Kant's Humanity Formula: Why a Person Is Never Merely a Tool (2026)

Opening: The Problem Behind a Familiar Formula

Welcome to the first lecture in this series on Kant's moral philosophy through Christine Korsgaard's constructivist interpretation. We begin with a sentence that is often quoted and just as often flattened into a slogan:

Act so that you use humanity, whether in your own person or in the person of any other, always at the same time as an end, never merely as a means. 1

The sentence is Kant's Formula of Humanity, one formulation of the categorical imperative in the Groundwork of the Metaphysics of Morals. It is tempting to hear it as a prohibition on using people. That cannot be right. I use the pharmacist as a means to obtain medicine; the pharmacist uses me as a means of earning a living. A student uses a lecturer's explanation to understand a text, and the lecturer uses the student's attention as part of the activity of teaching. Human cooperation is full of means-end relations.

The moral prohibition is narrower and more demanding: do not use a person merely as a means. The word matters because it identifies a failure in the relation between agents. The wrong is not that another person figures in my plan. The wrong is that I make another person's agency serve my plan while refusing to relate to that person as an agent who can set ends, assess reasons, and participate in the terms of the interaction.

Korsgaard's importance for this problem is that she asks a question beneath the formula. What makes a reason a reason for us at all? If morality is only an external set of commands, why should a reflective agent regard it as binding? Her Kantian constructivism reconstructs moral authority from the standpoint of practical reasoning itself. The aim is not to make morality optional. It is to explain why the structure of agency already commits us to standards that cannot be discarded whenever they become inconvenient.

From Desire to a Reason

Kant's moral theory starts from a difficulty about human action. We do not merely move toward objects; we act under descriptions. I do not simply raise my hand. I raise it to vote, to ask a question, or to signal that I need help. An action is guided by a maxim, a principle that describes what I am doing and why I am doing it.

This means that human action is reflective. We can step back from an impulse and ask whether it gives us a sufficient reason. Hunger may press me toward food, but I can ask whether I should steal it. Anger may make retaliation attractive, but I can ask whether retaliation is justified. Reflection opens a gap between being moved and having a reason.

That gap creates the problem of normativity. A desire can explain why I am inclined to act, but explanation is not yet justification. The fact that I want to humiliate someone does not, by itself, show that humiliation is permissible. The fact that an action would benefit me does not settle whether I may perform it. We need some standard by which motives can be assessed.

Korsgaard's constructivist strategy begins here. In the metaethical vocabulary she helped make prominent, constructivism does not treat moral truths as reports about a realm of values entirely independent of rational activity. It also does not reduce morality to whatever anyone happens to prefer. It asks what standards are generated by a correct procedure of practical reasoning, one that respects the conditions under which a rational agent can act and endorse principles at all. The Stanford Encyclopedia of Philosophy describes this Kantian position as a form of procedural realism: normativity comes from the standards of practical reasoning rather than from an external stock of moral objects. 2

The claim is therefore conditional in a precise sense: if I am to count as an agent, I must act on principles I can take responsibility for as reasons. But the conclusion is not merely conditional in its force. I cannot coherently claim the authority of agency when choosing and then reject every constraint that agency imposes. The question is whether the constraints are built into the activity I am already undertaking.

Practical Identity and Self-Legislation

A reflective agent does not encounter herself as a bare bundle of momentary desires. She understands herself under practical identities: as a friend, a citizen, a teacher, a researcher, a parent, or simply as a person who is answerable for what she does. These identities are not just social labels. They organize what counts as a reason for her.

If I regard myself as a friend, a promise made to a friend has a significance that it would not have in an otherwise identical encounter with a stranger. If I regard myself as a scholar, accuracy in presenting an argument is not merely an efficient way to gain approval. It is part of what I take myself to be doing. A practical identity gives shape to deliberation by making some considerations reason-giving.

Korsgaard's point is not that every identity automatically generates a valid obligation. A role can be corrupt, oppressive, or internally incoherent. The question is whether I can reflectively endorse the identity and the principles that follow from it. In the Sources of Normativity, Korsgaard presents reflective endorsement as a way of asking what we are committed to when we take something to be a reason. The Tanner Lectures page identifies that 1992 lecture as the source later developed in her work on normativity. 3

This is where Kant's notion of autonomy becomes decisive. Autonomy does not mean that I may choose any rule I like. It means that the law governing my action must be one I can regard as valid from the standpoint of rational agency. Heteronomy occurs when the principle of action is borrowed from something that cannot itself justify the principle: an appetite, a threatened punishment, a desire for status, or an authority treated as an object of fear rather than as a source of reasons.

Self-legislation is therefore not private legislation. I do not legislate a law for myself in the sense of inventing a personal exception. I test my maxim as a principle that could have authority for rational agents as such. The universal law formulation and the Formula of Humanity are two ways of making this demand visible. The former asks whether I can will my maxim as a universal law. The latter asks whether my maxim respects rational nature, in myself and in others, as an end in itself. Kant presents these formulations as equivalent expressions of the same fundamental moral law. 4

What "Merely as a Means" Excludes

Consider a false promise. I need money, so I promise to repay a loan even though I have already decided not to do so. My plan depends on the other person's expectation that I am making a genuine promise. I am not simply interacting with another agent under terms she can assess. I am manipulating her deliberation so that her choice becomes a mechanism for my end.

The problem is not that she is causally involved in my success. The problem is that the principle of my action cannot be disclosed as a principle in a relation among agents. If she knew the maxim, she could not accept the proposed transaction as the transaction I represent it to be. My action attempts to control her agency by withholding the information she would need to exercise it.

This gives us a useful, though not exhaustive, test. When my plan makes another person's agency part of its means, can I relate to that person as someone who could understand and authorize the terms of the interaction? Payment, cooperation, and ordinary dependence can be consistent with respect. Deception, coercion, and exploitation are wrong when they bypass the other's rational participation and reduce the person to an instrument.

The phrase "at the same time as an end" prevents a second mistake. Respecting another as an end does not require abandoning my own ends or treating every other person's project as decisive. It requires recognizing that the other has ends of her own, and that her capacity to form and pursue them places a constraint on how I may pursue mine. Respect is therefore not the same as benevolence, admiration, or emotional warmth. It is a structure of practical recognition.

Kant's examples in the Groundwork extend the point beyond interpersonal deception. He discusses suicide as a case in which a person treats the humanity in her own person as a means to escape suffering; he discusses the false promise as a case of using another; and he considers both the refusal to develop one's capacities and the refusal to assist others. 1 The examples are controversial, especially when applied to self-destruction, disability, or severe distress. Their philosophical function is nevertheless clear: the Formula of Humanity governs how we relate to our own agency as well as how we relate to the agency of others.

Korsgaard's Constructivist Reconstruction

Korsgaard's distinctive contribution is to connect the formula to the conditions of reflective action. We value objects, projects, and outcomes by taking them as reasons within deliberation. Yet the activity of valuing is performed by an agent who must be able to regard the activity as her own. This does not make the self a valuable object in the same way as a preferred outcome. It means that the capacity for reflective agency is the standpoint from which values and reasons become intelligible to us.

The argument can be set out in four steps:

To act, I must adopt some maxim under which my movement counts as an action rather than as an event that merely happens to me.
To adopt a maxim reflectively, I must be able to regard myself as the author of the principle on which I act.
Authorship cannot be reserved for my own agency while treating every other agent's agency as disposable. The same capacity that gives my deliberation authority is present in other rational beings.
Therefore, the standpoint of agency commits me to principles that protect rational agency as an end, not merely as a resource for whatever end I happen to have.

This is not an argument that people are valuable because they are useful. It runs in the opposite direction. Usefulness is assessed from the standpoint of agents pursuing ends. The agent cannot then treat the capacity that makes ends and assessment possible as if it had only instrumental value.

The constructivist does not need to say that moral properties are hidden objects waiting to be detected. The claim is instead that practical reasoning has a constitutive structure. A principle that destroys the conditions for treating oneself and others as responsible agents undermines the very standpoint from which it is offered as a reason. This is why Korsgaard's reading is more than a modern gloss on the slogan "be nice to people." It explains why respect is a requirement of rational consistency.

Still, the reconstruction faces a serious objection. Why should a constitutive condition of agency be moral? Perhaps I can acknowledge that I must use principles to act and still decide that I care only about my own success. The constructivist answer is that this decision cannot be made from nowhere. To deliberate about whether agency matters is already to occupy the standpoint of an agent who is answerable for a principle. The issue is not whether an external force will compel compliance. It is whether the proposed principle can survive the reflective test that makes it a principle of action rather than a rationalization after the fact.

That answer may not satisfy every critic. It leaves open disputes about the exact transition from agency to morality, about whether constitutive standards can generate categorical reasons, and about how much substantive content follows from formal reflection. These are not defects to hide. They are the pressure points that make Korsgaard's interpretation philosophically valuable. It takes the question of moral authority seriously enough to risk an argument.

A Working Distinction for Moral Judgment

We can now return to the original formula with a sharper vocabulary. Ask three questions of a proposed maxim:

What end is the agent pursuing, and what principle connects the action to that end?
Does the principle depend on another person's capacity to choose, communicate, or pursue ends?
Does the principle leave that person in a position to understand and assess the terms of the interaction, or does it bypass and manipulate that capacity?

The second question does not condemn the action. The third identifies the moral danger. A contract, a lesson, a medical consultation, and a political campaign can all involve one person's activity serving another's purpose. They become morally defective when the person whose agency is being enlisted is denied the information, freedom, standing, or reciprocal recognition needed to participate as an agent.

This distinction also changes how we think about self-respect. Respecting humanity in oneself is not a demand to maximize one's achievements or protect one's image. It is a refusal to make one's own rational capacities answerable only to appetite, fear, or convenience. A person may fail, change careers, abandon a project, or need help without thereby forfeiting dignity. The moral question concerns whether she is being treated, by herself or by others, as someone whose reasons and ends matter.

Korsgaard's constructivist reading thus turns the Formula of Humanity into a lesson about practical authority. We do not first discover ourselves as agents and then decide whether agency deserves respect. Our ability to deliberate, to endorse reasons, and to take responsibility for maxims already places us within a normative space. Moral law is the demand to inhabit that space without claiming its privileges for oneself while denying them to others.

Closing: The First Commitment

The first commitment of Kantian ethics is not the promise that moral life will be easy. It is the claim that a person is never just material available for someone else's plan, including her own plan when she treats her agency as expendable. The Formula of Humanity names this commitment in the language of ends and means. Korsgaard explains its authority by returning us to the reflective structure of action: reasons must be reasons for an agent, and the standpoint that makes this possible cannot be reserved for one agent alone.

In the next lecture, we will examine the Formula of Universal Law more closely. The question will be how a maxim fails when it cannot be willed as a universal law, and whether that test is best understood as a logical contradiction, a practical contradiction, or a test of what an agent can coherently endorse.

Kant's Humanity Formula: Why a Person Is Never Merely a Tool