Positions on Free-will can be divided up into three main kinds, according to their attitude to three theses:
Incompatibilism, (I): If Determinism is true, then we’re never free.
Free-will, (F): We are at least sometimes free.
Determinism, (D): Each state of the universe U1 uniquely determines each future state U2.
Since all three theses can’t be true, positions are defined by which of the theses are rejected. The rejection of (I), but endorsement of (F) and (D) is commonly known as Compatibilism. Rejection of (F), with affirmation of (I) and (D) is sometimes called Pessimism. Finally1, denial of (D), and the holding of (I) and (F) as true is Libertarianism, which will be my quarry in this post.
The common ground between Pessimism and Libertarianism can be merged under the following conditional, specifying the falsity of (D) in the case of free action2..
If a subject S1 is free with respect to action A in a state of the universe U1 then it is possible that S2 performs A in U2, and possible that S2 refrains from A in U2.
.. where Pessimists deny the consequent and the antecedent with it, and Libertarians affirm both. Beyond this, Libertarianism can be subdivided into two kinds. I’ll call ‘Moderate Libertarianism’, that view which allows for shadings of probability in free action, but denies that the influences on action necessitate; and I’ll call ‘Radical Libertarianism’ the view which denies any prior influence at all, taking the alternatives in free action to be equiprobable. My first argument, below, attempts to draw a contradiction from Radical Libertarianism and plausible assumptions about causality and responsibility.
First Argument: Radical Libertarianism
1. If a subject S1 is free with respect to action A in a state of the universe U1, then both S2‘s performing A and refraining from A in U2 are equiprobable, given U1 (Assumption).
2. Some S1 is free with respect to A in U1 (Assumption).
3. So, both S2‘s performing A and refraining from A in U2 are equiprobable, given U1.
4. If both S2‘s performing A and refraining from A are equiprobable, given U1, then S2‘s performing or refraining from A in U2 is not caused by U1 or any of its parts.
5. So, S2‘s performing or refraining from A in U2 is not caused by U1 or any of its parts.
6. S1 is a part of U1.
7. So, S2‘s performing or refraining from A in U2 is not caused by S1.
8. If S2‘s performing or refraining from A in U2 is not caused by S1, then S1 is not responsible for S2‘s performing or refraining from A in U2.
9. So, S1 is not responsible for S2‘s performing or refraining from A in U2.
10. A subject S1 is not free with respect to action A in a state of the universe U1 if S1 is not responsible for S2‘s performing or refraining from A in U2.
11. S1 is not free with respect to action A in U1.
12. S1 both is and is not free with respect to A in U1.
13. So one of our two assumptions must be false. That is, either 1 or 2 (or both) are false.
14. If Radical Libertarianism is true, 1 and 2 are true.
15. Therefore, Radical Libertarianism is false.
A few notes for comprehension: 1 and 2 are assumptions to be discharged when we reach 13 (their contradictoriness is the point, and not the demise of the argument); The numbers tagged to S and U are time-markers, so that S1 and U1 are to be understood as simultaneous; lastly, A is not given a time-marker, but if it had one it would be A2 throughout (I thought it confusing to have an A2 without an A1, so have left it out).
One minor worry is that, by taking S1 to be a part of U1, by taking a subject to be part of a state of the universe, I’m assuming materialism. The worry is misplaced because I intend “universe” to be all-inclusive, having no preference for material over psychic entities. A second objection concerns premise 4: why assume that the equiprobability of action/non-action amounts to a lack of cause? A few answers might do here, mine is that the equiprobability of mutually exclusive and exhaustive alternatives is a paradigm of random chance, and some entity or event counts as a cause only if it shifts the probability distribution from pure chance.
Another worry is whether 8 is true, but I have to admit, I cannot imagine any alternatives. It is entirely plausible to think that responsibility depends upon causal relations, even if one thinks that responsibility is not just a matter of causality. Finally, I need to address the objection that, although S1 is not responsible for S2‘s performing or refraining from A, S2 is. If what I’ve said about causality being a prerequisite for responsibility is correct, then for S2 to be responsible for S2‘s performing or refraining from A, it must be the case that S2 is the cause of S2‘s performing or refraining from A. However, causes precede their effects, and so S2 must be ruled out as a candidate cause.
If the above argument is sound, then Radical Libertarianism is not an option. It is not an option because if we are responsible for any of our actions, we are responsible for those of our actions which are free, whereas Radical Libertarianism leads us to deny such actions are caused and so to deny that we can be responsible for them.
Second Argument: Moderate Libertarianism
The alternative is Moderate Libertarianism. On this view some actions can be more probable than others, actions can be more probable than non-actions. Since we’ve already seen that there is a causal constraint on the ascription of responsibility, we can interpret this alternative more narrowly, as the view that all our free actions are caused, but that the causes do not necessitate, but only incline. The causes, then, are probabilistic.
I take the notion of probabilistic causation to be entirely in order. Unfortunately, it seems that Moderate Libertarianism falls to a similar line of reasoning to Radical Libertarianism. The argument:
1. If a subject S1 is free with respect to action A in a state of the universe U1, then S2‘s performing A is caused by S1, or S2‘s refraining from A is caused by S1.
2. If S1 is a cause of S2‘s performing A, then it cannot be a cause of S2‘s refraining from A.
3. Suppose S1 is a cause of S2‘s performing A.
4. Then, S1 is not a cause of S2‘s refraining from A.
5. If S2‘s refraining from A in U2 is not caused by S1, then S1 cannot be responsible for S2‘s refraining from A in U2.
6. S1 cannot be responsible for S2‘s refraining from A.
7. If Moderate Libertarianism were true, then S1 is at least sometimes responsible for S2‘s refraining from A, if it obtained.
8. Moderate Libertarianism is false.
I’ve taken a couple of short-cuts for brevity’s sake. Premise 1 builds in the causal constraint we found free action to require, and uses “cause” in the broad sense of including both probabilistic and (if you like) necessitating causes. 2 states that a cause cannot incline to or necessitate one outcome, whilst simultaneously inclining to or necessitating the exact opposite outcome. (Note that the same event/object can be a precondition to either of to exclusive outcomes, but that this does not amount to its being a cause). 3 takes one of the two options given in 1, and 4 is derived from 2 and 3.
Premise 5 is a consequence of 8 from the previous argument. Premises 4 and 5 yield 6. 7 states what I take to be a distinctive part of the Libertarian position, that we are at least sometimes responsible for both courses of action available to us, which implies that we are responsible for unlikely actions we take, as well as the obvious ones. 8 deduces that Moderate Libertarianism is false. I leave out the alternative option to 3, “Suppose S1 is a cause of S2‘s refraining from A”, to avoid repetition, but the argument would obviously run just as well with it.
To sum up, the problem with both forms of Libertarianism is that they implicitly deny a causal connection between a subject and her actions. Radical Libertarianism is most egregious in this respect, since it prohibits responsibility for any action or non-action. But Moderate Libertarianism is not much better, since it does not allow that a subject is responsible for her actions or non-actions against inclination (and then what is libertarian free will worth?). Both kinds of Libertarianism being failures in this way, I conclude that Libertarianism itself is a failure.
1 Other options are logically possible, but these are far and away the popular entrenchments.
2 I’ll follow Plantinga in focusing on free action as the manifestation of free will: “What is relevant to the Free Will Defense is the idea of being free with respect to an action. If a person is free with respect to a given action, then he is free to perform that action and free to refrain from performing it; no antecedent conditions and/or causal laws determine that he will perform the action, or that he won’t. It is within his power, at the time in question, to take or perform the action and within his power to refrain from it.” ~ Alvin Plantinga, God, Freedom, and Evil (1974), p. 29.