Freedom (of the Will) Resolved
When confusion about freedom of the will reaches middlebrow prose it is time to execute an intervention, or harm will disseminate from academic marshland (where it is scarcely to generate any significant aftermath) to the minute portion of the general population that can still read and write—and might feel ennobled, rather than confused, by perusing a chatty account of otherwise forbidding philosophical matters. This is why, though having devoted a whole book to sorting out the topic,1placeholder I have now decided that a more concise treatment is in order—not to mention one that does not cost the monstrous amounts demanded by university presses.
The occasion for my outburst is a review published in The New Yorker2placeholder of the long-winded essay Determined: A Science of Life Without Free Will,3placeholder by the neurobiologist Robert M. Sapolsky. The author of the review is Nikhil Krishnan, an Oxford Ph. D. who is currently teaching at the University of Cambridge. And the essay’s main points are usefully summarized in his text. First, says Krishnan paraphrasing Sapolsky, “our best contemporary science has established the truth of determinism,” by which is meant causal determinism: the thesis that “everything that happens is the inevitable consequence of the laws of nature and what the universe was like once upon a time.” Second, with specific reference to any human (presumed) action, the truth of causal determinism entails that “[w]e’re bound to do what we in fact do”: (in Sapolsky’s own words now) “we have no free will at all.” But then, third (and, again, Sapolsky speaking), no one “has earned or is entitled to being treated better or worse than anyone else.” From which, fourth and last, Sapolsky concludes that we should adopt attitudes “less punitive and more forgiving.”
True to his Oxford education, Krishnan makes ample use in his article of one of Oxford’s philosophical glories, P. F. Strawson, and in particular (without mentioning it—after all, this is middlebrow) of his paper “Freedom and Resentment.”4placeholder There Strawson argues that we can adopt two different standpoints to the behavior of others, or even our own: the participant and the objective. We adopt the former when we see others, or ourselves, as appropriate objects of such reactive attitudes as sympathy or resentment; we adopt (or, perhaps more truly to his aims, shift into) the latter when we judge them partly or wholly incapacitated—as children, mentally deranged, or subject to constraints—to act in ways deserving of those attitudes. Both standpoints are legitimate, and neither of them is induced in us by the belief in a universal thesis like causal determinism: they are constitutive (more so the first one, it would seem) of what it is to be a human, a social animal, living in society. Entirely forfeiting the first standpoint in favor of the second one, which is essentially what Sapolsky recommends we do, would make the outcome unintelligible as a system of human relationships. Of all this Krishnan offers a faithful précis—perhaps belatedly celebrating the sixtieth anniversary of the original publication of Strawson’s piece (its fiftieth was deferentially hallowed with an Oxford UP tome)—and closes with an edifying appeal: “We needn’t follow the skeptics to the conclusion that the best morality would be no morality at all to recognize that our current morality remains a work in progress.”
The place to start my attack is to point out that the problem of freedom (of the will, but not only of the will) is not a moral one, or one that has anything to do with the structure of human society or sentiments (a word that Strawson regrets has, or had, fallen out of favor). It is a metaphysical problem: it has to do with the conceptual framework mobilized in thinking about the world, not with the world itself. If the conceptual framework that underlies contemporary science is committed to causal determinism, a scientist will be necessarily led by it to investigating the causes of anything that happens, including anything anyone does, and will not stop at the remark that this is what that person (unimpeded, unconstrained, not deranged, psychologically mature) wanted to do. She will ask why that is what he wanted; and, one question after another, will soon find herself roaming a spatiotemporal terrain in which he is not even present—the terrain haunted by his ancestors, or Adam and Eve, or the Big Bang. At that point, the poor “willing agent” will have been declassed to an obtuse little wheel in a gigantic mechanism. To talk about social practices and attitudes, within this framework, is just fine, except that it has no explanatory purchase. It would be like adopting a “participant” standpoint toward the main character in Frank Capras’s Mr. Smith Goes to Washington, and sympathizing with his naïve struggle against a corrupt political environment, without realizing, or appreciating, that all his words and moves were written by someone (the prolific screenwriter Sidney Buchman). Similarly, when I sympathize with my neighbor, or resent him, it is useful to realize, and appreciate, the fact that his behavior, and my sentiments, were “written” by ancient history, with no contribution on our part—and that that’s where the problem lies. Invoking Strawson, Krishnan says that Sapolsky’s approach does not amount to solving a problem but to changing the subject; it would seem, however, that it is people like him and Strawson who are forgetting what the subject is, perhaps because they never got it in the first place.
A little historical perspective might help. Strawson wrote at a time when “metaphysics” was a bad word, and the meager tools at his disposal to counter the neopositivistic onslaught mounted against it were those offered by British ordinary-language philosophy.5placeholder Better gifted than his peers, and endowed with the brilliant philosophical imagination he features in the second chapter of Individuals6placeholder, easily the best metaphysics book written in English in the twentieth century (though under the protective epithet of “descriptive” metaphysics), he was also part of a brief, intense, ultimately disappointing Kant revival. But, in his seminal The Bounds of Sense7placeholder, he drew the line of Kant acceptance at transcendental idealism: What is good about the first Critique is the Strawson-labeled “principle of significance, […] one with which empiricist philosophers have no difficulty in sympathizing,” and on whose basis “there can be no legitimate, or even meaningful, employment of ideas or concepts which does not relate them to empirical or experiential conditions of their application” (16). What is bad is the “disastrous model” (21) proffered by transcendental idealism, whose “doctrines […] are undoubtedly the chief obstacles to a sympathetic understanding of the Critique” (22). So we are treated to a softer, gentler, more acceptable Kant, on whom Strawson has performed what John Findlay approvingly called a “Transcendental Excision”8placeholder (capitalized, no less), to see “the patient” rebound “from the operating table in an access of renewed vitality”9placeholder for having been liberated of all that “myth-obsessed nonsense.”10placeholder Which is unfortunate, because if one lets transcendental idealism go, and the Copernican revolution with it, all one is left with, of a “Kantian” approach, is vain talk of standpoints and attitudes, which has dominated analytic versions of compatibilism (the attempt at making determinism and freedom compatible) and does not let us see past the social manners chronicled by Strawson and Krishnan—as well as, with much more verve, depth, and sensitivity, by a host of writers who, we are absolutely clear about it as we read them, make up what their characters do and what emotions they feel, hence leave them no freedom in the matter, however decent or indecent they might (be made to) be, and however much we might sympathize with them or resent them. In philosophy, the crucial antecedent of this deflationary account is not Kant but Aristotle, who liquidates the problem of freedom early in the Nicomachean Ethics by saying that one is free if not forced or acting in ignorance, and then delivers as sophisticated and perceptive a treatment of social psychology (the sort of thing Strawson does) as we are to find anywhere.
Let us put aside all this nonsense—I am not sure whether myth-obsessed or not—and address the issue at the level adequate to it. On one side of Kant’s revolution (the one he rejects, and asks us to overcome) is what he calls transcendental realism: the conceptual framework in which what is basic is the concept of an object, or thing, or substance. Aristotle’s framework, to bring up again that most obvious archetype, where a substance is in the primary sense, and everything else that is (a quality, a quantity, a relation, …) is by being an attribute (a quality, a quantity, a relation, …) of a substance. In this framework, you start thinking of the world as a stock of objects, which, in Bishop Butler’s immortal words, are what they are and not another thing—which includes: they have the qualities, quantities, relations, … they have and no others—and go on to define what else you need (experiences, knowledge, delusions) based on that stock. If you decide that a given object has come to pass, or an event that involves it has occurred, because of the causal impact of some other objects or events, that is the end of that: that is how you decided the world is, and your decision can only be rescinded if it was wrong. You may talk about social acceptability, feelings, and expectations all night long, but that would be a pathetic diversion, and good for you if your adversary is someone like Sapolsky, who should stick to his scientific endeavors and not venture into exhortations to adopt more forgiving stances: within the position he is speaking for, what stance one has is as much determined as anything else is, and will or will not be adopted if it has to—the exhortations themselves will or will not have an effect if they have to—, hence those “philosophers” whose only professional mode consists of looking for holes in the views of others will have an easy time with him to finding themselves the winners in such “arguments.” (For an illustrious model of the sorry predicament Sapolsky has thus walked into, think of Diderot’s Jacques le fataliste.)
Turn to the post-revolutionary transcendental-idealist framework. Here the basic concept is that of a representation (Vorstellung), and objects are constructed—not with hammer and sickle, or with psychic processes, but conceptually constructed, defined: it is specified what it is to be an object—by applying categorial criteria of objectivity. In this story I told many times, the wrinkle that counts for present purposes is that the construction can happen in multiple, equally legitimate ways. Causation is a key category: nothing uncaused can be tolerated as part of the (or better, in light of what is coming, a) world. But causation and other categories can parse the data (the representations) differently, giving rise to different worlds, in the plural, in which different or even the same objects show up differently arranged. The determinism on which contemporary science is founded is never denied, but the fact is: there is room for overdetermination, and for genuine compatibilism, not the social-manners variety exploited to no avail by Strawson and Krishnan. I can admit that something I did is exhaustively explained by reference to physical, physiological, or other empirical causes that long antedated my existence on this planet, thus making contemporary scientists like Sapolsky happy, and then find that it can also be explained by reference to a cause that acted freely.
How is that? Suppose you are playing a game of chess, and make a move M. You can explain M as an effect of physiological factors: neurons firing in your brain, electrical impulses traveling through your nervous system, your hand moving, … Or of psychological factors: your being an aggressive or conservative player, your knowing your opponent’s past history as a player, … With both explanations, you are going to find yourself, sooner or later, beyond the narrow confines of your life: your physiology and psychology are seamlessly connected with any number of other factors external to yourself which conditioned their action. You can also, however, explain M as a rational move: the one that could be rationally proved the best one under the circumstances. If that were so, the explanation would stop at the rational proof, and you, as a rational being, could claim that it makes M your own, in addition to M being physically, physiologically, or psychologically determined. Rationality is a harbinger of autonomy, of being the originator of one’s law: a version of freedom11placeholder long applied to God (who can only ever act for the greatest good, but in spite of this necessity is also ever acting without any external influence, as a law to himself). And, because rationality demands, in the name of the universality of reason, the elision of any idiosyncratic motivations, acting autonomously also amounts to abiding by the categorical imperative, which Kant believes to be the foundation of ordinary morality. To be rational, to be free, and to be moral are therefore shown by him to be conceptually equivalent. And, as an interesting side remark (and a vindication of an earlier parenthesis), though this analysis is applicable to human behavior, it does not end there: in principle, any natural event could be regarded as rational, moral, and free on the same grounds—which explains why, when freedom first enters Kant’s critical discourse, in the first Critique,12placeholder it is freedom, period, not freedom of a (human) will.
There are serious complications on the way to making this suggestion workable, and there are consequences of it not everyone will approve. Here I can only give a general idea of both, indeed of one of each. Begin with the complication.
As opposed to a game of chess, that is, to an artificially limited context of concerns (which is why “rational machines” like computers have been mastering it better than any humans for the last twenty years), even a single move in human behavior has indefinite complexity, most of it flying under the radar of consciousness. Kant says, famously:
“In fact, it is absolutely impossible by means of experience to make out with complete certainty a single case in which the maxim of an action otherwise in conformity with duty rested simply on moral grounds and on the representation of one’s duty. It is sometimes the case that with the keenest self-examination we find nothing besides the moral ground of duty that could have been powerful enough to move us to this or that good action and to so great a sacrifice; but from this it cannot be inferred with certainty that no covert impulse of self-love, under the mere pretense of that idea, was not actually the real determining cause of the will; for we flatter ourselves by falsely attributing to ourselves a nobler motive, whereas in fact we can never, even by the most strenuous self-examination, get entirely behind our covert incentives.”13placeholder
The only alternative to constitutional uncertainty about the moral value of an action, hence about its manifesting freedom, is concluding it to be irrational. In the absence of such a dismal verdict, we can continue to hope to be on the right track—an affect for which Kant had prepared the ground at the end of the first Critique when he wrote:
“All interest of my reason (the speculative as well as the practical) is united in the following three questions:
1. What can I know?
2. What should I do?
3. What may I hope?”14placeholder
Since the categorical imperative spells out the conditions of my duty and freedom, and I know that I will never be able to establish my having satisfied those conditions, I may—indeed, in a way, to be faithful to reason, I ought to—hope for such a felicitous outcome.
And now to the consequence many will find hard to accept. It should already be clear from the above that only rational, moral behavior can be free, hence be constituted of true actions. Irrational, immoral behavior is only subject to an explanation in terms of the causal determinism that rules the empirical sciences. There is no freedom to do evil. Reason looks upon evil “acts” the same way it would look upon a metastasis eating up an organism or a dripping faucet.
This thesis must be carefully unpacked, and distinguished from others it might be confused with. It does not say that evil does not exist, that it is only a defect of being, as Plotinus did; on the contrary, human nature is, according to Kant, the seat of radical evil—governed by irrational contingencies and impulses that one cannot expect (but one can hope) will eventually align themselves with reason’s commands. What it does say is that evil is banal, as Hannah Arendt did, or stupid: not in the sense that the people who commit evil are psychologically banal or stupid (of Eichmann, Arendt says “He was not stupid”15placeholder), but in the sense that there is no independent conceptual content, no plan if you wish, to their committing evil—they are just playing out their natural impulses as an earthquake is playing out the vicissitudes of plates and faults. And, sure enough, it does say that one ought to deal with evil with an objective, dispassionate mindset, as Sapolsky says we ought to do with all human behavior; but only because here the language of “ought to” is reason’s, which is independent of what people in fact do and how they react to what they are told—reasonably or otherwise. It is like saying what move one ought to make in a chess game, independently of what move one in fact makes, and of whether or not one makes it after being told what the right move is.
What makes this consequence unappealing to many is that a drive for vengeance is one of the impulses that govern the natural world, and in the throes of it many will not be satisfied until they draw blood from their offenders. We can only rationally hope that humans will find a way out of vengeance, but there is no prospect in sight that this is going to happen any time soon—human nature has never looked more like the seat of radical evil than it does now. Also, many will not approve of a freedom that, despite treating them the way God has been treated, will not let them “freely” choose what dessert to eat, or what neighbor to murder or not murder: one in which they will do, and choose to do, whatever they are determined to, and can only conjecture that it be the right (free, rational, moral) option. But, pending a bunch of details I cannot go into, they must admit that it is a solution of a metaphysical problem—how is determinism compatible with freedom?—, which is not even in sight within the garrulous “compatibilism” of the likes of Strawson and Krishnan. Here, if Mr. Smith does the right thing in standing up to political cronyism, he is free, despite being written as a character by Buchman, as much as I would be if I did the same despite being written by ancient history: “fictional” and real characters are on a par—indeed all characters, and objects, are in a deep sense fictional.16placeholder I should also add, in closing, that, despite tons of writing, highbrow and lowerbrow, this is the only real solution of what Kant called a “difficult problem” and about which he said:
“Some still let themselves be put off by this subterfuge [of saying that one is free if one does what one wants] and so think that they have solved, with a little quibbling about words, that difficult problem on the solution of which millennia have worked in vain and which can therefore hardly be found so completely on the surface.”17placeholder
Ethics Vindicated: Kant’s Transcendental Legitimation of Moral Discourse (New York: Oxford University Press, 2007).
In the issue of November 13, 2023.
London: Penguin, 2023.
Originally a lecture, published in Proceedings of the British Academy 48 (1962), 187-211, then reprinted in the collection by the same title (London: Methuen, 1974).
Note that, though countering the onslaught to some extent, he is partial to its typical moves. Thus, repeatedly in his paper, he alleges that the thesis of determinism has for him no definite meaning—he does not know what it is—, a preferred tool in the neopositivistic kit.
London: Methuen, 1959.
London: Methuen, 1966.
“Kant and Anglo-Saxon Criticism,” in Proceedings of the Third International Kant Congress, edited by L. W. Beck (Dordrecht NL: Reidel, 1972), 128-148; here 130.
Other popular versions of freedom are ruled out by causal determinism: the freedom of indifference (you can do anything you want) because causes that long predate you have determined the one thing you will do, and the freedom of spontaneity (you do what you want) because what you do and what you want are equally determined.
Translated by Paul Guyer and Allen W. Wood (Cambridge: Cambridge University Press, 1998). See the Third Antinomy, 484-489.
Groundwork of the Metaphysics of Morals, in Practical Philosophy, translated by Mary J. Gregor (Cambridge: Cambridge University Press, 1996), 43-108; here 61.
Critique of Pure Reason 677.
Eichmann in Jerusalem: A Report on the Banality of Evil (New York: Penguin, 1994) 287.
See my “Transcendental Idealism at Work with Fictional Objects and Names,” Epoché Magazine 63, June 2023.
Critique of Practical Reason, in Practical Philosophy, 139-271; here 217.