Utilitarianism is often accused of being compatible with extreme injustices. If by tormenting a minority, a great amount of happiness is generated for a large majority, this situation would be acceptable or even desirable. I've read Bentham's "An Introduction to the Principles of Morals and Legislation" a while ago, and I'm skimming through Mill's "Utilitarianism", but I cannot find any direct response to this claim. Am I missing something they wrote? Or is there perhaps another philosopher who has defended utilitarianism against this attack?

  • So far, you have not sketched much of an attack utilitarianism would have to be defended against. Maybe you could give one or more serious real world examples.
    – user14511
    Commented Jun 6, 2020 at 18:48
  • Are you asking specifically about act utilitarianism (what Bentham argued for) as opposed to rule utilitarianism (closer to what Mill argued for)? This section of a longer article addresses this issue for act utilitarianism--a form of 'direct consequentialism' in the article's terminology--but it also notes that forms of 'indirect consequentialism' (like rule utilitarianism) are a way to avoid such difficulties.
    – Hypnosifl
    Commented Jun 6, 2020 at 23:49
  • Situations where "majority happiness" is generated by "tormenting a minority" are usually too contrived to matter, and often overlook alternatives where the same or greater happiness can be generated without the "tormenting". And if a situation is detached from reality enough it is not surprising that what "feels" like the "extreme injustice" comes out as right. Alternatively, one can directly subtract the cost for the "tormenting" from the majority's utility, and reverse the balance. See SEP, Consequentialism.
    – Conifold
    Commented Jun 7, 2020 at 2:27
  • @CanBeSaidClearly consider McCloskey's "sheriff scenario", for example. en.wikipedia.org/wiki/Utilitarianism#The_Sheriff_Scenario
    – Ariel
    Commented Jun 7, 2020 at 7:57

3 Answers 3



To fix ideas, utilitarianism is here taken to be the moral theory that requires the maximisation of the total net utility of all members of society or the maximisation of the average per capita utility of all members of society. These formulations have different implications for the number of persons affected but neither recognises individual claims of justice that prevent maximisation.

The putative clash with justice

For ordinary moral thinking it would be unjust - an extreme injustice - to send an innocent person to jail or execution even if doing so maximised utility in either of the above ways. This is a stock example of the putative clash between considerations of utility and considerations of justice. Another might centre on the elderly. It would seem unjust to euthanise all elderly unproductive people even if this maximised total net utility or average per capita utility.

Back to the formulas

Both formulas contain the clause, 'of all members of society'. It is clear that you do not maximise according to either formula if you omit entirely the utility of one member of society, namely the individual who is sent to jail or execution, or the class of elderly unproductive people. If this point is accepted, one widely-claimed clash between utility and justice is removed.

Revised formula

But nothing is that easy in philosophy. What, the suggestion may be made, if we use a different and historically more authentic formula, that of increasing the 'sum total of happiness' [utility], which we find in J.S. Mill's Utilitarianism (1863), ch. 2: https://www.utilitarianism.com/mill2.htm? Conditions are easily imaginable in which the sum total of utility is increased by sending an innocent person to jail or execution or by euthanising the class of the unproductive elderly.

J.S. Mill on justice

Since Mill has entered the picture, it may be useful to consider his theory of justice in Utilitarianism, ch. 5. Mill envisages the possibility that in extreme circumstances it may be necessary - utilitarianly justifiable - to postpone justice to utility but that such emergency ethics are a distorted background for a general theory of justice.

Central to Mill's account of justice is the idea of a right grounded in utility. Certain rights are notable for 'the extradordinarily important and impressive kind of utility which is concerned' (https://www.utilitarianism.com/mill5.htm). These rights 'concern the essentials of human well-being more nearly' than any others (ibid.). They are twofold: the right to right to security and the right to liberty - where 'liberty' includes the right to self-development outlined in On Liberty, ch. 2.

These rights are not absolute since 'particular cases may occur in which some other social duty is so important, as to overrule' them (https://www.utilitarianism.com/mill5.htm). But all else equal (i.e. in the absence of moral emergency) the fact that they 'stand higher in the scale of social utility' (ibid.) than other moral considerations protects the innocent person from being sent to jail or execution and the unproductive elderly from being euthanised since this would harm or destroy their vital rights to security (in the case of the elderly) and liberty (in the case of the innocent person). In calculating the 'sum total of happiness', rights pertaining to justice must be satisfied before other elements of happiness are taken into account.

The nature of what I've termed a 'moral emergency' and Mill calls 'particular cases' is not entirely clear from Mill's text but I think Mill's position, at least in this area of his moral theory, can be best explained if we represent him as a rule-utilitarian. Given a plurality of rules, there will always be the possibility of a clash between rules and we cannot say in advance that the rule of justice will always have priority.

Other perspectives

I have concentrated on Mill in considering how utility and justice can be reconciled but there are other approaches on which a reconciliation cannot be reached. One such is John Rawls' critique of utilitarianism in A Theory of Justice (1971). I have not included Rawls because, much as I admire his work, his theory of justice is as controversial as the utilitarianism which he attacks.


To be honest, I believe we have a very solid case for always acting in the best interests of others -- every time, all the time. I'll list a few reasons below, but they are not completely independent. Rather they reinforce/enable each other, making the whole greater than the sum is of its parts. And, hopefully, more convincing because while it makes a lot of sense, it is (tho unsurprisingly) not exactly mainstream.

  1. Prisoner dilemma.

When each individual pursues their own self-interest, the outcome is worse than if they had cared about the other guy. And you don't have to break the law for a chance to face it in real life. Far from it, it's fairly typical in social interactions.

  1. I'll show you the Golden Rule...

Actually, though :) What we know as the Golden Rule might be just a shadow of the real thing. As instant a Karma as it is inevitable, it comes in the form of our relationship with ourselves. We think we are special, every one of us. And maybe we are, but not when it comes to our relationships¹ -- we treat ourselves just like we treat others.²

This is very close to the concept referred by Zulu's Ubuntu or Javanese Guyub (there are many of pre-civilization words that are notoriously hard to translate simply because we are oblivious to some very important truths... Greek lógos or Vedic ātman are better-known examples of such words).

That is why, for example, we should be careful passing our judgment on others -- because we will judge ourselves just as harshly.

Or everyone seems to struggle with loving themselves. But what if it is not about love but compassion? And what if developing self-compassion simply means learning compassion to others? To me, it sounds like a much more reasonable goal, especially given that...

  1. We are all in this together.

I think it was never "us" vs "them". It's always good in ourselves vs evil in ourselves. And it was never supposed to be like that,³ but "civilization" takes a heavy toll on us, and it starts early, so ensuring that we develop unhealthy coping mechanisms, of which one is particularly troublesome.

Call it our dark side (or Shadow, or alter-ego). Originally purposed to be a fail-safe (if even), it was our panic response, the last-ditch desperate attempt on the part of person's subconsciousness to save their butt by pushing their rational self (their values, their sense of right & wrong, and whatever else the human in them cares about) out of the way and in the backseat, putting their evil alter-ego in charge.

One reason it is problematic is that due to the aforementioned childhood conditioning it grows up way more awake, more powerful, and more intrusive than it should have ever been.

Even psychology finally got a pretty good picture of it taken from the opposite (the evidence-based?) side. They had to analyze painstakingly the past research of different personality disorders thought responsible for the so-called dark triad of the human psyche, lookng for the traits present in every side of the triad. That's how they found the common denominator at its heart.

Needless to say, they decided on calling it "D-Factor". And, needless to say, I appreciate it as the next guy. However, I do think they missed(?) an opportunity to announce the evidence-based definition of evil. And, as everything true evil it is beautifully simple:

   Evil is a "tendency to maximize one's own utility safety at the expense of others". And that's exactly what the panic response does!

Disabling a person's human side allows it to maintain a laser focus on its goal -- reaching that emergency 🅴🆇🅸🆃 sign at any cost, with no regard for other people's wellbeing and safety.

And that is not even the worst part. It actually almost natural play devil's advocate here -- like, sure, it's evil. But at least it's on your side, right?

That's what pretty much everyone believes. And some obviously worse than others, but pretty much everyone falls for that lie.

Truth is, our dark side is not our friend. A panic response, it's our fear that gives it the energy, that makes it powerful. Even we give it the benefit of the doubt. Even if we assume that it is not simply using you as its food supply, it makes no difference. Even if it sympathetic to our plight, it cannot simply give us what we want. As long as we let it play us around, or outright bully us, it will have no confidence in our judgment, and no respect for our ability to take care of ourselves.

And as long as it's in charge, that's where it wants us to be. Anxious, depressed -- both to secure its food supply, and curb our enthusiasm a bit, thank you very much.

But you what? That is still not the worst part. The worst part is that such an arrangement is no stable at all. I mean being reasonably unhappy, but not too miserable. It doesn't take too much pressure for people to start taking it on each other, their dark sides scrambling to return the favors (because what do they know? they are damn neural nets, they can't think!). And before we know it, it's spiraling out hands -- and it can get pretty dark.

And that's why we should be easy on each other.

... and as if that wasn't enough, here is the man himself:

“Be kind, for everyone you meet is fighting a hard battle.”― Socrates


¹ Which in itself might be an evolutionary adaptation to the prisoner dilemma.

² No, not even your friends.. the other others! People who we don't particularly like, or about as much as we like ourselves.

³ We are meant to be fully rational, sharing universal knowledge of the objective reality that we all share, and, thus, being in agreement all the time on everything. No conflicts, no power hierarchy, no fighting for the top spot, and no evil (see Paleolithic warlessness). The way we used to be for most of human history (until things went horribly wrong during the transition to agriculture circa 8,000-10,000 BC).

Yes, everyone, but most importantly our enemies, as long as we clear what love doesn't mean 😍🥰🤩, ok? Love is understanding. Understanding is love. And, to that end, we can not possibly like everyone, but we must love them regardless.


To start with a given case: a common argument for Hiroshima/Nagasaki is that this act (these actions) shortened the war, saving a net balance of lives. So from a vague utilitarian perspective, an [in my opinion] incredibly unjust action seems right. However, the utilitarian has a good response to this case: one might object to the picture of causality in question. Normally, we think of "A causes B" as requiring a relatively close connection in time. So one might say that "the end of the war" was temporally too disconnected from the actual H/N bombings to count as an actual "effect" of them. (Think: people had to observe the aftermath, relay this info to the Japanese higher-ups, they had to make decisions, etc. all before "the war ended." And note moreover that we actually are saying the two separate bombings can be thought of as "one" act/cause (or suitably integrated set of these, or whatever).)

Note then that this carries over in general: when the case is "cutting up one patient to save five others by transplants," this is not temporally disjoint directly between "cutting apart one patient and saving five others," since gaining the organs for transplant isn't the same thing as performing the transplants.

These issues then lead into how utilitarians concatenate local actions over time, which has lead to the issue of "infinite ethics" (I will quote Bostrom's "Infinite Ethics" here, who frames the matter in terms of aggregative consequentialism, which intuitively covers forms of utilitarianism with supposedly unjust implications):

When we gaze at the starry sky at night and try to think of humanity from a “cosmic point of view”, we feel small. Human history, with all its earnest strivings, triumphs, and tragedies can remind us of a colony of ants, laboring frantically to rearrange the needles of their little ephemeral stack. We brush such late-night rumination aside in our daily life and analytic philosophy. But, might such seemingly idle reflections hint at something of philosophical significance? In particular, might they contain an important implication for our moral theorizing?

If the cosmos is finite, then our own comparative smallness does not necessarily undermine the idea that our conduct matters even from an impersonal perspective. We might constitute a minute portion of the whole, but that does not detract from our absolute importance. Suppose there are a hundred thousand other planets with civilizations that had their own holocausts. This does not alter the fact that the holocaust that humans caused contributed an enormous quantity of suffering to the world, a quantity measured in millions of destroyed lives. Maybe this is a tiny fraction of the total suffering in the world, but in absolute terms it is unfathomably large. Aggregative ethics can thus be reconciled with the finite case if we note that, when sizing up the moral significance of our acts, the relevant consideration is not how big a part they constitute of the whole of the doings and goings-on in the universe, but rather what difference they make in absolute terms.

The infinite case is fundamentally different. Suppose the world contains an infinite number of people and a corresponding infinity of joys and sorrows, preference satisfactions and frustrations, instances of virtue and depravation, and other such local phenomena at least some of which have positive or negative value. More precisely, suppose that there is some finite value ε such that there exists an infinite number of local phenomena (this could be a subset of e.g. persons, experiences, characters, virtuous acts, lives, relationships, civilizations, or ecosystems) each of which has a value ≥ ε and also an infinite number of local phenomena each of which has a value ≤ (‒ ε). Call such a world canonically infinite. Ethical theories that hold that value is aggregative imply that a canonically infinite world contains an infinite quantity of positive value and an infinite quantity of negative value. This gives rise to a peculiar predicament. We can do only a finite amount of good or bad. Yet in cardinal arithmetic, adding or subtracting a finite quantity does not change an infinite quantity. Every possible act of ours therefore has the same net effect on the total amount of good and bad in a canonically infinite world: none whatsoever.

Aggregative consequentialist theories are threatened by infinitarian paralysis: they seem to imply that if the world is canonically infinite then it is always ethically indifferent what we do. In particular, they would imply that it is ethically indifferent whether we cause another holocaust or prevent one from occurring. If any non-contradictory normative implication is a reductio ad absurdum, this one is.

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .