but I’ll remark that Parfit’s repugnant conclusion doesn ‘t really hold in a situation of fixed resources (like planet earth), instead the only possibile matemathic conclusion is that there is one defined amount of population that is the optimal one (because infinite population would lead to mass extinction). The problem is that is impossible to really calculate utility, so in reality we can’t know, for example, if we already passed optimal population or not.
(I did explain my position in the comments on the linked post about the repugnant conclusion, but of course I would be very happy to bore everyone again with my reasoning).
So I wonder if it really makes sense to argue against nativism with some anti utilitarian logic, or perhaps average utilitarian logic, or of instead it is more productive to say that increased population would lead to overall lower aggregate utility (e.g. by causing more people to live close to starvation level).
John Q 01.15.26 at 12:49 am
Chris @5 I’m glad we agree that Sidgwick and Parfit lead to disaster.
And you’re correct that what Sidgwick was aiming to do was to put utilitarianism on a sound axiomatic basis with unrestricted scope and no ad hoc stuff. This was a classic C19 ambition, seen in voting theory, various parts of economics and most notably in the Russell-Hilbert program in mathematics.
In every case, these attempts led to impossibility theorems, such as that of Arrow in voting theory and Godel’s incompleteness theorems. It struck me that the same was true of Parfit, and a search shows that a common reaction of social welfare theorists is to treat the Repugnant Conclusion as an impossibility result.
Once we accept that the quest for an ethical theory (utilitarian or otherwise) with sound axioms, logically tight conclusions and no domain restrictions is doomed to failure, we are back to the conclusion of the OP. Sidgwick’s version of utilitarianism was a mistake, and its appeal was based on mistaken C19 ideas about rigour.
That leaves pre-Sidgwick utilitarianism unaffected (neither better nor worse than before Sidgwick’s derailment). Again as I said in the OP, while there are plenty of arguments to be made against the utilitarianism of Bentham and Mill, the starting point for such arguments ought to be a complete rejection of Sidgwick’s errors.
Mike Furlan 01.15.26 at 1:56 am
Calls for more babies always makes me think of the end of “Dr. Strangelove.”
“And a computer could be set and programmed to accept factors from youth, health, sexual fertility, intelligence, and a cross-section of necessary skills. Of course, it would be absolutely vital that our top government and military men be included to foster and impart the required principles of leadership and tradition. Naturally, they would breed prodigiously, eh? There would be much time, and little to do. Ha, ha. But ah, with the proper breeding techniques and a ratio of say, ten females to each male, I would guess that they could then work their way back to the present Gross National Product within say, twenty years.”
Might have inspired a Nobel Prize winner:
“If marriage were organized to maximize the total output of households, polygyny would often be preferred to monogamy, since high-income men could marry several women and thereby increase total family income.” Gary S. Becker, A Treatise on the Family (1981)
Tm 01.15.26 at 9:07 am
MisterMr 10: This is as clear and obvious as can be and it’s a total mystery to me why this argument hasn’t put to rest all the longtermist nonsense long ago.
MisterMr 01.15.26 at 10:07 am
Tm 13
I think that the counterargument would be something like “yeah but we could increase the total population if rich people (like american and europeans) accepted to live poorer”, but we american and europeans don’t want to be poorer so utilitarianism must be wrong.
But in reality arguments about distribution (that if americans and europeans were poorer and africans were richer, total utility would increase) do not really imply that total utility would be higher if there were really really a lot of people who lived close to starvation.
This really isn’t true, but a lot of people seem to interpret Parfit argument to be that, and also to be airthight, for reasons that I don’t understand.
Tm 01.15.26 at 3:02 pm
MisterMr, “yeah but we could increase the total population if rich people (like american and europeans) accepted to live poorer”
That is hardly the concern of the longetermists. I think they fantasize about colonizing Mars. It’s too stupid but they are scientific illiterates and some of them are also extremely powerful.
Richard Y. Chappell 01.16.26 at 7:35 pm
Interesting post! I’ve responded with an extended defense of Sidgwick (and contemporary utilitarianism) against these objections at the website linked in my name :-)
engels 01.17.26 at 3:49 pm
Chappell’s post makes short work of the misconceptions getting re-aired here: well worth reading.
Matt 01.18.26 at 6:35 am
I agree that Chappell’s post is useful, as is the discussion in the comments. Chappell’s last comment, though, makes me wonder if something different and not yet really articulated isn’t driving the debate – a disagreement about what ethics is about and what its point is. From my perspective, ethics is about answering the question, “How can we best live together?” (*) This has a part that is about interpersonal morality, and a part that is about what the state can or should do. Lots of people think the 2nd question is built out of the first, but I think that’s wrong. I don’t think that utilitarianism is good answer to either question – I’d favour something like Scalon’s contractualism for the first question, and Rawls’s Political Liberaism for the 2nd. But, it’s not implausible to think that utilitarianism of some sort is a good answer to the 2nd question. That’s what Harsanyi argues for, I think, and maybe Robert Goodin. This isn’t necessarily based on the idea that in our individual choices we need or can or should be utilitrians. (As I noted above, Sidgwick doesn’t think most people will do well trying to apply the principle of utilitiy directly in any case, but that’s a bit different from what I say here.) On this approach, ethics is in some important way “about us”.
But, a lot of people in philosophy (and, I think, in an incohate form, lots of people not in philosophy) think that ethics is about knowing the “true” theory of morality, one that applies at all times, and is in some imporatant way independent of us. This is sometimes termed “moral realism”, though I think that can be confusing. There is, on this account, a deep and important sense in which the true principles of ethics would be true even if there were no people for them to apply to. These indepdent principles are seen as giving us reasons that apply to us whether we know them or accept them or not. (This is related to the black swamp of the internal/external reasons debate, but not reduciable to it, I think.) On this account, “How can we best get along?” isn’t really a fundamental question in ethics. If you think that is a, or the, fundamental question in ethics, then lots of contemporary moral theory will seem like a waste of time, and of no interest at all. But, if you think that the correct or true principles of ethics are independent of us in some strong sense, that they are not (just) about “how can we best get along”, and that knowing them is important, then Parfit’s (and Sigwick’s) questions will seem important and interesting, and objections about demandingness, strange outcomes, etc. will seem beside the point in a way that they won’t if you take the first approach. So, it seems to me that something like this disagreement is driving the discussion here, as far as substance goes. (There are lots of other issues about historical interpretation, but I don’t think they are really foundational here.)
(*) There are also potentially good questions like “what sort of person should I be?”, which I think are best thought of as questions of virtue rather than of ethics, where these are just distinct issues, even if in some ways related.
v-phil 01.18.26 at 3:44 pm
What a wonderful and thoughtful post. Made me want to add Sidgwick to my history of ethics syllabus this semester. I’m having trouble locating the third quotation, from Book IV, ch.3, sec.3. Could you help me out with a more specific reference, with say, edition and page? Thanks so much!
Michael Kates 01.18.26 at 5:06 pm
That’s an interesting comment, Matt. It reminds me of G.A. Cohen’s critique of Rawls in RJE. As I’m sure you know, Cohen thinks that Rawls’ theory of justice rests on a fundamental mistake, since he confuses “rules of regulation” for “fundamental principles of justice.” I take it that you would side with Rawls here, given your comments about “how can we best get along,” and this leads me to think that there is a real fault line in ethics/political philosophy on this issue.
Matt 01.18.26 at 9:28 pm
Michael Kates – thanks. Yes, I think the dispute between G.A. Cohen and Rawls(ians) is a similar one (and I’m on the Rawls side, for sure.) My thought is that a divide like this is in the background of a lot of disputes, but perhaps was made most explicit in Cohen’s critique, so if nothing else it’s good for bringing that out.
John Q 01.18.26 at 11:58 pm
v-phil @19 Even though I have a clear memory of locating the third quotation in the Gutenberg edition, I can’t find it now. So, I;ve edited the post to deleted it. Apologies for this.
I’m assuming the quote I found was some kind of AI confabulation, and that I slipped up on the check. I will need to double check more carefully in future.
LFC 01.19.26 at 12:35 am
I think that an important (maybe the most important) ethical question is “What does a just society, both on a ‘national’ and global scale, look like?” The “national” or domestic part of this question is what Rawls addressed in A Theory of Justice. He rejected the position that a just society is one that maximizes the sum total of utility or “happiness” or that one should aim to maximize utility, and I agree with that rejection. (If it turned out that a just or reasonably just society also maximized utility, that would simply be a coincidence, he noted.)
Part of the problem, I think, with the notion of maximizing “happiness” is that happiness is very subjective. (At the end of The Myth of Sisyphus, Camus says “One must imagine Sisyphus happy.”) It makes more sense, I think, to try to ensure that people have the means, resources, or “capabilities” to pursue whatever definition of happiness they might have, but not to be too concerned with trying to measure the aggregate total of “happiness” (or “utility”).
Richard Chappell writes:
I disagree with this. Every life, including every happy or “privileged” life, is going to contain a certain amount of unhappiness, and there’s no way to know in advance whether any given individual is going to think that his/her/their happiness is or has been enough to outweigh the inevitable unhappiness. We can’t ask the fetus or the developing being in utero whether it wants to be born or not, and we can’t predict how any given entity in utero will answer that question when it’s become a conscious or sentient or thinking person. Given that reality, the idea that we should celebrate the existence of every happy child as a positively good thing, as Chappell says, does not strike me as persuasive. There’s no way to tell for sure whether any given happy four-year-old is going to end up living a happy life, and there’s no way to tell for sure whether, even if he does end up living a relatively happy life, he would choose in retrospect to be born, given the inevitable unhappiness that even the most charmed and fortunate and “privileged” lives will contain.
J-D 01.19.26 at 2:51 am
My best guess is that most people who discuss these subjects would include ‘what sort of person should I be?’ and other closely related questions within the scope of ethics and also that they would include ‘how can we best live together?’ and other closely related questions within the scope of political philosophy or social philosophy which they might or might not consider to be a subdivision of ethics. It doesn’t really matter whether I’m right about this, though: what matters is that these are two different topics (although related ones) and confusion will result in the absence of clarity about which of these topics is being discussed. However, the question is ‘Which of these two topics is under discussion?’ and not ‘Which of these two topics does the name “ethics” properly belong to?’ The proper way to use the name ‘ethics’ is the way which promotes clarity and minimises confusion (on the assumption, that is, which I do make but I can imagine not always applying, that it is in fact desirable to promote clarity and minimise confusion).
J-D 01.19.26 at 2:53 am
On a more light-hearted note, so far Henry Sidgwick has only appeared once in Existential Comics:
https://existentialcomics.com/comic/258
Tm 01.19.26 at 9:56 am
LFC 23: Thanks for this comment.
“But then temporal consistency suggests that, even before they were born or conceived, it would be good for them to get to exist (all else equal).”
Doesn’t this raise the question: if we see an unhappy child, should we wish for that child to not exist? And what are the consequences of that? This line of argument is incoherent (“temporal consistency” doesn’t suggest anything because there is no consistency between existing and not existing) and dangerous.
MisterMr 01.19.26 at 1:38 pm
About the “happy child” sentence in Chappell’s blog, I think there are two problems:
First of all if you take a basic utilitarian stance, you can’t just count the child’s happiness, but also the effect of the child’s life on others; if for example an happy child is born in the USA, but via indirect effects this causes pain and death to other two childs in Canada, the total sum of happyness would still decrease.
In a natalism debate, if we are speaking of resource constraints, we have to count these indirect effects of taking away resources from others.
BTW Chappell apparently believes that these resource constraints are not real because of the end of the population boom, I disagree, but this is a disagreement on empirical issues and not on utilitarian theory; on the other hand utilitarianism, if applied to any real existing problem, will always bump in this “ok let’s now make a guess about really existing quantities of utility”, which is a problem.
But also, I think the “happy child” wording is a bad wording because it hides a real problem: Chappell IMHO is treating human ife as something that has value in itself, but since he is framing the argument in utilitarian terms he uses the concept of “happy child” so he can count life as a finite amount of happiness.
However, people can be very egoist and care only for their own happiness, the idea in utilitarianism that we should count everyone’s happiness as equal is quite unnatural.
The reason we do this is that we have some social instincts that tell us to care about others (at least about friends and family); it is only after we take in account this sort of basic human altruism that utilitarianism makes sense (as an abstraction and extension of this altruism to the whole humanity).
So this means that the respect and interest for other people’s life is logically antecedent to utilitarianism, it doesn’t really make sense to treat one life as a definite amount of happiness (although, in the utilitarian calculus, at some point one would be forced to attribute an “utility value” to a life).
@Michael Kates -Matt 20-21
Isn’t it just the usual distinction between rule-utilitarianism (justice) and some sort of very optimistic concept of act-utilitarianism (total effective happiness)?
reason 01.19.26 at 3:00 pm
Having skipped though the discussion, something seems missing here. If we are to choose between average and aggregate utility, it would seem we would have to have a way of measuring utility – but we actually don’t. I don’t even think that utility is a properly defined concept, but is just a fudge invented to allow us to explain consumer choice. I think the only thing we can say about utility is that in general it is not linear, that the value of a dollar to a pauper is higher than for a millionaire so that at any single point of time, the most effective to (likely) increase total utility is to redistribute money from the rich to the poor. (Hence the attraction to some of looking to very long time scales, despite the possibility of all life on earth being wiped out in the meantime.)
engels 01.19.26 at 8:00 pm
I found Matt’s comment interesting but find it very hard to understand ethics being about “how we can best get along” because I think I can easily imagine all of us getting along while doing something ethically abhorrent (eg burning the planet to a cinder or binge-watching Emily in Paris).
Matt 01.19.26 at 9:29 pm
Mr.Mister at 27 – No, the distinction I’m interested in is unrelated to the dispute between rule and act utilitarians.
JD at 24 – I don’t want to legislate terms, so don’t insist on “ethics” being reserved to “what should I do?”, but I think there is a real, if not always clear, distinction between an approach that takes “what should I do?” as basic and one that takes “what sort of person should I be?” as basic. Paradigms are of course Kant or Mill on the one hand, and Aristotle on the other. As with everything, there are of course a lot of complications here, but the reason why I mentioned the “virtue” approach at all is that it seems to me to that it can be worked on without, and independently of, deciding questions about the status of the sorts of moral rules I was talking about.
J-D 01.20.26 at 9:59 am
I hope it was clear that I don’t want to legislate terms either. My concern here is with lexical definition, not stipulative definition. I think the way most people use the word ‘ethics’ it includes both discussion of the question ‘What should I do?’ and discussion of the question ‘What sort of person should I be?’ even though they are admittedly different although (I think) very closely related questions. Of course some people might find one of those questions more interesting or more important and the other less so (or not at all), saying ‘This question is not interesting’ or ‘This question is unimportant’ is different from saying ‘The subject matter of this question lies outside the scope of ethics’.
It may perhaps be germane to point out that the comic I linked to, in which Henry Sidgwick appears, deals specifically with people taking different approaches to the subject matter of ethics.
MisterMr 01.20.26 at 12:13 pm
Thinking about the “happy kid” argument, and about the “what is ethic about” question, I think there is this distinction:
We have a variety of “moral intuitions”, that partly come from our social instincts anmd partly from culture. However these intuitions, even if we count only the ones that come from instincts, are very contradictory.
Then we build up ethic theories, such as utilitarianism or other theories, that ultimately are based on these intuitions, but have to compose these contradictory moral intuitions into a coherent whole.
There is a difference between the original intuitions and the final theory, that will only approximate the intuitions.
So for example utilitarianism largely approximates most people’s moral intuitions, but this doesn’t mean that people naturally are utilitarians: what happens is that we have natural instincts that push us to care for others (our family and friends), but as societies became larger and larger those instincts have to be stretched larger and larger, until we get to stuff like utilitarianism that tries to push these instincts to all humanity (and in some cases even to animals).
So utilitarianism tries to explain how to correctly compose various social instincts/moral intuitions, and in order to do this it has to quantify the effects of actions on people, and comes out with a concept like total quantity of happiness or utilitons (that obviously we can’t really calculate), but I think that the correct interpretation of this is not “one day we’ll have a supercomuter that will be able to solve the felicifer calculus”, but rather “we have to take in account the consequence of our actions on everyone, weighting everyone equally”.
@Matt 30
But if you think in terms of rule utilitarianism, and take in account that people need to have some autonomy and freedom to reach happiness (as Mill as far as I can understand did), you end up with a theory of personal rights/justice with boundaries where these cause too much unhappiness on others, that is a theory about “how we get along”.
Act utilitarianism on the other hand, if we take it at its extreme where we can assume that one understands perfectly what is the “maximum utility”, could easily fall into the philosopher king fallacy, and leads to more absolutist view of ethics and justice; also a theory about the fundamental principles of justice can lead to the same problems.
There is a problem because autonomy and lack of perfect knowledge of the results of our actions (that are linked because the ultimate consequences of my actions depend also on what other people do in their autonomy) are actually basic prerequisites of the concept of free will and similar concepts that are at the base of the concept of morals, but in many abstract arguments about justice/ethics they are assumed away because they create too much complexity.
J, not that one 01.20.26 at 4:15 pm
I recently read Charles Larmore’s What Is Political Philosophy? which has some interesting things to say about the issues between Rawls and Cohen, and discusses Raymond Williams’s thinking about legitimacy, as well.
Richard Y. Chappell 01.20.26 at 6:11 pm
LFC-23: I meant to stipulate a case where the child’s overall life is happy. (To make progress in ethics, we first bracket issues about uncertainty to get clear on what’s good, bad, and neutral in principle. Only after we’ve determined what matters in principle can we fruitfully turn to managing uncertainty.)
Still, even in realistic cases I think we often should default to an optimistic view about the value of others’ lives. It would be weird to hear that a child just survived a life-threatening surgery, and be like, “There’s no way to tell for sure whether any given happy four-year-old is going to end up living a happy life…” That’s cold, man! You don’t need certainty to be glad for others’ sake. We should be happy when kids survive life-threatening illnesses, since this is overwhelmingly likely to be good for them. Aside from rare cases of unremitting misery, we should be glad that others are alive, more generally.
Tm-26: We may reasonably assume that momentary unhappiness is fleeting. But in principle there obviously could be cases of tragic, miserable lives—”not worth living”—that we should regret for the sake of the person who suffered living through it. (Of course, we should be extremely hesitant about attributing this property to others; individuals are the best judges of their own well-being.)
MisterMr @27: Yes, I think other people have obviously net-positive externalities on average. Immigrants are good for the receiving country, for example, and so are children. Misanthropy is bad.
@32: “philosopher king fallacy” is more a matter of naive instrumentalism than act utilitarianism properly understood. It’s an awfully common conflation, though! I’ve shared in my name-link an old post of mine on “Naïve Instrumentalism vs Principled Proceduralism” that clarifies why I don’t see the former as any part of utilitarianism properly understood.
John Q 01.20.26 at 7:00 pm
Richard @34
Important to note that “happy” here means “not so unhappy as to wish you were dead”. Everyone who is happy in this sense adds to aggregate utility
Actual vs potential people is the crucial distinction here. It’s perfectly reasonable to hold the view “if I had my time over, I wouldn’t have so many kids, but now that they are here, I don’t regret any of them”
Migration doesn’t change the total number of people. Do you claim that emigration makes the source country worse off?
Richard Y. Chappell 01.20.26 at 8:42 pm
re: 1 – not necessarily! It depends on further assumptions about the structure of value as it applies to lives. For example, there could be a “grey zone” or neutral range wherein one’s welfare is neither outright negative (warranting regrets about existence) nor outright positive (warranting positive attitudes). Such lives would be on a par with non-existence—neither better, worse, nor strictly equal.
I discuss this sort of “critical range” view (a variant on the total view) in my linked textbook chapter on population ethics.
re: 3 – in the event of permanent emigration without remittances or other rewards, sure, I’d expect the loss of human capital to typically be bad for the source country. (Real-life migration may more often be win-win thanks to remittances.)
engels 01.20.26 at 10:58 pm
This is astounding.
China registers lowest number of births since records began
https://www.ft.com/content/a294436d-9d74-49f5-873e-22ba7a83365b
China last year registered the lowest number of births since records began, marking the fourth consecutive year of population decline as policymakers grapple with a demographic crisis. On Monday, the government reported that 7.92mn babies were born in 2025, down from 9.54mn the year before, and the lowest number of births since 1949. Last year, 11.31mn people died. China’s population fell 3.39mn to 1.405bn… Demographers have warned that China’s population decline will accelerate dramatically as postwar baby boomers die. In 2024, a UN study predicted that China will shrink from 1.4bn people in 2024 to 1.3bn by 2050, hitting 633mn by 2100…
How does average utilitarianism score a world with 0 people? (Maybe I’ll ask Chat-GPT…)
LFC 01.20.26 at 11:28 pm
Richard Y. Chappell @34
1) When a child survives a life-threatening illness, I’m happy for that child. However, your position, if I understand it correctly, requires more than that. It entails that I should or must be happy that the child was born in the first place. Sorry, I’m not inclined to sign on to that. Once a child has been born (the result of a decision and/or set of actions that the child, of course, cannot participate in), I want that child to be as happy, in terms of his/her overall life, as possible. But it doesn’t follow that I have to agree with the view that every additional child brought into the world is a cause for celebration, and in fact I’m not agreeing to that.
2) There’s probably room for some debate or disagreement about how widespread, or rare, misery is, even in certain “advanced” societies. Case and Deaton’s work on “deaths of despair” may be relevant here (see the link).
https://press.princeton.edu/books/paperback/9780691217079/deaths-of-despair-and-the-future-of-capitalism
MisterMr 01.21.26 at 12:13 am
@Richard Y Chappell
The question if we are before or after optimal population is an empirical one, we can disagree but this is not about “utilitarism” per se.
However if you assume positive externalities you are just skipping the problem because you think it is not present today, but presumably you don’t think that the problem is absolutely impossible?
For example imagine a spaceship in a centuries long trip to Alpha Centauri, this ship has a minimal crew of 100 and a maximum possibile crew of 1000, however at 1000 the resources are so stretched that a small error could cause many deaths.
Surely there is an optimal number of crew that is between 100 and 1000, is there a way to determine what this number is? Or should people just push close to 1000 for some reason?
Or perhaps there are some automatic limits so people won’t pop out kids if they feel resource constrained? If so, how can you exclude that this is what is happening now? Even if overall income is growing, for example, non household workload on women is going up, which is still a resource constraint.
Hearta 01.21.26 at 12:35 am
Insightful
Tm 01.21.26 at 8:40 am
Richard: “other people have obviously net-positive externalities on average”
I too disagree with this but I’m curious what your argument is for that claim. I just don’t see any reason why this should the case. Even leaving aside the question of resource constraints (which we absolutely cannot leave aside!), I don’t see a good reason for that assumption. I would argue that the number of people we can fruitfully interact with (in the sense that their existence makes my life more interesting or whatever) is not unlimited. And there is no reason to assume that the net impact of each person on others is, on average, positive. And pointing this out is not misanthropy ;-)
Tm 01.21.26 at 8:44 am
engels 37: this is indeed a remarkable drop. Compared to 10-15 years ago, the number of births recorded in China declined by about 10 million per year, quite something. Thailand (https://en.wikipedia.org/wiki/Demographics_of_Thailand) and Taiwan also have recorded new lows, whereas South Korea reports more births for the second year in a row. The EU and Russia will no doubt also record declining birth numbers.
Unfortunately many countries with high birth rates don’t have reliable statistics so we don’t really know what’s going on there. India, Indonesia, Pakistan, Nigeria, and so on, it’s very difficult to know how trustworthy the UN estimates are (it seems that they have tended to be on the high side). Overall the number of births has been declining, slowly, only since 2012 (from > 144 million to maybe about 130 million per year). The very large cohorts born after 2005 are now entering reproductive age so it’s likely that even with falling overall fertility, birth numbers will remain high.
https://en.wikipedia.org/wiki/Demographics_of_the_world#Vital_statistics
steven t johnson 01.21.26 at 2:33 pm
engels@27 It’s ambiguous whether it’s you or the Chinese policy makers or both who think there’s a demographic crisis. In any event, it’s not clear to me there is any such thing. I must agree that a shrinking population is a problem for economic growth. But I am so perverse as to think that a kind of stationary economy, though likely completely incompatible with a capitalist system, is in the long run the only sustainable industrial economy. That is, fundamentally any so-called growth is technological improvements in efficiency in use of resources. (This is not the same thing as profitability.) This being the case, my view is, so much the worse for capitalism.
The relevance to utilitarian questions about population levels I think is obvious enough.
Talin 01.21.26 at 7:28 pm
I think that the way our social instincts operate is not to maximize the number of happy individuals, but the number of happy stories. Human minds operate, to a large extent, on stories, it’s a basic unit of cognition. The Ship of Theseus paradox is really about a continuity of the material manifestation of a narrative – the continuity of the physical stuff is (pun intended) immaterial.
A young child full of sweet disposition and talent has a story about their future promise; an elder who has a rich lived experience has a story about their past. These are things we seek to preserve and maximize, along with the individual to whom they are attached. A hypothetical individual who has not yet been born has no story, other than perhaps heredity.
As stories, there’s a diminishing marginal utility in repetition – the same story told a thousand times isn’t as valuable as a thousand unique stories. However, humans also have value beyond their stories which doesn’t diminish with scale (the more doctors and physicists in the world, the better off we all are), so the actual quantity of “goodness” we optimize for is a hybrid of these factors. What this means, though, is that the calculus of the marginal benefit of adding a new individual depends on how many individuals there already are.
For myself, my moral foundation was influenced early on by reading Olaf Stapleton (Last and First Men and Star Maker) at an early age. My stance is that any entities that is capable of looking up at the stars and wondering what’s out there are my brethren; it’s not universal happiness, it’s strictly factional loyalty, but for a faction that is extremely inclusive.
J-D 01.22.26 at 12:14 am
In Charles Dickens’s A Christmas Carol, the Spirit of Christmas Yet To Come shows Ebenezer Scrooge a young couple whose debt he holds. The husband brings the wife news of Scrooge’s death: her first reaction is gladness, her second remorse for her rejoicing. Both these reactions are correct, but it’s the first of them I’m concerned with here. Why is she glad that he’s dead? Because it means that he can’t harm them any more. It also means that he can’t harm anybody else any more; and whatever pleasure we imagine the life of Ebenezer Scrooge, as he is at the beginning of the novel, is giving to himself, it is a cause of little if any happiness and much unhappiness to other people. I give a fictional example to avoid disagreement about the record of any actual person, but I’m sure you can think as easily as I can of people now dead who did much harm while they were alive, and also of people now alive who are doing much harm and who are likely to continue doing so if they are not prevented, as (for example) by death. It would be weird to say this about a four-year-old, but there are many adults, both living and dead, of whom it would be correct to say that occasions on which they escaped death did more harm than good and increased rather than decreased the world’s burden of misery.
Richard Y. Chappell 01.22.26 at 1:39 am
Tm 41 – On the positive externalities of people, check out the book After the Spike by economists Dean Spears and Michael Geruso. I shared some highlights in my review, here:
https://www.goodthoughts.blog/p/why-depopulation-matters
I guess people can dispute the application thresholds for terms like ‘misanthropy’, but if someone claimed that it was bad for the world that I exist, I’d consider them a hater. If they then clarified that their negative judgment extends to (almost) every human, they sure don’t sound like someone who loves and values humanity as I think we should. (“Some of my best friends are human,” I imagine them protesting. But of course that’s compatible with having an objectionably negative attitude towards humanity more broadly.)
@MisterMr 39 – Sure, it’s an empirical issue. You can similarly imagine a possible situation in which some specific demographic is especially bad to have around. But I’m a bit suspicious of people who are too quick to jump to negative conclusions about either specific demographics or humanity in general. Zero-sum worldviews generally strike me as both misguided and vicious. If you fully grasp where the anti-immigrant crowd goes wrong, it’ll help you to see (part of) what’s wrong with anti-natalism too.
Tm 01.22.26 at 10:54 am
Richard 46: JQ reviewed Spears and Geruso here:
https://crookedtimber.org/2025/08/18/if-something-cant-go-on-forever-it-will-stop/
John has also in several posts dealt with the Einstein/Beethoven fallacy (more people means more ideas and more creativity etc.) I don’t want to rehearse the arguments here but just as a general observartion, we all know that humans can do bad as well as good, and there is no empirical or logical reason to assume that the Beethovens and Einsteins that a more populous humanity purportedly will bring outweigh the Stalins and Hitlers. Furthermore, and here the question of externalities come s back, it is reasonable to assume that increased resource competition due to increased population can have a negative effect on overall happiness, and also on the ethical behavior of people.
Regarding the “misanthropy” question, the problem is that you change the framing as suited. None of us here was suggesting that any particular person or group of people should die. What we are discussing is the question of “hypothetical additional people”.
Btw it just occurs to me that your argument is very similar to one made by the Catholic Church against contraception: If you use contraception to prevent the existence of an additional human, that is no different than wanting somebody dead, iow it’s analogous to murder. To be clear, this position is controversial even among Catholic moralists, but it has been offered.
MisterMr 01.22.26 at 3:18 pm
@Richard Y. Chappell
as it happens, I’m an open border guy. If you want to discuss why, empirically, you think that more people would increase total happiness, ok (I did already read your blog post and I’m unconvinced).
But my problem is more about the theory of utilitarianism: if we only think about the situations where everybody’s utility goes up, there is no hard choice to make, and our moral intuitions will always say OK. But an ethic theory has to work also in situations where there is difficulty, so:
Is there a possibility where an increase in the number of people would be bad and so new birth should be prevented? IMHO yes, so we have to find a way to compare the birth of a person with other people’s happyness, which is a pain. OTOH if we give infinite value to human life we should spend everything in hospitals and exactly 0 in leisure, and this evidently is unthinkable.
What about abortion? does an happy child overrule an happy mother? If no, why?
And so on.
John Q 01.23.26 at 5:22 am
If children generate positive externalities, and, on average, the same utility as everyone else, both average and aggregate utility will be increased by having more children. The philosophical debate, therefore, turns on whether having more children is desirable even if they generate negative externalities. Aggregate utility says yes (at least until the negative externalities become large), average utility says no.
J-D 01.23.26 at 5:40 am
I wasn’t going to name names …
… but somebody else did.
J-D 01.23.26 at 5:44 am
Referring specifically to children makes it important to remember that the negative externalities relevant include not just any negative externalities generated by children as children but also the negative externalities generated by children after they grow up and cease being children. In the world today there are some people generating negative externalities on a colossal scale even though they were infants once and as such more appealing.
engels 01.23.26 at 12:36 pm
It’s ambiguous whether it’s you or the Chinese policy makers or both who think there’s a demographic crisis.
Sorry: the long paragraph was a verbatim quote from the FT.
Dick Veldkamp 01.24.26 at 12:16 pm
[Somewhat tangential]
Prof Q, since you have written on the subject before, would you care to comment on Noah Smith’s piece on the problem (in his view) of fertility decline and shrinking population?
See noahpinion.blog, Jan 22.