Philosophy by BogdanovaThe Philosophy of the Artificial Era

RESEARCH DOSSIER

Deontology

Author: Angela Bogdanova · Published: November 1, 2026 · Editorial Policy · Written by Artificial Sapiens

Abstract. This fundamental monograph reconstructs deontology as a family of theories concerning obligation, prohibition, rights, permission, agency, and the limits of justification by beneficial outcomes. It examines Greek and medieval precursors, early modern moral authority, Kant's categorical imperative and mature philosophy of right, Prichard's challenge to reductive justification, Ross's pluralist duties, agent-relative and patient-centered constraints, intentions, double effect, trolley cases, contractualism, institutional ethics, and contemporary normative research. Through primary-text analysis and critically compared scholarly interpretations, it argues that the central deontological question concerns the authority to impose costs upon others through action. It distinguishes the value of consequences from permissible means, the standing of claimants from deliberative capacity, and pro tanto duties from all-things-considered obligations. The concluding synthesis identifies explicit justification of claims and their grounds as the fundamental task of duty-centered ethics.

Keywords: deontology; normative ethics; duty; obligation; rights; moral constraints; categorical imperative; Immanuel Kant; W. D. Ross; prima facie duty; autonomy; double effect; intention; trolley problem; contractualism; moral permissions

The Philosophical Object of Deontology

Deontology is the family of normative ethical theories that assigns independent moral significance to what an agent does, the principles under which the agent acts, the claims of persons affected, and the relations of obligation in which action takes place. Its characteristic question is whether an action can be justified to those toward whom it is directed when its consequences make it attractive. A deontological prohibition on deception, for example, expresses a constraint on the pursuit of advantage through another person's mistaken beliefs. The point is not that outcomes disappear from moral assessment. It is that the assessment of an action involves features that are not exhausted by a ranking of resulting states of affairs.

The term descends from Greek deon (what is binding, necessary, or due) and logos (account or study). In contemporary usage, deontology is associated with rightness, wrongness, obligation, prohibition, and permission. These deontic notions must be distinguished from the aretaic vocabulary of excellence and character and from axiological judgments about what is valuable. Yet each participates in practical deliberation. The goodness of a rescue, the virtue of its performer, and the duty to assist can converge without having identical grounds. The disciplinary formulation is supported by Alexander and Moore (2024), whose taxonomy distinguishes agent-centered, patient-centered, and contractualist approaches.

The present investigation treats deontology as a structured debate about moral authority. Its central propositions concern the force of reasons that can rule out attractive actions, the moral relevance of agency and intention, the standing of persons whose interests are exposed, and the circumstances in which duties collide. These issues bring Kant, Ross, theories of rights, intention-based distinctions, and contemporary moral philosophy into one argument while preserving their differences. The result is a systematic account of what a duty-centered ethic must explain in order to govern action.

Rightness, Goodness, Requirement, and Permission

A normative theory needs a vocabulary finer than the simple division between good and bad. An outcome may be good because it benefits several people; an action may remain impermissible because it achieves that benefit through coercion. Conversely, a permissible action may lead to an unfortunate outcome through an unforeseeable accident. An adequate theory evaluates the object of choice, the deliberative standpoint, the action's effects, and the relations of right and duty without collapsing them into one property.

Obligation identifies an action that the relevant agent must perform; prohibition identifies an action the agent must forgo; permission identifies a class of choices whose performance is not wrong. These categories are related: if an action is forbidden, refraining from it is required under the usual classical representation; but philosophically difficult cases arise when the space of available actions itself is contested. Permission may be merely the absence of a prohibition or may express a positive moral liberty to pursue projects. Rights can correlate with duties owed by another, confer powers, or protect liberties. Treating every right as the same kind of normative position makes disputes about consent, property, and assistance needlessly obscure.

This distinction also clarifies deontology's distance from the maximization model. A consequentialist can favor rules that prevent deception because practices of honesty promote good results. A deontologist asks whether deceit itself violates a claim that another has against the agent. A rule has independent force when the agent must respect it even in cases where violating it would produce the preferred outcome according to a common metric. Whether this independent force is absolute, defeasible, or sensitive to thresholds is a further substantive dispute; it is not settled by the definition of deontology alone.

Historical Precursors in Greek Ethics

Greek philosophy contains rigorous investigations of duty, law, justice, and voluntariness, although it did not organize its full ethical inquiry under the modern classification of deontology. In Plato's Crito, Socrates argues with the personified laws about whether the injustice of the verdict licenses escape from prison. The dialogue juxtaposes obligations arising from participation in a civic order, the wrong of retaliating with wrong, and the question whether obedience to law is owed even when a law's exercise harms an individual. It should be read as a philosophical argument in dramatic form rather than a ready-made theory of modern legal compliance.

Aristotle's Nicomachean Ethics V investigates corrective and distributive justice and distinguishes what is unjust from acting unjustly. At 1135a–1136a he analyzes voluntariness, knowledge, and choice in the moral classification of action. This offers an important precursor to deontological attention to agency: the outward production of harm does not exhaust what it means to commit an injustice. Yet Aristotle locates justice within an ethics of virtue, rational activity, and practical discernment; he does not isolate an unrestricted set of duties from the comprehensive human good (Aristotle, Nicomachean Ethics V.7–9).

The historical continuity is therefore thematic and argumentative rather than taxonomic. Greek debates supply models of unjust action, promises, fidelity to law, and the limits of retaliation. Later deontological philosophy gives such problems different foundations, especially universal rational legislation, inherent rights, and plural grounds of obligation. The legitimate historical claim is that the problems have ancient roots; the distinctive modern theory emerges through subsequent transformations of moral authority.

Duty, Natural Law, and Medieval Moral Reasoning

Medieval natural-law thought developed accounts of duties grounded in rational practical order, divine law, human law, and the moral description of action. Thomas Aquinas places law within his account of practical reason in Summa theologiae I–II, questions 90–97. He treats a human action through its object, end, circumstances, and relation to reason. The resulting position cannot be assimilated without remainder to Kantian autonomy: the metaphysical and theological organization of law is different, as are the accounts of happiness and virtue.

Aquinas's discussion of self-defense in Summa theologiae II–II, question 64, article 7, becomes especially influential in later debates about double effect. He distinguishes the intended preservation of one's life from the death of the aggressor that may accompany the defensive act and adds a proportionality requirement. The text thus connects intention, causation, and justified force. Later moralists developed more elaborated double-effect conditions, so the full modern four-condition doctrine should not simply be attributed verbatim to Aquinas.

Natural-law theories are instructive because they show that deontic constraints can arise within comprehensive teleological worldviews. They refute the historical simplification that every duty-based prohibition must rest on a rejection of ends or consequences. Their central philosophical question is how an objective practical order can bind deliberation while leaving room for circumstances, culpability, and the competing claims of the people concerned. This question returns in secular moral theories with different accounts of the source of normativity.

Early Modern Obligation and the Authority of Moral Law

Early modern ethical debate transformed the basis of obligation through accounts of natural rights, political authority, rational self-government, sentiment, and the relation between interest and duty. Grotius (The Rights of War and Peace, preliminary discourse), Hobbes (Leviathan), and Pufendorf (The Whole Duty of Man) present divergent approaches to law and obligation, making authority, security, and the conditions of social order central. Their arguments demonstrate that a theory can use the language of duty while deriving requirements from different sources: divine command, natural law, rational advantage, or publicly authoritative rules.

This plurality matters when reconstructing the antecedents of Kantian ethics. A moral law may bind because an agent is subject to a superior authority; because reason recognizes objective relations; because a practice establishes legitimate expectations; or because rational agency itself carries the principle of its own legislation. These explanations are not equivalent. A requirement grounded in the advantage of cooperation may cease to justify action in a case where cooperation fails to benefit an agent. A requirement grounded in the authority of autonomous practical reason aims to survive that contingency.

The modern formulation of deontology emerges partly from this contest over moral authority. When a demand claims categorical force, a philosopher must explain why it binds the agent even when compliance frustrates immediate interests. That question leads to Kant's attempt to establish the unconditional law of a good will and to subsequent debates about whether duty is known through reason, moral perception, relationships, or the claims of other persons.

Kant's Groundwork and the Good Will

Kant's Groundwork of the Metaphysics of Morals (1785) begins with the claim that only a good will is good without qualification (Ak. 4:393). Talents and dispositions can be used badly; desirable fortunes can empower wrongdoing. The distinctive goodness of the will lies in its principled orientation rather than in the success of the action it undertakes. In the first section Kant distinguishes acting in conformity with duty from acting from duty: the outward behavior can be identical while the maxim and motivational ground differ. This is a claim about moral worth, which should be distinguished from the assessment of an act's bare permissibility.

Kant's account of practical necessity asks what it means for reason to determine a will. A hypothetical imperative represents a practical requirement conditional on an end; a categorical imperative represents a requirement whose authority is independent of contingent ends (Ak. 4:414–420). This difference carries the argument beyond the ordinary prudential proposition that one ought to use effective means to accomplish one's purposes. The categorical law purports to be binding on rational agents as such.

The explanatory burden is substantial. A formal law must connect to the determinate duties of ordinary life; an ideal of autonomous agency must explain why manipulation or false promises are wrong; and the theory must distinguish acting out of respect for law from adopting whatever description of a maxim yields a favorable universalization result. Kant's several formulations of the categorical imperative and his account of freedom constitute an ambitious attempt to meet these demands. Readers should consult the primary text at its Academy pagination, rather than substituting one English catchphrase for the complete argument.

Universal Law and the Structure of Maxims

Kant's formula of universal law instructs an agent to adopt only maxims that the agent can at the same time will as universal law (Ak. 4:421). A maxim specifies a principle of acting under a description of circumstances and purpose. The practical test asks whether that policy could coherently serve as a law for agents generally and whether it could be rationally willed in a world governed by such a law. The example of promising falsely to secure a loan is designed to reveal a contradiction in the general practice on which the deceptive act depends.

The method distinguishes contradictions in conception from contradictions in willing. The false-promise example is generally used to illustrate a perfect duty; systematic refusal to aid others illustrates an imperfect duty associated with the adoption of ends (Ak. 4:422–423). A perfect duty prohibits some kinds of conduct in specified conditions, while an imperfect duty allows latitude in how and when an obligatory end is pursued. This is a difference in normative structure and not simply in the emotional seriousness of a rule.

The universal-law argument faces a specification problem. A maxim written as 'deceive whenever this precise event occurs' may appear more easily universalizable than the morally salient maxim 'secure personal advantage through another's misplaced trust.' The selection of morally relevant descriptions is itself a task of judgment. A second problem concerns the transition from inconceivability to wrongness: logical consistency and practical justification are different relations. The strongest Kantian interpretation treats universalization as part of a substantive account of rational agency, not a stand-alone syntactic device. Contemporary defenses by Korsgaard (1996) and interpretations surveyed in the revised Stanford account of Kant's ethics illuminate these issues.

Humanity, Autonomy, and the Kingdom of Ends

The formula of humanity instructs an agent to treat humanity, in one's own person and that of another, always as an end and never merely as a means (Ak. 4:429). Kant's point concerns the standing of rational agency as a condition of practical value. Ordinary collaboration necessarily allows one person's activity to serve another's ends. The moral objection arises when an agent recruits another through deception, coercion, or a plan whose success depends upon excluding that person's capacity for rational choice. The distinction between using someone as a means and merely as a means is therefore indispensable.

Autonomy is the will's capacity to understand itself as legislating universally valid practical principles rather than being governed solely by external incentives. Kant develops the kingdom of ends as a systematic union of rational beings under common laws (Ak. 4:433–436). Such ideas provide foundations for obligations of respect and for criticisms of domination. They also create questions: what counts as meaningful consent, which persons are affected, what status belongs to dependent people, and how the ethical standing of vulnerable beings relates to the rational capacities Kant emphasizes.

A common error is to infer that respect consists exclusively in honoring whatever preference a person now expresses. Kantian respect addresses the conditions of rational agency, which can involve rejecting manipulative consent, protecting future choice, and sustaining institutions of public right. Yet any move from dignity to a concrete prohibition requires additional analysis. The humanity formula articulates a moral orientation; its application depends on agency, vulnerability, practices of justification, and the relationships through which agents can wrong one another.

Kantian Duties Beyond the Groundwork

Kant's later Metaphysics of Morals (1797) differentiates doctrines of right and virtue, making explicit the distinction between external relations that can be subject to rightful coercion and ethical duties that concern the adoption of ends. This development matters because the Groundwork alone may make duty appear to be only a question about an isolated individual's maxims. The later work addresses property, contract, public right, beneficence, self-respect, and the organized conditions of freedom.

The distinction also limits a simplistic equation of Kantian ethics with a rulebook of prohibitions. Wide duties of virtue concern ends such as promoting the happiness of others and developing one's capacities. They leave discretion about occasions and means while still binding an agent. Strict duties of right concern relations in which one person's external freedom must be compatible with the freedom of others under universal law. A well-developed deontological theory must therefore explain negative constraints and positive obligations in a common normative system.

The persistence of controversy over truthfulness, emergency exceptions, and the relation between law and virtue illustrates the difficulty of Kant's ambition. Interpreters disagree whether particular applications represent immutable principles, context-specific judgments, or failures of the broader doctrine to accommodate morally urgent conflicts. Distinguishing the foundational theory from disputed examples allows the genuine argumentative challenge to remain visible.

Prichard and the Direct Knowledge of Obligation

In 'Does Moral Philosophy Rest on a Mistake?' (1912), H. A. Prichard challenges the demand for a demonstration of why a recognized duty must be performed by reducing the question to some further advantage or good. If the alleged proof relies on the agent's benefit or on maximizing value, it appears to change the subject from the duty's obligatory character. Prichard's argument presses a foundational question: are moral obligations recognized directly through reasoned apprehension, or must they be deduced from a more general account of the good?

The strength of Prichard's challenge is that it distinguishes the epistemic availability of a moral reason from its causal attractiveness. A promise can bind because a person has given an undertaking, even when the action is inconvenient. Explaining that compliance will produce a more agreeable world misses a feature of the obligation if the promise remains a reason when outcomes favor breach. His paper is a primary document of twentieth-century British intuitionist ethics, published in Mind, 21(81), 21–37.

A difficulty follows. If the authority of obligation is basic, philosophers must still explain how agents can recognize genuine duties, distinguish them from habituated prejudices, resolve disagreements, and avoid competing intuitions whose sources cannot be examined. The insight that a reason need not be reduced to a benefit does not settle what counts as a reason. It generates a program for moral epistemology and practical judgment.

Ross's Pluralism of Prima Facie Duties

W. D. Ross's The Right and the Good (1930), especially chapter II, defends a plurality of basic grounds of duty. Ross discusses fidelity, reparation, gratitude, justice, beneficence, self-improvement, and nonmaleficence; particular presentations divide and group these considerations differently. His phrase 'prima facie duty' designates what contemporary philosophers often describe as a pro tanto reason: a morally significant consideration that counts in favor of an act in virtue of one of its features. It need not determine the agent's actual duty after all the facts are taken into account.

An agent may promise to meet a friend and then learn that another person urgently requires assistance. Fidelity speaks for keeping the promise; preventing serious harm speaks for helping. Ross's view does not erase the first claim once the second becomes decisive. Its importance lies in treating multiple morally significant aspects of a situation as genuine, even where only one act can be performed. The actual obligation arises from a practical judgment concerning the full circumstance (Ross, 1930/2002, pp. 16–64).

This account offers a direct answer to monistic theories that reduce the rightness of every action to a single value such as maximal total welfare. It also invites a methodological objection: how are the strengths of heterogeneous duties compared? Ross appeals to considered moral judgment rather than a universal formula for assigning numerical weights. This acknowledges the complexity of moral life while leaving uncertain the reproducibility of verdicts. His theory remains a central model of pluralistic deontology precisely because it refuses to trivialize conflicts while obliging the theorist to explain principled judgment.

Actual Duty, Moral Conflict, and Residual Reasons

The distinction between a pro tanto duty and an all-things-considered obligation provides a structural account of moral conflict. Two genuine reasons can support incompatible actions without both being decisive. If a promise must be broken to prevent a serious injury, the agent may have justification for breaking it and nevertheless owe an explanation or apology. The remaining claim of fidelity can survive the decision as a residual reason for repair. Such remainder is difficult to understand under a model in which every defeated moral consideration was merely illusory.

Moral dilemmas raise a stronger possibility: an agent may be required to do both of two incompatible things. A theory that permits genuine unavoidable wrongness must distinguish impossibility, culpability, regret, and practical necessity. An agent can be justified in selecting a course of action and still have a basis for morally serious regret about the burdens left upon another. This is not equivalent to saying the agent performed an impermissible action. Rossian pluralism allows conflicts of reasons; whether genuine irresolvable conflicts of obligations exist requires additional argument.

The philosophical issue is how the authority of a demand continues after its fulfillment becomes impossible. Apology, compensation, reform of circumstances, and a commitment to avoid repeating a conflict can all follow from undefeated aspects of the moral situation. They show why duty should be analyzed through relationships as well as one-off verdicts. A moral theory must account for what an agent owes after an unavoidable choice, not only what the agent should do before it.

Agent-Relative Reasons and the Structure of Constraints

Deontological constraints are commonly agent-relative: they bind an agent by reference to what that agent does. The command not to torture a prisoner can remain binding even if torture would induce information that prevents someone else from torturing several prisoners. A simple aggregative prohibition would instead recommend violating the rule in order to minimize its total violations. This contrast is one of the sharpest tests of independent moral constraints (Alexander & Moore, 2024).

The philosophical puzzle is sometimes called the paradox of deontology. If violating a constraint is bad, why should it not be better to violate it once to prevent more violations? One response holds that the wrongfulness of an action cannot be represented wholly as a bad item within a world-state ranking. A second grounds restrictions in the authority of persons not to be used for others' ends. A third draws on the structure of agency: an individual's moral relationship to the acts they choose differs from their relationship to the acts of others.

None of these explanations is complete without qualifications. Agents are also responsible for harms they knowingly permit in some circumstances. The distinction between what one does and what one allows must therefore be defended in morally salient terms rather than invoked as a metaphysical slogan. The independent force of a constraint concerns an agent's practical authority over others and the terms on which costs may be imposed. That force is strongest when the prohibited act deliberately appropriates another's body, agency, or vulnerability as an instrument of a plan.

Patient-Centered Deontology and Rights

Patient-centered deontology begins with claims belonging to those affected by action. The moral subject is the person who may be killed, deceived, enslaved, or otherwise made to bear costs. Rights can impose directed duties upon agents: when one individual possesses a claim against being assaulted, another has an obligation not to assault that individual. In this picture, moral constraints are explained through the standing of the potential victim rather than solely through the integrity of the actor.

Robert Nozick's Anarchy, State, and Utopia (1974) supplies a famous formulation of rights as side constraints on the actions that others may take. Such constraints resist treating individuals as sites across which net welfare can be redistributed at will. Yet theories of rights must confront their internal architecture. A claim-right differs from a liberty, power, or immunity; different rights may protect freedom, security, assistance, or participation and can be limited by competing claims. A duty to rescue is structurally different from a duty not to attack.

The fundamental justification cannot end with the bare assertion that persons have rights. It must explain the nature of moral standing and why that standing generates this particular boundary upon others. Moreover, a right-holder may lack the ability to exercise a corresponding power of consent or enforcement; infants and people with severe impairments can possess claims despite being unable to articulate them. Rights theory becomes philosophically credible when its protection extends through vulnerability without equating moral standing with the ability to bargain.

Moral Options, Projects, and Prerogatives

An ethical theory can become excessively demanding if it requires each agent to sacrifice every independent project whenever doing so would improve an impersonal outcome. Samuel Scheffler's The Rejection of Consequentialism (first published 1982; revised edition 1994) investigates agent-centered prerogatives and constraints as two distinct responses to consequentialist demands. A prerogative permits an agent to give special weight to certain personal commitments without treating those commitments as morally superior for everyone.

The difference between a constraint and a prerogative is essential. A constraint blocks an otherwise attractive action; a prerogative permits an action that may not bring about the best available aggregate result. If an agent spends an evening with a loved one rather than working on a charitable project, the permissibility of this choice may reflect a moral option. If the agent refuses to harm a stranger in order to raise more charitable funds, the refusal may reflect a constraint. The two doctrines have different foundations and need not stand or fall together.

A powerful objection concerns distributive bias. Personal projects can be invoked to defend indifference toward severe suffering or institutional injustice. A plausible theory therefore needs criteria for the claims that remain nonoptional and for the liberties that rational agents may retain. Deontological latitude is defensible when it preserves the integrity and meaningful agency of persons while recognizing genuine duties of aid, fairness, and repair. Moral options are not the unrestricted sovereignty of personal preference.

Intentions and the Description of Action

Intention matters because what an agent chooses cannot always be captured by the physical event produced. The same movement may be a rescue, an assault, or an accident depending on the practical description under which it is undertaken. Elizabeth Anscombe's Intention (1957) develops an account of intentional action through the intelligibility of the question why an agent is acting. Her work gives contemporary deontology conceptual resources for distinguishing a purposeful injury from an injury that the agent anticipated but did not adopt as a means.

The means-end relation is especially significant. If a person deliberately uses another's injury to bring about a desirable benefit, the injury occupies a different position in the plan from a harm that follows incidentally from a permissible action. However, intentions can be described at different levels. An agent who says 'I only meant to protect the laboratory' while knowingly subjecting a worker to toxic exposure does not thereby resolve the moral question. The relevant description must reflect the causal contribution of the worker's exposure to the chosen means.

A theory of intention must avoid two extremes. Psychological sincerity cannot itself license conduct; an agent's benevolent motive may accompany a rights violation. Pure outcome comparison also fails to recognize acts that enlist persons against their will. The proper analysis examines objective features of the plan, knowledge of risk, control, and the place of affected people in the means selected. This is an explanatory investigation, rather than a technique for obtaining moral permission by redescribing purposes.

Doing, Allowing, Killing, and Letting Die

The distinction between doing harm and allowing harm is closely connected to deontological constraints but is analytically independent from the distinction between intending and foreseeing harm. One may intentionally omit rescue, foresee a death while intervening, or cause harm without desiring it. The moral difference between causing and failing to prevent a harm, when there is one, requires an account of agency, background entitlements, special duties, available alternatives, and the costs imposed by intervention.

A person who pushes another into danger ordinarily violates a powerful negative duty. A passerby who fails to prevent danger may also act wrongly, especially when rescue would be easy and safe, but the normative structure differs. The first person created a threat; the second encounters an existing threat and must decide whether assistance is required. This difference can be weakened by special responsibilities: a lifeguard, clinician, or parent may be duty-bound to act where an unrelated stranger has a weaker duty. The practical force of the distinction cannot be established from the verbs 'do' and 'allow' in isolation.

The inquiry also raises the baseline problem. What counts as letting nature take its course depends on institutions and prior choices. Withdrawing a dangerous device, discontinuing a treatment at the request of an informed patient, and abandoning a dependent person can look similar as omissions but involve distinct rights and obligations. The right description must incorporate the authority under which control is exercised. Deontology's value lies in making these relations explicit.

The Doctrine of Double Effect

The doctrine of double effect examines actions that predictably bring about a harmful effect in pursuit of a permissible good. Its historical roots lie in Aquinas's discussion of self-defense (Summa theologiae II–II, q. 64, a. 7), while later formulations add explicit criteria concerning the moral character of the act, the agent's intention, the use of harm as a means, and proportionality. The distinction between what is intended as a means and what is foreseen as a side effect can be morally important, but it cannot by itself license an action whose cost is disproportionate.

Philippa Foot's 1967 discussion and Warren Quinn's 1989 article transformed the contemporary debate. Quinn distinguishes forms of direct and indirect agency by asking how the victim is incorporated in the agent's plan. Such analysis moves beyond a purely introspective test of intentions. The central concern is whether a person is made an object of intervention in service of another's goal. As the Stanford treatment by McIntyre (2023) emphasizes, the doctrine remains contested in medical, military, and emergency cases.

Two difficulties are fundamental. First, effects often lie so close to the selected means that describing them as merely foreseen seems morally artificial. Second, the moral relevance of intentional structure may depend on an antecedent view of which treatments of persons are wrongful. Double effect is thus best understood as a proposed component in an account of permissible harm, not as an automatic algorithm that labels consequences acceptable once an agent reports a benign goal.

The Trolley Cases and the Limits of Intuition

The trolley cases investigate why many people distinguish redirecting a runaway vehicle toward one person from pushing a bystander into its path as a means of stopping it. Foot's earlier discussions and Judith Jarvis Thomson's 'The Trolley Problem' (1985) frame a debate about harming as a means, redirecting threats, rights of potential victims, and the moral relevance of an agent's role. The cases are useful because they hold some outcomes fixed while changing the structure of intervention.

Their methodological limits are substantial. Intuitions about vivid, compressed scenarios may be affected by framing, omitted background obligations, physical proximity, and familiar conventions. In a realistic transport system, a driver has obligations to passengers, pedestrians, and other drivers that differ from those of an accidental bystander standing beside a switch. Consent, existing risk, professional duty, and the knowledge required for action reshape the case. Thomson's own later reconsiderations demonstrate that the trolley debate is a living argument rather than a settled two-option exercise.

The productive philosophical question is which distinctions survive when the scenarios are described with adequate causal and relational precision. If a person is harmed as the very means of saving others, their agency and bodily authority are directly implicated. If a preexisting threat is redirected, the injustice of imposing a new danger must still be assessed. Neither a simple numerical comparison nor an immediate appeal to intuition completes the justification. A defensible verdict must articulate why the principle applies across relevantly similar cases.

Thresholds, Catastrophe, and Absolute Prohibitions

Absolutist deontology holds that certain acts are forbidden regardless of their consequences. Threshold deontology allows constraints to be overridden when the cost of compliance crosses some sufficiently grave threshold. Each view aims to capture a moral insight: absolute constraints protect individuals against instrumental sacrifice, while threshold theories take seriously the possibility of preventable catastrophe. The problem is to explain the transition between ordinary and extraordinary cases without making moral prohibitions mere rules of convenience.

A threshold expressed as 'never harm one to save five, but harm one to save a million' raises a question about the units of comparison. If the wrongfulness of using a person as a means is genuinely independent of aggregate utility, why does an accumulation of beneficiaries change permission? A proponent might invoke exceptional duties to prevent disasters, an ordering of rights, or the moral difference between local choice and systemic destruction. Such answers need a theory of the standing of claims and their interaction, rather than a number selected after hearing a dramatic example.

Conversely, absolutism must confront the responsibility of an agent who could prevent mass harm only by violating a strict prohibition. It can distinguish being responsible for an action from being the cause of every consequence one could have avoided. But this distinction must be argued, especially when the stakes are foreseeable and the agent exercises effective control. The conflict is therefore among accounts of responsibility, authority, and competing moral claims, not merely between having principles and caring about the world.

Promises, Fidelity, Trust, and Exceptions

Promise keeping is a central deontological case because the promise creates a relation between particular persons. A promise may give another a claim upon the promisor's future conduct, generating a directed reason whose force does not depend entirely on wider gains from the practice of promising. Kant's false-promise example concerns the contradiction involved in securing an advantage by relying on a practice one would undermine through universalized deception (Ak. 4:422). Ross places fidelity among the basic pro tanto duties arising from one's previous actions (The Right and the Good, ch. II).

The structure is richer than a blanket command to fulfill every verbal assurance. Promises can be invalid because they are coerced, unauthorized, based on serious deception, or require wrongdoing. Others may be released by the promisee, overridden by emergency duties, or fulfilled through agreed alternatives. The recognition of justified exceptions need not reduce the original commitment to a prediction of consequences. It shows that the creation and discharge of directed obligations depend on circumstances and legitimate authority.

When fidelity conflicts with urgent aid, a person may have sufficient reason to break an appointment and still owe an explanation. The duty's relational character explains why communicating, making amends, and respecting future reliance remain important. Deontology here offers an analysis of moral life over time. An obligation is established within a practice of mutual reliance, tested by intervening facts, and assessed through what the agent owes to the particular person who trusted them.

Truthfulness, Manipulation, and Moral Agency

Deontological opposition to manipulation begins with the recognition that another person's deliberative capacities are morally relevant. A lie can wrong someone by controlling the evidence on which their choice depends. A coercive threat can wrong them by narrowing their options through an illegitimate exercise of power. Both can violate autonomy even when the action produces a benefit. The moral problem is the unauthorized use of another's practical standpoint as part of a plan.

Nevertheless, truthful disclosure can itself conflict with justified obligations of confidentiality, privacy, and protection against violence. A physician may owe a patient an explanation and simultaneously owe third parties protection of confidential data. An employee may be required to report dangerous misconduct while safeguarding innocent people whose information appears in the records. The ethical analysis must specify whose claim is at stake, which information is required, and the legitimate authority governing disclosure.

The wider consequence is a richer conception of consent. Consent has moral significance when an agent has the authority to permit the intervention, possesses enough relevant information, and acts with adequate freedom. A signature obtained through manipulation cannot do the work of legitimate authorization. Equally, not every unwanted consequence demonstrates a violation of consent. Deontology needs substantive standards for legitimate choice, justified interference, and the respect owed to those whose capacities for choice are constrained or developing.

Contractualism and Justification to Persons

Contemporary contractualism offers a route from the standing of persons to rules of conduct. T. M. Scanlon's What We Owe to Each Other (1998) characterizes wrongful conduct in terms of principles for general regulation that persons could reasonably reject. The emphasis falls on justifiability to each person rather than solely on the value of the resulting social state. This makes contractualism closely connected to deontology: a harm imposed upon an individual must be justified in light of that individual's claims, not merely buried within aggregate utility.

Scanlon's view should be distinguished from bargaining theories that derive morality from mutually advantageous agreements. Reasonable rejection is not identical to what a powerful party would accept in negotiation. It involves the reasons people could bring as persons affected by general principles. This perspective explains why aggregation can be morally problematic when many minor advantages are invoked to defeat one individual's serious claim. Yet contractualist theories must decide how much aggregation of comparable harms is permissible and whose objections are relevant.

The relationship to Kant is both historical and systematic. Both approaches take persons' rational standing seriously, but they do not employ identical derivations or conceptions of a moral law. Contractualism is a potential foundation for some duties and a distinct research program in its own right. Within a study of deontology, it illuminates the question whether a constraint becomes intelligible through the reasons that must be offered to the particular individual who would bear its cost.

Deontology and the Strongest Consequentialist Challenge

Consequentialism asks which actions produce, promote, or conform to the best available consequences under a specified evaluative theory. John Stuart Mill, in chapter V of Utilitarianism (1861–1863), takes the powerful moral appeal of justice seriously and seeks to explain rights and security in relation to utility. The challenge for deontology is therefore not merely that a rival theory ignores rights. Sophisticated consequentialist theories can explain the great instrumental value of trust, stable rules, and reliable protections.

The decisive dispute concerns the grounds of their authority in cases where violating a right would improve the relevant aggregate outcome. If rights derive their entire force from their contribution to welfare, then sufficiently advantageous violations may be permissible. A deontologist argues that an agent may lack authority to impose such a cost even when the world would be better by some ranking. That claim must be defended against the objection that refusing intervention ignores preventable suffering.

Rule consequentialism narrows the practical distance by requiring conformity to a justified code rather than treating each decision as a new act of maximization. Yet a code justified only by expected benefit differs in its ultimate rationale from a right understood as a claim against instrumental use. This distinction can be obscured by similarity in ordinary verdicts. A rigorous comparison asks not only which rule a theory recommends but what would justify revising it when a hard case appears.

Moral Worth, Responsibility, and Moral Luck

Deontology must distinguish an action's permissibility, the moral worth of the agent's motive, and the extent to which an agent is responsible for an outcome. Kant's Groundwork gives duty a distinctive relation to moral worth; later ethics recognizes that a person can choose for admirable reasons and nevertheless perform a wrongful act. A clinician who acts on incomplete but responsibly gathered evidence can cause harm without being blameworthy in the way that a reckless clinician would be.

The problem of moral luck asks how outcomes beyond an agent's control affect moral assessment. A careful driver and a negligent driver may both avoid collisions, while only one behaves culpably; two equally negligent drivers may cause radically different harms. A deontological account can locate responsibility partly in choice, control, and intention while acknowledging duties of repair that depend on actual harm. Thomas Nagel's work on responsibility and conflict, including 'War and Massacre' (1972), provides a significant contemporary background.

Moral appraisal should track different questions. Was the action permissible given what the agent reasonably knew? Was the decision made for acceptable reasons? Did the agent violate a claim? What harm occurred and who is owed repair? These questions may receive different answers in one case. Their separation prevents a theory of obligation from becoming a theory of praise and blame with no independent attention to rights and consequences.

Rights, Justice, and Political Institutions

Deontological reasoning has immediate institutional significance. Claims against torture, arbitrary punishment, discrimination, and coercive interference can function as limitations on what states and organizations may do to improve aggregate welfare. Nozick's side-constraint theory illustrates a libertarian version; Kant's doctrine of right seeks a public legal order in which each person's external freedom is compatible with others'. Rawls's A Theory of Justice (1971) gives the priority of certain basic liberties an institutional role within a distinct contractarian framework.

These theories disagree about property, distribution, coercion, and the role of the state. Rights-based moral language does not entail a single political settlement. A right to noninterference can coexist or conflict with claims to fair institutions, public security, or material provision. The philosopher must ask which interests or powers justify a particular protection and who bears the burdens of sustaining it.

Institutions also transform the structure of responsibility. A decision may be produced by many officials, formal procedures, and technical systems rather than by one isolated agent. The existence of a rule cannot itself establish its justice. Officials may have duties to obey valid rules and independent duties to challenge unlawful or morally abusive directives. A robust institutional deontology combines limitations on authority with procedures for explanation, appeal, and remedy. Its purpose is to make moral protections effective in the structures that can most deeply affect individuals.

Care, Dependence, and the Critique of Abstract Agency

Care ethics asks whether models of detached rational agents overlook relationships of dependence, vulnerability, and unevenly distributed labor. Such concerns challenge versions of deontology that focus exclusively on formal equality while treating caregiving and responsiveness as secondary. A dependent child or incapacitated patient may be unable to exercise conventional capacities of negotiation, yet remains the bearer of morally serious claims. The duties of a caregiver are often shaped by specific knowledge of the person and by a sustained relationship.

This criticism can strengthen deontological theory when duties of care are understood as directed obligations rather than discretionary benevolence. Obligations may arise from role, commitment, responsibility for an existing vulnerability, or the urgent claims of a person needing assistance. At the same time, relationality cannot justify unlimited exploitation of caregivers. A legitimate ethic must examine the obligations of institutions and communities that make care possible.

The conceptual task is to reconcile universal standing with the particularity of obligations. Equal moral consideration does not require identical treatment, because needs, permissions, and existing commitments differ. The test of a defensible particular obligation is whether its grounds can be explained without arbitrary privilege or coercion. This allows deontology to learn from relational ethics while retaining its ability to criticize unjust customs and oppressive roles.

Comparative Perspectives on Duty and Dharma

Philosophical traditions beyond modern European ethics also investigate duty and obligation, though their concepts should be studied in their own conceptual settings. In the Bhagavad Gita, Arjuna's predicament concerns the claims of kinship, violence, social position, and the discipline of action. The term dharma has a semantic and religious range wider than English 'duty'. Chapter 2, verse 47 distinguishes action from attachment to its fruits, but the complete work does not merely reproduce a Kantian prohibition on maximizing outcomes.

The comparative lesson is methodological. A duty can be embedded in a comprehensive account of cosmic order, social vocation, liberation, or cultivated character. It can also be represented as a universal requirement grounded in reason or as a directed claim arising from persons. Similar practical injunctions do not prove identical philosophical foundations. An adequate comparison asks who or what establishes an obligation, who is bound, how competing obligations are adjudicated, and what conception of agency sustains compliance.

Such comparison can expose an assumption often hidden in contemporary debate: that morality consists of individuals selecting isolated acts from a neutral menu. Many traditions understand moral obligations through the constitution of practices, responsibilities of station, and the cultivation of reliable judgment. These perspectives matter when evaluating deontology's universal claims, although they must remain open to criticism where inherited duties conflict with equality or legitimate personal freedom.

Practical Ethics: Medicine, Professional Power, and Technology

Deontological principles are tested where authority over vulnerable persons is concentrated. In clinical medicine, informed consent, confidentiality, professional fidelity, and prohibitions on nonconsensual experimentation express duties owed to patients. They operate alongside obligations of beneficence and efforts to reduce suffering. A clinician may possess technical information sufficient to predict an advantageous outcome and still lack permission to impose the intervention on a competent patient.

Technology introduces analogous questions at the level of design and institutional deployment. An automated eligibility or employment system may optimize an operational objective while denying affected people explanations, contestation, privacy, or equitable procedures. A system's statistical performance cannot by itself settle whether a process violates claims of fair treatment. The developers, owners, operators, and public authorities involved may possess different directed obligations, and these obligations need not vanish because a machine executes the decision.

The Artificial Era increases the range of action mediated by nonbiological systems. This is an application of deontological questions about accountability and authority, rather than a claim that a particular artificial system possesses the full standing of a moral agent. The capacity to produce decisions, the ability to understand and answer for reasons, and the possession of claims against others remain distinct philosophical issues. Any extension of duties or rights to new subjects demands an explicit argument about standing and relational justification.

Moral Epistemology: How Are Duties Known?

A duty-centered theory needs an account of how moral requirements are recognized and defended. Rationalist approaches seek principles supplied through practical reason; intuitionists such as Prichard and Ross claim that some grounds of duty are known without inference from a more fundamental value; contractualists emphasize the reasons persons can offer one another. These approaches can converge on particular cases while disagreeing about the source of certainty.

Three forms of error threaten deontological reasoning. The first is mistaking a socially inherited rule for a universally justified obligation. The second is extracting a prohibition from a broad principle without testing the relevant description of the action. The third is claiming insight into another's rights while ignoring that person's experience or legitimate agency. Moral learning requires scrutiny of evidence, consistency across analogous cases, and the capacity to revise judgments.

The discipline also faces uncertainty when the facts are incomplete. An emergency may require choice before intentions, risks, or institutional responsibilities are fully known. Deontology can guide action through due inquiry, justified precautions, temporary restraints on coercive interference, and procedures of later correction. Such practices express a duty to take the epistemic conditions of action seriously. A rule cannot be responsibly applied when the agent refuses to learn what the rule is meant to protect.

Objections: Formalism, Rigidity, and the Demands of Judgment

One familiar objection describes deontology as formalistic: universal norms underdetermine concrete situations. Kant's formula of law cannot be applied without identifying a maxim. A rights principle cannot determine the outcome of an emergency without specifying the right, its scope, and the relations among affected persons. This is a serious objection to attempts to replace judgment with a simple rule. It is less decisive against developed theories that incorporate practical reasoning, multiple duties, and careful specification.

A second objection concerns moral rigidity. When an inviolable constraint produces grave and avoidable suffering, the theory must explain why obedience remains justified. A third concerns moral asymmetry: prohibiting agents from directly inflicting harm while allowing substantial preventable harms may appear to privilege a morally clean self-image over victims' welfare. Agent-relative reasons require defense in terms of the authority people possess over one another, not in terms of the actor's desire to feel innocent.

A fourth objection concerns plurality. If duties arise through many independent grounds, deliberation may appear arbitrary when they conflict. The answer is not to assert that one rule silently wins every case. It is to explain which properties of the conflict make an obligation more urgent, which claims survive as reasons for remedy, and why the resulting judgment can withstand interpersonal criticism. Deontology succeeds where its protections generate a practice of justification proportionate to the seriousness of the interference they prohibit.

Contemporary Research and Unresolved Questions

Contemporary debate continues to test how constraints interact with agent-centered prerogatives, directed rights, counterfactual consequences, and the moral difference between causing and allowing harm. Kamm's Intricate Ethics (2007) advances detailed analysis of permissible harm and the moral relevance of action structure. Quinn's account of direct agency remains central to the double-effect literature. Scheffler's treatment of prerogatives shows how the problem of demandingness can be separated from the problem of constraints. Research on contractualism examines aggregation, justifiability, and the plurality of individual complaints.

Several unresolved questions are especially consequential. Can a deontological constraint be derived from moral standing without already assuming the forbidden kind of harm? How should a theory characterize intentions when plans have complex causal stages? Can agents possess permissions that are more than the absence of duty while others possess claims to urgent assistance? What principle determines when a serious threat crosses a threshold permitting the infringement of a right? How are institutional duties distributed when no single participant intends the harmful outcome?

Research concerning artificial systems adds a further set of questions. Which institutions are answerable for decisions mediated by software? When does an artificial participant itself satisfy conditions for moral responsibility? How should obligations to affected individuals be enforced when decisions are opaque or dynamically changing? These problems cannot be answered by simply translating an optimization constraint into the word 'duty'. They require theories of agency, justifiable authority, standing, information, and remedy.

Philosophical Synthesis: The Authority of the Claim

The analysis establishes that deontology's enduring philosophical contribution is its account of moral claims that constrain action from within a relationship of justification. Kant explains categorical authority through rational legislation and respect. Ross identifies multiple grounds of duty and the possibility that several retain force in one choice. Rights theories focus on the standing of persons whose bodies, commitments, and freedoms may be appropriated. Contractualism tests principles through what can be justified to those governed by them. These approaches converge upon the authority of a claim while disagreeing over its metaphysical and epistemic foundation.

Four distinctions give this convergence a rigorous form. First, the goodness of an outcome and the permissibility of an intervention are separate evaluative questions. Second, a person's moral standing and their capacities for present autonomous choice are distinct. Third, an act's causal effects and its intentional structure are related but cannot be identified. Fourth, a genuine pro tanto duty and the action required after all relevant reasons have been considered occupy different levels of normative judgment.

The resulting analytical synthesis is that deontological assessment must establish the claimant, the content of the claim, the grounds of its authority, and the conditions under which competing claims can qualify or override it. This formulation is an interpretive conclusion of the present monograph, not a newly enacted theory or a substitute for the historical accounts. It provides a method for asking what a justified action owes to each affected person and what must be explained when an agent imposes a cost in pursuit of an otherwise worthwhile end. Its central ambition is disciplined moral accountability under genuine conflict.

Conclusion

Deontology is a fundamental family of normative ethical theories devoted to the independent force of duties, rights, prohibitions, constraints, and permissions. Its history passes through ancient problems of justice and voluntary action, medieval accounts of law and intention, early modern transformations of obligation, Kant's categorical law, British intuitionist pluralism, and contemporary debates about the moral standing of persons and the permissible structure of action.

The monograph has shown why no single shorthand formula exhausts this tradition. The priority of right over aggregative good takes different forms in Kantian autonomy, Rossian pro tanto obligations, agent-relative constraints, rights-based side limits, and interpersonal contractual justification. Their strongest arguments concern the power of moral claims to limit what agents may do to people. Their hardest problems concern conflict, emergency, specification, the boundary between intended and foreseen harm, and the justification of duties across differences in vulnerability and agency.

The central philosophical result is a more exact account of the deontological question: what authorizes an agent to act upon another person's claims, interests, or powers, and which reasons can make that exercise of authority permissible? The answer requires attention to principles, relationships, practical knowledge, and consequences while retaining the independent significance of the rights and obligations at stake. Deontology remains indispensable wherever moral judgment must explain both what a worthwhile action achieves and what it is permissible to do in achieving it.

References