This post is a reflection upon "Just Deserts" by Daniel Dennett and Greg Caruso. One can read the original exchange at Aeon by clicking the following link: https://aeon.co/essays/on-free-will-daniel-dennett-and-gregg-caruso-go-head-to-head
In their recent exchange published by Aeon, aptly titled "Just Deserts", the philosophers Greg Caruso and Daniel Dennett go head to head on the topic of free-will and its relationship to retributive justice. For those who have followed Dennett's storied history with the subject, from Elbow Room (his early defense of compatibilism) to his ongoing(?) debate with Galen Strawson, the prospect of yet another round might appear unwarranted or, even worse, tiresome. Thankfully, Caruso proves to be an able sparring partner, eagerly honing in on the implications of Dennett's position for social policy in a clear and precise manner.
In their recent exchange published by Aeon, aptly titled "Just Deserts", the philosophers Greg Caruso and Daniel Dennett go head to head on the topic of free-will and its relationship to retributive justice. For those who have followed Dennett's storied history with the subject, from Elbow Room (his early defense of compatibilism) to his ongoing(?) debate with Galen Strawson, the prospect of yet another round might appear unwarranted or, even worse, tiresome. Thankfully, Caruso proves to be an able sparring partner, eagerly honing in on the implications of Dennett's position for social policy in a clear and precise manner.
For the sake of full disclosure, I should state at the outset that I have always found myself largely convinced by the compatibilist camp on these issues, though I have frequently been disappointed with Dennett's particular style of characterizing it. His go-to examples (repeated in this conversation) such as "the Moral Agents Club" and the analogy to sports seem, by my lights, to place free-will on the foundations of social construction, rather than admit its existence as a substantial, metaphysical phenomenon. To his credit, Caruso provides an effective deconstruction of Dennett's metaphors on this score, repeatedly forcing Dennett to clarify the precise nature of his philosophical commitments.
Nevertheless, I could not help coming away with several questions concerning the choice of topics discussed. According to Caruso, "the notion of basic desert", serves as his primary target over the course of the discussion. He takes this notion to serve as the core criterion for determining whether or not an agent proves "truly deserving of blame and praise, punishment and reward" in the retrospective sense that ignores any utilitarian benefit that may be provided in the long run. With this emphasis in mind, Caruso poses the following question to Dennett: "My question, then, is whether the kind of desert you have in mind is enough to justify retributive punishment?"
I find it curious that Caruso focuses on this point in order to challenge compatibilism, primarily because I can't see how the success of Dennett's position turns on it, regardless of how he replies. For one, it seems as though a robust conception of free-will can remain intact even if it were merely necessary to justify retributivism without being sufficient by itself. One might hold that, even if individuals are truly deserving in virtue of making free choices, we should nevertheless refrain from punishment on consequentialist grounds. Indeed, such a view was explicitly propounded by the Athenian, Diodotus, regarding the question of capital punishment in his speech before the Athenian Assembly:
Though I prove them ever so guilty, I shall not, therefore advise their death, unless it be expedient... I consider that we are deliberating for the future more than for the present... All, states and individuals, are alike prone to err, and there is no law that will prevent them; or why should men have exhausted the list of punishments in search of enactments to protect them from evildoers? It is probable that in early times the penalties for the greatest offences were less severe, and that, as these were disregarded, the penalty of death has been by degrees in most cases arrived at, which is itself disregarded in like manner. Either then some means of terror more terrible than this must be discovered, or it must be owned that this restraint is useless... (Thucydides 1968:152)
Caruso is quick to emphasize that his position does not entail being a skeptic about punishment as an institution. As he notes, "free-will skeptics typically point out that the impositions of sanctions serve purposes other than punishment of the guilty: it can also be justified by its role in incapacitating, rehabilitating and deterring offenders." What he fails to properly address, however, is the fact that a parallel argument can be run by free-will realists against punishment by recourse to similar means. My point is not that such arguments succeed or that retributivism is necessarily misguided. I only wish to point out one can consistently reject retributivism without also rejecting a robust conception of free-will and moral desert. Given that this is the case, I remain puzzled as to why Caruso finds it such a pressing issue in the exchange.
Nevertheless, one might argue, even if punishment can be rejected by the free-will realist, it still remains the case that whether or not a person is in fact worthy of praise or blame in any "backward-looking" sense is a question which Caruso can still reject. This gets to the heart of the difference between Dennett's view and his own. For Caruso, the fact that "the particular reasons that move us, along with the psychological predispositions, likes and dislikes, and other constitutive factors that make us who we are, themselves are ultimately the result of factors beyond our control" is one that undermines any legitimate appeal to the kind of responsibility he takes to be required for a person to be truly praiseworthy or not.
Before proceeding, it is worth asking what would make praise and blame appropriate for a free-will realist in the first place. On the common-sense view of things, what we are doing when we praise someone, for instance, is expressing our approval of their having done the right thing. Crucially, we do not praise someone simply because they did something that just happened to be right. Rather, someone is worthy of praise when they do the right thing because it is the right thing to do. In other words, we praise them precisely because they are adequately responsive to the right kinds of reasons when acting. Of course, again, it must be asked whether or not being responsive to the right reasons is sufficient for being praiseworthy. Must it not also be the case that the reasons in question are responded to freely?
On this front, I suspect that Dennett would agree with me that the question does not make sense. To say that people are free requires nothing beyond the claim that "in general people are reasonable, are moved by reasons, can adjust their behaviour and goals in the light of reasons presented to them." When individuals adjust their behavior in light of the right reasons, they are praiseworthy. When they adjust in light of the wrong reasons, they are blameworthy.
Caruso is wrong to claim that the particular reasons that move us are beyond our control. If this were the case, we would not be moved by them at all. He is right, however, to say that "constitutive factors that make us who we are, themselves are ultimately the result of factors beyond our control." The difficulty Caruso faces is simply that this fact provides no obvious reason to reject the reality of free-will. The fact that I am embodied is also a constitutive fact that makes me who I am over which I had no control, yet this fact poses no obstacle for free-will realism. If Caruso wishes to pose such a difficulty, he will have to substantiate the further claim that the reason I chose to act in the way that I did was beyond my control. There is no compelling reason to believe that this is the case.
In his essay "Wide Causation" (1997), Stephen Yablo explores whether or not we can ever truly say that physical events are subject to mental influence. The difficulty for the view that they are comes in the form of what Yablo refers to as "the argument from below". According to this argument, given that a story can be told regarding the causal underpinnings of our behaviour in wholly physical terms, it would seem as though there is simply no room for our mental life to play any causal role. Unfortunately for the free-will skeptic, whether or not this principle is appropriate in some circumstances, Yablo argues, it is not applicable if the relevant higher-order cause and its physical underpinnings stand in a determinable/determinate relation, as mental causes plausibly do with respect to brain states.
It is here that Yablo finds a powerful wedge for the believer in mental causation by way of the "proportionality principle" According to this principle, it is not enough that something be causally sufficient for its effect in order to hold pride of place as the "true cause". If this were the case, there would be no grounds for preferring the physical brain state to the specific mental state it realizes since both are causally sufficient for their effects. Instead, it is also required that the determinate or determinable in question be proportional to its resulting effects in a way that the state it realizes or is realized by is not.
To take a simple example, Yablo considers the case of pain. In a specific case in which a specific brain state realizes a specific pain, it is obvious that the brain state is causally sufficient for any of the effects that follow. However, in light of the proportionality principle, it is not at all clear that the brain state should be considered as the cause of its bearer's behaviour instead of the pain. Imagine that one of the subsequent effects is a grimace by the subject. In this case, Yablo points out that "I would still have grimaced even if my pain had occurred in a different microphysical way. Whereas the issue of how I would have behaved had the brain state occurred in the pain's absence cannot even be raised, because the brain state includes the pain." In this scenario, although the determinate brain state is causally sufficient for all resulting effects, it is not clear that it is appropriately proportional to them. One of the effects (the grimace) is more appropriately explained by the presence of pain than the presence of its realizing brain state. Consequently, it is not at all clear that the brain state is the true cause of that subject's behavior, rather than the pain.
It is here that Dennett can (and perhaps does, though not in the same words) help himself to an account of behavioural causation that fully allows for human action to be caused on the basis of reasons. When he claims that "autonomy is something one grows into, and this is indeed a process that is initially entirely beyond one’s control", one could plausibly interpret Dennett as claiming that it is only once the realized mental states which are sufficient for our behavior are the proportional cause of such actions that we can truly be said to have free-will. I suspect that Yablo and Dennett would find each other to be powerful allies in developing a model of precisely this kind.
The purpose of this excursion was not to argue that Yablo is completely right, or that nothing more is required for a proper explanation of our ability to act on the basis of reasons. I only highlight Yablo's essay as a shining example of the vast array of options which the free-will realist has at their disposal. When I act on the basis of a reason, Caruso might say that this action is itself the result of a vast array of causes which I have no knowledge of or control over. The free-will realist can grant all of these claims without conceding to Caruso's further claim that my actions cannot truly be said to be the result of my own free-will. In short, even if Caruso is ultimately correct, he still has all his work ahead of him if he wishes to show the reality of free-will to be a sham.
Before concluding, I would like to address one issue raised by Caruso that I find extremely insightful as a reflection regarding the extent to which praise and blame are appropriate even if one recognizes the reality of free-will. Addressing Dennett's claim that the differences between individuals and their choices average out to such an extent that we can reasonably hold each of them responsible for their actions as independent agents, Caruso retorts:
Luck does not average out in the long run. Those who start from a disadvantaged position of genetic abilities or early environment do not always have offsetting luck later in life. The data clearly shows that early inequalities in life often compound over time rather than average out, affecting everything from differences in health and incarceration rates to success in school and all other aspects of life... as Levy puts it: ‘We cannot undo the effects of luck with more luck’. Hence the very actions to which compatibilists point, the actions whereby agents take responsibility for their endowments, either express that endowment (when they are explained by constitutive luck) or reflect the agent’s present luck, or both. Either way, responsibility is undermined.
Here Caruso points to a reality that, to my mind, Dennett has completely failed to adequately address. Even if our free decisions are properly explained by our free-will (rather than the chemical underpinnings of our brain, for instance), it still might seem that we are left with a free-will that is wholly irrelevant in a world of unwanted emotions and systematic inequality. Rather than take these as a reason to deny human freedom, I view these undeniable facts as examples of phenomena that can only be properly addressed once we recognize the reality of free decisions.
If we take free-will as our starting point, there is nothing stopping us from admitting that there are powerful forces at play which influence our choices in sub-optimal conditions. What does it mean to be in "sub-optimal conditions"? It means that one is in a situation where the human ability to rationally recognize and act upon reasons is interfered with by factors external to this rational process. When these factors are systemic, such as when public institutions are systemically biased against a specific race, this fact must be reflected in the ways we praise and blame. Similarly, if an individual is severely depressed, there is an interfering force which affects their ability to properly recognize and act upon reasons. We must recognize the relevance of these factors to judgments of desert without making the unwarranted leap to the conclusion that these factors undermine the very notion of desert itself.
What is the free-will skeptic to say in such cases? Unlike the free-will realist, they cannot say that the reason these factors matter is because they effect our ability to act freely. Rather, they are forced to view these as examples of the unfortunate truth "that the lottery of life is not always fair, that luck does not average out in the long run..." But this is to miss the whole point of what makes such inequalities reprehensible in the first place. Consider the free-market. The ideal of the free-market was never to create an egalitarian society. Rather, it was to create a society of just inequality in which any differences of outcome were the result of free decisions made on the part of consenting individuals.
The systemic inequalities of racism and ableism are not just wrong because they are inequalities (although inequality may be intrinsically wrong on its own). They are wrong because they are unjust. Yet, according to the free-will skeptic, what means is there for drawing out the deep difference between these two forms of inequality? If there were no free-will, the unequal results of a utopian free-market and the society we actually live in are just two different rolls from the same pair of dice. It is only once the profound distinction between free action and choices beyond our control is clarified that a meaningful grasp of these injustices can be established. Perhaps Caruso will reply that we can draw the difference in the forward-looking terms of what consequences these two visions of reality would lead to in the future, but that would be to overlook a moral asymmetry which I believe is undeniable and cannot be grounded upon the monolithic reasoning of consequentialism.
"Just Deserts" is an entertaining an enlightening debate between two learned scholars on the nature of moral responsibility. Daniel Dennett remains at his strongest when highlighting the reasons why his conception of free-will remains the only one worth wanting. Nevertheless, I believe he continues to falter when he appeals to "the ‘forward-looking benefits’ of the whole system of desert" in order to justify a phenomenon of the present. Likewise, Caruso proves to be a welcome arrival to a long and storied debate when he presses Dennett to recognize the underlying commonality between their means of justifying the retributive policies of a community. Although I have raised many points of disagreement between myself and these two participants, I wholeheartedly recommend their discussion to anyone interested in free-will and the proper role of desert.
-Shane Wagoner
I find it curious that Caruso focuses on this point in order to challenge compatibilism, primarily because I can't see how the success of Dennett's position turns on it, regardless of how he replies. For one, it seems as though a robust conception of free-will can remain intact even if it were merely necessary to justify retributivism without being sufficient by itself. One might hold that, even if individuals are truly deserving in virtue of making free choices, we should nevertheless refrain from punishment on consequentialist grounds. Indeed, such a view was explicitly propounded by the Athenian, Diodotus, regarding the question of capital punishment in his speech before the Athenian Assembly:
Though I prove them ever so guilty, I shall not, therefore advise their death, unless it be expedient... I consider that we are deliberating for the future more than for the present... All, states and individuals, are alike prone to err, and there is no law that will prevent them; or why should men have exhausted the list of punishments in search of enactments to protect them from evildoers? It is probable that in early times the penalties for the greatest offences were less severe, and that, as these were disregarded, the penalty of death has been by degrees in most cases arrived at, which is itself disregarded in like manner. Either then some means of terror more terrible than this must be discovered, or it must be owned that this restraint is useless... (Thucydides 1968:152)
Caruso is quick to emphasize that his position does not entail being a skeptic about punishment as an institution. As he notes, "free-will skeptics typically point out that the impositions of sanctions serve purposes other than punishment of the guilty: it can also be justified by its role in incapacitating, rehabilitating and deterring offenders." What he fails to properly address, however, is the fact that a parallel argument can be run by free-will realists against punishment by recourse to similar means. My point is not that such arguments succeed or that retributivism is necessarily misguided. I only wish to point out one can consistently reject retributivism without also rejecting a robust conception of free-will and moral desert. Given that this is the case, I remain puzzled as to why Caruso finds it such a pressing issue in the exchange.
Nevertheless, one might argue, even if punishment can be rejected by the free-will realist, it still remains the case that whether or not a person is in fact worthy of praise or blame in any "backward-looking" sense is a question which Caruso can still reject. This gets to the heart of the difference between Dennett's view and his own. For Caruso, the fact that "the particular reasons that move us, along with the psychological predispositions, likes and dislikes, and other constitutive factors that make us who we are, themselves are ultimately the result of factors beyond our control" is one that undermines any legitimate appeal to the kind of responsibility he takes to be required for a person to be truly praiseworthy or not.
Before proceeding, it is worth asking what would make praise and blame appropriate for a free-will realist in the first place. On the common-sense view of things, what we are doing when we praise someone, for instance, is expressing our approval of their having done the right thing. Crucially, we do not praise someone simply because they did something that just happened to be right. Rather, someone is worthy of praise when they do the right thing because it is the right thing to do. In other words, we praise them precisely because they are adequately responsive to the right kinds of reasons when acting. Of course, again, it must be asked whether or not being responsive to the right reasons is sufficient for being praiseworthy. Must it not also be the case that the reasons in question are responded to freely?
On this front, I suspect that Dennett would agree with me that the question does not make sense. To say that people are free requires nothing beyond the claim that "in general people are reasonable, are moved by reasons, can adjust their behaviour and goals in the light of reasons presented to them." When individuals adjust their behavior in light of the right reasons, they are praiseworthy. When they adjust in light of the wrong reasons, they are blameworthy.
Caruso is wrong to claim that the particular reasons that move us are beyond our control. If this were the case, we would not be moved by them at all. He is right, however, to say that "constitutive factors that make us who we are, themselves are ultimately the result of factors beyond our control." The difficulty Caruso faces is simply that this fact provides no obvious reason to reject the reality of free-will. The fact that I am embodied is also a constitutive fact that makes me who I am over which I had no control, yet this fact poses no obstacle for free-will realism. If Caruso wishes to pose such a difficulty, he will have to substantiate the further claim that the reason I chose to act in the way that I did was beyond my control. There is no compelling reason to believe that this is the case.
In his essay "Wide Causation" (1997), Stephen Yablo explores whether or not we can ever truly say that physical events are subject to mental influence. The difficulty for the view that they are comes in the form of what Yablo refers to as "the argument from below". According to this argument, given that a story can be told regarding the causal underpinnings of our behaviour in wholly physical terms, it would seem as though there is simply no room for our mental life to play any causal role. Unfortunately for the free-will skeptic, whether or not this principle is appropriate in some circumstances, Yablo argues, it is not applicable if the relevant higher-order cause and its physical underpinnings stand in a determinable/determinate relation, as mental causes plausibly do with respect to brain states.
It is here that Yablo finds a powerful wedge for the believer in mental causation by way of the "proportionality principle" According to this principle, it is not enough that something be causally sufficient for its effect in order to hold pride of place as the "true cause". If this were the case, there would be no grounds for preferring the physical brain state to the specific mental state it realizes since both are causally sufficient for their effects. Instead, it is also required that the determinate or determinable in question be proportional to its resulting effects in a way that the state it realizes or is realized by is not.
To take a simple example, Yablo considers the case of pain. In a specific case in which a specific brain state realizes a specific pain, it is obvious that the brain state is causally sufficient for any of the effects that follow. However, in light of the proportionality principle, it is not at all clear that the brain state should be considered as the cause of its bearer's behaviour instead of the pain. Imagine that one of the subsequent effects is a grimace by the subject. In this case, Yablo points out that "I would still have grimaced even if my pain had occurred in a different microphysical way. Whereas the issue of how I would have behaved had the brain state occurred in the pain's absence cannot even be raised, because the brain state includes the pain." In this scenario, although the determinate brain state is causally sufficient for all resulting effects, it is not clear that it is appropriately proportional to them. One of the effects (the grimace) is more appropriately explained by the presence of pain than the presence of its realizing brain state. Consequently, it is not at all clear that the brain state is the true cause of that subject's behavior, rather than the pain.
It is here that Dennett can (and perhaps does, though not in the same words) help himself to an account of behavioural causation that fully allows for human action to be caused on the basis of reasons. When he claims that "autonomy is something one grows into, and this is indeed a process that is initially entirely beyond one’s control", one could plausibly interpret Dennett as claiming that it is only once the realized mental states which are sufficient for our behavior are the proportional cause of such actions that we can truly be said to have free-will. I suspect that Yablo and Dennett would find each other to be powerful allies in developing a model of precisely this kind.
The purpose of this excursion was not to argue that Yablo is completely right, or that nothing more is required for a proper explanation of our ability to act on the basis of reasons. I only highlight Yablo's essay as a shining example of the vast array of options which the free-will realist has at their disposal. When I act on the basis of a reason, Caruso might say that this action is itself the result of a vast array of causes which I have no knowledge of or control over. The free-will realist can grant all of these claims without conceding to Caruso's further claim that my actions cannot truly be said to be the result of my own free-will. In short, even if Caruso is ultimately correct, he still has all his work ahead of him if he wishes to show the reality of free-will to be a sham.
Before concluding, I would like to address one issue raised by Caruso that I find extremely insightful as a reflection regarding the extent to which praise and blame are appropriate even if one recognizes the reality of free-will. Addressing Dennett's claim that the differences between individuals and their choices average out to such an extent that we can reasonably hold each of them responsible for their actions as independent agents, Caruso retorts:
Luck does not average out in the long run. Those who start from a disadvantaged position of genetic abilities or early environment do not always have offsetting luck later in life. The data clearly shows that early inequalities in life often compound over time rather than average out, affecting everything from differences in health and incarceration rates to success in school and all other aspects of life... as Levy puts it: ‘We cannot undo the effects of luck with more luck’. Hence the very actions to which compatibilists point, the actions whereby agents take responsibility for their endowments, either express that endowment (when they are explained by constitutive luck) or reflect the agent’s present luck, or both. Either way, responsibility is undermined.
Here Caruso points to a reality that, to my mind, Dennett has completely failed to adequately address. Even if our free decisions are properly explained by our free-will (rather than the chemical underpinnings of our brain, for instance), it still might seem that we are left with a free-will that is wholly irrelevant in a world of unwanted emotions and systematic inequality. Rather than take these as a reason to deny human freedom, I view these undeniable facts as examples of phenomena that can only be properly addressed once we recognize the reality of free decisions.
If we take free-will as our starting point, there is nothing stopping us from admitting that there are powerful forces at play which influence our choices in sub-optimal conditions. What does it mean to be in "sub-optimal conditions"? It means that one is in a situation where the human ability to rationally recognize and act upon reasons is interfered with by factors external to this rational process. When these factors are systemic, such as when public institutions are systemically biased against a specific race, this fact must be reflected in the ways we praise and blame. Similarly, if an individual is severely depressed, there is an interfering force which affects their ability to properly recognize and act upon reasons. We must recognize the relevance of these factors to judgments of desert without making the unwarranted leap to the conclusion that these factors undermine the very notion of desert itself.
What is the free-will skeptic to say in such cases? Unlike the free-will realist, they cannot say that the reason these factors matter is because they effect our ability to act freely. Rather, they are forced to view these as examples of the unfortunate truth "that the lottery of life is not always fair, that luck does not average out in the long run..." But this is to miss the whole point of what makes such inequalities reprehensible in the first place. Consider the free-market. The ideal of the free-market was never to create an egalitarian society. Rather, it was to create a society of just inequality in which any differences of outcome were the result of free decisions made on the part of consenting individuals.
The systemic inequalities of racism and ableism are not just wrong because they are inequalities (although inequality may be intrinsically wrong on its own). They are wrong because they are unjust. Yet, according to the free-will skeptic, what means is there for drawing out the deep difference between these two forms of inequality? If there were no free-will, the unequal results of a utopian free-market and the society we actually live in are just two different rolls from the same pair of dice. It is only once the profound distinction between free action and choices beyond our control is clarified that a meaningful grasp of these injustices can be established. Perhaps Caruso will reply that we can draw the difference in the forward-looking terms of what consequences these two visions of reality would lead to in the future, but that would be to overlook a moral asymmetry which I believe is undeniable and cannot be grounded upon the monolithic reasoning of consequentialism.
"Just Deserts" is an entertaining an enlightening debate between two learned scholars on the nature of moral responsibility. Daniel Dennett remains at his strongest when highlighting the reasons why his conception of free-will remains the only one worth wanting. Nevertheless, I believe he continues to falter when he appeals to "the ‘forward-looking benefits’ of the whole system of desert" in order to justify a phenomenon of the present. Likewise, Caruso proves to be a welcome arrival to a long and storied debate when he presses Dennett to recognize the underlying commonality between their means of justifying the retributive policies of a community. Although I have raised many points of disagreement between myself and these two participants, I wholeheartedly recommend their discussion to anyone interested in free-will and the proper role of desert.
-Shane Wagoner
No comments:
Post a Comment