Reference, error, naturalism

This chapter reinforces the HPC natural-kind view about free will by considering its advantages over other approaches in more detail, including Manuel Vargas’s revisionist approach. It is argued that Vargas’s view founders on a dilemma, which the natural-kind view escapes: Either Vargas’s approach is non-descriptivist, like the natural-kind view, in which case it is not revisionist; or it is revisionist, in which case it is not clear how it is descriptivist. Moreover, this chapter argues that the natural-kind view serves Vargas’s purposes better than his own view does. Finally, the chapter compares the natural-kind view to standard compatibilist, libertarian, and hard-incompatibilist views. Among all of these views, it is argued that only the HPC view is sufficiently methodologically naturalistic in its approach.

Conclusion

This concluding chapter summarizes the central claims of the book. Additionally, it argues that the HPC natural-kind view about free actions has the resources to address various empirical threats to free will. For example, Neil Levy has argued that recent findings about how implicit biases affect actions threatens free will and moral responsibility. However, the natural-kind view defuses this threat, including Levy’s version of it. The chapter also shows how the natural-kind view can shed light on emerging questions about whether artificially intelligent agents might ever act freely or be responsible for their actions, and if so in what sense. Finally, the chapter sketches some findings indicating that folk thinking may actually assume something like the natural-kind view.

Methodology and free will

The existence question asks whether actual human agents have free will. By contrast, the compatibility question asks whether free will is compatible with determinism. Most philosophers hope that answering the compatibility question will subsequently help to answer the existence question. However, standard approaches to free will tend to result in empirical and dialectical stalemates about how to analyze the concept of free will. At the very least, these approaches leave us unable to definitively establish the application or reference conditions of the concept, and so they make it difficult to answer not only the compatibility question, but also the existence question. The main goal of Chapter 1 is to survey a number of problems with standard approaches to free will, which rely heavily on intuitions. The aim in doing so is to motivate an alternative approach, which does not rely on intuitions.

Introduction

This chapter situates the natural-kind view defended in the book in relation to standard views about free will. Typically, standard approaches assume a broadly descriptivist theory of reference, according to which the concept of free will refers (and so free will exists and we act freely) just in case it is associated with presuppositions that are (mostly) satisfied by actual human behaviors. On the natural-kind view, by contrast, the presuppositions associated with the concept do not have to be satisfied in order for reference to succeed (or for free will to exist). According to the natural-kind view, moreover, even if people’s free-agency phenomenology influences the reference conditions of the concept, the phenomenology supports both the natural-kind view and compatibilism.

Is free-agency phenomenology accurate?

This chapter argues that free-agency phenomenology might be accurate or veridical even if determinism is true, in which case it is compatibilist. Indeed, free-agency phenomenology might be veridical even if it has libertarian presentational content and assuming the truth of determinism. Compatibilists can concede, at least for argument’s sake, that free-agency phenomenology is in one sense libertarian. Yet they should insist that it is in another sense compatibilist. Consequently, even if libertarian descriptions of free-agency phenomenology are apt, there is still a sense in which this very phenomenology might be veridical, assuming determinism. This verdict undermines a motivation for libertarianism, since it removes any presumption in favor of libertarianism based on phenomenology. It also supports the natural-kind view about free will.

Preserving preservationism (about free will)

This chapter addresses the question of how phenomenology might influence the reference conditions of the concept of free will. For descriptivists about reference, if the presentational content of free-agency phenomenology is libertarian, then descriptively libertarian reference conditions for the concept might be inherited from the phenomenology. In that case, eliminativism about the concept and denialism about free will would be true, assuming determinism. However, Gregg Caruso has maintained that even on a non-descriptivist and apparently preservationist and realist approach to the conceptual question, such as the natural-kind view, if the phenomenology has libertarian presentational content, then eliminativism and denialism are also true, once we assume determinism. Relying on the view about free-agency phenomenology developed in Chapter 4, this chapter provides a non-descriptivist defense of both preservationism and realism about free will, against Caruso’s claims. The chapter also considers Shaun Nichols’s discretionist position about free will.

Free-agency experience and belief

Accumulating evidence indicates that (1) people tend to presuppose indeterminism as required for free will, and (2) people’s free-agency experiences suggest that they have indeterministic free will. However, most deny that (3) people’s presupposition of indeterminism has its source in their experience. This chapter explains how (3) might be true. It does so by appeal to the phenomenon of prospection, which is the mental simulation of future possibilities for the purpose of guiding action. The resulting view fills in at least some of the details of the HPC natural-kind view defended in Chapters 2 and 3, and also some details of the dual-content view about free-agency phenomenology defended in Chapters 4 and 5. It also links this view about phenomenology to the natural-kind view.

Naturally free action

Adopting the alternative approach motivated in Chapter 1, this chapter argues that free will is a natural kind, by relying on the influential idea that natural kinds are homeostatic property clusters (HPCs). The resulting HPC natural-kind view about free will answers the existence question positively: free will exists and we act freely. Moreover, it does so without directly addressing the compatibility question, although the view favors compatibilism over libertarianism. The chapter also rebuts a prominent objection to natural-kind views about free will, including the HPC view. Finally, the HPC view builds on Andrew Sims’s recent view that agents are a natural kind and it yields an appealing alternative to standard approaches as well as to recent revisionist approaches to free will and moral responsibility.