QuestionsAre we free?Being free is having the will you want to have
Are we free?Free without absolute free will

Being free is having the will you want to have

Second-order desires (the unwilling addict)

Harry Frankfurt 1929 · 2023One can be responsible without having been able to do otherwise; being free is wanting what one wants to want.
See profile
What it claims

One can have a free will in a determined world: freedom of the will is the agreement between the will one has and the one one wants to have. Frankfurt notes that his conception is neutral with regard to determinism.

Why

We can want one desire rather than another: being free is having the will we want to have, even in a determined world.

In short

"Being free is having the will you want to have" is an argument by Harry Frankfurt ("Freedom of the Will and the Concept of a Person", 1971) on the question "Are we free?" (answer: free without absolute free will). Its claim: One can have a free will in a determined world: freedom of the will is the agreement between the will one has and the one one wants to have. Frankfurt notes that his conception is neutral with regard to determinism. According to AskPhilosophy's logic check (Z3 solver), the conclusion does follow from its 3 premises. AskPhilosophy presents 3 objections, 3 of them with a documented reply, and the original text behind each premise, checked word for word.

"Freedom of the Will and the Concept of a Person", 1971
Valid

One drug addict who hates his addiction and another who is happy with his take the same drug for the same physical reason. Yet we sense that they are not free in the same way. In 1971, Harry Frankfurt draws an idea from this: freedom of the will does not consist in being able to do otherwise, but in having the will one wants to have. And this can be true in a world where everything has causes.

P1A person does not only have first-order desires (wanting to do this or that). They can also have second-order desires: wanting to have a certain desire. When they want a certain desire to be their will, that is, the one that moves them to act, Frankfurt speaks of a "second-order volition".
P2Having freedom of the will is being free to want what one wants to want, that is, to have the will one wants. Just as freedom of action is doing what one wants, freedom of the will is wanting what one wants to want.
P3The unwilling addict would like not to take the drug and wants that desire to be his will. But the desire for the drug is stronger and moves him: his will is not the will he wants. He is "helplessly violated by his own desires": his will is not free.
P4 · implicit premiseWhether a person's will agrees with their second-order volitions is a fact about the structure of their desires. This fact can hold even if everything is caused in advance, and it does not require that one could have done otherwise.
Therefore one can have a free will in a determined world: freedom of the will is the agreement between the will one has and the one one wants to have. Frankfurt notes that his conception is neutral with regard to determinism.

click a premise to reject it

If you accept every premise, logic forces you to accept the conclusion. So the debate is about the premises.
Checked by a logic solver: the premises cannot all be true while the conclusion is false.
Without the implicit premise: Without P4, one could accept the definition and still say that the agreement between orders of desires is never real if everything is determined, for instance because freedom requires being able to do otherwise. Incompatibilists say roughly this.
The argument is valid. P1 to P3 define freedom of the will as an agreement between the will and second-order volitions, and P4 says this agreement can exist even in a determined world. The conclusion follows. The objections target P2 (why would the second order have authority?) and the conclusion (what is this agreement worth if someone else manufactured it?).
Analogy. Two passengers are on the same train. One chose the destination, the other was put aboard by force. The train goes to the same place, but one is travelling and the other is being carried. For Frankfurt, the difference between the two addicts is of the same kind: the path is the same, but one wants to go there and the other endures it.

Objections

3

Each objection is a conversation: click to see who replies and where the debate stands.

G
Gary Watson 1975 · attacks premise 2

Frankfurt assumes that second-order desires speak for the person, because we fear that first-order desires may not. But why would the same doubt not return at the second order? Why stop at that level, and not at the third, the fourth, without end? A higher-order desire is still a desire: nothing gives it more authority.

Frankfurt replies
Harry Frankfurt 1971 · Frankfurt's reply (1971 text, written before the objection)
Frankfurt himself sees the risk of an endless series of ever higher-order desires. He proposes to end it without cutting it off arbitrarily: when a person identifies decisively with one of their desires, this commitment "resounds" throughout the potentially endless series of higher orders, and the question of a still higher order no longer arises.
G
Gary Watson 1975
Watson does not keep the hierarchy of desires: for him, the question is whether the person's desires agree with their values, that is, with their judgment about what is good to do, rather than with their higher-order desires.
The debate goes on (our summary) Frankfurt developed his answer later (for example in "Identification and Wholeheartedness", 1987), speaking of wholehearted commitment. His critics think that "decisive identification" is just one more act, and that nothing explains why this act would have the authority that the others lack. The debate is not closed.
A
Alfred Mele 1995 · attacks the conclusion

Ann acquired her preferences and values in the normal way. Beth was manipulated by neuroscientists to have exactly the same ones, and after the manipulation she reflectively endorses them. On Frankfurt's account, Beth has the agreement between her orders of desires. Yet, says Mele, she has only an "ersatz self-government", because her values were imposed on her. A definition that looks only at the structure of the moment therefore lets through cases where one is not free.

Frankfurt replies
M
Michael McKenna 2004 · "Hard-line" reply
One can hold firm: Beth acquired her values in a strange way, but everyone acquires their values in ways that are not fully under their control. If Beth is responsible for her actions, she is like everyone else.
The debate goes on (our summary) Frankfurt's supporters must therefore choose: add a historical condition (as Fischer and Ravizza do in 1998, for whom an agreement produced by brainwashing does not make one responsible), or accept that Beth is as free as we are. The first route moves away from the simplicity of Frankfurt's idea; the second shocks many readers.
D
Derk Pereboom 2001 · attacks the conclusion

The four-case argument. Plum kills White while satisfying the conditions that compatibilists propose. In case 1, neuroscientists manipulate him directly; in case 2, they have programmed him; in case 3, his upbringing and culture fix his character; in case 4, he is a normal human in a determined world. Pereboom claims there is no relevant difference between these cases. Plum is not responsible in case 1, so he is not responsible in case 4. A will that agrees with its volitions is not enough for freedom if everything is determined.

Frankfurt replies
S
"Soft-line" compatibilists (Fischer, Mele, Demetriou) 2004
In case 1, the manipulation undermines Plum's responsibility for some other reason than the mere fact that his behaviour is determined. This stops the generalisation from case 1 to the later cases.
M
Michael McKenna 2008 · "Hard-line" reply
One can instead argue, on compatibilist grounds, that Plum is responsible in case 4, and extend this conclusion to the earlier cases since he meets the same conditions there.
The debate goes on (our summary) Both replies have costs. The soft line must explain which difference matters between case 1 and case 4 without contradicting itself; the hard line must accept that even case 1 leaves Plum responsible, which Pereboom finds absurd. Pereboom refined his argument in 2014 to answer the soft line. The debate is open.