PhronesisThe kernel, granted
Take the seed at its strongest and the tradition backs it. The will has more than one layer.
Harry Frankfurt drew the structure. In "Freedom of the Will and the Concept of a Person" (The Journal of Philosophy 68, no. 1, 1971) he separates the orders of desire. A first-order desire wants some object: the food, the drug, to leave. A second-order desire is a desire about a desire — wanting to want, or wanting not to want, some first-order thing. And among these there is a sharper case, the second-order volition: wanting a certain desire to be one's will — "the desire that moves him effectively to act," the one that carries him "all the way to action." Not just wanting to have a desire, but wanting a particular desire to be the one that wins.
On this account the mark of a person is having second-order volitions. Frankfurt names the contrast case a wanton: an agent who has first-order desires but "who [is] not a person because, whether or not he has desires of the second order, he has no second-order volitions." The wanton is moved by whatever desire is strongest and takes no stake in which one that is — "he does not care about his will." A person cares.
The classic pair makes it concrete. Two addicts feel the same craving. The unwilling addict has a second-order volition that his desire for the drug not be his will — he is carried to the needle by a desire he disowns, and so acts against his own will. The wanton addict forms no volition either way; he does not care which of his conflicting desires moves him. Same craving, different will — because one has a hierarchy with a stake in it and the other has only the crowd of first-order pulls. So in a person — an agent with second-order volitions — a volitional hierarchy exists. It is real, and it is what distinguishes willing from mere urge.
The over-reach, corrected
"A hierarchy exists" is true. It over-reaches the moment it is taken to mean the hierarchy settles which desire is really you.
Gary Watson pressed exactly this in "Free Agency" (The Journal of Philosophy 72, no. 8, 1975). Two objections, and the note is stronger for meeting them. First: why is the second-order desire more truly the agent than the first? A higher desire is still a desire — one more force in the psychic crowd. If a first-order pull can be a force other than the agent's own, so can a second-order one; height alone confers no authority to speak for the person. Second: why stop at the second order? If a desire needs a higher desire to endorse it, the higher one needs a higher one still — a third order to ratify the second, a fourth to ratify the third — and the ladder never terminates. Adding a floor above only moves the question up a floor.
Both objections point at one thing: the ascent cannot ground itself. Climbing to a higher want does not tell you which want is the ground, because the climb can always go one rung further, and no rung is authoritative merely by being higher.
What fixes the top
The regress stops not at a highest tier but where the agent identifies.
Frankfurt's later move answers Watson from inside the will. In "Identification and Wholeheartedness" (in The Importance of What We Care About, Cambridge University Press, 1988) the series is terminated not by a further desire but by the agent's decisive, wholehearted commitment to a desire — taking it as one's own "without reservation," so that no question at a higher order stays open. And in The Reasons of Love (Princeton University Press, 2004), his fullest statement of it, he names the limit: volitional necessity, the condition of being unable to bring oneself to will otherwise — the things and people one loves, where the constraint's source is "not in some alien intrusion, but in the will itself." What ends the climb is not a taller ladder but a place the agent rests: identification, not ascent.
Watson's own answer differs, and the difference is live. He grounds free agency not in a higher desire but in the agent's values — a valuational system (what one judges good and worth pursuing) distinct from the motivational system (what one merely happens to be moved by). Charles Taylor drew the same line as strong evaluation: assessing desires themselves in qualitative terms of worth — higher and lower, noble and base — not on a single scale of how much one wants ("What Is Human Agency?", 1977, reprinted in Human Agency and Language, Cambridge University Press, 1985). Whether the ground is care or value is a real dispute. What both sides share is the shape of the answer: authority comes from something the ascent alone cannot supply, not from a further floor.
The precise claim
Here is the line, corrected and bounded. A volitional hierarchy exists — layered desire with reflective endorsement is real — but it does not justify itself. Its top is not the highest rung you can climb to; it is the level you identify with, the desire you wholeheartedly take as your own, the one you cannot bring yourself to will otherwise. The hierarchy has floors. What makes one floor the ground is not its height.
What makes one floor the ground is love, not height.
This is how the seed is kept rather than overturned. "A volitional hierarchy exists" survives intact; what is corrected is only the silent extra claim that existing is enough to ground the will. It is not. The structure is real; the top is chosen.
A framework reading
One passage, offered plainly as a proposal, not a redefinition of the math. The endorsed level reads like the will's ground state — the configuration the will returns to and rests in, rather than one more displaceable layer above it. On this reading identification is exactly what makes a level a ground: a rung you climb past can always be climbed past again, but a level you rest at is where the returning stops. The ascent gives you the ladder; identification gives you the floor. (The displacement and ground-state language belongs to the published framework; this borrows the shape, not the equations.)
This is kin to Love Is a Verb, the same identification aimed at a person: love as the chosen, repeated return toward a bond. And lighter kin to The Introspection Ceiling — you may not have accurate access to which desires you truly endorse, so the hierarchy's top can be hard to report even where it is fixed.
The limit
This is philosophy of action and moral psychology, and the central dispute is open. Whether identification and wholeheartedness fully answer Watson, or whether agency must instead be grounded in values or strong evaluation, is unsettled — critics press that "decisive identification" either relocates the authority problem or smuggles in exactly the evaluative standing Watson said was missing, while the value account faces objections of its own. This note takes no winner. What it asserts is the shared, safer claim on both sides: the hierarchy exists but needs a ground the ascent alone cannot give.
And the seed's scope is narrow. Not every willing thing has the full hierarchy — wantons, non-human animals, and very young children have first-order desires without second-order volitions. Read the seed as "a volitional hierarchy exists in persons," not "in every willing thing."
Kin to Love Is a Verb — the will as chosen return, aimed at a person — and The Introspection Ceiling, with Love, Calculated Grammatically nearby and the sheet.
Rests on: Frankfurt, "Freedom of the Will and the Concept of a Person," The Journal of Philosophy 68, no. 1 (1971): 5–20 (first- and second-order desires, second-order volitions, the wanton, personhood); Watson, "Free Agency," The Journal of Philosophy 72, no. 8 (1975): 205–220 (the regress and no-special-authority objections, and the valuational-vs-motivational grounding); Frankfurt, "Identification and Wholeheartedness," in The Importance of What We Care About (Cambridge University Press, 1988), 159–176, and The Reasons of Love (Princeton University Press, 2004) (identification, wholeheartedness, volitional necessity); Charles Taylor, "What Is Human Agency?" (1977), in Human Agency and Language: Philosophical Papers 1 (Cambridge University Press, 1985), 15–44 (strong evaluation). The Frankfurt–Watson dispute — desire-hierarchy with identification versus grounding in values — is a live and unsettled debate; this note presents it as open. What is proposed is only the bounded claim that the hierarchy exists but is not self-grounding, and the framework reading of the endorsed level as the will's ground state — offered to be argued with, not a proven result.