Free Will Foretold: Review of Peter B. Clark on Compatibilism

Tom Clark
Book Title: 
The Compatibility of Free Will and Determinism
Book Author: 
Peter B. Clark

In 1963, a very smart senior at Reed College wrote a thesis prefiguring arguments of contemporary philosophers who claim that genuine human agency is compatible with determinism. It stands as a trenchant critique of libertarian free will, helping to naturalize our self-conception in light of science.  

Free Will Foretold

Review of

The Compatibility of Free Will and Determinism


Peter B. Clark


The Perceived Threat of Determinism to Human Agency

Despite efforts over decades by compatibilist philosophers – those who hold that determinism is compatible with free will and moral responsibility – determinism is still widely perceived as the enemy of human freedom. How can we be free and responsible if, as is possibly the case, we are the fully caused products of genetic and environmental factors we didn’t choose? And if the causal story of our creation is perhaps leavened by a random swerve here and there at the microphysical level, how can that add to our self-formation, our control over behavior, and our responsibility for it?

About sixty years ago as of this writing, one Peter B. Clark (Peter the Younger, PtY), a smart senior at Reed College and future economist at the International Money Fund, produced a thesis aiming to reassure us about this. Now brought to light, (Kindle edition here) it stands as an astute argument that, properly understood and put in its place, determinism poses no threat to human freedom and responsibility. Philosophy being what it is, there has of course been lots of continuing debate in the years since. As PtY puts it, given disagreements about the meaning of key terms “…philosophy is not a science; a proof in philosophy does not necessarily imply that anything has been proved.” His monograph thus remains a good exposition of much that’s still central to the free will debate, and his arguments pass the test of relevance, anticipating those put forth by today’s philosophers of agency.[1] It is therefore meet and right that his thesis be given its due as a cogent precursor of current thinking about free will. Not to mention that this precocious undergraduate had a good sense of style and clear argumentative structure, making for relatively straightforward reading despite the necessary technicalities. It could easily pass for graduate level work. How he managed at such a tender age to so skillfully command and disambiguate the concepts that populate the free will debate is, Peter the Elder (PtE) now admits, somewhat of a mystery. But the stewardship of philosopher Hugo Bedau, who signed off on the thesis, no doubt contributed.

Why It Matters: Moral Responsibility in Question

PtY locates the practical importance of the debate where many now agree it resides: in worries about moral responsibility.[2] If, as is conceivably the case, science eventually shows human behavior to be explicable in terms of reliable causal relations at various levels (physical, chemical, biological, psychological), then it seems to leave the person as an originator of action out of a job, calling into question commonsense ideas of credit and blame. If in an actual situation I couldn’t have acted otherwise given all the causes in play, how am I free and perhaps deserving of praise or censure? The potential personal and social ramifications of the free will debate are therefore considerable, encompassing deep disagreements about such things as retributive punishment, economic inequality, and just deserts.[3] As PtY points out in his introduction, the hope of compatibilists (who think free will and moral responsibility are compatible with determinism) is to persuade libertarians (who think determinism threatens freedom and responsibility) that in fact there is no incompatibility or threat here. Even if our behavior is physically necessitated, PtY will argue, we can still legitimately hold each other morally responsible. The arguments to come are thus mainly directed at libertarians, among whom are arguably the majority of non-philosophers, although a distinct minority of philosophers as well.[4] But he says determinists too sometimes stray from a proper interpretation of their outlook. He also rightly observes that the truth of determinism is not at issue: “…if it is true that determinism does not negate man’s freedom, then there is no need to find out whether determinism is true.” But since it might be true, the libertarian worry, and any determinist misconceptions, must be addressed in order to keep the world safe for moral responsibility.

Scientific Determinism

As described in the introduction (Chapter 1), the determinism that upsets libertarians is the idea that if human behavior follows scientifically described exceptionless natural laws, then those laws, when applied to a given situation, logically entail that only a single behavior is possible. Therefore, it seems that if determinism is true, one’s actions, whatever they were, were unavoidable. And to act freely and be morally responsible, one must have been able to avoid behaving as one did, that is, able to act otherwise than one did, what PtY calls the avoidability condition for moral responsibility: “That we do hold people responsible only for those acts which could have been avoided is a fact of our social life.” The challenge from science to human freedom thus looms: “So far precise laws with a high predictive value have been developed mainly in the physical sciences. However, there is no reason to suppose that such laws cannot be developed for human behavior.”

In a typically nuanced and careful discussion, PtY notes that the determinism in question – what he calls scientific determinism – is that of the logical relations between terms of a scientific theory which then are applied in describing human action:

If determinism is understood in this sense, i.e., as a term referring to the logically necessary connections between the terms of a scientific theory, then (in the formal mode) an event may be said to be determined if and only if its description is deducible from a set of expressions describing the initial conditions of a system in terms of a scientific theory. Throughout this thesis an event, choice, action, etc., will be said to be determined in that sense of “determined” defined here.

Why We Needn’t Escape Determinism

With the potential conflict between determinism and avoidability (and thus moral responsibility) introduced, Chapter 2 critiques elements of the libertarian position on free will, with extensive quotations from its then major proponents, in particular C.A. Campbell, its “most outspoken defender.” The basic libertarian contention is that “…a person is free if and only if his actions and choices cannot be predicted and explained by scientific laws...”  It seems to the libertarian that only by being exceptions to determinism – by having contra-causal powers – do we have the ability to do otherwise that then makes us free and responsible. What’s properly meant by the ability to do otherwise will figure centrally in chapter 4, but here PtY offers an insightful analysis of what he takes to be faulty libertarian intuitions, showing them to be naturalistically implausible. Unfortunately (I’d say), they persist widely among the folk (my term for non-philosophers) and are still defended by some in the academy.

One is that determinism somehow robs the person of causal power, which is instead attributed to the antecedent conditions that explain (by having caused) the person, their motives, decisions, and actions. To be a free and responsible libertarian agent, the person must not be fully traceable to such conditions, but ultimately self-caused in some respect. But as PtY points out, the fact that the agent herself might be determined (and the science of human behavior grows apace) does not entail that she does not determine her choices. A consistent application of determinism locates causal power within the individual as well as outside, which meets what PtY suggests is the other main requirement for moral responsibility besides avoidability: the self-determination condition for a free choice. We are, in natural fact, determinators of action – we are robust proximate sources of behavioral control – even if it turns out we ourselves are fully explained in terms of a deterministic scientific description. Determinism might rule out alternative possibilities in actual situations (as opposed to counterfactual, see below re Chapter 4), but it doesn’t strip the agent of the causal powers and control that produce effects in those situations. The belief that it does, and the resulting intuition that determinism entails fatalism, is one factor blocking acceptance of a naturalistic causal understanding of ourselves.[5]

Another misconception harbored by libertarians is that our feeling of freedom – of choosing between what seem genuinely open possibilities – is good evidence that we somehow evade or transcend determinism and that there is a controlling self independent of motives and desires that arbitrates among them when making choices. Against what he calls “the appeal to introspection,” PtY observes: “The necessity of the deterministic system which the libertarian attempts to disprove is logical necessity, and this kind of necessity cannot be disproved by examining one’s feelings.” This anticipates what a respected philosophical expert, Alfred Mele, says about libertarian free will (“deep openness”) in his 2022 guide to free will for non-experts: “What I’m saying is that the difference between deep openness and its absence isn’t the kind of thing that can be felt. We sometimes do feel uncertain about what we will do. But we can have exactly that feeling even if the universe is deterministic.” (Mele 2022, p. 142) The basic epistemic precept PtY and Mele are getting at here is that one’s subjective experience shouldn’t be taken as a reliable indicator of what’s objectively the case.

PtY also makes an important and not uncontested claim about our essential nature which bears on conceptions of free will: “The point is that we don’t think of a person or a self as something over and above, and independent from the desires, habits, motives and dispositions of the person or self.” This is non-dual music to my naturalist ears, but evidence coming from Xphi research (see note 4) suggests many folks might harbor at least a latent, sometimes supernatural dualism concerning the self, e.g., of a soul or spirit that rides herd on desire.[6] If compatibilism requires giving up such a self, then it might be a tough sell in the wider culture. Dualism about the self and the belief that it has contra-causal capacities aren’t likely to disappear anytime soon.

Why Determinism Isn’t a Universal Excuse

After this bracing critique of libertarian intuitions, Chapter 3 proceeds to dispatch a further common misreading of determinism: that by explaining behavior as governed by laws it somehow excuses wrongful or criminal acts. One might suppose that, if determinism logically necessitates that a particular act occurred given the conditions and causal relations in play, then the agent couldn’t have done otherwise, so can’t be morally responsible, and thus must be excused. But in looking at the excuses commonly adduced in a criminal law textbook, and in analyses of tort liability, PtY finds no mention of determining conditions as a defense that negates culpability. But he also takes a careful look, in light of our actual conceptions of an excuse, at the faulty reasons that might lead us to imagine that determining conditions count as excuses. One is that determinism is akin to compulsion, that it somehow undermines the intentional, voluntary aspect of behavior originating in an agent’s character and motivations – the self-determination condition for moral responsibility. But, as PtY says, “to show that an action was determined would not enlighten us about the circumstances of the action in such a way that would influence us in our allocation of responsibility,” for instance, whether it was done intentionally and voluntarily as opposed to under duress. He points out that the logical necessity of determinism is quite other than the necessity involved in compulsion, which has to do with being forced to act against one’s intentions and desires, or with lacking the capacity to resist doing what one should not have done. Determinism leaves the distinction between voluntary action and compulsion perfectly intact.

The intuition that determinism might excuse wrongful acts has been dubbed the “fundamental psycho-legal error” by philosopher of law Stephen Morse (Morse 2004, Moore 2014). Morse concludes, as did PtY decades before him, that “All behavior may be caused in a physical universe, but not all behavior is excused, because causation per se has nothing to do with responsibility” (Morse 2004). Agents, albeit possibly fully caused to be who they are and act as they do, are themselves usually (not always) the reasons-sensitive and consequence-anticipating causers of action, and as such are justifiably held responsible as a means to “guide goodness,” as Morse puts it.

Could You Have Done Otherwise?

Still, the intuition might persist that determinism negates the avoidability condition. This is the central concern of chapter 4, a detailed analysis of what we mean, or should mean, by could have done otherwise.  In the most demanding section of his thesis, PtY compares at length the views of philosophers, including J. L. Austin, P. H. Nowell-Smith, Gilbert Ryle, and John Canfield, on the meaning of can. He concludes that the best reading of it consistent with logic and ordinary usage is that you could have done something if you had the ability and opportunity to do it, even if you didn’t actually do it. The thought experiment he uses, “Austin’s putt,” originally described in J. L. Austin’s classic paper “Ifs and Cans” (Austin 1956) and Nowell-Smith’s paper of the same title (Nowell-Smith 1960), is also deployed by Daniel Dennett years later in his book Freedom Evolves (Dennett 2004, p. 75). Dennett uses it to reach the same conclusion: we shouldn’t suppose determinism entails unavoidability, what he calls inevitability.

What does it mean, they ask, to say that Austin on a particular occasion could have made the (easy) putt, even though he missed it? On a deterministic reading, that of physical necessity, it looks as if he couldn’t have done otherwise given all the conditions in play; but both PtY and Dennett argue that this isn’t the correct perspective to take when asking what he could have done. If you can regularly sink a putt at a certain distance, and nothing is preventing you from exercising that ability (so you have the opportunity to exercise it), then you could have sunk it even if you didn’t on a particular occasion. We can likely explain why you didn’t, e.g., a momentary lapse of attention, which, had it not occurred, would have resulted in you holing the ball. This means it wasn’t unavoidable or inevitable that you missed: you likely would have holed it had the lapse not occurred. In general, as PtY puts it, “To say that someone could have avoided doing something is to say that he had the ability and the opportunity to do something else.” Determinism doesn’t entail that neither the ability nor opportunity were present on the occasion when you didn’t make the putt, even though a causal analysis could likely explain why you didn’t. More consequentially, morally responsible agents are those that have the general abilities – rationality, sensitivity to and knowledge of norms, standard capacities of anticipation, deliberation, and self-control, all of which Dennett calls moral competence – to do the right thing, even if on a particular occasion they don’t. Such competence makes them justifiably subject to responsibility practices involving praise and blame which make it more likely they will exercise their ability to do the right thing on future occasions.

This sense of could have done otherwise, what PtY and Dennett want us to endorse, is called by philosophers conditional or counterfactual: I likely would not have cheated on the test had the conditions been somewhat different, for example had the proctor been standing over my shoulder. Had that been the case, I wouldn’t have cribbed from my friend’s worksheet since my abilities include sensitivity to surveillance and knowledge that cheating is wrong. So, in that sense it wasn’t inevitable or unavoidable that I cheated, and this sense is compatible with the determinism that may have held in the actual situation. However, there’s also what’s called the unconditional sense of doing otherwise, which has it that given conditions exactly as they were, I might not have cheated and thus could have avoided cheating. This sort of avoidability, it seems, is incompatible with determinism since on determinism there’s no reason to suppose I would have done otherwise given the conditions and causes in play in the actual situation. But it’s this sense – our being local exceptions to causal necessity – that the libertarian, whether in the academy or on the street, thinks is crucial in establishing moral responsibility (see above re chapters 2 and 3). According to PtY, as well as Dennett and most naturalist philosophers, we are not such exceptions, and they would say we don’t need to be exceptions be held morally responsible. We only need to be morally competent, and most of us are.

Compatibilism Meets Free Will Skepticism in the 21st Century

An important conclusion PtY draws from his analysis is that “Whenever a philosopher says that the fact that an action is determined means that the agent could not have acted otherwise, he should be very careful to indicate the special meaning he is giving to ‘could’.” Good advice for anyone discoursing about free will. But, as noted above, this supposedly special meaning undergirds the widespread libertarian intuition among the folk, and among some philosophers, that determinism poses a threat to moral responsibility. They think we need to be unconditionally free – ultimately self-created first causers in some sense – to be justly held responsible (Nadelhoffer et al. 2020). Currently, a growing band of free will and moral responsibility skeptics (Waller 2011, Focquaert et al., 2020, Caruso 2021, Pereboom 2021, Caruso & Pereboom 2022) leverage this intuition to argue against what they call basic desert responsibility, that which is used to justify punishment in the retributive sense, where no socially beneficial consequence of punishment need obtain, e.g., public safety, deterrence, or moral reform.[7] Determinists, they argue, need to show why morally competent agents deserve non-consequentialist harsh treatment if, as is likely the case, they were fully caused (determined) to become who they are and behave as they do, and where indeterminism, should it play a role, can’t add to anyone’s control or responsibility. So, although PtY presents a strong case against libertarianism, his compatibilist determinism, still very much the majority position in the academy, is being challenged as an insufficient basis for retribution, one of the dominant justifications for punishment in the United States (Focquaert et al. 2020, Caruso 2021).[8] He could not, sixty years ago, have anticipated the increasingly public questioning of libertarian free will coming from a scientific, naturalistic perspective that’s now underway (Harris 2012, Sapolsky 2017, Sapolsky 2023). Not only is retribution under pressure from such questioning, but also desert-based justifications for increasingly extreme social and economic inequality: are billionaires and the top 1% self-made in a way that makes them deeply deserving of their riches, or are they, perhaps, simply lucky in their causal antecedents? More generally, it challenges what’s called the just world belief: that people usually get what they deserve in life (e.g., rags instead of riches, prison instead of liberty) because they could have unconditionally done otherwise, but simply chose not to.[9]

These challenges, momentous as they are, do not upset the good sense that abides in PtY’s thesis and that still typifies compatibilist reassurances about being free and responsible. Were determinism the case (and indeterminism doesn’t help) we do, as he argues, have agential capacities that underwrite the self-determination condition for free, morally responsible action, and that make it the case (for most of us) that we could have done otherwise in the conditional, counterfactual sense, meeting the avoidability criterion thus defined. Peter the Elder can justifiably look back and admire the work of his personal predecessor in articulating a naturalistic and logically compelling case that determinism doesn’t threaten the practical basis, and necessity, for holding each other responsible, even if certain responsibility practices can be called into question. Those following the debate on free will and moral responsibility, especially those engaged in the philosophical arguments, will find much here not only of historical interest, but of current relevance as the science of human nature intrudes on libertarian intuitions. Determinism, should it be true, doesn’t diminish us, and Peter the Younger was ahead of the curve, for his age and in his era, in showing why.

TWC  10/2023


Austin, J. L. 1956. Ifs and Cans. Proceedings of the British Academy 42, pp. 109-152.

Bourget, D. & Chalmers, D. 2020. The 2020 PhilPapers Survey.

Caruso, G. D. & Dennett, D. C. 2021. Just Deserts: Debating Free Will. Cambridge: Polity Books. Reviewed here

Caruso, G. D. & Pereboom, D. 2022. Moral Responsibility Reconsidered. Cambridge: Cambridge University Press.

Caruso, G. D. 2021. Rejecting Retributivism: Free Will, Punishment, and Criminal Justice. Cambridge: Cambridge University Press. Reviewed here.

Clark, T. W. 2001. Against retribution: review of Placing Blame by Michael S. Moore. Naturalism.Org

Clark T. W. 2005. Re: Morse and retribution. Naturalism.Org

Clark, T. W. 2008. Don’t forget about me: avoiding demoralization by determinism. Naturalism.Org.

Dennett, D.C. 2004. Freedom Evolves. New York: Penguin Books.

Fischer, J. M. et al., 2007. Four Views on Free Will. London: Wiley-Blackwell. Reviewed here.

Fischer, J. M. forthcoming. Moral responsibility skepticism: semiretributivism. Reply to this here.

Focquaert et al. 2020. Justice Without Retribution: Interdisciplinary Perspectives, Stakeholder Views and Practical Implications. Neuroethics Special Issue V13 (1).

Harris, S. 2012. Free Will. New York: Free Press.

Hirstein, W. et al. 2018. Responsible Brains: Neuroscience, Law, and Human Culpability. Cambridge: MIT Press.

Kelly, E. I. 2018. The Limits of Blame: Rethinking Punishment and Responsibility. Cambridge: Harvard University Press.

Knobe, J. & S. Nichols 2017. "Experimental Philosophy," The Stanford Encyclopedia of Philosophy (Winter 2017 Edition), Edward N. Zalta (ed.).

Mele, A. R. 2022. Free Will: An Opiniated Guide. New York: Oxford University Press.

Moore, M. S. 2014. Stephen Morse on the fundamental psycho-legal error. Criminal Law and Philosophy V10, pp. 45–89.

Morse, S. J. 2004. New neuroscience, old problems. Cerebrum. Dana Foundation.

Morse, S. J. 2004. Reasons, results, and criminal responsibility. Illinois Law Review.

Nadelhoffer, T., Yin, S., & Graves, R. 2020. Folk intuitions and the conditional ability to do otherwise. Philosophical Psychology, 33(7), 968–996.

Nowell-Smith, P.H. 1960. Ifs and Cans. Theoria, 26, pp. 85-101.

Pereboom, D. 2021. Wrongdoing and the Moral Emotions. Oxford: Oxford University Press.

Sapolsky, R. M. 2017. Behave: The Biology of Humans at Our Best and Worst.  New York: Penguin Books.

Sapolsky, R. M. 2023. Determined: A Science of Life Without Free Will. Reviewed here.

Waller, B. N. 2011. Against Moral Responsibility. Cambridge: MIT Press. Exchange with Clark, Dennett, Waller on this book here.

Waller, B. N. 2020. Beyond moral responsibility to a system that works. Neuroethics 13, 5–12.

Wisniewski et al. 2019. Free will beliefs are better predicted by dualism than determinism beliefs across different cultures. Plos One.



[1] For debate among advocates of major positions on free will – libertarianism, compatibilism, revisionism, and skepticism – see Fischer et al., 2007.

[2] See for example Pereboom 2021 and Caruso & Pereboom 2022.

[3] See for example Waller 2011, Kelly 2018, Hirstein et al. 2018, Caruso 2021, and Caruso & Dennett 2021.

[4] Using surveys and focus groups, philosophers have studied beliefs about free will and moral responsibility in the general population over the last two decades in what’s called experimental philosophy (“Xphi”). Findings suggest that belief in libertarian free will (or that determinism poses a threat to free will) is common, but also that, depending on how survey questions are framed, non-philosophers also have compatibilist intuitions. Vigorous debate in the Xphi community on methods and the interpretation of data continues, about which see Knobe & Nichols 2017, section 2.2.  A 2020 survey of philosophers (Bourget & Chalmers 2020) found that about 19% accept or lean toward libertarianism and nearly 60% accept or lean toward compatibilism.

[5] Fatalism is the thesis that no matter what one does, a particular unavoidable future awaits, effectively removing the agent as a causal influence in how events unfold. Determinism keeps the agent’s causal role intact, in which case what we do matters in determining future events, see Clark 2008.

[6] See for example Wisniewski et al., 2019.

[7] Note that other non-desert senses of moral responsibility survive on their views, about which see Waller 2011, Pereboom 2021, and Caruso & Pereboom 2022. John Martin Fischer, a compatibilist who endorses what he calls semi-retributivism, has noted the increasing influence of free will and moral responsibility skeptics (Fischer forthcoming).

[8] Compatibilists need not endorse retributivism (Daniel Dennett does not) but some of considerable prominence do, in particular Michael S. Moore in his book Placing Blame (reviewed in Clark 2001), Stephen J. Morse (Morse 2004, Clark 2005) and John Martin Fischer (forthcoming).

[9] About belief in a just world and its social consequences, see Waller 2020.