5 Conclusion and Considerations for Future Work

Being, World, and Loneliness

as its they-self, wherein it is weighed down into the lonely mood by refusing others as such, and judging them instead as practically useful objects. In solitude, we acknowledge that others are there and that we are among them; in loneliness, we acknowledge that others are there and that we are apart from them; in the mood of loneliness, we do not acknowledge that others are there at all, because we are so completely submerged in the material (and digital) world of equipmental things. Others lose their “otherness” and manifest to us as hollow shells. This peculiar appearance of “otherless” others is a sign of Dasein’s fleeing from itself, because it is contradictory—Dasein is also always Mitsein, so can never really be alone.

To finally bring our investigation of loneliness to a close: I posit that loneliness, perceived through such a lens and understood in the Heideggerian sense of attunement, is Dasein’s Being-alone-in-the-world, marked by a fleeing in the face of its own existential quality of Being-with that arises through dispersal into the “they.” This conceptualization differs starkly from that of both Hannah Arendt and Hans-Georg Gadamer, who describe loneliness as something that happens to the individual rather than something that the individual is.

I feel as though there is considerable room for further clarifying some of the ideas I have presented in this paper. Most striking to me are two in particular: firstly, it seems that there is room for further interpretation of my conception of loneliness as attunement in the context of Heidegger’s notion of the ready-to-hand (Zuhandenheit), i.e., of the utility-oriented way in which Dasein proximally and for the most part discovers and relates to objects in the world. (In this view, we do not usually encounter objects in the mode of detached, scientific contemplation/observation, but instead as pieces of equipment for us to use alongside other equipment in order to achieve our goals.) It appears that, in loneliness, this readiness-to-hand is exaggerated to the point of compulsivity; it is no longer a point of description that human beings are involved in equipment for the carrying out of objectives, but an ideological norm that this ought to be the case with all things, including other humans. This coincidence between readiness-to-hand and the transformation of others into equipmental objects under loneliness is not something that has fully been fleshed out here, but the trappings of a more thematic connection are surely evident.

Secondly, there is the apparent connection between the mood of loneliness and

64

Being, World, and Loneliness

the capitalistic orientation of society; in retrospect, we might say that Gadamer was on the right track when he understood loneliness in relation to structures of the division of labor and class. The “they,” as I conceive it with regards to loneliness, demands that we envelop ourselves entirely in our pragmatic activities, refuse our intrinsic relation to others, concern ourselves strictly with objects and tools, (possibly?) as a product of the alienating conditions and expectations associated with contemporary late-stage-capitalist milieu. But this thesis, of course, is only speculative. Substantial elaboration would be necessary to make it worth taking seriously.

65

AB S T R A C T

Although some metaphysicians conceive of their proper roles as discoverers and disputants of the literal, fundamental facts of the world, there are numerous reasons to be skeptical of this conception. The apparent inability of metaphysicians to make progress on resolving their debates is one such reason. In the free will debate, this inability to resolve is the result of incompatibilist free will theorists proposing semantically incompatible theories which rely on modal facts. It is unclear whether these modal facts are accessible or even relevant to the free will debate, and thus we may be skeptical that incompatibilists are engaged in a productive dispute. Compatibilism seemingly provides a way around this obstacle by grounding the truth of free will in concepts such as “free” and “compelled” rather than potentially inaccessible modal facts, allowing compatibilist theories to be evaluated in terms of broader normative conceptual schemes. Metalinguistic negotiation provides guidance for this move away from epistemically metaphysical questions by distinguishing between canonical and non-canonical disputes, revealing that not only can non-canonical disputes provide guidance for normative schemes generally, they also may be more suited to this task than their canonical counterparts. When the canonical content of disputants’ utterances hinge on potentially inaccessible modal facts, it may be advantageous to reframe these disputes as metalinguistic negotiations and thus evaluate competing metaphysical theories as part of a broader normative conceptual scheme.

67

Reconceptualizing the Free Will Debate as a Metalinguistic Negotiation

To the participants engaged in a metaphysical debate, it may seem as though they are disputing the literal, fundamental facts of reality. Hard determinists and libertarians, for example, make very different claims about the modal facts of action and free will. While hard determinists claim that our actions are necessitated by past conditions and thus we do not have free will, libertarians reject this assertion and maintain that, in virtue of having “regulative control” (Fischer) when we act, meaning “the ability to do otherwise,” we do in fact have free will. These two incompatibilist views seem to agree that whether we have free will is dependent upon the modal facts of action, and so it may be tempting to describe the proper role of free will disputants as discovering and interpreting these modal facts in order to reach a decisive conclusion.

However, as Amie Thomasson (2016) points out, metaphysical questions (and questions of modality and ontology in particular) are often “epistemically metaphysical,” or epistemically mysterious, in the sense they “can be answered neither by direct empirical methods nor by conceptual analysis” (2). It may be, then, that viewing the task of metaphysicians as “discovering especially deep or fundamental facts about the world,” a conception which Thomasson dubs “heavyweight metaphysics,” is misguided (1). If Thomasson’s conception of heavyweight metaphysics is accurate, and incompatibilist theories are predicated upon epistemically metaphysical (and thus inaccessible) modal facts of action, then we may have stronger reasons to be skeptical of the incompatibilist foundation than we do for believing either side of the incompatibilist debate.

Does this mean that all hope is lost for the free will debate and that disputants are simply wasting their time? Likely not. Although incompatibilist theories appear to rely on epistemically metaphysical (and specifically modal) facts, compatibilist theories reject the incompatibilist scheme and propose that free will ought to be evaluated independently of modal facts. I see this step from incompatibilism to compatibilism as potentially moving away from heavyweight metaphysics, but also similar to the move that Plunkett & Sundell make in their paper, “Disagreement and the Semantics of Normative and Evaluative Terms” (2013), to assert that noncanonical disputes can express genuine disagreement as metalinguistic negotiations. Metalinguistic negotiations are non-canonical disputes in which participants tacitly advocate for a particular “deployment of linguistic representations” (3) rather than expressing incompatible semantic content. I argue that since compatibilist theories

68

Reconceptualizing the Free Will Debate as a Metalinguistic Negotiation

ground the truth of the free will debate in concepts such as “free” or “compelled,” metalinguistic negotiation is exemplified by the compatibilist debate. Conversely, since incompatibilists propose theories which appear to rely on incompatible semantic content, they appear to engage in a canonical dispute. If we are right to be skeptical of heavyweight metaphysics’ ability to decisively conclude what the relevant modal facts are, then we ought to consider reframing the free will debate as a metalinguistic negotiation. The goal of this paper is to demonstrate that, by reframing the free will debate in terms of metalinguistic negotiation, it may be more philosophically productive to shift the free will debate away from incompatibilism and towards compatibilism, exploring free will issues explicitly through the lens of normative conceptual theories.

In Section I, I give a broad overview of the free will debate, summarizing the basic positions of hard determinism, libertarianism, and compatibilism. In Section II, I discuss the role of modal facts in the free will debate, arguing that certain epistemic mysteries surrounding modal facts seem to challenge the legitimacy of heavyweight metaphysics-style free will theories, i.e incompatibilist theories. Given this, I argue in Section III that since compatibilist theories do not require making the epistemically problematic commitments required by incompatibilist theories, compatibilism allows us to coherently shift the free will debate away from epistemically metaphysical modal facts and towards a more fruitful conceptual analysis by reconceptualizing disputants as engaged in a metalinguistic negotiation.

Sider (2005) gives a general overview of the free will debate, summarizing three basic positions: hard determinism, libertarianism, and compatibilism. First, he identifies the tension between two seemingly incompatible (but not obviously unreasonable) theses: (1) causal determinism and (2) free will. For this essay I borrow Plunkett’s (2015) broad definition of hard determinism. He describes causal determinism as “roughly the thesis that the facts about the way the world is, in combination with the facts about the laws of nature, fully determine all of the facts about the future” (855). Causal determinism is generally taken by incompatibilists to mean that if our actions are necessitated by prior events then we cannot have free will; this is the core position of hard determinism.

Additionally, as incompatibilists, hard determinists believe that causal determinism prevents us from making cogent ascriptions of responsibility. If it turns out

69

Reconceptualizing the Free Will Debate as a Metalinguistic Negotiation

that human actions are events similar to falling dominoes, merely the inevitable result of prior conditions, then libertarian beliefs about choice and free will must be incorrect. If true, the implications of causal determinism are potentially worrisome for our moral schemes, particularly in relation to ascriptions of moral responsibility. As Sider points out, it seems strange to blame Hitler for the 1939 invasion of Poland if we accept that his actions were necessitated by prior conditions, but we probably want to retain our ability to blame people for their destructive behavior. Of course, this is not a metaphysical argument against hard determinism, but rather a pragmatic consideration we ought to be aware of. After all, if we are deeply committed to our moral intuitions but they seem incompatible with hard determinism, then it may be pragmatically sound to advance a theory of free will that allows us to maintain ascriptions of moral responsibility.

The libertarian view, often described as the antithesis to hard determinism, claims to allow ascriptions of moral responsibility by positing the free will thesis. The basic claim of libertarianism is as follows: whereas the external laws of nature are the sole mechanism of causality in cases involving dumb objects (like a falling domino), free will allows people to choose how they act such that they are responsible for their own actions. Sider points out that this has potentially worrisome conceptual implications as well, insofar as the libertarian view seems to exclude the possibility of a “complete” psychology or theory of the mind which can fully predict human behavior. Additionally, libertarianism may also exclude the possibility of an “all encompassing physics” (a physical theory that can fully predict particle behavior), and it may be hasty to assume that human minds and action are (or rather, would be) exempt from such a physical theory.

The third basic view that Sider describes, compatibilism, rejects that the causal determinism thesis and the free will thesis are incompatible (127). Sider describes compatibilism as allowing that deterministic or necessitated actions are free if caused “in the right way,” but he acknowledges the difficulty in articulating what exactly this means (130). Compatibilism may be conceptually valuable if it allows us to preserve both causal determinism and free will, but it also seems to invite a problem of vagueness. For example, are my actions still caused “in the right way” if I would have acted differently in a marginally distinct, counterfactual scenario? Even if we grant that a free action is one appropriately caused by the person and ignore questions of context or extent, it seems that we run into definitional woes. After all, the notion of being caused “in the right way” seems to rely on what does and does not constitute being “compelled,” but how we ought to answer the question of what

70

Reconceptualizing the Free Will Debate as a Metalinguistic Negotiation

constitutes compulsion seems to rely on an unclear conceptual distinction between compulsion and necessity. If compatibilists accept the causal determinism thesis then they accept that all actions are necessitated. But if uncompelled actions are free on the compatibilist view and compelled ones are not, the distinction between free necessitated action and unfree compelled action may seem arbitrary.

In other words, compatibilism may have us drawing lines in the sand. Sider gives the example of brainwashing and moral transformation to illustrate this problem (134). Take a good person (whatever this means to you) and imagine them placed in two scenarios: in the first, they are brainwashed into becoming evil (again, whatever this means to you) by a mad scientist, and in the second, through no fault of their own, they fall in with the wrong crowd and naturally become evil. Can responsibility be similarly ascribed to the evil actions of the brainwashed person and the “wrong crowd” person? Both seem to act according to prior conditions, but Sider suspects that compatibilist accounts of free will might ascribe responsibility differently in these two cases based on competing notions of compulsion and freedom.

Having outlined these three basic positions in the free will debate - hard determinism, libertarianism, and compatibilism - I will now discuss the metaphysical commitments required by the two incompatibilist positions, hard determinism and libertarianism. In his essay, “The Incompatibility of Free Will and Determinism” (1974), Peter Van Inwagen states the following: “the concept of free will should be understood in terms of the power or ability of agents to act otherwise than they in fact do. To deny that men have free will is to assert that what a man does do and what he can do coincide.” This “power or ability” is known as “regulative control,” and on Van Inwagen’s incompatibilist conception, whether we have free will is determined by the modal facts of action or the “truth of ” regulative control: the libertarian view, “I freely chose my own actions because it is a basic fact that I could have acted differently,” or hard determinism, “I did not freely choose my own actions because it is a basic fact that I could not have acted differently.” Regardless of whether Van Inwagen’s conception is the metaphysically correct one, it grounds the truth of free will and thus the incompatibilist free will debate in the modal facts of action.

Because hard determinism and libertarianism are both incompatibilist views, it seems the incompatibilist free will debate would be settled by the discovery of

71

Reconceptualizing the Free Will Debate as a Metalinguistic Negotiation

relevant modal facts. If it were demonstrated that human action is completely determinate or, despite our best predictive efforts, still at times indeterminate, we could reject hard determinism or libertarianism in favor of its counterpart. If we could “test” for regulative control then we could determine the relevant modal facts and settle the free will debate. However, as Thomasson (2016) points out, we may have reasons to be skeptical of grounding the free will debate in the modal facts of action. For one, there appear to be “familiar and formidable epistemic problems” that prevent us from really knowing what the modal facts are. After all, we are not actually able to test for regulative control, at least not yet. We can make speculations about what such a test might consist of or reveal, but if our best method of exploring modal facts is considering how well different theories handle thought experiments and imagined cases, we should probably be skeptical that these theories really communicate “deep or fundamental facts about the world” (2). Thus, viewing the modal facts of action as especially relevant to the free will debate may be theoretically unproductive, and in doing so disputants may render themselves incapable of making epistemically sound claims.

Furthermore, while science generally makes progress on convergence and thus seems like it’s really engaged in truth discovery, the theories of disputing metaphysicians instead tend to diverge. So instead of ever being settled, metaphysical debates seem to grow increasingly complex, “with no agreement even on what might resolve [them]” (Thomasson, 2017). Thus, at least part of the problem with incompatibilism is made clear by Thomasson’s critique of heavyweight metaphysics: incompatibilist free will theories seemingly rely on unknowable facts, and attempts to ground these facts appear speculative at best, with no consensus as to how we might actually discover them. So, if the modal facts of action are potentially inaccessible and perhaps not even relevant to the free will debate, then what are our other options? One potential answer is given by compatibilists, who propose that free will theories need not rely upon facts like whether we have regulative control, but instead in a conception of agency compatible with causal determinism. Glannon (1999) gives an account of such a theory, writing that the crucial feature of free will is “not that we have the ability to choose and do otherwise, but that we acquire our reasons [for acting] autonomously and act on them in an ... uncompelled way” (188). Glannon (in light of the theory’s original authors, Fischer and Ravizza) refers to this notion as “guidance control.” Guidance control theorists claim to have a distinct advantage over regulative control theorists insofar as they conceive of “what explains and justifies attributions of responsibility to persons” independently of the modal

72

Reconceptualizing the Free Will Debate as a Metalinguistic Negotiation

facts of action. And, given our skepticism that metaphysicians have access to the relevant modal facts, guidance control may indeed be theoretically advantageous. Although I do not argue for or against guidance control theory on its metaphysical merits, I believe Glannon makes the right move in shifting the free will debate away from modality since doing so allows free will disputants to move away from an epistemically metaphysical, heavyweight metaphysics-style debate and towards a metalinguistic negotiation.

Having discussed the tenuous relationship between incompatibilism and modality, it seems appropriate to turn this discussion towards what we want a free will theory to do. This may help articulate why the compatibilist free will debate should be evaluated as a metalinguistic negotiation. On the incompatibilist scheme, the truth of free will seems to function as little more than a reflection of modal facts like whether we have regulative control. However, in addition to accurately reflecting certain modal facts, a free will theory ought to provide a normative scheme, or provide guidelines for “how we should treat people (including ourselves)—when and to what extent we should hold people responsible, punish, praise or blame them, feel guilt, regret or pride, and so on” (Thomasson, 2017). Free will disputants generally share the view that if we have free will, then we are morally responsible agents, which I won’t dispute. Thus, incompatibilist/regulative control theories ask what the modal facts are and then attempt to provide a normative scheme based on these facts - if we have the ability to act otherwise, then we have free will and are morally responsible agents; without regulative control, we do not have free will and therefore cannot be morally responsible agents.

However, as Thomasson (2016) demonstrates with her critique of heavyweight metaphysics, we have good reasons to be skeptical that we can ever really know the relevant modal facts. It seems unfruitful, then, for free will disputants to engage in a dispute involving these modal facts if we want a free will theory to provide us with a normative theory. This is where the compatibilist debate as a metalinguistic negotiation may avoid going in the same epistemically worrisome direction as incompatibilists: instead of asking a free will theory to reflect the modal facts and provide a yes/no answer on questions of free will and moral responsibility, compatibilism may position us to formulate a free will theory which appropriately reflects our intuitions about human agency while preserving causal determinism.

73

Reconceptualizing the Free Will Debate as a Metalinguistic Negotiation

Recall that participants in a metalinguistic negotiation are said to advocate for a particular “deployment of linguistic representations,” meaning roughly that, rather than advancing competing logically incompatible propositions or disputing via the literally expressed content of their words (which Plunkett & Sundell call a canonical dispute [3]), disputants in a metalinguistic negotiation tacitly advocate for competing normative conceptual theories, or how and when we should use our words and concepts (3). If I understand them correctly, Plunkett & Sundell would describe the two major incompatibilist camps of the free will debate, libertarians and hard determinists, as engaged in a canonical dispute. After all, if incompatibilists agree that free will is a direct function of whether we have regulative control, the disagreement among them is on the literal “facts of the matter,” i.e, whether we in fact have regulative control.

Plunkett & Sundell offer the following case as an example of metalinguistic negotiation in their paper, “Disagreement and the Semantics of Normative and Evaluative Terms.” Imagine two people disputing whether the racehorse Secretariat is an athlete. Speaker A might say to speaker B, “Secretariat is an athlete,” eliciting the response from speaker B, “No, Secretariat is not an athlete.” Should this interaction be described as a “genuine disagreement”? Plunkett & Sundell point out that there is “little reason to think that [this dispute] concerns straightforward factual matters about the topic at hand,” since the actual features of Secretariat (physical qualities, accomplishments, etc.) are mutually understood (16). So, rather than describing the conversation between speakers A and B as expressing a genuine disagreement, we might be tempted to describe their dispute as “merely verbal,” “in the sense that each disputant can be charitably interpreted as speaking a language in which what she says is true” (Thomasson, 2017). Since the literal semantic content that speaker A expresses when uttering the term ‘athlete’ differs from the literal semantic content uttered by speaker B, we might conceive of them not really disagreeing over anything substantive, and instead uttering trivial truths based on their conception of the term “athlete.” Since speakers A and B appear to “mean different things” by their words, there may be a question of whether they can even express a genuine disagreement. This formulation that disagreement requires speakers to “mean the same things” by their words may seem initially compelling - after all, we might want to describe genuine disagreements in terms of “incompatible contents,” or literal expressions which directly, logically contradict one another (Plunkett & Sundell call disputes of this kind canonical), and it is hard to see how “merely verbal” disputes (which Plunkett & Sundell call non-canonical)

74

Reconceptualizing the Free Will Debate as a Metalinguistic Negotiation

involve this kind of contradiction (12).

But something about this picture seems wrong: if speakers engaged in noncanonical disputes perceived themselves as doing nothing more than uttering trivial truths in their own language, it is hard to see what would motivate such a dispute in the first place. Yet this type of dispute is incredibly common (4), and whether we believe that one speaker employs a “better concept” than the other (whatever this may mean) seems to have real consequences. For instance, if two people dispute what really counts as “murder” as part of a broader moral discussion, the conception of “murder” that prevails may inform the basis of legal proceedings, legislation, human rights, etc. Plunkett & Sundell account for the ubiquity and felt-depth of some non-canonical disputes by reconceptualizing certain non-canonical disputes as metalinguistic negotiations (3). In a metalinguistic negotiation, speakers engaged in a non-canonical dispute genuinely disagree about the proper metalinguistic usage of an expression (what should or should not count as an “athlete,” or “murder,” etc.). Thus, contrary to the view that non-canonical disputes are merely verbal and thus do not express genuine disagreement, participants in a metalinguistic negotiation appear to evince genuine disagreement via their incompatible beliefs regarding how and when certain concepts ought to be deployed.

I will now discuss how metalinguistic negotiation might provide guidance for evaluating competing positions within the free will debate. As previously noted, there are potential conceptual advantages and disadvantages for each free will theory discussed in this paper. Hard determinism, while preserving the causal determinism thesis, seemingly precludes our ability to make cogent ascriptions of moral responsibility; libertarianism, while preserving our ability to make cogent ascriptions of moral responsibility, seemingly excludes the logical possibilities of “complete” psychological and physical theories, and thus potentially rivalizes itself with science. Furthermore, as incompatibilist theories, both hard determinism and libertarianism fail to escape the epistemic worries of heavyweight metaphysics. By grounding the truth of free will in the seemingly inaccessible modal facts of action, we are left skeptical that incompatibilist theories are capable of taking the next step to provide a normative scheme. Compatibilist theories, conversely, ground the truth of free will in terms of whether or not a person’s actions result from the reasons that caused them in the right way. This shift away from the modality of action potentially allows us to avoid the epistemic problems of incompatibilism, but it also appears to invite a different sort of conceptual challenge - that of pervasive vagueness.

While outlining compatibilism in the first section of this paper, I suggested that

75

Reconceptualizing the Free Will Debate as a Metalinguistic Negotiation

the vagueness of guidance control was a potential downside of compatibilism. I repeat this suggestion here: a notion of being caused in the right way seems to rely on what does and does not constitute compulsion, but the distinction between which actions are compelled (not free) and which actions are merely necessitated (free) may be hard to locate. If compatibilists accept the causal determinism thesis, they accept that a person’s actions are necessitated by whatever prior events caused them. But, if uncompelled actions are free on the compatibilist view and compelled ones are not, the distinction between which actions are necessitated without being compelled so as to preclude free will may seem arbitrary.

Lacking a clear method as to how we ought to differentiate between free and compelled actions given that all actions are necessitated, I stated that compatibilism asks us to draw lines in the sand. However, I now admit that I am not convinced this vagueness is a conceptual weakness, and, considering Plunkett & Sundell’s conception of metalinguistic negotiation, it may even be a conceptual strength, especially if we want a free will theory to provide a normative scheme.

Recall the epistemic problems of heavyweight metaphysics: since certain metaphysical questions (especially within the realms of ontology and modality) rely on seemingly inaccessible facts, it is unclear if making progress on these questions is possible. The debate between incompatibilist theories, then, in virtue of the “truth of the matter” relying on currently unknown (and possibly unknowable) modal facts of action, seems to have gone wrong at the very start. Can metalinguistic negotiation save the incompatibilist debate? In the following section I attempt to demonstrate that the incompatibilist debate is a canonical dispute while the compatibilist debate, in addition to being non-canonical, is a metalinguistic negotiation. I borrow heavily from David Plunkett’s Which Concepts Should We Use? (2015) to guide this demonstration. Following this, I discuss how the incompatibilist debate gets “stuck” and thus fails to provide a normative theory while the compatibilist debate avoids this obstacle. Plunkett’s aim in this section of What Concepts Should We Use? is to demonstrate that the debate between compatibilists and incompatibilists is a metalinguistic negotiation, which is distinct from my goal to show that the debate among compatibilists is a metalinguistic negotiation. However, I believe that his formulation of the evidence as to whether a dispute is a metalinguistic negotiation is nonetheless applicable. First, let’s look at Plunkett’s four features or pieces of evidence (A-D) for a dispute being a metalinguistic negotiation, edited for concision:

76

Reconceptualizing the Free Will Debate as a Metalinguistic Negotiation

(A) There is good evidence that the linguistic exchange is a dispute. That is: there is good evidence that it is a linguistic exchange that appears to express a disagreement. (B) There is good evidence that the dispute really does express a disagreement. (C) There is good evidence that speakers in the dispute mean different things by (at least) one of the terms in that dispute. There are different things that might provide such evidence for a given term. (D) There is good evidence that the disagreement expressed in the dispute, insofar as there is one, isn’t just about descriptive information about what a word does mean, or how it is used (847).

Next, let’s look at Plunkett’s imagined dispute between an incompatibilist (speaker A) and a compatibilist (speaker B), which I’ll call interaction X:

(A) It is part of the essence of free will that only agents that are capable of fully causing their own actions have free will. This means that their actions cannot be fully determined by events over which they have no control, such as the events of the past. (B) No, that is wrong. The nature of free will is such that agents can have free will even if they are not capable of that kind of self-determination. What matters is that we can hold them morally responsible in the right way, which we can do even if they don’t have that kind of capacity for self-determination that you just described. And that is a good thing, given that we in fact don’t have that kind of self-determination (855).

Plunkett evaluates the preceding interaction step by step, considering features A-D. He says of features A and B that interaction X appears to be a dispute which evinces genuine disagreement, which I’ll take for granted. There seems to be clear evidence that interaction X satisfies feature C, that speakers “mean different things” by their words. If speaker A uses the term “free will” to describe something of which incompatibilism is true while speaker B uses the same term to describe something of which compatibilism is true, then it seems that the two speakers necessarily mean different things by their words. Regarding feature D, it seems that the disagreement in interaction X does not concern merely descriptive information about what a term means or how it is used, since we can imagine the debate between compatibilists and incompatibilists persisting in spite of broad agreement that “free will” refers to the human capacity to perform free/uncompelled action. Thus, in virtue of appearing to satisfy features A-D, there is strong evidence that interaction X is a metalinguistic negotiation.

I will now attempt to formulate two interactions similar to Plunkett’s, the first between two incompatibilist disputants (interaction Y) and the second between two compatibilist disputants (Interaction Z). I will evaluate each of these interactions using Plunkett’s features, A-D, to determine whether they appear to be metalinguistic negotiations.

In interaction Y, libertarian speaker A and hard determinist speaker B dispute the truth of free will and its implications:

77

Reconceptualizing the Free Will Debate as a Metalinguistic Negotiation

(A) In virtue of having regulative control over our actions, we are capable of freely choosing how to act. Thus, we have free will and are morally responsible agents. (B) It is not the case that we have regulative control over our actions. Since our actions are necessitated by facts about the present world and past conditions, we cannot have free will and thus are not morally responsible agents.

This interaction does appear to be a dispute that evinces genuine disagreement, satisfying features A and B. Regarding feature C, it does not seem as though speakers mean different things by their words. When speaker A says we do have free will, it is the precise inverse of speaker B’s assertion that we do not have free will. If “regulative control” is P and “free will” is Q, then speakers A and B agree that P ⊃ Q as well as ∼ P ⊃ ∼ Q, the difference in their assertions being what truth values P and Q actually possess. Since this dispute does not satisfy feature C, it is irrelevant to evaluate feature D. Therefore, since speakers appear to mean the same thing by their words, this dispute appears to be canonical and thus cannot be a metalinguistic negotiation.

In interaction Z, two compatibilists dispute whether the behavior of person S is compelled or free. Assume that speakers A and B have the same level of knowledge regarding person S:

(A) Since person S fell in with the wrong crowd through no fault of their own, their actions are not free. Rather, their actions were compelled through conditioning in a violent environment and thus cannot be judged as evil. (B) No, although it wasn’t the fault of person S that they fell in with the wrong crowd, they identified with the causes of their subsequent actions and understood their effects. They were not compelled by a brain disorder, for example, and so not only are their actions free, they are evil.

In interaction Z, speakers A and B appear to engage in a dispute which evinces genuine disagreement, satisfying features A and B. Regarding feature C, speakers A and B do appear to mean different things by their words. There are three terms that are deployed differently in this interaction: “free,” “compelled,” and “evil.” Since speakers A and B have the same level of knowledge about person S, interaction Z appears to satisfy feature D as well; there is seemingly no reason to believe that speakers A and B would consider their dispute resolved if they came to a descriptive consensus about the meanings of “free,” “compelled,” and “evil.” Thus, since interaction Z satisfies features A-D, there is especially strong evidence that it is a metalinguistic negotiation.

I chose to include the terms “free,” “compelled,” and “evil” in my formulation of interaction Z because I believe these terms help demonstrate how the compatibilist debate avoids the obstacle that incompatibilism faces when attempting to

78

Reconceptualizing the Free Will Debate as a Metalinguistic Negotiation

provide a normative scheme. In interaction Y, it is not clear whether speaker A or speaker B should be believed, as audiences to this dispute do not have access to the truth value of P, i.e whether we have regulative control. Therefore, speaker A and speaker B’s assertions that we are/are not morally responsible agents cannot be evaluated on their conceptual merits. However, looking at interaction Z, there does not appear to be a similar obstacle. While audiences may not know whether to believe speaker A or B based on interaction Z alone, one can imagine two different sets of features of person S, one on which speaker A’s formulation would turn out true and another on which speaker B’s formulation would turn out true. Thus, how we ought to respond to the actions of person S, whether person S is imprisoned, blamed, provided counseling, etc., is based on how the terms “free,” “compelled,” and “evil” appear to relate to person S and their actions. On the incompatibilist scheme, the truth of moral responsibility cannot be evaluated because the truth value of P is inaccessible. On the compatibilist scheme, conversely, whether speaker A or speaker B’s deployment of terminology is more appropriate for informing how we ought to respond to the actions of person S determines the normative course. Furthermore, while the incompatibilist scheme asks us to start with modality and move towards a normative theory, compatibilism reveals that there is another option: we can begin with normative intuitions, develop terms which allow us to appropriately describes these intuitions, and then decide how to use these terms as part of a broad and ever-evolving dialectic. Thus, compatibilism allows us to preserve both causal determinism and our moral intuitions as a metalinguistic negotiation.

Earlier, I identified the problem of heavyweight metaphysics-style debates as requiring that disputes be canonical to express genuine disagreement, I now argue that some metaphysical debates can be “fixed” by rejecting this requirement. The problem of heavyweight metaphysics-style debates, which I take to be the tendency for metaphysicians to engage in canonical debates relying upon potentially inaccessible modal facts, should be understood in terms of how to progress when a dispute seems “merely verbal.” As I have mentioned, it may be one’s intuition to dismiss merely verbal disputes as not expressing a genuine disagreement due to speaker’s utterances being semantically compatible. As Plunkett (2015) mentions, “Philosophers often tend to privilege canonical disputes when thinking about how disagreements are expressed (including, crucially, when thinking about their own disputes” [836]). However, due to the ubiquity, felt-depth, and real consequences borne of some non-canonical disputes, it is more accurate to conceive of some noncanonical disputes as metalinguistic negotiations. This reconception reveals that

79

Reconceptualizing the Free Will Debate as a Metalinguistic Negotiation

non-canonical disputants can express genuine disagreement, not by the literal semantic content of their utterances, but rather in their tacit prescriptions of how a term ought to be used. Since incompatibilist theories assert literal incompatible semantic contents in the form of theories based on seemingly inaccessible facts (i.e, whether we do/do not have regulative control), we can describe the incompatibilist debate as canonical, but philosophically unproductive. However, despite being non-canonical, the compatibilist debate allows the preservation of the causal determinism thesis and the development of normative theories.

Thus, our intuition on which metaphysical disputes are worth having reverses: non-canonical disputes initially appear unproductive as “merely verbal” and therefore trivial, but metalinguistic negotiation reveals that these interactions can be highly fruitful by revealing disputant’s incompatible beliefs regarding how to properly use certain words and concepts. Meanwhile, canonical metaphysical disputes initially appear productive on the heavyweight metaphysics conception, but Thomasson’s critique of heavyweight metaphysics reveals the difficulty of evaluating epistemically metaphysical theories. It is revealed, then, that the apparent vagueness of compatibilist theories may be viewed as a conceptual strength: compatibilist theories, in virtue of grounding the “truth” of free will in concepts independent from modality, are theoretically flexible. I have argued that compatibilism demonstrates the following: free will theories need not preclude our metaphysical commitments nor our moral intuitions, and not only can the free will debate be non-canonical, it may be advantageous to pursue free will issues explicitly through competing normative conceptual schemes.

Although some metaphysicians conceive of their proper roles as discoverers and disputants of the literal, fundamental facts of the world, there are numerous reasons to be skeptical of this conception. The apparent inability of metaphysicians to make progress on resolving their debates is one such reason. In the free will debate, this inability to resolve is the result of incompatibilists proposing semantically incompatible theories which rely on potentially inaccessible modal facts. Thus, we may be skeptical that incompatibilists are engaged in a productive dispute. Compatibilism seemingly provides a way around this obstacle by grounding the truth of free will in concepts such as “free” and “compelled” rather than potentially inaccessible modal facts, allowing compatibilist theories to be evaluated in terms of broader normative

80

Reconceptualizing the Free Will Debate as a Metalinguistic Negotiation

conceptual schemes. Metalinguistic negotiation provides guidance for this move away from epistemically metaphysical questions by distinguishing between canonical and non-canonical disputes, revealing that not only can non-canonical disputes provide guidance for normative schemes generally, but they may be more suited to this task than their canonical counterparts. When the content of disputants’ claims hinges on potentially inaccessible or irrelevant modal facts, it may be advantageous to reframe these disputes as metalinguistic negotiations and thus evaluate competing metaphysical theories as part of a broader normative conceptual scheme.

81