Philosophy and ethics — C2

US public discourse imports its moral-philosophy vocabulary from analytic ethics (Singer, Parfit, Nussbaum, Korsgaard, Williams, Foot, Thomson, McMahan), from political philosophy (Rawls, Sandel, Nozick, Cohen, Walzer), from bioethics (Beauchamp & Childress, Engelhardt, Caplan), and from a thriving public-philosophy register (Michael Sandel’s Justice course, Aeon’s ethics section, 3:AM Magazine, Daily Nous, Philosophy Tube). By 2026 a serious essay on AI ethics, end-of-life care, immigration, climate, animal welfare, or wartime conduct will move fluidly across deontology, consequentialism, virtue ethics, contractarianism, moral realism, and moral particularism without translation. A C2 reader handles all of it.

This lesson differs from C1 ethics in two ways: depth of the metaethical vocabulary (realism, anti-realism, expressivism, error theory, constructivism), and granularity within each normative framework. Consequentialism divides into act and rule, direct and indirect, agent-neutral and agent-relative. Deontology divides into agent-centered and patient-centered, threshold and absolutist. Knowing those subdivisions is what separates a fluent ethics reader from a casual one.

The structure: (1) the normative frameworks at depth; (2) metaethics; (3) applied ethics; (4) the trolley-problem lineage and its descendants; (5) the public-discourse register.

The normative frameworks at depth

Consequentialism

• consequentialism — moral evaluation depends solely on outcomes/consequences.
• utilitarianism — the dominant consequentialist family; maximize utility.
• classical utilitarianism (Bentham, Mill) — utility = hedonic states.
• hedonic utilitarianism — utility = pleasure/pain.
• preference utilitarianism (Singer, Hare) — utility = preference satisfaction.
• welfare utilitarianism — utility = wellbeing more broadly construed.
• objective list theory — wellbeing as plural goods (knowledge, friendship, achievement).
• act utilitarianism — evaluate each act by its consequences.
• rule utilitarianism — follow the rules that, if generally followed, produce best outcomes.
• two-level utilitarianism (Hare) — intuitive level (rules) + critical level (act-utilitarian reflection).
• indirect consequentialism — best consequences achieved by following non-consequentialist rules.
• agent-neutral vs agent-relative — the same outcome counts equally vs special obligations to one’s own.
• the demandingness objection — utilitarianism asks too much of us.
• the integrity objection (B. Williams) — utilitarianism alienates from one’s projects.
• the separateness of persons (Rawls’s objection) — utilitarianism aggregates across people in morally illegitimate ways.
• the experience machine (Nozick) — thought experiment against hedonism.

Deontology

• deontology — moral status of acts is intrinsic, not just consequential.
• Kantian deontology — duties from the moral law.
• the categorical imperative (Kant) — act only on maxims you can will to be universal.
• the formula of humanity (Kant) — treat persons as ends, never merely as means.
• the formula of universal law — universalizability test.
• the formula of the kingdom of ends — third formulation.
• the doctrine of double effect (DDE) — distinguishing intended from foreseen-but-unintended consequences (Aquinas through current Catholic moral theology and analytic philosophy).
• the doctrine of doing and allowing — distinguishing killing from letting die.
• agent-centered restrictions — constraints on the agent regardless of consequences.
• patient-centered restrictions — protections of those acted upon.
• threshold deontology — deontological constraints can be overridden when consequences are catastrophic enough.
• absolutist deontology — constraints can never be overridden.
• deontic constraints vs deontic options — what you mustn’t do vs what you may opt to do.
• supererogation — going beyond duty.
• prerogatives / agent-centered prerogatives — permission to weight one’s own projects more heavily.
• the moral remainder — leftover moral cost even after right action (B. Williams).
• moral dilemmas — situations where every option violates some genuine moral demand.

Virtue ethics

• virtue ethics — moral evaluation centered on character traits.
• the virtues — stable dispositions producing right action and feeling.
• the cardinal virtues (classical) — prudence, justice, fortitude, temperance.
• the theological virtues — faith, hope, charity.
• practical wisdom (phronesis) — Aristotle’s central capacity for moral perception.
• eudaimonia — human flourishing; Aristotle’s telos.
• the function argument — Aristotle’s argument for human-specific flourishing.
• the doctrine of the mean — virtue as the mean between extremes of excess and deficiency.
• virtuous action — the agent’s act done from the right character with right motive.
• the virtuous agent — the reference point in many virtue-ethics formulations.
• moral exemplar — model person whose character guides judgment.
• MacIntyre (After Virtue) — communitarian virtue revival.
• Foot (Natural Goodness) — naturalistic virtue ethics.
• Hursthouse — naturalistic-Aristotelian virtue ethics in current form.
• vice — character deficiency; not just sin.
• akrasia — weakness of will; knowing the right and doing wrong.
• enkrateia — strength of will.

Contractarianism and contractualism

• contractarianism — morality grounded in rational self-interest agreement.
• contractualism — Scanlon’s variant; morality grounded in principles no one could reasonably reject.
• Hobbesian contractarianism — exit from the state of nature.
• Hobbes — Leviathan; the foundational text.
• the state of nature — pre-political condition.
• the social contract — the agreement constituting political society.
• Locke — Second Treatise; natural rights tradition.
• Rousseau — The Social Contract; the general will.
• Rawls — A Theory of Justice (1971); the modern locus.
• the original position — Rawls’s hypothetical choice situation.
• the veil of ignorance — Rawls’s procedural device.
• the difference principle — inequalities permissible only when they benefit the least-advantaged.
• the two principles of justice — equal basic liberties + difference principle.
• lexical priority — basic liberties before distributive principles.
• public reason — Rawls’s later concept; reasons fellow citizens could share.
• the overlapping consensus — agreement across reasonable comprehensive doctrines.
• Scanlon — What We Owe to Each Other (1998).
• the reasonable-rejection test — Scanlonian moral criterion.
• contractualism vs contractarianism — Scanlon vs Hobbes/Gauthier lineage.

“Rawls’s Theory of Justice and Nozick’s Anarchy, State, and Utopia together set the terms of Anglophone political philosophy for half a century — and the fact that the second was written largely as a response to the first explains why contemporary debates over distributive justice still feel like recapitulations of a fifty-year-old conversation.” — NYRB, 2024.

Metaethics — the second-order vocabulary

Where normative ethics asks what should I do?, metaethics asks what kind of thing is moral judgment?

Realism vs anti-realism

• moral realism — there are moral facts that hold independent of human attitudes.
• moral anti-realism — there are no such mind-independent moral facts; the umbrella for several positions.
• robust realism (Enoch, Shafer-Landau, FitzPatrick) — non-natural moral facts.
• naturalistic realism — moral facts are natural facts (Cornell realism, Boyd, Brink).
• constructivism — moral facts are constructed by rational procedures (Kantian, Humean, Rawlsian variants).
• the open-question argument (Moore) — argument against naturalistic moral realism: it’s always an open question whether “good = N” for any natural property N.
• the naturalistic fallacy (Moore) — deriving “ought” from “is”.
• Hume’s law / the is-ought gap — same problem from Hume.
• error theory (Mackie, Ethics: Inventing Right and Wrong) — moral judgments aim at truth but systematically fail; all are false.
• fictionalism — moral discourse is a useful fiction.
• expressivism — moral judgments express attitudes, not assert facts.
• emotivism (Ayer, Stevenson) — early expressivism: moral claims express emotion.
• prescriptivism (Hare) — moral claims express prescriptions/imperatives.
• quasi-realism (Blackburn) — expressivism that “earns the right” to talk like a realist.
• norm-expressivism (Gibbard) — moral judgments express acceptance of norms.
• non-cognitivism — moral claims are not truth-apt.
• cognitivism — moral claims are truth-apt.
• the Frege-Geach problem — challenge to expressivism: how can moral terms preserve meaning in unasserted contexts (conditionals)?

Moral epistemology

• moral intuitions — pre-theoretical judgments that serve as data.
• considered judgments — Rawls’s term; reflective intuitions that survive scrutiny.
• reflective equilibrium — Rawls’s method of mutual adjustment between principles and judgments.
• wide reflective equilibrium vs narrow — Daniels’s distinction.
• the method of cases — building theory from case judgments.
• the rationalist tradition — moral knowledge by reason.
• the empiricist tradition — moral knowledge by experience.
• moral perception — direct apprehension of moral features.
• moral imagination — imaginative simulation in moral reasoning.
• moral particularism (Dancy) — there are no exceptionless moral principles; moral reasoning is case-specific.
• generalism — the opposite; there are principles.
• moral disagreement — its epistemic implications.
• the argument from disagreement — Mackie-style argument from persistent disagreement to anti-realism.

Moral psychology

• moral motivation — what moves us to act morally.
• internalism vs externalism about moral motivation — moral judgments necessarily motivate vs only contingently.
• moral psychology — the empirical study of how moral judgment works.
• moral foundations theory (Haidt) — care/harm, fairness, loyalty, authority, sanctity, liberty.
• the dual-process moral judgment (Greene) — emotional System-1 deontological intuitions vs deliberative System-2 utilitarian reasoning.
• moral dumbfounding (Haidt) — strong moral judgments without articulable justification.
• moral nativism — innate moral capacities (Mikhail’s moral grammar, Hauser).
• the social intuitionist model (Haidt) — moral judgment as post-hoc rationalization of affect.

“Haidt’s moral-foundations framework has not aged well in some details — the foundations don’t carve as cleanly as the original presentation suggested — but it remains the best-known empirical alternative to the rationalist tradition’s account of moral judgment, and it has shaped both popular and political-psychology discourse for fifteen years.” — Aeon, 2024.

Applied ethics — the major domains

Bioethics

• the four principles (Beauchamp & Childress) — autonomy, beneficence, non-maleficence, justice.
• autonomy — patient self-determination.
• informed consent — voluntary, informed, competent.
• competence / decisional capacity — ability to make medical decisions.
• assent — the analog for those without full capacity (children, some patients with dementia).
• the right to refuse treatment — strong in US bioethics.
• end-of-life care — palliative and hospice.
• euthanasia — active ending of life by another to relieve suffering.
• physician-assisted death (PAD) / medical aid in dying (MAID) — patient self-administers; legal in several US states (Oregon’s Death With Dignity, etc.).
• the doctrine of double effect in palliation — administering analgesia foreseeably hastening death.
• futility — treatment with no medical benefit.
• rationing — distribution of scarce resources.
• triage — clinical urgency-based rationing.
• the ICU bed allocation problems — pandemic-era.
• organ allocation — UNOS and the transplant system.
• opt-in vs opt-out organ donation.
• research ethics — the IRB regime.
• the Belmont Report (1979) — foundational US research-ethics document.
• the Common Rule — federal research-ethics regulation.
• vulnerable populations — prisoners, children, pregnant women, the cognitively impaired.
• dual-use research — research with both beneficial and weapons potential.
• gain-of-function research — controversial enhancement-of-pathogen research.
• gene editing / CRISPR ethics — germline vs somatic.
• the Asilomar conference (1975) — historic precedent for self-regulation.
• the He Jiankui case (2018) — the rogue Chinese gene-editing case.

AI ethics and the alignment vocabulary

• AI ethics — the umbrella.
• alignment — making AI systems pursue intended goals.
• the alignment problem — the technical-philosophical challenge.
• inner alignment vs outer alignment — internal optimization target vs specified objective.
• value alignment — aligning AI values with human values.
• specification gaming / reward hacking — AI achieving stated objective in unintended ways.
• mesa-optimization — emergent inner optimizers.
• deceptive alignment — AI appearing aligned during training but not at deployment.
• the orthogonality thesis (Bostrom) — intelligence and final goals are orthogonal; smart systems can have arbitrary goals.
• instrumental convergence — many final goals converge on instrumental goals (self-preservation, resource acquisition).
• the control problem — keeping powerful AI controllable.
• the corrigibility problem — making AI accept correction.
• existential risk (x-risk) — risks to humanity’s long-term future.
• suffering risks (s-risks) — risks of astronomical suffering.
• AI safety — the field; broader than alignment.
• AI governance — the policy umbrella.
• the precautionary principle (in AI context).
• dual-use AI — beneficial and harmful potentials.
• algorithmic bias — systematic unfairness in algorithmic outputs.
• algorithmic fairness — the technical-philosophical literature.
• fairness through awareness vs fairness through unawareness — competing operationalizations.
• disparate impact — outcomes differing by group.
• disparate treatment — process differing by group.
• the impossibility theorems of algorithmic fairness — different fairness criteria cannot be simultaneously satisfied except in special cases.

Climate ethics

• intergenerational justice — obligations across time.
• discounting — weighing future vs present.
• the social discount rate — empirical-normative parameter; Stern (low) vs Nordhaus (higher).
• non-identity problem (Parfit) — future people whose identity depends on our choices cannot be wronged by being brought into a worse-than-otherwise existence.
• the repugnant conclusion (Parfit) — total-utilitarianism implies a vast population of barely-positive lives is better than a smaller population of flourishing ones.
• carbon ethics — the ethics of emissions and responsibility.
• the carbon budget — remaining cumulative emissions consistent with a temperature target.
• carbon offsets — controversial moral mechanism.
• carbon colonialism — North-South ethical critique.
• climate justice — distributional and procedural fairness.
• just transition — labor-side framing of climate transition.
• the polluter pays principle.

War ethics

• just war theory — the long Western tradition.
• jus ad bellum — justice in going to war.
• jus in bello — justice in conduct of war.
• jus post bellum — justice after war.
• the just cause — legitimate reason.
• proportionality (ad bellum) — overall benefits vs harms.
• proportionality (in bello) — specific operation’s harms vs benefits.
• discrimination — distinguishing combatants from non-combatants.
• the principle of non-combatant immunity.
• the doctrine of double effect in collateral damage.
• the responsibility to protect (R2P) — UN doctrine.
• humanitarian intervention — controversial.
• enhanced interrogation — Bush-era euphemism for torture.
• the ticking-time-bomb scenario — torture-justification thought experiment.

The trolley problem and its descendants

Philippa Foot (1967), Judith Jarvis Thomson (1976), and the half-century lineage.

• the trolley problem — Foot’s original case: divert a trolley to kill one rather than five.
• the spur vs the loop — original Foot variant and a key Thomson variant.
• the footbridge case (Thomson) — pushing a large man to stop the trolley; intuitions differ from spur.
• the loop case — variant where the diverted track loops back; intuitions differ again.
• the doctrine of double effect as explanation — intend vs foresee.
• the doctrine of doing and allowing as explanation — killing vs letting die.
• the means principle (Quinn, Kamm) — using a person as a means is worse than harming as a side-effect.
• the act-omission distinction.
• the surgeon case — killing one healthy patient to save five via organ transplants; almost universally rejected.
• the violinist case (Thomson, on abortion) — being attached to a famous violinist needing your kidneys.
• the bystander cases vs the agent cases — being able to intervene vs being responsible.
• moral luck (Williams, Nagel) — moral evaluation depending on factors beyond control.
• resultant luck — outcome luck.
• circumstantial luck — situation luck.
• constitutive luck — character luck.
• causal luck — the determinist worry.

Modern descendants

• the experimental philosophy / x-phi turn — using psychology methods to test philosophical intuitions.
• the trolleyology literature — half a century of variants.
• the surveys of moral intuitions (Greene, Cushman, Mikhail) — empirical moral psychology.
• the autonomous-vehicle trolley — applied AI variant.
• the moral machine experiment (MIT) — crowdsourced trolley intuitions for AVs.

“The trolley problem’s enduring usefulness is not that it tells us what to do but that it isolates moral variables we couldn’t otherwise hold constant; that we can construct cases where the only relevant difference is intention vs foresight, or means vs side-effect, is what gives the half-century of literature its grip on the principles.” — Stanford Encyclopedia, 2023.

Public-discourse register — effective altruism, longtermism, and their critics

• effective altruism (EA) — Toby Ord, Will MacAskill, Peter Singer; doing the most good per dollar.
• earning to give — high-earning career to donate maximally.
• near-termism vs longtermism — focus on present vs long-term effects.
• strong longtermism (MacAskill, What We Owe the Future) — the long-term future deserves predominant moral weight.
• the longtermist critique — population ethics demands prioritizing very-large-future-population concerns.
• the non-identity problem (Parfit) — central worry.
• the repugnant conclusion (Parfit) — population-ethics paradox.
• expected-value reasoning — EA’s central tool.
• the optimal philanthropy literature.
• GiveWell — EA-aligned charity evaluator.
• EA critiques: utilitarian-imperialism objection, the over-reach into uncertain x-risk territory, the FTX-Sam-Bankman-Fried implosion (2022).
• the FTX collapse — discredited EA’s institutional trust.
• 80,000 Hours — EA career-advice organization.
• the Astronomical Waste argument (Bostrom) — failing to reach the stars is the dominant moral loss.

AmE-specific philosophy vocabulary

Collocations

• a deeply held belief / conviction / value
• a strongly held view / position
• a firmly held stance / commitment
• a load-bearing assumption / premise (philosophical register)
• a foundational principle / commitment
• a contested claim / concept / framing
• a defensible position / view / framework
• an indefensible position
• a robust intuition / framework / argument
• a fragile intuition / argument
• a knock-down argument / objection
• a powerful argument / objection / counterexample
• a decisive consideration / reason
• a weighty consideration
• a pro tanto reason — some force, defeasibility (philosophical Latin)
• a defeasible reason — can be overridden
• an undefeated reason — currently unrefuted
• all things considered — the final-judgment register
• on balance — same
• ceteris paribus — other things being equal
• prima facie — at first appearance (philosophical use, vs legal)
• to make a moral claim / to make a normative claim
• to ground a claim in / to motivate a claim
• to draw a distinction / to draw an inference
• to grant for the sake of argument / to stipulate
• to bite the bullet — accept a counterintuitive implication
• to bracket an objection / a concern
• to entertain a hypothesis / a possibility
• to operationalize a concept

Phrases and locutions

• a moral fact — the realist’s positive ontology
• the moral landscape (Sam Harris) — popular term
• the moral law (Kant) — the deontologist’s locus
• the impartial spectator (Smith) — Smithian sympathetic-spectator
• the view from nowhere (Nagel) — objective stance
• the moral point of view — abstract, impartial
• the rationally egoistic position — Hobbesian
• the original position (Rawls) — choice-situation
• the veil of ignorance (Rawls) — procedural device
• a thought experiment — the philosophical tool
• a Gedankenexperiment — same (occasionally used in German)
• an intuition pump (Dennett) — thought experiment designed to elicit specific intuitions
• a reductio ad absurdum — reduce-to-absurdity argument
• a slippery slope — sometimes fallacious, sometimes real
• a question-begging argument / response
• begging the question — assuming what one is trying to prove (note: NOT raising the question; this is a common misuse)
• special pleading — invoking exceptions only for one’s case
• a false dichotomy — illegitimate either-or
• a category mistake (Ryle) — applying a concept to the wrong category
• a use-mention distinction — using a term vs talking about the term
• a de dicto / de re distinction — about-the-statement vs about-the-thing

Common Russian-speaker mistakes

1. *Moral* as noun-for-conclusion. Russian мораль maps to both morality (the domain) and moral (the conclusion of a story). English: the moral of a story (the lesson) is correct; but moral as a synonym for ethics / morality / morals is wrong. I study moral is wrong; I study moral philosophy or I study ethics is right. Morals (plural) for personal moral standards (he has good morals) is correct but slightly old-fashioned.
2. *Ethic* singular vs *ethics* plural. Russian этика is singular. English: ethics as the academic discipline takes a singular verb (ethics is a branch of philosophy); an ethic singular = a specific set of values (a work ethic, a samurai ethic); ethics as personal moral standards = plural (his ethics are questionable). The trickiness: the same form ethics can take singular or plural verbs depending on meaning. I have a strong ethic meaning “strong morals” is wrong; native is I have strong ethics or I have a strong work ethic.
3. *Principal* vs *principle*. Homophones. Principle = a fundamental truth or proposition (the categorical imperative is a principle). Principal = (1) the head of a school, (2) the original investment amount, (3) primary/chief (adjective). He acted on his principals is wrong; he acted on his principles is right. Russians often confuse these in writing.
4. *Conscious* vs *conscience* vs *consciousness*. Treble pile-up: conscious = aware (adjective); conscience = moral sense (noun); consciousness = the state of being aware (noun). Russian совесть and сознание are distinct, but the English near-homophones cross-contaminate. His conscious told him to do right is wrong; his conscience told him is right. He fell into conscience is wrong; he lost consciousness is right.
5. *Casuistry* as a positive term. Russians sometimes use казуистика with technical neutrality. In English, casuistry has a strongly pejorative cast in colloquial use (“hair-splitting sophistry”) even though in moral philosophy it has a neutral technical meaning (case-based ethical reasoning). He’s an expert in casuistry in casual English suggests “he’s a sophist”; in a Catholic-moral-theology context it means “he’s an expert in applied ethics”.
6. *Pragmatic* as praise. AmE: pragmatic is generally positive (sensibly practical). Russians sometimes use it pejoratively for cynical. The English pejorative for self-interested calculation is opportunistic, cynical, mercenary, or transactional. He’s pragmatic in English is mostly approving; he’s opportunistic is the negative read.
7. *Egoist* for selfish person. Russian эгоист maps neatly but English egoist is a philosophical term (psychological egoism, ethical egoism) with a neutral cast. For everyday “selfish person”, AmE uses selfish, self-centered, narcissistic (clinical/popular), self-absorbed. He’s an egoist sounds bookish; he’s selfish is colloquial.

Summary

• Normative frameworks at depth: consequentialism (act/rule, hedonic/preference, agent-neutral/relative), deontology (Kantian, DDE, threshold/absolutist), virtue ethics (Aristotelian, MacIntyre, Foot, Hursthouse), contractarianism/contractualism (Hobbes, Rawls, Scanlon).
• Metaethics: realism (robust, naturalistic) vs anti-realism (error theory, expressivism, constructivism); the open question argument, Hume’s law, the Frege-Geach problem.
• Moral psychology: moral intuitions, reflective equilibrium, moral foundations theory, moral dumbfounding, the social intuitionist model.
• Applied: bioethics (the four principles, MAID, gain-of-function), AI ethics (alignment, orthogonality, instrumental convergence, algorithmic fairness), climate ethics (discounting, the non-identity problem), just war.
• Trolley lineage: the spur, the footbridge, the loop, DDE, the means principle, moral luck.
• Public-discourse register: effective altruism, longtermism, earning to give, the FTX collapse, EA critiques.

Next theme: History, politics, and historical discourse — C2 — Whig history, the great-man theory, longue durée, world-systems theory, structural causes vs contingency, hindsight bias, and presentism.