1. Thomist Libertarians are Committed to Peter van Inwagen’s Mysterianism

Peter van Inwagen’s mysterianism about free will consists of the belief that there exists a powerful case that free will is impossible and that free will undeniably exists.Footnote ² Van Inwagen believes that there are compelling and seemingly bullet-proof arguments that free will is incompatible determinism and that free will is incompatible with indeterminism. Given that determinism and indeterminism exhaust the available options about the way reality is, it seems that free will cannot exist. Nonetheless, human beings are in some way ‘hard-wired’ to believe that free will exists.Footnote ³ Our daily experience tells us that free will obviously exists, that people are responsible for their actions, and that most human beings possess some kind of agency. So, he concludes that, ‘free will remains a mystery - that is, that free will undeniably exists and that there is a strong and unanswered prima facie case for its impossibility’.Footnote ⁴ Mysterianism can be assessed in relation to three propositions: (1) free will is incompatible with determinism, (2) free will is incompatible with indeterminism, and (3) free will exists.Footnote ⁵ Mysterians believe that there is a seemingly unanswerable case that the first two propositions are true, believe that the third proposition is true, and believe that hopefully either (1) or (2) is false. Libertarian mysterians such as van Inwagen tend to believe that (2) is false.Footnote ⁶ Compatibilist mysterians such as Noam Chomsky tend to believe that (1) is false.Footnote ⁷

Thomist libertarians are committed to mysterianism. First, Thomist libertarianism affirms, simply by virtue of being libertarian, the proposition that freedom is incompatible with determinism. So, Thomist libertarians believe that there is a good case that freedom is incompatible with determinism. This claim seems true. Second, Thomist libertarianism is committed to the proposition that freedom is incompatible with indeterminism. Thomist libertarianism is committed to this proposition because it cannot defeat the intuitive strength of van Inwagen’s Replay argument which shows that freedom is incompatible with indeterminism. So, Thomist libertarians must believe that there is a good case that freedom is incompatible with indeterminism. In conjunction with the previous point, Thomist libertarians must believe that there is a powerful case that freedom is impossible. This claim seems controversial. Many Thomist libertarians have argued that free will is compatible with indeterminism. By implication, these philosophers would believe that mysterianism is false. On the other hand, some Thomist libertarians and Thomist compatibilists have argued that Thomist libertarianism is open to mysterianism or ‘leads to’ mysterianism.Footnote ⁸ So, it isn’t clear to many that this second claim is true. Third, Thomist libertarianism, simply by virtue of being libertarian, affirms the third proposition that free will exists. This claim seems true. Finally, Thomist libertarianism, simply by virtue of being libertarian, affirms that freedom is compatible with indeterminism. So, Thomist libertarians must believe that that (2) is false. Since Thomist libertarians seem to clearly believe that a good case can be made that free will is incompatible with determinism, believe free will is compatible with indeterminism, and believe that free will exists, all that is left is to demonstrate the more contested claim that Thomist libertarians must believe that there is a good case that free will is incompatible with indeterminism, and by extension, a cumulative powerful case that free will is impossible. To begin this demonstration, I intend to first present some recent accounts of Thomist libertarianism.

2. Some Accounts of Thomist Libertarianism

There are two kinds of Thomist libertarianism that are offered in the recent literature. There are ‘Intellectual Libertarians’ and ‘Voluntary Libertarians’. I borrow this division from Tobias Hoffman and Cyrille Michon, who describe themselves and Scott MacDonald as intellectual libertarians and David Gallagher and others as voluntary libertarians.Footnote ⁹ First, there is the intellectual libertarian kind. Some intellectual libertarians are Elizabeth Anscombe, Tobias Hoffman, Cyrille Michon, Steven J. Jenson, and Scott MacDonald.Footnote ¹⁰ According to intellectual libertarians, an action is freely willed if it proceeds from a free choice. A choice is free if and only if it is made on the basis of the intellect’s undetermined judgment. The fount of free choice lies in the indeterminacy of the intellect’s power of making judgments about what to do in practical matters. For example, Scott MacDonald writes, ‘If Aquinas’s account is defensible, it preserves genuine indeterminacy in human agency while at the same time securing a necessary connection between an agent’s free choices and her reasons for acting. He does this, in effect, by identifying the locus of the indeterminacy essential to free human activity in reason rather than the will’.Footnote ¹¹ MacDonald describes Aquinas’s brand of libertarianism as being, ‘build around the around the intriguing notion of reason-based indeterminacy [emphasis added]’.Footnote ¹² MacDonald’s claim is that Aquinas is a libertarian who believes that the indeterminacy of free choice is grounded in the agent’s intellect. When an agent deliberates, she sometimes sees multiple courses of action to take in a given situation, all of which are reasonable. So, the intellect’s judgment as to what course of action to take is not necessitated by the available reasons at hand. Therefore, the intellect’s judgment that one ought to take up one course of action as opposed to another is indeterminate and free.

Another example of Intellectual Libertarianism is Hoffman and Michon who argue that according to Aquinas, ‘What makes human choices contingent is that, while the premises of a practical inference might remain unchanged and its soundness undisputed, the practical inference can be defeated’.Footnote ¹³ A human choice is a free choice in part because it is not determined by any reasons that the agent possesses. They argue that,

What is essential to the intellectual libertarian account is that the root of free will lies in the indeterminacy of the intellect’s activity. Although Hoffman and Michon may part with MacDonald and others about the manner in which the agent finally settles on her choices, they all seem to converge on the thesis that the locus of free will is grounded in intellectual indeterminacy. I will engage with Tobias Hoffman and Cyrille Michons’ 2017 paper ‘Aquinas on Free Will and Intellectual Determinism’, as representative of intellectual libertarianism.

Second, there is the voluntary libertarian kind of Thomist libertarianism. Some voluntary libertarians are Stephen Wang, David Gallagher, Eileen C. Sweeney, Jacques Maritain, and Eleanore Stump.Footnote ¹⁵ According to voluntary libertarians, an action is free if it proceeds from a free choice. A choice is free if and only if it is made on the basis of the will’s undetermined assent. The fount of free choice lies in the indeterminacy of the will’s power of assenting to the intellect’s judgments about what to do in practical matters. Stephen Wang writes that a free choice is possible, because ‘there are multiple practical truths, multiple possible acts, which all make sense in different ways. Let’s call them different lines of reasoning. Reason cannot decide between them, since reason is the very faculty that has brought them to light. It is up to the will to prefer one way of reasoning and acting. This is free choice’.Footnote ¹⁶ According to Wang, reason cannot come to a necessary conclusion about what course of action to take in any concrete situation. So, the will must introduce itself and assent to a course of action. Put more concretely, Wang says, ‘If reason itself cannot determine what is to be done, what does? If in questions of human action “the judgment of reason may follow opposite courses, and is not determinate to one”, what finally determines that a certain judgment be made? Aquinas believes it to be the will’.Footnote ¹⁷ So, the will plays a fundamental role in Aquinas’s thought as the undetermined and final arbiter of what choice to make.

Another example of voluntary libertarianism is offered by David Gallagher who writes that,

According to Gallagher, a free choice requires rational reflection and judgment. However, the intellect’s judgments are not sufficient to explain a free and indeterminate choice because the intellect’s judgments are themselves not free. So, Aquinas appeals to the will as a source of indeterminacy in order to account for free and indeterminate choice. The act of choice is an autonomous act of the will that is not necessitated by the intellect’s activity. Gallagher makes this point clearer when he says that ‘the act of the will by which that judgment is free is also the act of will found in the choice, or better, the act of will which is the choice’.Footnote ¹⁹ It is ultimately the will which has the final say about what choice to make. In sum, according to voluntary libertarians a free choice is grounded in the indeterminacy within the will’s power of assenting to the intellect’s judgments. I will engage with Stephen Wang’s recent work, Aquinas and Sartre: On Freedom, Personal Identity, and the Possibility of Happiness, as representative of voluntary libertarianism.

The point of difference between intellectual libertarians and voluntary libertarians is over what accounts for the freedom of choice. Hoffman and intellectual libertarians claim that a choice is free and indeterminate because the intellect’s practical judgment is free and indeterminate. On the other hand, Wang and voluntary libertarians claim that a choice is free and indeterminate because the will’s assent is free and indeterminate. Now, I proceed to present analytic philosopher Peter van Inwagen’s so-called Replay argument that free will is incompatible with indeterminism.

3. Peter van Inwagen’s Replay Argument from ‘Free-will Remains a Mystery

Van Inwagen’s Replay argument will be presented in two steps. First, it is presented how the argument purports to show that freedom and indeterminism are incompatible. Second, it is presented how van Inwagen responds to the libertarian’s agent-causation reply to the argument. Van Inwagen begins to argue that freedom and indeterminism are incompatible with the following thought experiment,

Van Inwagen supposes that if the probabilities of Alice telling the truth or lying appeared to an observer to both be around 0.5 after one thousand replays, then the observer would become increasingly convinced that the outcome of further replays would be a matter of mere chance. Van Inwagen continues the argument with,

The function of this thought experiment is to produce a strong intuition that Alice was not able to tell the truth and not able to lie. Neither actions are in Alice’s control, because the outcome seems to be a mere matter of chance, luck, or otherwise sheer indeterminacy. And if Alice’s action at t ₁₀₀₂ is a matter of chance, then all of Alice’s actions in the t series are a matter of chance. If an agent’s action is a matter of chance, then the agent’s action is clearly not in the agent’s control. And if an agent’s action is not in her control, then it seems that the agent’s action is not a free action. Therefore, free will and indeterminism are incompatible.

Some libertarians offer an ‘agent-causation’ response to the Replay argument. An agent-causal libertarian may argue that an act may be free and undetermined, if the act was ‘agent-caused’. Van Inwagen briefly describes agent causation as follows,

The libertarian may argue that the concept of agent causation would defeat the Replay argument in the following manner,

Van Inwagen directly applies this agent-causal response to the replay scenario,

Van Inwagen then offers a response to the agent causation proposal,

Van Inwagen surmises that introducing agent causality to the picture fails to defeat the intuitive strength of the Replay argument. It fails to defeat the Replay argument’s intuitive strength, because the concept of agent causation does nothing to assure the observers that what Alice will do at t ₁₀₀₂ is in her control. It doesn’t seem that the concept of agent causation is even relevant to the problem posed by Replay argument. Therefore, agent causation does not defeat the intuitive strength of the Replay argument. Next, I intend to consider how well some recent accounts of Thomist libertarianism fare against the Replay argument. I contend that they do not fare very well.

4. Recent Accounts of Thomist Libertarianism Cannot Defeat the Replay Argument

In order to defeat the intuitive strength of the Replay argument, a libertarian account of free will must be able to make sense of how Alice’s undetermined action at t ₁₀₀₂ is in her control and is not a matter of chance. Both kinds of Thomist libertarianism cannot make sense of how Alice’s action at t ₁₀₀₂ is in her control and is not a matter of chance for several reasons.

Firstly, intellectual libertarians cannot explain how Alice’s action at t ₁₀₀₂ is in her control because they cannot explain how Alice both freely and indeterministically chooses an action at t ₁₀₀₂. Recall that intellectual libertarians believe that an action is free if it proceeds from a free choice and that a choice is free if it is made on the basis of the intellect’s undetermined practical judgment. Regarding the formation of a free choice, Hoffman and Michon write that,

The authors claim that what finally causes an agent to conclude some practical judgment and choice, as opposed to some other practical judgment and choice, cannot be fully explained. However, the authors note that what causes an agent to conclude some practical judgment cannot be a matter of chance. So, the authors propose that what allows a partially unexplainable choice to be free and not a matter of chance, is that the agent is the agent-cause of herself coming to make that choice. The agent being a special cause of the choice and subsequent action for some reasons, is what makes the agent’s action free and distinguishable from chance events. The authors’ proposal is very similar to the agent causation solution that van Inwagen responds to in his Replay argument.

Once again consider the replay scenario of Alice. Suppose that in Alice’s case, that Alice possesses some reasons for telling the truth and possesses some reasons for lying. Suppose that Alice is sometimes the agent cause of herself settling on the reasons that it is best to tell the truth and thus tells the truth four hundred and eight times, and that Alice is sometimes the agent cause of herself settling on the reasons that it is best to lie and thus lies four hundred and ninety-three times. Suppose that we are in a position to observe t ₁ through t ₁₀₀₁. As the replays unfold, it begins to seem intuitively true that whether Alice is the agent cause of herself settling on the reasons to tell the truth or to lie is a matter of chance and not in Alice’s control. What could Hoffman and Michon’s intellectual libertarianism tell the observers that would defeat their strong intuition? It seems that they would explain that Alice is in control of what reasons she settles on at t ₁₀₀₂, because she will be the agent cause of whatever reasons she settles on at t ₁₀₀₂ and her subsequent action.

However, this explanation seems unilluminating. The explanation doesn’t seem to offer any helpful information that would allow the observers to see how Alice is in control of her choice at t ₁₀₀₂. It doesn’t seem that making Alice the agent-cause of whatever choice she makes at t ₁₀₀₂ sheds any light on how Alice is in control of her choice at t ₁₀₀₂. This is so, because the ground-floor probabilities of Alice telling the truth or lying at t ₁₀₀₂ are both around .5, even if Alice agent-causes all her choices in the t series. Intellectual libertarians concede that what ultimately moves the agent to make a choice is inexplicable. However, whether this inexplicable movement is the result of an event or an agent seems to not touch upon the strong intuition that this movement is a matter of chance. It seems as though Alice’s actions are a matter of chance whether or not Alice is the agent cause of her choices and actions. Additionally, it doesn’t seem clear what explanation intellectual libertarianism could offer, given their commitments to the existence of free will and indeterminism. Therefore, the intellectualist kind of Thomist libertarianism cannot defeat the intuitive strength of the Replay argument.

Secondly, voluntary libertarians cannot explain how Alice’s action at t ₁₀₀₂ is in her control because they cannot explain how Alice both freely and indeterministically chooses an action at t ₁₀₀₂. Recall that voluntary libertarians believe that an action is free if it proceeds from a free choice, and that a choice is free if it is made on the basis of the will’s undetermined assent. Wang writes with regard to attempts at explaining free choice and action that,

Wang writes further,

Wang claims that what finally moves the agent to make a choice is the agent’s will. However, Wang also claims that the agent’s choice cannot be explained. The choice is caused by an unanticipated and undetermined act of the will.

Once again consider the thought experiment of Alice. Suppose that Alice may exercise her will to choose to tell the truth or exercise her will to choose to lie. Suppose that Alice exercises her will to choose to tell the truth four hundred and eight times and that Alice exercises her will to choose to lie five hundred and ninety-three times. Suppose that we are in a position to observe t ₁ through t ₁₀₀₁. As the replays unfold, it begins to seem intuitively true that whether Alice wills to tell the truth or to lie is a matter of chance and not in Alice’s control. What could Wang’s voluntary libertarian account tell the observers that would defeat their strong intuition? It seems that Wang’s account would simply concede that there is nothing to tell the observers that would assure them that Alice is in control of her choices. Indeed, it seems Wang ardently works to say that it is impossible to tell the observers something that would make them think that Alice is in control of her will. In fact, Wang seems to say that even Alice herself would not know what she will decide at t ₁₀₀₂! However, by conceding to the Replay argument, Wang’s account is unable to defeat the strong intuition that Alice’s choice and action at t ₁₀₀₂ is a matter of chance. So, Wang’s account is unable to show how Alice’s free will is compatible with indeterminism. Therefore, the voluntary libertarian kind of Thomist libertarianism cannot defeat the intuitive strength of the Replay argument. If the intellectual and voluntary libertarian kinds of Thomist libertarianism cannot defeat the intuitive strength of the Replay argument, then Thomist libertarians are committed to the belief that there is a powerful case that free will is impossible. Therefore, Thomist libertarians are committed to mysterianism.

It is important to note that this conclusion is not necessarily an objection or criticism against Thomist libertarianism in toto. Perhaps there are forms of Thomist libertarianism which can be developed in such a way that can adequately respond to the Replay argument. However, the aforementioned recent accounts of Thomist libertarianism do seem unable to adequately respond to the Replay argument. Perhaps these aforementioned forms of Thomist libertarianism can be modified in such a way to incorporate some potential responses to the Replay argument which non-Thomist libertarians and analytic philosophers have offered. However, it remains to be seen the degree to which some non-Thomist responses can be faithfully incorporated into Thomism. Thomism and strands within analytic philosophy, in some cases, are not prima facie easily compatible. For example, Thomism’s commitment to agent-causality precludes event-causal type responses offered by analytic philosophers like Laura Ekstrom.Footnote ³¹ Additionally, Thomism’s commitment to divine foreknowledge and divine sovereignty raises further concerns about adopting responses offered by Lara Buchak which include beliefs in brute indeterminacy or unknowable ‘chanciness’.Footnote ³² Thirdly, many accounts of Thomist libertarianism – and St. Thomas Aquinas himself - attempt to explain the nature of free action in such a way that adopting the view that free acts could be inexplicable, like Meghan Elizabeth Griffith, would constitute a costly concession.Footnote ³³ Thomist libertarianism potentially could integrate some non-Thomist responses to the Replay argument into their accounts, but such work remains to be done. Hopefully, the present paper can motivate these attempts at faithful ecumenical philosophy.

I believe that Thomist libertarianism has room for mysterianism and that the above conclusion can be used to support such a position. In the next section, I will attempt to carve out a position for a libertarian form of mysterianism in Thomistic philosophy.

6. Mysterianism is a Legitimate Position in Thomistic Philosophy

I believe that the above conclusion may enrich and expand Thomistic philosophy by carving out a new position in the logical space of Thomism. I argue that (libertarian) mysterianism can be a specific type of either intellectual or voluntary Thomist libertarianism. Traditional intellectual and voluntary libertarians tend to affirm that there is a good case that free will is incompatible with determinism and that there is a good case that free will is clearly compatible with indeterminism. As a result of this affirmation, traditional intellectual and voluntary libertarians are saddled with having to answer the various arguments given by philosophers against the compatibility of free will and indeterminism. On the other hand, the mysterian libertarian affirms that there is a good case that free will is incompatible with indeterminism, but hopefully believes that (2) is false. Van Inwagen states at the end of his investigations into the Replay argument that, ‘I confess I believe there is something wrong with this argument. (I expect I believe this because I fervently hope that there is something wrong with it.) [emphasis added]’.Footnote ³⁴ The propositional attitude of hope simpliciter doesn’t involve being saddled with having to answer the various arguments given by philosophers against the compatibility of free will and indeterminism, because to hope that (2) is false is a far weaker attitude than to believe that (2) is false. Elizabeth Jackson offers a relevant description of the epistemic component of hope when she writes,

So, the mysterian’s belief that (2) is false is complex and resists a complete treatment within the confines of this section. In one sense, it is grounded in a ‘fervent’ hope that free will is possible, presumably for the sake of practical reasons related to moral responsibility and agency. In another sense, it is grounded in the evidence that free will clearly exists and so must be possible, thereby indicating that (2) must be false. The kind of propositional attitude involved in the mysterian’s belief that (2) is false is more complicated than more normative kinds of belief, suggesting one mark of difference between mysterians and non-mysterian libertarians. The mysterian’s belief that (2) is false tends to involve additional components related to matters of practical rationality, whereas the non-mysterian libertarian’s belief that (2) is false tends to not involve these additional components.

Mysterian libertarians in the Thomist tradition can be either intellectual libertarians or voluntary libertarians. The belief that (2) is false is underdetermined with respect to the essential claims of intellectualism and voluntarism. So, a mysterian libertarian can be either an intellectual libertarian or a voluntary libertarian. A mysterian who is an intellectualist would concur with Hoffman and Michon that in free choice, the intellect’s indeterminate practical judgment is the source of free choice. Similarly, a mysterian who is a voluntarist would concur with Wang and Gallagher that in free choice, the will’s indeterminate assent is the source of free choice. What distinguishes a mysterian intellectualist or voluntarist from a non-mysterian intellectualist or voluntarist, is that the mysterian intellectualist or voluntarist believes that there exists a powerful case that free choice is most likely incompatible with indeterminism and therefore impossible, but (hopefully) believes that free choice is compatible with indeterminism. The mysterian believes that there is a powerful case that free will is impossible, but nonetheless firmly believes that free will exists. He believes that there is some flaw in the compelling arguments against free will’s existence, specifically the arguments which purport to show that free will is incompatible with indeterminism. On the other hand, non-mysterians of both stripes deny altogether that there is a powerful case against free will’s existence. So, a libertarian mysterian, in the footsteps of Peter van Inwagen, could be at home in the Thomist tradition in either the intellectualist or voluntarist camps as a special type of both respective species.