1 Introduction

This paper explains how a libertarian who endorses ACT, an agent-causal theory providing the grounding conditions for the performance of an action for which the agent is morally responsible, can solve a problem constitutive luck poses for moral responsibility. It employs the following terminology.

“[A] being who A-s at t is basically morally responsible for A-ing only if there is another possible world with the same past up to t and the same laws of nature in which, at t, he does not A” (Mele, 2013, 238) (emphasis added). To be basically morally responsible for performing A, a human agent, H, must exercise free will, “the strongest sense of control over [one’s] actions necessary for moral responsibility” (McKenna & Pereboom, 2016, 6). By assumption, if, in addition to exercising free will in performing A, H can understand A’s moral significance, then H is morally responsible for performing A. Indeterminism is the denial of determinism, the view that “for any given time, a complete statement of the facts about that time, together with a complete statement of the laws of nature, entails every truth as to what happens after that time” (Fischer & Ravizza, 1998, 14). Compatibilism is the denial of incompatibilism, the view that “the type of global deterministic causation described by determinism is metaphysical[ly] incompatible with—precludes, undermines, destroys, makes impossible” (Mickleson, 2019, 232) H’s free will. Broad incompatibilism is “the view that both free will and [moral] responsibility are incompatible with determinism” (Clarke, 2003, 12). According to libertarianism, incompatibilism and indeterminism are true, and H can exercise free will. And ACT-endorsing libertarianism is the view that: (i) broad incompatibilism is true; (ii) libertarianism is true; (iii) ACT provides the correct conditions for the performance of an action for which H is morally responsible; and (iv) H is sometimes morally responsible for performing actions because H sometimes satisfies ACT’s conditions.

Assume H performs A at t. Can H be morally responsible for performing A, given H’s constitutive luck? Such luck includes H’s lack of control over “the kind of person [H is], where this is not just a question of what [H] deliberately do[es], but of [H’s] inclinations, capacities, and temperament” (Nagel, 1979, 28). H’s constitutive luck includes H’s lack of control over H’s character and decision-making capacities. When H performs A at t, H has an internal condition at t, “something specified by the collection of all psychological truths about [H] at [t] that are silent on how [H] came to be as [H] is at [t]” (Mele, 2019, 6). H’s internal condition at t is largely due to H’s character and decision-making capacities. So, since H’s constitutive luck includes H’s lack of control over H’s character and decision-making capacities, H’s constitutive luck is largely responsible for H’s lack of control over H’s internal condition at t.

According to Galen Strawson’s Basic Argument, H’s constitutive luck prevents H from being morally responsible for performing A at t:

  • (1) Nothing can be causa sui—nothing can be the cause of itself.

  • (2) In order to be truly morally responsible for one’s actions one would have to be causa sui, at least in certain crucial mental respects.

  • (3) Therefore nothing can be truly morally responsible (Strawson, 1994, 5).

If (2) is true, then H cannot be morally responsible for performing A at t because H lacks sufficient control over H’s internal condition at t, which is due in no small part to H’s character and decision-making capacities. According to (2), for H to have control over H’s character and decision-making capacities sufficient for moral responsibility, H must possess a power that, as (1) asserts, nothing can have, the power to be causa sui regarding its own character and decision-making capacities.Footnote 1 “[I]f [H] is to be truly responsible for how [H] acts, [H] must be truly responsible for how [H] is, mentally speaking—at least in certain respects” (Strawson, 1994, 6). And to be truly responsible for that, “[H] must have consciously and explicitly chosen to be the way [H] is, mentally speaking, in certain respects, and [H] must have succeeded in bringing it about that [H] is that way” (Strawson, 1994, 6). But then, a regress precludes H’s moral responsibility for performing any action:

When an agent acts, she acts because of the way she is. But to be morally responsible for acting, the agent must then be morally responsible for the way she is, at least in key mental respects. But if an agent is to be morally responsible for the way she is in those key mental respects, she must be responsible for the way she is that resulted in those mental respects. This reasoning generates a regress, which indicates that finite beings like us can never satisfy the conditions on moral responsibility. (McKenna & Pereboom, 2016, 265)

The regress entails that H cannot be causa sui in respect to H’s character and decision-making capacities, for H, a finite being, cannot consciously and explicitly choose that H’s constitutive properties are a certain way and bring it about that they are that way “all the way back.” According to the Source Model of free action, the Basic Argument’s regress might seem to entail that H’s constitutive luck precludes H’s free will and, thereby, H’s basic moral responsibility for performing A at t. On the Source Model, H’s exercising free will in performing A at t “essentially involves [H]’s having a certain role in the causal history of [A], a role that secures [H]’s control by being in some way [A’s] causal origin” (McKenna & Pereboom, 2016, 146). To be the Source Model’s free-will-exercising causal origin of A, H must have sufficient control over H’s internal state at t, which H could have only if H were causa sui in respect to H’s character and decision-making capacities. H’s constitutive luck “keeps [H] from satisfying the source condition on free will” (Mickleson, 2019, 233) because such “freedom‐relevant sourcehood can only be achieved through an act... of ex nihilo self‐creation” (Mickleson, 2019, 246).

In what follows, Sect. 2 considers a compatibilist solution to the constitutive luck problem for moral responsibility posed by the Basic Argument. A compatibilist can try to solve this problem without worrying about basic moral responsibility because, according to compatibilism, if determinism is true, then H can be morally responsible for performing A at t even though H cannot be basically morally responsible for performing A at t. Section 3 describes features of ACT that explain how H can be basically morally responsible for performing A at t while providing the basis for a solution to the Basic Argument’s constitutive luck problem. And Sect. 4 presents that ACT-endorsing libertarian solution, which enables a libertarian believing in basic moral responsibility to address the Basic Argument’s constitutive luck problem.

2 A Compatibilist Solution to the Constitutive Luck Problem for Moral Responsibility

According to the Basic Argument’s constitutive luck problem for basic moral responsibility, to be the Source Model’s free-will-exercising causal origin of A, H must have sufficient control over H’s internal state at t, which H could have only if H were causa sui in respect to H’s character and decision-making capacities. One compatibilist solution explains why H has enough control at t for H to exercise free will at t even though H is not causa sui in respect to H’s character and decision-making capacities:

[I]t seems to me that if someone does something because he wants to do it, and if he has no reservations about that desire but is wholeheartedly behind it, then—so far as his moral responsibility for doing it is concerned—it really does not matter how he got that way. One further requirement must be added to this: the person’s desires and attitudes have to be relatively well integrated into his general psychic condition. Otherwise they are not genuinely his, but are merely disruptive intruders on his true nature. As long as their interrelations imply that they are unequivocally attributable to him as his desires and his attitudes, it makes no difference—so far as evaluating his moral responsibility is concerned—how he came to have them. (Frankfurt, 2002, 27)

H can exercise free will at t even though H is not causa sui in respect to H’s character and decision-making capacities. So far as evaluating whether H is morally responsible for performing A at t, it makes no difference how H came to be in the internal condition H is in at t. In particular, it makes no difference that H came to be in the internal condition H is in at t without being causa sui with respect to H’s character and decision-making capacities. Even though nothing can be causa sui with respect to its character and decision-making capacities, if determinism were true, then H would, on this compatibilist picture, exercise free will at t in virtue of H’s brain realizing a deterministic reasons-responsive guidance-control mechanism that causes A at t. And since H also can understand A’s moral significance, H can be morally responsible for performing A at t, even though determinism precludes H from being basically morally responsible for performing A at t.

3 Selected Features of ACT

ACT is an agent-causal actual-sequence account of the performance of an action for which H is morally responsible. This section describes features of ACT that explain why H can be basically morally responsible for performing A at t while providing the basis for a solution to the Basic Argument’s constitutive luck problem for moral responsibility. ACT explains the performance of an action for which the agent is morally responsible by providing that explanandum’s grounding conditions. And some of ACT’s grounding-condition facts are about a causal sequence that must obtain for H to be basically morally responsible for performing an action.

ACT adopts a particular take on the Source Model of free action. “[I]n describing anything as an act there must be an essential reference to an agent as the performer or author of that act... in order even to know that it is an act” (Taylor, 1966, 109). “There must... not only be this reference to [H] in distinguishing [H’s] acts from all those things that are not acts, but it must be a reference to [H] as an active being” (Taylor, 1966, 111). “The only conception of action that accords with our data is one according to which people—and perhaps some other things too—are sometimes, but of course not always, self-determining beings... beings that are sometimes the causes of their own behavior” (Taylor, 1992, 51).

To put ACT’s take on the Source Model of free action differently: Since an agent is a thing that can perform actions, an agent must be an “active being” to be the causal source of an action, free or otherwise. On Sect. 2’s compatibilist Source Model position, H is a human animal that performs an action insofar as H’s brain realizes an event-causal process that causes H to move a certain way. Compare the questionable idea that a particular local ambient environment is a thing that performs a tornado insofar as a part of that environment realizes an event-causal process that causes it move a certain way.

ACT rejects the idea that H could be an agent performing an action insofar as H is caused to move the way Sect. 2’s compatibilist Source Model picture suggests. ACT would similarly reject the idea that, when a tornado occurs, a local ambient environment could be an agent performing an action. A tornado is an event that (i) is not an action performed by an agent and (ii) is caused insofar as part of a local ambient environment realizes an event-causal process that causes the local ambient environment to move a certain way. According to ACT, Sect. 2’s compatibilist Source Model picture’s claim that H performs A at t insofar as a part of H—H’s brain—realizes an event-causal process that causes H’s body to move is false because “[r]andom acts and caused acts alike seem to leave us not as the... originators of action but as an arena, a place where things happen, whether through earlier causes or spontaneously” (Nozick, 1981, 292). ACT posits that, because H must cause H’s behavior when H freely acts, a correct analysis of one of H’s free actions must include an essential reference to H qua cause of H’s behavior when H freely acts.

According to ACT, when H performs A at t, H exercises free will if and only if H satisfies three strong-control conditions at t: a causal source condition, a consciousness condition, and a reasons-responsiveness condition. Because ACT’s reasons-responsiveness condition is the primary basis of ACT-endorsing libertarianism’s solution to the Basic Argument’s constitutive luck problem for moral responsibility, the following discusses ACT’s causal source condition and ACT’s consciousness condition only briefly.

3.1 ACT’s Causal Source Condition

H’s actions are almost entirely event-causal processes that begin with an uncaused substance-causal process, itself a causally complex event involving H’s brain. The initiating uncaused causally complex event that begins A has the form (H’s causing e), where e is a token event involving parts of H’s brain (see O’Connor, 2000). (H’s causing e) at t’s beginning begins H’s performing A at t.

H, a substance, causally settles a matter at t if and only if the matter is open to resolution in multiple ways at every moment up until t, and, at t, H resolves that matter in a determinate way by exercising irreducibly substance-causal power. H’s causally settling a matter presupposes “a question capable of being resolved in different ways at all times up until a certain moment—the moment of settling—at which” H “causes it to become resolved in one particular way” (Steward, 2012, 39). Unless H causally settles, at t, that e occurs as and when e does down to every microphysical detail, H is not performing an action at t. Whenever H acts, H causally settles numerous matters about e’s occurrence down to the microphysical level by exercising enough control to be performing an action, even when H lacks basic moral responsibility for doing so. H satisfies the causal source condition for free action when H begins the process that A is by exercising down-to-the-microphysical-level, causal-settling-control over e.Footnote 2

3.2 ACT’s Consciousness Condition

For H to be exercising free will when H performs A at t, ACT requires H to be conscious at t in the sense of not being asleep or in a similar state of automatism. According to a psychological concept of mind, the mental “[i]s the causal or explanatory basis for behavior. A state is mental in this sense if it plays the right sort of causal role in the production of behavior, or at least plays an appropriate role in the explanation of behavior” (Chalmers, 1996, 11). Being awake is a psychological state that “can plausibly be analyzed... in terms of an ability to process information about the world and deal with it in a rational fashion” (Chalmers, 1996, 26). H satisfies ACT’s consciousness condition at t if H is awake at t and able to exercise this type of strong-control by processing information about the world and dealing with it in a rational fashion. When asleep or in a similar state of automatism, H cannot exercise this type of strong-control and, therefore, cannot exercise free will.

3.3 ACT’s Reasons-Responsiveness Condition

ACT’s third strong-control condition is the primary basis of ACT-endorsing libertarianism’s solution to the Basic Argument’s constitutive luck problem for moral responsibility. All H’s free actions are reasons-responsively performed, and all H’s reasons-responsively performed actions result from a sequence, IR1HR2e, featuring two token causal relations, R1 and R2, which cannot be reduced to or realized in token event-causal cause-and-effect relations governed by deterministic or probabilistic event-causal laws of nature. I designates some of H’s propositional intentional states, and e designates the brain event discussed above in connection with ACT’s causal source condition.

R1 is a causal influence relation. H causes e under the causal influence of I such that H can exercise a distinctive type of strong-control, which the Stoics attributed to human agents:

In the Stoic theory of agency, a mature human agent normally has the power to freely and voluntarily assent to, dissent from, or suspend judgment with regard to any proposal for action suggested by its motivational and doxastic states. Its source is the rational, ruling, part of the soul—the hegemonikon. This is a conception of an agent having the executive power to determine which of her motivational and doxastic states will result in action, and which will not. (Pereboom, 2014, 57)

ACT attributes to H such an executive “independence of the causal efficacy of... motivational and doxastic states” (Pereboom, 2014, 57) such as I.Footnote 3 Because neither R1 nor R2 can be reduced to or realized in token event-causal cause-and-effect relations covered by deterministic or probabilistic laws of nature, at t, the laws of nature cannot ground any stable long-run probability that H causes e under the causal influence of I. The nonexistence of any nomologically grounded long-run probability reflects H’s executive independence at t from the causal influence of H’s motivational and doxastic states, including those designated by I.

To illustrate H’s executive causal independence from I, imagine that H deliberates about what to do and decides, at t, to perform φ later.Footnote 4 Assume that H’s deciding to perform φ later is a morally sub-par decision. The morally sub-par action, A, that H performs at t in this illustration is the act of forming the intention to φ later. I designates those of H’s beliefs and desires causally operative in H’s deliberation. Brain event e is the intention to φ later that H forms at t in virtue of substance-causing e at t. The uncaused causally complex event (H’s causing e) is the action, A, that H performs at t.

As with any actual-sequence account of reasons-responsively performed action, ACT’s account must fill the “gap” between I and A with a “non-deviant” causal chain such that I’s propositional content entails H’s practical reasons for performing A. The sequence IR1HR2e is ACT’s gap filler. As IR1HR2e’s central causally operative element, H must be able to comprehend and reason based on the propositional contents of I. Those contents are H’s practical reasons, at t, for deciding to φ later. When, at t, H decides to φ later in response to (that is, for) those reasons, H’s deciding to φ later is (H’s causing e), the uncaused causally complex event (the morally sub-par action), A, that H performs at t under I’s causal influence.

According to ACT, although I causally influences H when H deliberates and makes a decision, IR1H is not reducible to or identifiable with any cause-and-effect relation(s). Because H is a substance and all effects are events, H is not in the correct category to be an effect. Nor does I causally contribute to e as, say, Billy’s throwing blue paint causally contributes to a green spot’s formation in the following:

Billy and Suzy splash paint on the wall. Billy throws blue paint. Suzy throws yellow paint. A green spot forms on the wall where the blue paint and yellow paint run together. Billy’s throw and Suzy’s throw are joint causes of the green spot. Though each throw causes the appearance of the green spot, it does so only in virtue of the other throw. (Johansen, 2014, 1)

If event X causally contributes to effect Y in this sense, then there must be an entirely event-causal chain running from X to Y. There is such a chain from Billy’s throw to the green spot’s formation. Downstream from Billy’s throw, token events only are causally efficacious. But there is no such chain running from I to e, as H, and only H, substance-causes e under I’s causal influence.

According to ACT, if, in contrast, H could respond to reasons in virtue only of H’s brain realizing a reasons-responsive guidance-control mechanism operating in accordance with deterministic or probabilistic event-causal laws of nature, then H would lack the reasons-responsive type of strong-control necessary to exercise free will. H’s brain realizing such a token event-causal guidance-control process leaves no room for H to exercise the executive causal independence that ACT requires for H to be exercising the sort of reasons-responsive strong-control necessary for free action.

4 ACT-Endorsing Libertarianism’s Solution to the Basic Argument’s Constitutive Luck Problem for Moral Responsibility

Recall that, according to the Basic Argument’s constitutive luck problem for basic moral responsibility, to be the Source Model’s free-will-exercising causal origin of A, H must have sufficient control over H’s internal state at t, which H could have only if H was causa sui in respect to H’s character and decision-making capacities. As Sect. 3 explains, ACT’s reasons-responsiveness condition for free action requires H to exercise a sort of reasons-responsive strong-control at t to perform an action at t for which H is basically morally responsible. As also explained, H’s internal condition at t is specified by the collection of all psychological truths about H at t that are silent on how H came to be as H is at t. Among those psychological truths are facts about how much causal influence various elements of I have on H at t. Those “amounts of causal influence” facts are silent about: (i) their being in no small part due to H’s character and decision-making capacities at t and (ii) H’s not being causa sui with respect to H’s character and decision-making capacities.

Assuming H is mature enough at t, by then, H’s character has developed sub-par and on-par aspects. In Sect. 3’s illustration of reasons-responsive strong-control, because H can exercise executive causal independence from I’s various elements, H can respond at t to the causal influence from H’s character’s sub-par aspect and make the morally sub-par decision H makes at t—the decision to φ later. Or H can respond at t to the causal influence from H’s character’s on-par aspect and not make that morally sub-par decision at t. H’s ability to do either at t does not depend on whether the amount of causal influence on H at t from H’s character’s sub-par side is less than, equal to, or greater than the amount of causal influence on H at t from H’s character’s on-par side. In this sense, H satisfies ACT’s reasons-responsiveness strong-control condition for free will at t by exercising a character-independent type of reasons-responsive strong-control at t. When H exercises ACT’s reasons-responsive character-independent “strong-control” in performing A at t, H does so in virtue of A causally resulting from the actual sequence IR1HR2e obtaining at t’s beginning. Additionally, according to ACT-endorsing libertarianism, H can be basically morally responsible for performing A at t because neither R1 nor R2 are governed by any deterministic or long-run-probabilistic laws of nature.

For the foregoing reasons, in addition to explaining why agents can be basically morally responsible for what they do, ACT-endorsing libertarianism can solve the Basic Argument’s constitutive luck problem for moral responsibility. This result is significant because it enables libertarians believing in basic moral responsibility to address that problem.