Sam Harris’s first book, The Moral Landscape: How Science Can Determine Human Values, sought to derive moral from scientific principles.1 In Free Will, he argues that science has shown that freedom of the will does not exist. Harris takes up one of the richest, most complex, and most interesting debates in the history of Western philosophy, and succeeds in trivializing it beyond recognition.

Free Will reflects the same refusal to grapple with the deeper philosophical questions that one finds across much of the philosophy of science today. Recall John Cornwell’s judgment about Daniel Dennett:

The most striking gap in Breaking the Spell is its lack of humanistic commentary from anthropology, aesthetics, and confessional literature. … Breaking the Spell is an insidious book; not because it breaks taboos by asking uncomfortable questions of religion, nor because its author is an ardent atheist, but because it is written by a brilliant philosopher who betrays his academic standards by proceeding from emotive, ill-informed prejudice.2

The claim that human beings never act freely implies that their behavior is determined. Defending the thesis of determinism is far more difficult than Harris acknowledges, and this for two reasons: the natural world does not support it, and our own existence makes the claim implausible, if not absurd.


Determinism tempts in part because one imagines that the law of the excluded middle must hold for future events. Aristotle formulates the argument in general terms: either it is true now that a sea battle will take place tomorrow, or it is false. If it is true now, nothing can be done to prevent it, and if false, nothing can be done to ensure it. What place is left for freedom of the will?

Assuming middles are excluded means assuming that indeterminism is false. But do the details of the various causal histories that we study actually establish determinism? Undeniably there are causal patterns, some of which recur with great regularity and predictability; but that point was never at issue. What is contested is whether all future events are determined: this being given, that must follow. No account of causality by itself implies very much about determinism. This is, after all, one of the consequences of Hume’s discussion of cause and effect: one cannot deduce that all events—future or, for that matter, past events—are determined simply by analyzing cause and effect.

Already we can see the danger of confusing determinism with its cognate concepts. Neither features of specific logical systems (the law of the excluded middle) nor philosophies of modality (possibility and necessity) suffice to establish that the future is determined. Baruch Spinoza supposed that causality and rational entailment are the same thing, but they remain distinct, Spinoza notwithstanding. Determinism as a metaphysical idea (in the sense of Laplace, perhaps) and determinism in the sense of a deterministic theory are equally distinct. But determinism and predictability are not identical: chaotic systems fail to be predictable but may still be deterministic; and we sometimes make successful predictions about horse races, say, or the lottery, that reflect no deterministic scheme whatsoever. Nor, finally, should determinism and fatalism be equated: numerous people manifest an attitude of resignation even when outcomes are undecided, whereas many determinists, Harris included, have taken pains to insist that their deterministic theories need not produce fatalism.

Indeterminism is a necessary, but not a sufficient condition for freedom of the will. The motion of a particle or the evolution of a system may be indeterministic, but it in no way follows that the particles are making free decisions as they go. Nevertheless, the necessity condition does hold. If there were a strictly deterministic system—a being given at t1, b must follow at t2—then it would make no sense to affirm that an agent at t1 could decide to bring about either b or not-b at t2.


None of the scientific details that advocates recite as they tell causal stories suffice to decide the issue. Clearly there are phenomena in the natural world that are highly regular: so regular that we call them laws. In the universe as Isaac Newton conceived it—and in the regions of the actual universe that approximate Newtonian systems—f = ma, and bodies move toward each other as predicted by the inverse square law. Language quickly becomes sloppy—even here, even so. Is it true to say that a meteor traveling through space is not free to deviate from Newton’s Second Law? The term “law” encourages the illicit picture of bodies being compelled into regular behaviors, as if they were motivated by fear of punishment.

Ours is not a purely Newtonian universe. At speeds close to c, in the neighborhood of black holes, and at microphysical scales, Newton’s laws no longer hold. The regularities in quantum physics are probabilistic, as are the realities they describe.

Why then side with strict determinism?

Newtonian determinism is useful in some contexts, but stochastic laws are required for the vast majority of natural systems, because their patterns lie at the level of statistical regularities. Laws play an even smaller role in living systems. Scientists generally explain them by describing their structure and function. Biologists don’t compute the body’s homeostasis or its resistance to viruses. When it comes to the common assumption that nothing alien is truly human, even Darwin does not suffice; history and cultural studies become essential. Social psychologists describe our behavior in groups, based on character traits we share with others. Yet even they throw up their hands when it comes to idiosyncratic features of individuals. “Mary is just like that,” we say.

“You simply can’t predict what she will do next.”

Note also that what is supposed to be determined varies. Is determinism a thesis about states, waves, or particles? About events or acts? About things or their mathematical representations? About individuals or collectives? For Newton, it is particular bodies, each with a specified mass. Schrödinger's equation, which is deterministic, describes the propagation of a wave, whereas statistical mechanics involves probabilities over large groups of particles. Biology would be deterministic only if the behavior of organisms were deterministic, either as a whole or in all of their parts.

Faced with the complexity and unpredictability of Mary’s actions, it seems impossible to know that every one of her actions is determined. Each variant of determinism is law-based; which has Mary in its iron grip? All of them, or just physics? Given de facto unpredictability, it will be difficult to make the case for any of them. If Harris answers physics, then which branch of physics; why Newton and not Heisenberg?

The dynamics of character development appear to be neither random nor derived from the deeper laws of physics. They give every appearance of being influenced by intentions and beliefs. Human action, Thomas Aquinas observed, is inclined but not impelled. Our ordinary use of language provides overwhelming confirmation that this is so, a point often stressed by Noam Chomsky. If in this respect, human action stands alone, then it cannot be deduced, or refuted, by instances of randomness or determination elsewhere in the world. Recall H. A. Prichard’s classic argument that moral philosophy rests on the mistake “of supposing the possibility of proving what can only be apprehended directly by an act of moral thinking.” The only solution, he argues in his conclusion, is to realize

the self-evidence of our obligations, i.e., the immediacy of our apprehension of them. … [I]f we do doubt whether there is really an obligation to originate A in a situation B, the remedy lies not in any process of general thinking, but in getting face to face with a particular instance of the situation B, and then directly appreciating the obligation to originate A in that situation.3

The Universe According to Laplace

An intelligence which knows at a given instant all forces acting in nature, as well as the momentary positions of all things of which the universe consists, would be able to comprehend the motions of the largest bodies of the world and those of the smallest atoms in one single formula, provided it were powerful enough to subject all data to analysis. To it, nothing would be uncertain; both future and past would be present before its eyes.4

The discoveries of the next 150 years undermined the picture of the universe that Laplace presupposed when he imagined his famous demon. The nineteenth-century study of temperature, energy, and entropy forced physicists to be satisfied with statistical mechanics, as in Ludwig Boltzmann’s explanation of the second law of thermodynamics. Henri Poincaré’s work on the three-body problem led him to discover the existence of systems that are both deterministic and chaotic. The laws of quantum mechanics are deterministic, but their interpretations, according to the Born rules, are probabilistic. What is more, quantum phenomena are inherently indeterminate. It is not that there is an exact state of affairs that we do not know.

There is no deterministic state of affairs to be known.

Refuting free will means establishing across the board that all the systems relevant to human behavior are deterministic. No one has done that; no one has the faintest idea how it might be done; the issue is barely coherent.

Philosophers of mind like Harris can argue that the specialists are wrong about their own theories. Or they can maintain that indeterminacy is erased by the time one reaches macrophysical scales. Recent physics undercuts attempts to draw qualitative lines between micro- and macro-phenomena. One thinks of the laser cooling methods that now allow physicists to study the quantum behavior of gases, for which William Phillips and his colleagues received the 1997 Nobel Prize in Physics. In 2011, using the laser cooling technique, scientists from the California Institute of Technology and the University of Vienna succeeded in cooling a mechanical object consisting of billions of atoms to its lowest possible energy state, at which point “it behaves according to the laws of quantum mechanics.”5

Harris and company might object that, even if indeterminacies infringe into parts of the macrophysical world, they have no place in brain dynamics. This, too, is a difficult position to defend: the electrochemical dynamics in brain synapses occur at scales at which quantum effects cannot be excluded. Even if the brain were a deterministic system, as deterministic as a set of billiard balls in action, the denial of free will would not follow.

All that we would know is that it is not within the brain that indeterminism resides.

Science As It Should Be

Hundreds of plants and animals make up even a simple ecosystem, and thousands of protein interactions are involved in the operations of a single cell. With some 86 billion neurons and some 1014 neural connections, the human brain is the most complex system we have yet encountered. Surely the thoughts and attitudes that are somehow correlated with the brain, and the actions motivated by conscious reflection, will not turn out to be less complex. It’s thus strange that, as soon as Harris turns to the cognitive states and beliefs of the large-brained Homo sapiens, all complexity evaporates. One supports science and is a determinist, we are told, or one opposes science and affirms free will.

Harris’s appeal to unsophisticated arguments is especially perplexing because it comes at a time when scientists are acquiring new empirical data about human cognition and the functions of human beliefs. Jonathan Haidt has recently identified six cognitive attitudes that interact in complex ways, creating distinctive cognitive fingerprints for the different approaches to belief-formation and ethical decision making.6 In Unto Others, Elliott Sober and David Sloan Wilson describe complexities in the development of sociality and morality across evolutionary time.7 Similarly, the cognitive science of religion now makes available significant new studies of the functions of religious beliefs and communities.

There are many arguments in the philosophical literature about whether religious ideas have added to the fitness of human groups.8 Are beliefs that enhance fitness more, or less, likely to be true? Alvin Plantinga has argued that, if naturalism and Darwinian evolution are both true, then we have no reason to believe in our cognitive faculties, and therefore no reason to think that naturalism is true.9

But no reference to these recent debates, or to the empirical data to which they are responding, can be found in Harris’s case against free will.

Why not?

Harris Lite

Harris was first thrust onto the public stage while still a graduate student thanks to two books, The End of Faith and Letter to a Christian Nation. They are not examples of subtle, complex philosophy: “[w]ords like ‘God’ and ‘Allah’ must go the way of ‘Apollo’ and ‘Baal,’ or they will unmake our world,” or “[i]t is therefore in the very nature of faith to serve as an impediment to further inquiry.”10 Perhaps it should not surprise us that when he finally turns to the philosophy of mind, the same light-headed tone emerges. Like so many others in the grip of an illusion, Harris imagines that he is able to see beyond it:

[W]e can’t make sense of [our sense of free will] in scientific terms … [W]e feel that we are the authors of our own thoughts and actions. However, I think that this mystery is itself a symptom of our confusion. It is not that free will is simply an illusion—our experience is not merely delivering a distorted view of reality. Rather, we are mistaken about our experience. Not only are we not as free as we think we are—we do not feel as free as we think we do. Our sense of our own freedom results from our not paying close attention to what it is like to be us. The moment we pay attention, it is possible to see that free will is nowhere to be found, and our experience is perfectly compatible with this truth.11

The quality of this argument is representative of the whole. There is no enduring self, we are told: no active subject. External behavior is accessible to the naturalist’s observation, as are brain scans and the electrochemical processes they reveal. No other kind of evidence is admitted. Natural science is the sole standard for knowledge because all knowledge is scientific.

These are the background assumptions upon which Harris’s conclusions rest. It follows from them that “the attribution of agency is always in error.”12 Human beings can act as if they consciously intend to do things, and do them because they intend them, but this fiction cannot ever be taken as a true account of human action.13 In possession of the truth, Harris can afford to make at least one magnanimous gesture to common sense: “[f]or most purposes, it makes sense to ignore the deep causes of desires and intentions—genes, synaptic potentials, etc.—and focus instead on the conventional outlines of the person.”14

In Society

Enter a crowded room. Pay attention. You perceive social hierarchies. You recall different degrees of connection and intimacy with the others present. You recall common experiences, previous conflicts, shared anecdotes, shared beliefs and values. Both consciously and unconsciously, you mold your words and actions so that they are appropriate. Because others are doing the same thing, an ever more complex set of reciprocal relations emerges. Some of your beliefs stay constant; others may evolve in significant, even life-changing ways during the course of a single evening.

It is hard to imagine any adequate philosophical account of our own existence that fails to offer some constructive account of what constitutes a person. The more metaphysical traditions have linked human agency to an otherworldly soul or eternal essence;15 idealists in the Greek, Neoplatonic, and German traditions have shown the irreducible role of subjectivity; phenomenologists have analyzed the nature of intentionality and consciousness as fundamental phenomena of personal and interpersonal experience;16 and analytic philosophers have worked in the same mine.17 The collection of concepts associated with personal existence has served as an orienting point for most of the history of philosophy.18 One can declare the entire history a mistake and undertake to do philosophy from scratch—a sort of pseudo-scientific equivalent of Derridean deconstruction. But then one should not be surprised to find that one lacks the very concepts that are essential for making sense of human life.


Agency is fundamental to free will; and it can be studied and described, albeit not in terms of physics or the neurosciences alone. Free agency is part of a rich web of concepts and capacities.

We would be lost without them.

Agency cannot be parsed in terms of gene expression or synaptic potentials alone. Synaptic potentials take a particular state at a specific time in a particular place in the brain. By contrast, intentions are temporally indeterminate and are not physically localizable; to suggest otherwise is to return to the absurdity of locating mind-body interactions in the pineal gland.

The further down the reductionist ladder one descends, the fewer resources one finds for explaining human intentionality. The serious student of nature looks for explanatory frameworks with which to comprehend the observed data. Intentional states are most plausibly explained in terms of personal agents and their psychological, social, cultural, and intellectual worlds.

It is too facile to write off human intentionality by equating it with an anti-scientific dualism. The natural world contains many causal relations and causal dynamics; that event a causes event b is one thing in quantum mechanics, and another in solid state physics, chemistry, and animal behavior. One has to oversee, or to negate, this rich empirical variety in order to claim that there are only two choices: either b follows from a through causes that can be expressed in terms of the laws of physics, or b is a chance event, hence unexplainable.

Natural laws are not imposed on phenomena. Certain types of natural systems, such as the propagation of a quantum wave or the orbit of the planets, exhibit such strict regularities that we can predict their future states. The biosphere also exhibits regularities, but far fewer of them can be predicted. Human beings also exhibit regularities, but they are of yet another kind: patterns of purpose, character, and intention.

There is nothing anti-naturalistic or anti-scientific in acknowledging that many human actions are influenced by introspection, conscious awareness, and reflection. Reflection involves selection from among a set of possible alternatives based on one’s perception of what is the more reasonable thing to think or do. There is no reason not to construe reasoning as a sort of causal influence, as long as the concept of a cause does not exclude mental processes.

Engaging in a lengthy process of reflection in order to deny that one has the ability to reflect represents a painful case of self-contradiction.

Freedom is inherently connected to the will, intentions, and character of an actor. It presupposes moral responsibility. If one denies intentionality, or if one arbitrarily limits causality to electrochemical interactions, one has acquired a conceptual system too impoverished to encompass free agency.

A human being has the capacity to form mental representations, to comprehend culturally transmitted concepts, freely to enhance and alter them, and to use them in new ways. We live in complex social worlds that societies and civilizations have constructed, transmitting ideas across multiple generations through speaking, writing and art, through science, philosophy and religion. Our daily experience is molded far more by these worlds of ideas and values than by anything else.19

On Not Finding Free Will

Philosophical arguments can err as much by omission as by the fallacies they commit. If one rules out the concepts necessary for describing and understanding human intentionality, as Harris does, then one will have no means for speaking of personal agency. Being unable to speak of it, one will claim that it does not exist. If there is no agency or will, then there is no free will. The assumptions being given, the conclusion follows.

If Harris believes he can demonstrate that no intentionality or free agency could exist in a world that also includes physical regularities, then he should supply the argument. He may be convinced that all of nature is strictly deterministic—a being given, b must follow—but he has not even begun to establish this thesis. Until he does, the case against free will has not been made.

  1. Sam Harris, The Moral Landscape: How Science Can Determine Human Values (New York: Free Press, 2010). 
  2. John Cornwell, review of Daniel Dennett, Breaking the Spell, Sunday Times of London, February 19, 2006. 
  3. H. A. Prichard, “Does Moral Philosophy Rest on a Mistake?Mind, N.S., vol. 21 (1912). 
  4. Quoted in Henry Margenau, Scientific Indeterminism and Human Freedom (Latrobe, PA: Archabby, 1968), 3. Laplace’s own position was more complex: “We are so far from knowing all the agents of nature and their diverse modes of action that it would not be philosophical to deny phenomena solely because they are inexplicable in the actual state of our knowledge. But we ought to examine them with an attention all the more scrupulous as it appears more difficult to admit them” (Pierre-Simon de Laplace, Essai philosophique sur les probabilités (Paris: 1814), 50.). 
  5. See Kimm Fesenmaier, “Caltech Team Uses Laser Light to Cool Object to Quantum Ground State,” Caltech Media Relations, October 5, 2011. 
  6. Jonathan Haidt, The Righteous Mind: Why Good People are Divided by Politics and Religion (New York: Pantheon Books, 2012). For an equally nuanced treatment of religious themes by one who writes as a Jewish atheist, see Haidt, The Happiness Hypothesis: Finding Modern Truth in Ancient Wisdom (New York: Basic Books, 2006). 
  7. Elliott Sober and David Sloan Wilson, Unto Others: The Evolution and Psychology of Unselfish Behavior (Cambridge, MA: Harvard University Press, 1998). 
  8. See Robert Wright, The Moral Animal: Why We Are the Way We Are: The New Science of Evolutionary Psychology (New York: Vintage Books, 1995); Jeffrey Schloss and Michael J. Murray, eds., The Believing Primate: Scientific, Philosophical, and Theological Reflections on the Origin of Religion (New York: Oxford University Press, 2009). 
  9. Alvin Plantinga first published his evolutionary argument against naturalism in 1993; it has been reprinted and expanded in James Beilby, ed., Naturalism Defeated? Essays on Plantinga’s Evolutionary Argument against Naturalism (Ithaca: Cornell University Press, 2002). 
  10. Sam Harris, The End of Faith: Religion, Terror, and the Future of Reason (New York: W.W. Norton & Co., 2005), 14, 45–46. 
  11. Sam Harris, Free Will (New York: Free Press, 2012), 95. 
  12. Sam Harris, Free Will (New York: Free Press, 2012), 50. 
  13. Here one sees how strongly Dennett’s intentional stance has influenced Harris. Daniel Dennett, Freedom Evolves (New York: Viking, 2003). 
  14. Sam Harris, Free Will (New York: Free Press, 2012), 90. 
  15. Cornelia J. de Vogel, The Concept of Personality in Greek and Christian Thought, Studies in Philosophy and the History of Philosophy (Washington: Catholic University of America Press, 1963). 
  16. Michael Theunissen, The Other: Studies in the Social Ontology of Husserl, Heidegger, Sartre, and Buber, trans. Christopher Macann (Cambridge, MA: MIT Press, 1984). 
  17. Among the important analytic treatments of the person—to mention just a few of the major titles—see Bernard Williams, Problems of the Self (Cambridge: Cambridge University Press, 1973); Sidney Shoemaker, Self Knowledge and Self Identity (Ithaca: Cornell, 1963); Derek Parfit, Reasons and Persons (New York: Oxford University Press, 1984); John Perry, The Problem of the Essential Indexical and Other Essays (New York: Oxford University Press, 1993); Amelie Rorty, ed., The Identities of Persons (Berkeley: University of California, 1979); Harold Noonan, Personal Identity (New York: Routledge, 1989); Peter Unger, Identity, Consciousness and Value (New York: Oxford University Press, 1990); and Carol Rovane, The Bounds of Agency: An Essay in Revisionary Metaphysics (Princeton: Princeton University Press, 1998). 
  18. Michael Carrithers, Steven Collins, and Steven Lukes eds., The Category of the Person: Anthropology, Philosophy, History (Cambridge: Cambridge University Press, 1987). 
  19. See John R. Searle, The Construction of Social Reality (New York: Free Press, 1995).