20.02.07 · philosophy / ethics

Consequentialism in depth: act vs. rule utilitarianism, preference satisfaction, effective altruism

stub3 tiersLean: nonepending prereqs

Anchor (Master): Singer, P. — The Most Good You Can Do (2015)

Intuition Beginner

Consequentialism judges actions by their outcomes. The right action is the one that produces the best consequences — the most good for the beings affected. The most influential form is utilitarianism: maximize happiness and minimize suffering for everyone capable of experiencing either.

Jeremy Bentham (1748-1832) treated pleasure and pain as the only intrinsic goods, and he proposed a "hedonic calculus" measuring them along axes such as intensity, duration, certainty, and extent. His student John Stuart Mill (1806-1873) feared this reduced humans to the level of pigs content at a trough. Mill's answer: pleasures differ in kind, not merely amount. Intellectual, moral, and aesthetic pleasures rank higher than bodily ones. "It is better to be Socrates dissatisfied than a fool satisfied."

Two versions split the tradition. Act utilitarianism evaluates each action on its own — do whatever maximizes good in this exact situation, even if that means lying or breaking a promise. Rule utilitarianism instead asks which general rules, if widely followed, would maximize good, then binds you to those rules even when a single breach looks helpful. This act-versus-rule divide runs through every later debate about what consequentialism demands and permits.

Peter Singer's effective altruism brings these ideas down to earth. If saving a child from a shallow pond is obligatory, then donating a few thousand dollars to stop a child dying of malaria should be equally so. The movement urges us to give where each dollar does the most verifiable good — measured by lives saved or suffering averted, not by emotional proximity or vivid imagery.

Visual Beginner

Picture three frameworks lined up. Act utilitarianism inspects each choice in isolation: will this lie, this promise, this donation produce the most good right now? The calculator runs on every decision, every time, with no rule exempt from recalculation.

Rule utilitarianism steps back. It asks which rules — tell the truth, keep your promises, do not steal — would make the world best if nearly everyone internalized them. You follow the rule even when breaking it once might help, because the rule's value lies in its general acceptance, not in any single application.

Preference utilitarianism changes the currency. Instead of tallying pleasure, it asks whether beings get what they want — their preferences satisfied — across humans and animals alike. The diagram contrasts the three aggregation strategies against the same population.

Each panel answers the same question — "what makes an act right?" — with a different level of description: the individual act, the social rule, the satisfied preference.

Worked example Beginner

Singer's shallow-pond case is the gateway to effective altruism. Imagine walking past a child drowning in a shallow pond. Saving her costs you only wet clothes and a missed appointment. Almost everyone agrees you are morally required to wade in. The cost is small; the benefit — a child's life — is enormous.

Singer's move: distance and number do not change the moral picture. Children die every day from poverty-related causes that a few thousand dollars would prevent. If you would rescue the child in the pond, you should give comparably small amounts to save children you cannot see. Refusing to count proximity as morally relevant is the engine of the argument.

Act utilitarianism pushes the logic to its ceiling: keep giving until the last dollar you retain does as much good in your hands as it would in the hands of the poorest person you could help. That ceiling feels impossibly high. The demandingness objection — is morality really this costly? — becomes the central worry, and it sets up the deeper debates of the intermediate and master tiers.

Rule utilitarianism offers a softer landing. A rule like "give a meaningful fraction of your income to effective charities" is one most people could internalize without collapsing their lives; its general adoption would do enormous good, yet it stops short of the act-utilitarian demand to give to the point of marginal equivalence. The same example, read through the two lenses, yields two different duties.

Check your understanding Beginner

Exercise (easy, multiple choice).

The core difference between act and rule utilitarianism is:

A. Act utilitarianism follows fixed moral rules; rule utilitarianism calculates per case. B. Act utilitarianism assesses each action by its direct consequences; rule utilitarianism follows rules whose general adoption would maximize good. C. Act utilitarianism counts only the agent's happiness; rule utilitarianism counts everyone's. D. The two theories are identical under different names.

Hint

One theory asks "what outcome does this act produce?" The other asks "what would happen if this rule became standard?"

Answer

Option B. Act utilitarianism runs the calculation fresh on each action; rule utilitarianism evaluates rules by the consequences of their general internalization, then binds you to the optimal rule-set even where a one-off breach would help.

Exercise (easy, short answer).

In two or three sentences, state Mill's "higher pleasures" doctrine and why Mill thought it answered the charge that utilitarianism is a "doctrine worthy only of swine."

Hint

Mill agreed pleasures differ, but he denied they differ only in quantity. What second dimension did he introduce, and who did he trust to rank pleasures?

Answer

Mill held that pleasures differ in kind as well as amount: intellectual, moral, and aesthetic pleasures are qualitatively higher than merely bodily ones. Competent judges — those who have genuinely experienced both — reliably prefer the higher kinds, which is why "it is better to be Socrates dissatisfied than a fool satisfied." This lets utilitarianism value a rich human life without abandoning the claim that happiness is what matters.

Exercise (easy, multiple choice).

Singer's shallow-pond argument is designed to establish that:

A. Distance and number reduce our obligation to help distant strangers. B. If we are obligated to save a nearby drowning child at small cost, we are comparably obligated to help distant children we could save at small cost. C. Only states, not individuals, have duties to the global poor. D. Morality never demands more than the law requires.

Hint

Singer explicitly rejects moral relevance of proximity. What follows if the child in the pond and the child abroad are morally on a par?

Answer

Option B. Singer argues the only morally relevant features are the badness of the outcome and the cost of preventing it — not physical or emotional distance. If the pond rescue is required, a donation of comparable cost that achieves a comparable good is required too.

Exercise (medium, short answer).

Why might someone find act utilitarianism "too demanding"? Give the strongest version of the worry you can in three or four sentences.

Hint

Consider what the theory says you must do with every spare dollar, every spare hour. Is there any room left for personal projects?

Answer

Act utilitarianism seems to demand that you keep redirecting resources toward whoever benefits most until you and the neediest stranger are at the margin equally placed. That leaves little room for friendships, hobbies, career ambitions, or family commitments that do not maximize aggregate good. The worry is not that helping is bad, but that a morality with no agent-centered prerogative is alienating — it treats your life as a neutral conduit for utility rather than as your own.

Formal definition Intermediate+

Consequentialism has a precise value-theoretic core, and the major disputes — act versus rule, hedonic versus preference, total versus average — are best read as choices within one formal structure.

Definition (Consequentialist criterion). A moral theory is consequentialist when the deontic status of an action depends only on the value of its outcome. Fix a feasible action set $A$ , an outcome map $o : A \to O$ , and a value ranking $V : O \to R$ . The right actions are exactly the optima

ar g a \in A max V (o (a)) .

What separates consequentialist theories is the choice of the currency $V$ and the level at which $V$ is applied — to individual acts or to rules.

Definition (Act utilitarianism). Let $u_{i} (a)$ denote the well-being of individual $i$ under action $a$ . The act-utilitarian right-making property selects

a^{*} \in ar g a \in A max i = 1 \sum n u_{i} (a) .

Each action is judged by its direct contribution to aggregate well-being, with no rule insulated from recalculation ^{[Smart 1973]}.

Definition (Rule consequentialism). A moral code $C$ is a set of action-guiding rules. Let $U (C)$ denote the expected aggregate well-being were $C$ internalized by most members of a society over time. The rule-consequentialist selects the optimal code

C^{*} \in ar g C max U (C),

and an action is right if and only if it is permitted by $C^{*}$ ^[Hooker]. Smart's "rule worship" objection charges that rule consequentialism is merely act consequentialism with an unnecessary detour; Hooker replies that the level shift is principled, because the rules that maximize welfare when internalized are not the rules an act-utilitarian would endorse case by case.

Definition (Preference utilitarianism). Replace hedonic well-being $u_{i}$ with $s_{i} (a)$ , the degree to which person $i$ 's informed preferences are satisfied under $a$ . Social value becomes $W = \sum_{i} s_{i} (a)$ , now extended to every being whose preferences count — including non-human animals ^[Singer]. Hare and Singer prefer this currency because it respects what beings actually want rather than imposing an external account of "pleasure."

Definition (Social welfare function — total vs average). For a population of $n$ with well-being vector $(u_{1}, \dots, u_{n})$ the two classical aggregators are

W_{total} = i = 1 \sum n u_{i}, W_{avg} = \frac{1}{n} i = 1 \sum n u_{i} .

They agree when $n$ is fixed; they diverge sharply once population size is itself a variable, which is the entry point to Parfit's population ethics.

Counterexamples to common slips. First, consequentialism is not egoism: the value function ranges over every affected being, and the agent's own well-being carries no special weight in the aggregation. Second, "maximize the good" is not a policy of always pursuing the most pleasant option: the good being maximized is aggregate, often requiring personal sacrifice. Third, rule consequentialism is not a synonym for rule-of-thumb act consequentialism: a rule-consequentialist can be committed to a rule even when no act-utilitarian calculation would endorse the particular act it prescribes, because the commitment is to the code whose general internalization is optimal.

Key argument — Singer's pond argument and the demandingness objection Intermediate+

Singer's "Famine, Affluence, and Morality" (1972) ^[Singer] is the canonical reconstruction of the argument that drives effective altruism. Its power lies in deriving an extremely demanding conclusion from premises that are hard to deny.

Premise 1 (Badness of suffering). Suffering and death from shortage of food, shelter, and medical care are bad.

Premise 2 (Strong principle). If we can prevent something bad from happening without thereby sacrificing anything of comparable moral importance, we ought to do it. The principle makes no exception for proximity, for the number of other people who could help, or for whether the sufferer is nearby.

Premise 3 (Efficacy of aid). By donating to effective aid agencies we can prevent some of that suffering and death, and the cost to us is not of comparable moral importance to the death of a child.

Conclusion. We are morally obligated to donate to aid agencies, and to keep donating up to the point at which what we sacrifice reaches the moral weight of what we prevent — roughly, down to the level where our marginal dollar does as much good kept as given.

The argument is valid. Nearly everyone grants Premise 1. The action is in Premise 2, the strong principle, and in Premise 3's claim that aid genuinely works. Singer offers a weaker substitute for Premise 2 — preventing bad outcomes without sacrificing anything morally significant — which yields a softer duty but still far more than common-sense morality condones.

The demandingness objection. The conclusion asks for a radical redirection of resources and life plans. Scheffler's agent-centered prerogatives (1982) respond that morality may permit each person to give disproportionate weight to her own projects, so that the cost of helping need not rise to the moral level of the harm prevented before one is allowed to stop. Wolf's "moral saints" worry (1982) adds that a life optimized for impartial beneficence leaves no room for the qualities — humor, style, particular attachments — that make a life worth living. Singer's reply reframes the burden: the ability to do enormous good cheaply is an opportunity, and scope sensitivity — letting the number of lives at stake actually matter — is a feature, not a defect.

Bridge. Singer's argument builds toward the population ethics of the master tier and appears again in Parfit's non-identity problem, where the very beings whose suffering Premise 1 condemns might not exist absent the policies that harm them. The foundational reason the pond argument is so unsettling is that it accepts no morally relevant difference between the nearby child and the distant one; putting this together with rule consequentialism and preference satisfaction, the central insight is that consequentialism relentlessly aggregates — and aggregation is exactly where the deepest objections (Williams on integrity, Rawls on separateness of persons, Taurek on whether numbers count) strike back.

Exercises Intermediate+

Exercise 1 (easy, argument reconstruction).

Reconstruct Smart's "rule worship" objection to rule utilitarianism. State what rule utilitarianism prescribes, what Smart thinks an act utilitarian should do instead, and why Smart calls the rule utilitarian's adherence "worship."

Hint

If breaking a rule once would produce more good than following it, what does the act utilitarian do? What does the rule utilitarian do? Why does Smart think the rule utilitarian is honoring the rule for its own sake?

Answer

The rule utilitarian follows the rule that belongs to the code whose general internalization maximizes welfare, even when, in a particular case, breaking that rule once would produce more good. The act utilitarian instead calculates directly and breaks the rule. Smart calls the rule utilitarian's stance "rule worship" because it treats the rule as a constraint even when one knows that suspending it this once would make the world better — as though the rule had intrinsic authority rather than derivative, consequence-based authority. The objection presumes that the only reason to follow a rule is the goodness of its consequences, so knowingly obeying a suboptimal rule in a particular case is irrational.

Exercise 2 (medium, application).

A government must allocate a fixed health budget between two programs. Program A extends the lives of 100 patients with a chronic disease by five years each. Program B cures 1000 cases of a curable tropical disease, restoring each recipient to a full life. Using the total-utilitarian aggregator $W_{total} = \sum_{i} u_{i}$ , argue which program should be funded. Then state one respect in which a rule consequentialist might reach a different conclusion.

Hint

Compare the aggregate well-being gained. Then ask: what rule, if generally internalized by health systems, would maximize welfare — and might that rule protect the chronically ill for reasons of trust or insurance?

Answer

On total utilitarian grounds Program B dominates: it produces a much larger aggregate well-being gain (1000 full lives restored versus 500 life-years extended), so $W_{total}$ is maximized by funding B. A rule consequentialist might nonetheless endorse a rule protecting the chronically ill — for instance, "honor continuing care commitments" — on the grounds that a society in which such commitments are reliable generates more welfare over time through reduced fear, higher trust in institutions, and wider risk-taking (knowing one will not be abandoned). The act-utilitarian calculation favors B case by case; the optimal internalized code may favor A.

Exercise 3 (medium, comparison).

Contrast how a hedonistic utilitarian and a preference utilitarian would evaluate a person who, after reflection, prefers a life of demanding creative struggle that yields little pleasure. Each perspective should receive two or three sentences.

Hint

The hedonist ranks lives by the balance of pleasure over pain. The preference utilitarian ranks lives by whether informed preferences are met. Where do they come apart?

Answer

Hedonistic utilitarian: the artist's life looks comparatively low in value, because its daily texture is frustration and fatigue with only intermittent reward; whatever the artist's commitment, the hedonic ledger is thin. A hedonist would judge that a more tranquil life would have been better for this person.

Preference utilitarian: the artist's informed preferences are satisfied by the struggle itself — the difficulty is part of what was wanted — so preference satisfaction is high even though pleasure is not. The two theories diverge precisely because "good for someone" can mean "feels good" or "gets what they want," and these come apart for lives organized around difficult goods.

Exercise 4 (hard, critique).

State the strongest objection you can to Premise 2 (the strong principle) of Singer's pond argument. Then give the most forceful Singerian reply, and indicate where the exchange remains unresolved.

Hint

Consider whether morality must be impartial in the way the premise assumes, and whether an agent is permitted to count her own interests for more than one $n$ -th.

Answer

Objection (agent-centered prerogative, after Scheffler). Premise 2 demands strict impartiality between the agent's interests and strangers' interests, but a plausible morality must permit each person to give her own projects, relationships, and well-being some independent weight — otherwise the theory treats the agent as a mere utility-conduit and is, as Williams argues, alienating. The prerogative need not be large; it need only exist, so that one is allowed to stop giving before the marginal cost equals the marginal benefit.

Singerian reply. Once we grant that the suffering is bad and that we can prevent it cheaply, the burden falls on the objector to say why proximity, identity, or mere preference should exempt us. Agent-centered prerogatives look like licensed partiality — a permission to let strangers die so that we may enjoy conveniences. Scope sensitivity demands that numbers count; the prerogative licenses us to discount them.

Unresolved residue. The exchange turns on whether impartiality is constitutive of morality or merely one value among others. That is not a question Singer's argument can close from within; it requires the independent case for or against impartiality that deontology (the next unit) makes its centerpiece.

Effective altruism and longtermism Master

Effective altruism (EA) is the applied consequentialism of the past fifteen years. Singer's The Life You Can Save (2009) and The Most Good You Can Do (2015) ^[Singer] argue that giving should be guided by evidence of cost-effectiveness rather than by emotional pull. GiveWell-style evaluations rank interventions by lives or well-being saved per dollar: the Against Malaria Foundation (insecticidal bed nets), deworming pills, and unconditional cash transfers recurrently top the list because each averts a large burden of suffering at low cost. MacAskill's Doing Good Better (2015) and the 80,000 Hours project extend the framework from donation to career choice, arguing that the most impactful decision a talented person makes is which problem to work on.

Longtermism pushes the same logic across time. Bostrom's "Astronomical Waste" (2003) argues that the expected value of humanity's future — potentially trillions of flourishing lives across billions of years — dwarfs present-day concerns, so even small reductions in existential risk dominate the consequentialist calculus. Ord's The Precipice (2020) catalogues the risks (nuclear war, engineered pandemics, unaligned artificial intelligence, climate collapse) and argues that reducing them is the highest-leverage moral work available. On this view, the pond argument generalizes: if the distant future contains vast numbers of beings whose suffering we can cheaply avert, their claims are no less pressing for being temporally remote.

The criticisms bite from two directions. The first is structural: EA's focus on measurable outcomes may neglect systemic and political change, favoring charity that soothes symptoms over transformation that removes causes — a tension sometimes glossed as relief versus revolution. Berkey presses this charge, arguing that EA's methodological individualism and fondness for randomized controlled trials render it technocratic, unable to confront the power structures that generate poverty in the first place. The second is epistemic: longtermism in particular rests on speculative forecasts about the far future, and Torres argues that the license to act on astronomical expected value can be morally dangerous, licensing almost any present sacrifice in the name of vast hypothetical futures. The defender's reply is that uncertainty cuts both ways — expected-value reasoning under deep uncertainty still favors action when the downside of inaction is existential — but the hubris charge remains live.

The deepest tension inside EA is the demandingness inheritance. EA takes over Singer's pond argument wholesale, so it inherits the agent-centered-prerogative and moral-saints objections. The movement's practical answer — earn to give, pledge a fraction of income, pursue high-impact careers — is essentially a rule-consequentialist compromise: a code most people could internalize, whose general adoption would do enormous good, without the act-utilitarian demand to give to the point of marginal equivalence.

Williams, integrity, and the separateness of persons Master

Bernard Williams's half of Utilitarianism: For and Against (1973) ^[Williams] mounts the most influential internal critique of consequentialism. His central charge is that utilitarianism alienates the agent from integrity.

The Jim and the Indians case. Jim stumbles upon a bandit captain about to execute twenty innocent villagers. The captain offers Jim a guest's privilege: if Jim himself shoots one villager, the other nineteen go free; if Jim refuses, all twenty die. Act utilitarianism says Jim must shoot — one death produces less disvalue than twenty. Williams argues this verdict distorts moral psychology. For Jim to shoot is for Jim to become, through his own intentional hand, the author of a death that is his project in a way the twenty deaths are not. Utilitarianism, by treating Jim's agency as a neutral valve for steering outcomes, effaces the distinction between what one does and what one merely allows.

Negative responsibility. Williams targets the utilitarian doctrine that you are as responsible for what you allow as for what you do. If you refuse to shoot and twenty die, the utilitarian says those deaths are on your ledger. Williams rejects the picture: a person's commitments, projects, and integrity constitute the standpoint from which she acts, and a theory that overrides those whenever the numbers demand it does not leave her with a life that is recognizably her own.

The separateness of persons. Rawls presses a structurally allied objection: classical utilitarianism "does not take seriously the distinction between persons." It aggregates utilities across distinct individuals as seamlessly as a single person aggregates gains and losses across the stages of her own life, but there is no actual being who enjoys the aggregate — only separate persons, each with her own life to lead. Nozick sharpens the point with the utility monster: a being that converts resources into utility vastly more efficiently than others would, under utilitarianism, receive everything, leaving everyone else worse off — a reductio of unfettered aggregation.

Aggregation and whether numbers count. Taurek's "Should the Numbers Count?" (1977) denies that numbers matter as such: when you can save one or five, each person has an equal claim on your aid, and you may permissibly flip a coin rather than save the five. Consequentialists insist the five have a stronger aggregate claim. Scanlon's contractualism offers a middle path: what matters is not raw aggregation but the strength of each individual's complaint against the chosen principle. Kamm and Otsuka refine the terms on which aggregation is permissible. The separateness-of-persons worry is the common root: consequentialism's willingness to trade one person's good for another's is precisely what its rivals find unacceptable, and what its defenders must justify rather than assume.

Population ethics: Parfit's repugnant conclusion Master

Parfit's Reasons and Persons (1984) ^[Parfit] transformed population ethics by showing that the innocent-looking axioms of total utilitarianism generate deeply counterintuitive verdicts about how many people there should be.

The repugnant conclusion. For any possible population of very happy people, there is a larger population — each of whose members has a life barely worth living — whose total well-being is greater. Total utilitarianism therefore ranks the vast, barely-worth-living population above the smaller, deeply flourishing one. The conclusion feels repugnant: it seems to recommend replacing a world of richly happy lives with a teeming world of marginal ones merely because there are more of them.

The non-identity problem. Many policies that harm future generations change who will exist: a risky energy policy may cause harms a century hence, but the people harmed would not have existed without that policy. No particular person is made worse off than they otherwise would have been — they owe their existence to the very policy that shortens their lives. Standard person-affecting morality, which condemns only acts that make someone worse off, cannot reach these cases. Parfit concludes that morality must be impersonal: it can be bad that worse-off people exist, even when the policy harms no one who would otherwise have been better off.

The mere addition paradox. Adding happy people to a population, without making anyone worse off, seems to make things better; yet iterating "mere addition" plus Pareto-style reasoning drives you step by step to the repugnant conclusion. Each step looks acceptable; the destination does not.

Average vs total, and the escape attempts. Average utilitarianism avoids the repugnant conclusion but pays a different price: killing below-average people raises the average, and adding very happy people to a still-happier population can lower the average. Parfit's own later work (On What Matters) sought a "Theory X" that respects our intuitions about both quantity and quality without collapse; the search remains open. Negative utilitarianism (Popper, Smart) prioritizes the reduction of suffering over the creation of happiness, but risks a pathological implication: if minimizing suffering is the sole aim, a world with no sentient life — or a painlessly ended world — might rank best. Attfield and others reply that the view can be formulated to block this, but the worry illustrates how sensitive population axiology is to small changes in the aggregator.

Rule consequentialism, motives, and animal ethics Master

Two-level moral thinking. Hare's Moral Thinking (1981) resolves the act-versus-rule tension by splitting moral cognition into two levels. At the intuitive level we follow everyday rules — do not kill, keep promises, tell the truth — because following them reliably produces good outcomes and because constant calculation is impossible. At the critical level, when intuitions conflict or circumstances are extraordinary, we reason as act utilitarians. The architecture parallels Kohlberg's stage theory in psychology and lets utilitarianism honor both the flexibility of act consequentialism and the stability of rule-governed common-sense morality.

Rule consequentialism as a standalone theory. Hooker's Ideal Code, Real World (2000) defends rule consequentialism not as a detour from act consequentialism but as a distinct criterion: an act is right if and only if it is permitted by the code whose internalization by most people would maximize expected aggregate well-being. Hooker argues this code converges substantially with common-sense morality — prohibitions on killing the innocent, requirements of fidelity and gratitude, a duty of rescue — and that it handles demandingness more plausibly than act utilitarianism, because no one is required to internalize a code that would devastate her own life. The standing challenge is whether rule consequentialism collapses back into act consequentialism: if the justification of every rule is its expected-welfare payoff, why not recalculate case by case? Hooker's answer is that internalizing a code has systematic benefits — predictability, coordination, trust — that case-by-case calculation would destroy.

Consequentialism about motives. A separate question is what motives a consequentialist should cultivate. Driver and Adams develop motive consequentialism: the right motive is the one whose general possession produces the best consequences, even when acting on it occasionally yields a worse outcome than a better-informed motive would have. The worry is the inverse: wrong motives can produce right actions, and a theory that evaluates only outcomes seems silent on whether the agent deserves credit. The two-level view resolves this by evaluating motives at the intuitive level and outcomes at the critical level.

Animal ethics. Singer's Animal Liberation (1975) applies preference utilitarianism across species: the capacity for suffering, not species membership, grounds moral consideration, so the routine use of animals in factory farming and experimentation is, on consequentialist grounds, a catastrophe of suffering. "Speciesism" is the unjustified privileging of human interests merely because they are human. Regan's The Case for Animal Rights (1983) reaches allied conclusions by a non-consequentialist route, grounding strong animal rights in being "subjects-of-a-life." Francione's abolitionist theory presses that welfarist reforms — bigger cages, humane slaughter — entrench the property status of animals rather than dissolving it, and so are condemned even on broadly consequentialist grounds if they slow the path to abolition. The disagreement between Singer (reduce suffering now), Regan (respect inherent rights), and Francione (abolish the institution) reproduces, within animal ethics, the broader act-versus-rule and consequentialist-versus-deontologist fault lines.

Connections Master

Theories of justice 20.02.01 is the direct prerequisite. Rawls's separateness-of-persons objection to utilitarian aggregation and the maximin-versus-expected-utility dispute with Harsanyi are the bridge from this unit's aggregation debates back to distributive justice. Consequentialism supplies the target Rawls and Nozick both define themselves against.
Rights theory 20.02.02 (pending) connects through the standing consequentialist worry that rights are at best rules of thumb within a deeper utility calculus. Mill's On Liberty defends rights on consequentialist grounds; Nozick and Dworkin reject that grounding as unstable. The integrity and separateness-of-persons objections in this unit are the seed of those rights-based replies.
Moral dilemmas and the trolley problem 20.02.04 (pending) is the laboratory where act-versus-rule and aggregation debates are stress-tested. Taurek's "should the numbers count" and the Kamm-Scanlon exchange on aggregation appear there in concrete case form.
The good life and eudaimonia 20.02.05 (pending) connects through the well-being currency dispute: hedonism, preference satisfaction, and objective-list theories of the good are the value functions over which consequentialism optimizes. Nozick's Experience Machine originates there and is deployed against hedonistic utilitarianism here.
Ethics of artificial intelligence 20.02.06 connects through alignment and the orthogonality thesis: instrumental convergence is itself a consequentialist argument about what any sufficiently capable optimizer will do, and the demandingness of longtermism feeds directly into AI existential-risk discourse.
Deontology and Kantian ethics 20.02.08 pending (pending, successor) is the natural successor. Williams's integrity objection, the separateness of persons, and the aggregation debates all set up the deontologist's insistence that certain acts are wrong regardless of consequences, and that persons must be treated as ends in themselves.

Cross-domain to economics: social-choice theory (Arrow, Sen) formalizes the aggregation problem this unit raises in philosophical form, and Harsanyi's social-welfare theorem is the formal counterpart of the total-versus-average dispute.

Historical and philosophical context Master

Consequentialism's modern history opens with Bentham's An Introduction to the Principles of Morals and Legislation (1789) ^[Mill], which announced that nature has placed mankind under the governance of two sovereign masters, pain and pleasure, and built the hedonic calculus to make good on the claim. Bentham was a reformer: the calculus was meant to rationalize law, prisons (the panopticon), and social policy by replacing inherited authority with measured outcomes.

Mill's Utilitarianism (1863) both popularized and qualified the doctrine ^[Mill]. Against the "doctrine worthy only of swine" charge, Mill introduced the qualitative distinction among pleasures and the test of competent judges; against the charge that utilitarianism is godless, he grounded the ultimate desirability of happiness in the empirical fact that people do desire it. Mill is often read as a rule utilitarian (especially in On Liberty's defense of rights as utility-maximizing protections), which makes him a natural ancestor of Hooker's contemporary rule consequentialism.

The twentieth-century revival came through Moore's Principia Ethica (1903), which argued that "good" is unanalyzable and attacked "naturalistic" definitions of welfare, and through the careful act-utilitarian formalism of Smart and Hare. Smart's contribution to Utilitarianism: For and Against (1973) ^[Smart] crystallized the act-utilitarian position and the "rule worship" critique; Williams's companion essay supplied the integrity objection that has shaped normative ethics ever since. Hare's two-level theory (Moral Thinking, 1981) was the mediating response, splitting act and rule reasoning across cognitive levels.

The decisive late-twentieth-century event was Parfit's Reasons and Persons (1984) ^[Parfit], which relocated the hardest problems from interpersonal ethics to population axiology. The repugnant conclusion, the non-identity problem, and the mere addition paradox showed that the aggregation intuitions driving classical utilitarianism are mutually inconsistent, and Parfit's unfinished search for "Theory X" set the agenda for a generation of work on population ethics (Arrhenius, Greaves, Cowen).

The twenty-first century brought consequentialism into applied practice. Singer's Animal Liberation (1975) had already made preference utilitarianism the philosophical engine of the animal-rights movement; his "Famine, Affluence, and Morality" (1972) became, decades later, the founding text of effective altruism. GiveWell (2007), 80,000 Hours, and the Open Philanthropy Project turned Singer's argument into institutions. Bostrom's "Astronomical Waste" (2003) and Ord's The Precipice (2020) extended the logic to existential risk and longtermism, while MacAskill's What We Owe the Future (2022) made the philosophical case for counting the far future heavily in present deliberation.

The contemporary field is marked by three live disputes: the demandingness inheritance (how much can morality ask?), the epistemic humility debate (how much should astronomical but uncertain futures weigh?), and the structural critique (does EA's measurement bias it toward measurable, symptomatic relief over systemic change?). Each reproduces, at larger scale, the act-versus-rule and integrity objections that defined consequentialism's twentieth-century theory.

Bibliography Master

Bentham, J. — An Introduction to the Principles of Morals and Legislation (T. Payne, 1789; Oxford ed. 1996).
Mill, J. S. — Utilitarianism (Parker, Son, and Bourn, 1863).
Mill, J. S. — On Liberty (Parker, 1859).
Sidgwick, H. — The Methods of Ethics (Macmillan, 1874; 7th ed. 1907).
Moore, G. E. — Principia Ethica (Cambridge University Press, 1903).
Hare, R. M. — The Language of Morals (Oxford University Press, 1952).
Hare, R. M. — Moral Thinking: Its Levels, Method, and Point (Oxford University Press, 1981).
Smart, J. J. C. & Williams, B. — Utilitarianism: For and Against (Cambridge University Press, 1973).
Parfit, D. — Reasons and Persons (Oxford University Press, 1984).
Parfit, D. — On What Matters (Oxford University Press, 2011).
Singer, P. — "Famine, affluence, and morality", Philosophy & Public Affairs 1(3), 229-243 (1972).
Singer, P. — Animal Liberation: A New Ethics for Our Treatment of Animals (Harper & Row, 1975).
Singer, P. — The Life You Can Save (Random House, 2009).
Singer, P. — The Most Good You Can Do (Yale University Press, 2015).
Scheffler, S. — The Rejection of Consequentialism (Oxford University Press, 1982; rev. ed. 1994).
Wolf, S. — "Moral saints", The Journal of Philosophy 79(8), 419-439 (1982).
Taurek, J. M. — "Should the numbers count?", Philosophy & Public Affairs 6(4), 293-316 (1977).
Hooker, B. — Ideal Code, Real World: A Rule-Consequentialist Theory of Morality (Oxford University Press, 2000).
Driver, J. — "The virtues of ignorance", in Morality and the Emotions (Oxford, 2001).
Adams, R. M. — A Theory of Virtue (Oxford University Press, 2006).
Regan, T. — The Case for Animal Rights (University of California Press, 1983).
Francione, G. L. — Animals, Property, and the Law (Temple University Press, 1995).
MacAskill, W. — Doing Good Better (Avery, 2015).
Bostrom, N. — "Astronomical waste: the opportunity cost of delayed technological development", Utilitas 15(3), 308-314 (2003).
Ord, T. — The Precipice: Existential Risk and the Future of Humanity (Hachette, 2020).
Kamm, F. M. — Morality, Mortality, Vol. I (Oxford University Press, 1993).
Scanlon, T. M. — What We Owe to Each Other (Harvard University Press, 1998).
Rawls, J. — A Theory of Justice (Harvard University Press, 1971).
Nozick, R. — Anarchy, State, and Utopia (Basic Books, 1974).
Mulgan, T. — Demandingness (Routledge, 2018).