Is Kantian Deontological Ethics Convincing?
Kantian deontological ethics is not convincing
- Kantian deontological ethics holds that the moral worth of an action derives entirely from the will of the agent — specifically, from whether the agent acts out of respect for the moral law rather than from inclination or desire. Kant argues that this moral law takes the form of a categorical imperative: an unconditional command of reason binding on all rational beings regardless of their circumstances. He offers three formulations: the universal law, the formula of humanity, and the kingdom of ends. - Kant's ambition is to ground morality entirely in pure practical reason — to derive moral content, moral motivation, and moral bindingness from rationality alone, without appeal to desire, feeling, or contingent interest. This is a demanding project, and its difficulty is what the three objections I will consider each expose from a different angle. - I will consider three objections — problems with the universal law formula, the value of loving motives, and Foot's hypothetical imperatives objection — and argue that together they expose the same underlying failure. - I will conclude that Kantian deontological ethics is not convincing.
Section 1: Pure Formalism Cannot Generate Moral Content
- O'Neill's amendment handles both objections with some success. The false negative is cleanly resolved, and the false positive appears to be blocked. The formula is more nuanced than critics assume. - However the false positive resolution is less secure than it appears. The result depends entirely on how specifically the maxim is formulated. There is no principled guidance on what level of specificity is appropriate, and a sufficiently unusual lying maxim could always be constructed to pass the test. O'Neill handles the examples given but does not close the problem in principle. - The deeper issue survives O'Neill's amendment: she shows the formula can be refined to avoid certain misclassifications, but refinement is not the same as explanation. What O'Neill does not provide is a principled account of why formal universalisability should track moral status at all. The formula tells us about the logical structure of maxims — it says nothing about their content. That a maxim can be universalised without contradiction does not, on its own, show it is morally permissible; and that it cannot does not show it is immoral. The false positive problem makes this concrete: the formula only generates the right result when the maxim is formulated at an appropriate level of specificity, but the judgement about what level is appropriate is itself a substantive moral judgement — one that cannot come from the formula itself. The moral content is being imported by whoever formulates the maxim, not derived from the universalisability test. In this instance pure formalism cannot do the work Kant assigns it. - Furthermore, O'Neill's amendment is no longer purely formal in Kant's own sense — her refined test requires a judgement about which maxims are genuine candidates for duty, which presupposes moral content the formula was supposed to generate. The cost of O'Neill's move is that it quietly abandons the ambition of grounding moral content in pure form alone, which is precisely what the Kantian project requires.
Kant: The Universal Law Formula
- Kant's first formulation of the categorical imperative requires that we act only on maxims we can universalise without contradiction. A maxim fails the test if universalising it produces a contradiction in conception — as with lying promises, where universal adoption would undermine the very practice of promising — or a contradiction in will, where universalising produces a state of affairs rational beings could not consistently will, as with refusing to develop one's talents. The formula provides a precise, a priori procedure for determining moral duty that applies equally to all rational beings, independent of any particular desires or consequences.
: False Negatives and False Positives
- The formula generates false negatives: some non-universalisable maxims are morally neutral rather than wrong. Consider the maxim 'I will sell chocolate but not buy it.' Its universalised counterpart — everyone sells but no one buys — is contradictory, since selling presupposes buyers. Yet the maxim is plainly morally neutral. The formula incorrectly classifies it as impermissible. - The formula also generates false positives: some universalisable maxims are immoral. Consider the highly specific maxim 'I, who have ten letters in my surname, will make lying promises in the bank on Tuesdays.' Because so few people fit this description, universal adoption would never undermine the practice of promising. The formula permits it — yet it clearly involves lying and should be forbidden. Both objections point to the same underlying problem: the universalisability test is a purely formal procedure that tracks logical structure, not moral substance — and formal consistency alone cannot determine what is morally right or wrong.
O'Neill: O'Neill's Kantian Reply
- O'Neill argues the false negative objection misreads the formula. A maxim that fails the universalisability test is not thereby immoral — it is merely not obligatory. O'Neill refines the test: if neither the maxim nor its contrary can be universalised, acts conforming to either are permissible. Applied to the chocolate maxim — neither 'I will sell but not buy' nor 'I will buy but not sell' can be universalised — the maxim is correctly classified as permissible. The false negative is resolved. - O'Neill's amended test also appears to handle the false positive. The contrary of the lying surname maxim — 'I, who have ten letters in my surname, will not make lying promises on Tuesdays in the bank' — can straightforwardly be universalised. So on O'Neill's test the original maxim is impermissible, which is the correct result.
Section 2: The Reason/Feeling Distinction Is Unstable
- Kant's response has genuine force. The desire to ground morality in something objective and universal is a serious philosophical motivation, and the worry that emotion-based morality would be irreducibly subjective is not easily dismissed. - However the sharp reason/feeling distinction Kant depends on is less stable than he needs it to be. Aristotle argues that virtuous emotional dispositions — courage, compassion, love — are not merely lucky accidents but achievable through habituation and rational moral education. If reliable emotional motivation is possible through rational cultivation, the boundary between reason and feeling blurs precisely where Kant needs it to be sharpest. A person who visits a friend from genuinely cultivated love is not acting arbitrarily — they are acting from a stable, rationally developed disposition. Kant's reliability argument loses its force. - Being wrong about motivation does not destroy the rest of the theory — the objection targets what must accompany right action, not the procedure for identifying it, so the deontological framework survives. The conclusion is incompleteness, not falsity. But the instability of the reason/feeling distinction matters beyond this section: Kant's entire project depends on keeping reason and desire sharply separate, because if moral content cannot be derived from pure reason alone it must draw on something else — and once that concession is made, the project of grounding morality entirely in rationality is already in trouble. The failure in Section 3 is the direct consequence.
Stocker / Williams: The Value of Loving Motives
- Kant holds that only actions performed from duty — out of respect for the moral law — have moral worth. Actions performed from inclination, sympathy, or love may happen to conform with duty but carry no moral value in themselves. Consider a shopkeeper who charges customers fairly: if they do so because honesty is good for business, they act in accordance with duty but not from it, and their action carries no moral worth. - This objection presses more directly on what Kant is committed to than the formula problems in Section 1. Kant's motivational account depends entirely on a sharp distinction between pure practical reason on one side and feeling, inclination, and desire on the other. Moral worth belongs exclusively to the first. Stocker and Williams argue this produces an alienated and unnatural picture of moral life. Williams argues that a friend who visits you in hospital but confesses they came only because duty required it — not out of any genuine care — has had 'one thought too many.' A virtuous person acts from love and friendship without consulting moral laws. Moral worth, on this view, can attach to loving motivation, and Kantian ethics cannot account for this.
Kant: Kant's Response — Emotions are Unreliable
- Kant can reply that moral worth cannot depend on feelings because they are unreliable, variable, and subjective. If morality were grounded in feeling, different people with different emotional dispositions would face different moral obligations — destroying the objectivity and universality morality requires. Duty is stable and available to all rational beings regardless of emotional constitution. - Kant is not claiming emotions are bad — only that they cannot ground moral worth. The moral credit belongs to the maxim acted on, not the emotional state accompanying it.
Aristotle: Aristotle Shows Emotions Can Be Rationally Cultivated
- Kant's argument that emotions are too unreliable to carry moral value is undermined by Aristotle's account of virtue. Aristotle argues that we can rationally cultivate virtuous emotional dispositions over time — courage, compassion, and love are not merely lucky accidents but achievable through habituation and moral education. - If reliable emotional motivation is possible through rational cultivation, Kant cannot argue that emotions are inherently too variable to be morally significant. A person who visits a friend in hospital from genuinely cultivated love is not acting arbitrarily — they are acting from a stable disposition they have rationally developed. Kant's reliability argument therefore loses much of its force.
Section 3: Moral Bindingness Cannot Be Derived from Rationality Alone
- Kant's reply has initial appeal — the connection between rationality and morality is a deep philosophical claim and not obviously false. - However the reply fails because the instrumental/full rational agency distinction is stipulated rather than derived. Kant defines full rational agency in a way that already includes moral responsiveness and then concludes that all rational beings are bound by the moral law — but this is circular. The conclusion is built into the definition. The argument does not show that amorality is irrational; it redescribes amorality as a failure of rationality by fiat. History provides abundant apparently rational amoralists — people capable of complex reasoning and consistent action who show no moral interest whatsoever. Kant can respond that such people are merely instrumentally rational, but this now looks like a stipulation designed to protect the theory rather than a genuine argument. - Furthermore, even granting that most people have some interest in being moral, this establishes only that moral imperatives are hypothetical imperatives most people happen to have reason to follow — which is precisely Foot's point. Williams' internal reasons thesis deepens this: a reason for action must connect to something in the agent's existing motivational set. There are no external reasons that bind agents regardless of their desires — and an unconditional categorical imperative is exactly the kind of external reason Williams' account rules out as incoherent. Kant's reply is not merely weak — it is the kind of move Williams shows to be unavailable in principle. - Hume's is-ought gap delivers the structural diagnosis: you cannot derive an ought from an is, and Kant's project of deriving moral obligation from facts about rationality is precisely this derivation. The circularity in Kant's reply is not an accidental error that better arguments could fix — it is symptomatic of the impossibility of the project itself. No revision of Kant's reply can succeed because the project of grounding unconditional moral bindingness in pure rationality alone cannot be completed without importing moral assumptions at the start. - Foot's objection therefore succeeds. And its success is the culmination of what Sections 1 and 2 already showed: Section 1 revealed that pure formalism cannot generate moral content without smuggling in substantive assumptions; Section 2 revealed that the reason/feeling distinction the project depends on is unstable; Section 3 reveals that moral bindingness cannot be derived from rationality without circularity. These are not three independent problems — they are three expressions of the same underlying failure. The Kantian project of grounding morality entirely in pure practical reason is more ambitious than any version of the theory can sustain.
Foot: Foot's Hypothetical Imperatives Objection
- Philippa Foot argues that Kant's categorical imperative is not genuinely categorical at all — that moral imperatives, like rules of etiquette, are hypothetical, conditional on whether the agent cares about being moral. - Foot draws the comparison with etiquette: 'you should not ask an acquaintance how much they earn' has a categorical appearance but only binds you if you want to follow the rules of etiquette. You do not act irrationally by disregarding it. Foot argues moral rules work the same way: they bind you only if you have a personal interest in being moral. The amoral person is not acting irrationally by ignoring moral rules, any more than someone who ignores etiquette is. All moral imperatives are therefore hypothetical: 'you should tell the truth' means 'you should tell the truth if you want to be moral.' - This does not merely point to a problem with one component of the theory. It strikes at the foundational claim of the entire Kantian project — that morality can be grounded in pure practical reason as an unconditional demand on all rational beings. If moral imperatives are hypothetical, that project fails at its root.
Kant: Kant's Reply — Rationality Entails Moral Interest
- Kant can respond that all rational beings necessarily have an interest in being moral because rationality itself requires respect for the moral law. A being with no interest in morality would not be a rational agent in the full sense — they would be exercising merely instrumental rationality, the capacity to reason about means to given ends, but not full rational agency, which requires the capacity to act on universalisable principles. The amoral person is not indifferent to morality but failing to exercise their rational nature fully.
- The formula problems are genuine but partially answerable through O'Neill's amendment — the objection does not conclusively defeat the theory. But the deeper issue survives: the formula produces the right results in paradigm cases without being able to explain why, suggesting moral content is being imported rather than derived from pure reason. - The motive objection partially succeeds: Aristotle's account of cultivated virtue destabilises the reason/feeling distinction Kant depends on. The deontological framework survives — the objection targets accompaniment not procedure — but the boundary between reason and desire that the entire project requires has been shown to be less sharp than Kant needs. - Foot's objection is fatal, and its fatality is the culmination of what the previous two sections already showed. Williams' internal reasons thesis confirms that Kant's reply is unavailable in principle; Hume's is-ought gap shows the circularity is structural rather than incidental. One might argue the theory is partially convincing — the formula of humanity captures something genuinely important about the intrinsic dignity of persons, and this insight survives even if the categorical imperative cannot be grounded as Kant claims. But a theory whose foundational principle does not exist in the form it claims cannot be convincing overall: the formula of humanity derives its force from the categorical imperative, and without a genuinely categorical imperative it loses the unconditional character that gives it philosophical bite. Kantian deontological ethics is not convincing.