“All the world’s a stage, and all the men and women merely players.”

— Shakespeare

Humans love stories.  We tell each other the stories of our lives, in which we are not merely players reading a script but also the authors.  As authors we make choices that influence the plot and the other players on the stage.  Free will can be understood as our capacities both to make choices—to write our own stories—and to carry them out on the world’s stage—to control our actions in light of our choices.

What would it mean to lack free will?  It might mean we are merely puppets, our strings pulled by forces beyond our awareness and beyond our control.  It might mean we are players who merely act out a script we do not author.  Or perhaps we think we make up our stories, but in fact we do so only after we’ve already acted them out.  The central image in each case is that we merely observe what happens, rather than making a difference to what happens.

How might neuroscience fit into the story I am telling?  Most scientists who discuss free will say the story has an unhappy ending—that neuroscience shows free will to be an illusion.  I call these scientists “willusionists.” (Willusionists include Sam Harris, Jerry Coyne, Jonathan Bargh, Daniel Wegner, John Dylan Haynes, and as suggested briefly in some of their work, Stephen Hawking and Richard Dawkins.) Willusionists say that neuroscience demonstrates that we are not the authors of our own stories but more like puppets whose actions are determined by brain events beyond our control.  In his new book Free Will, Sam Harris says, “This [neuroscientific] understanding reveals you to be a biochemical puppet.” Jerry Coyne asserts in a USAToday column: “The ineluctable scientific conclusion is that although we feel that we’re characters in the play of our lives, rewriting our parts as we go along, in reality we’re puppets performing scripted parts written by the laws of physics.”

There are several ways willusionists reach their conclusion that we lack free will.  The first begins by defining free will in a dubious way.  Most willusionists’ assume that, by definition, free will requires a supernatural power of non-physical minds or souls:  it’s only possible if we are somehow offstage, beyond the causal interactions of the natural world, yet also somehow able to pull the strings of our bodies nonetheless. (For example, Read Montague.)  It’s a mysterious picture, and one that willusionists simply assert is the ordinary understanding of free will.  Based on this definition of free will, they then conclude that neuroscience challenges free will, since it replaces a non-physical mind or soul with a physical brain.

But there is no reason to define free will as requiring this dualist picture.  Among philosophers, very few develop theories of free will that conflict with a naturalistic understanding of the mind—free will requires choice and control, and for some philosophers, indeterminism, but it does not require dualism.  Furthermore, studies on ordinary people’s understanding of free will show that, while many people believe we have souls, most do not believe that free will requires a non-physical soul.  And when presented scenarios about persons whose decisions are fully caused by earlier events, or even fully predictable by brain events, most people respond that they still have free will and are morally responsible.   These studies strongly suggest that what people primarily associate with free will and moral responsibility is the capacity to make conscious decisions and to control one’s actions in light of such decisions.

But willusionists also argue that neuroscience challenges free will by challenging this role for consciousness in decision-making and action.  Research by Benjamin Libet, and more recently by neuroscientists such as John Dylan Haynes, suggests that activity in the brain regularly precedes behavior—no surprise there!—but also precedes our conscious awareness of making a decision to move.  For instance, in one study neural activity measured by fMRI provided information about which of two buttons people would push up to 7-10 seconds before they were aware of deciding which to push.

If such early brain activity always completely determines what we do before our conscious thinking ever comes into the picture, then this would suggest we lack free will, because our conscious thinking would happen too late to influence what we did—an audience rather than author.  But the data does not show that brain activity occurring prior to awareness completely causes all of our decisions.  In the study just described, the early brain activity correlates with behavior at only 10% above chance.  It is not surprising that our brains prepare for action ahead of time and that this provides some information about what people will do.

Of course, improved brain imaging technology will likely provide increasingly precise predictions of future behavior.  But here’s my prediction:  the more complex the decisions and behavior, the more likely such predictions will be based on information about the very neural processes that are the basis of conscious deliberation and decision-making.

Once we assume that all mental processes have neural correlates, then whether consciousness plays a role in our complex behavior turns on whether the neural correlates of conscious processes occur at the right time and place to influence behavior.  It’s unlikely that the neural processes involved in complex deliberations, planning, and self-control play no role in behavior.  Instead, there is evidence that conscious and rational thinking can play an important causal role in complex behavior.  If we give up the mysterious picture of our conscious selves being offstage, then we can give up the threatening image of our brains pulling the strings while we helplessly watch.

One reason it is easy to move from the assumption that neural processes cause behavior to the presumption that consciousness does nothing is that neuroscience still lacks a theory to explain how certain types of brain processes are the basis of conscious or rational mental processes.  Without such a story in place, it is easy to assume that neuroscientific explanations supersede and bypass explanations in terms of conscious and rational processes.  But that conclusion is unwarranted.  Explanations in organic chemistry do not explain away life; they explain life.  A more complete scientific theory of the mind will have to explain how consciousness and rationality work, rather than explaining them away.  As it does, we will come to understand how and when we have the capacities for conscious and rational choice, and for self-control, that people ordinarily associate with free will.  These are the capacities to reflect on our desires and reasons, to consider which of them we want to motivate us, and to make efforts to act accordingly—or as Roy Baumeister explained in his recent post, to habituate ourselves to make choices that accord with our reflectively endorsed goals.

By understanding how the most complex thing in the universe—the human brain—works, we can better understand our capacities to make choices and to control our actions accordingly.  On this telling of the tale, neuroscience can help to explain how free will works rather than explaining it away.

Now, if one insists that free will requires that we have an impossible ability to make choices beyond the influence of anything, including our own brains—or to make choices for no reason at all—then you will be disappointed by the story I am telling.  Here, willusionists like Sam Harris and I agree that we cannot have what is impossible.  Our choices do not arise from nothing any more than an author’s stories arise from nothing, but our choices do influence the way our stories unfold.

Nonetheless, fascinating research suggests that our conscious reasoning and planning is not pulling the strings as much as we tend to believe.  We are subject to biases and influences beyond our awareness, and we sometimes confabulate or rationalize our behavior. But our stories are not always fiction.  Other research suggests that our deliberations and decisions can have significant causal influences on what we decide and do, especially when we have difficult decisions to make and when we make complex plans for future action.

Free will is not all-or-nothing.  It involves capacities that we develop as we mature, but that have limitations.  Recognizing that people have differing degrees of free will can help us better determine when, and to what extent, people are responsible for their actions, and are deserving of praise or blame.  Indeed, where it really matters—legal responsibility—it is most useful to understand free will as a set of capacities for reasoning and self-control which people possess to varying degrees and have varying opportunities to exercise.

In this respect, neuroscience and other sciences of the mind can play an important role by providing new insights into our capacities for rationality and self-control, as well as their limitations.  We do not write our stories from scratch, but within the context of a complicated world of influences and interactions, our tales are not “full of sound and fury, signifying nothing.”

Questions for the discussion:

  • What do you think free will is?
  • Do you think free will is all-or-nothing or that we possess and exercise free will to varying degrees?  How much free will do we have?
  • Is free will necessary to deserve praise and blame for one’s actions?  If so, how much are people responsible for their actions and their situations?
  • Is free will inconsistent with a naturalistic worldview—with an understanding of our conscious minds as physically instantiated in our brains?

Discussion Summary

We’ve seen that people understand “free will” to mean different things and that people think our having free will would require different things.  I think the best way to define “free will” is (roughly):  “the set of powers or capacities for making choices and controlling actions that an agent needs to be morally responsible for her choices and actions.”  I think this definition accords with the way most people, and most philosophers, understand free will, and I think it is also theoretically useful.  That is, it provides a useful target for philosophical analysis—what are those capacities and what would limit or eliminate them?—and then for scientific study.  Once we pick out the relevant capacities, we can study:  how they are instantiated in humans (if they are), to what degree humans (as a species) possess them, to what degree (individual) humans possess them and exercise them in particular actions, and what might help us develop these capacities.  Free will, as defined here, seems to require that free actions can be influenced by rational deliberation and conscious choice.  On the conceptual side, how should we understand these capacities and the type of causal influence they need to have for our actions to count as free and responsible?  On the scientific side, how do our brains implement these capacities and what prevents them from playing a causal role in action?

Other definitions of “free will” are out there, and some people will forever be dissatisfied with any definitions that do not include non-physical souls, indeterminism, agent-causal or “contra-causal” powers, and/or the ability to act outside of the purview of the laws of nature.  It’s unlikely I will convince such people; after all, most of the age-old philosophical debates have focused on whether these “libertarian” powers are required for free will and moral responsibility.  But my studies on people’s intuitions about free will suggest that the main reason determinism or naturalism appear to challenge free will is because people understand them to mean that our conscious selves play no role in our behavior.  I think that as the sciences of the mind advance, they will demonstrate—and delineate—what role our conscious selves play in action.  In doing so, they will help us see how a naturalistic free will works.

New Big Questions:

1. If we lack powers to act outside of the laws of nature or to make indeterministic choices, does that mean we lack free will?

2. If we have capacities to make conscious and rational choices, and to control our actions in light of these choices, is that enough for us to have free will?

3. What can neuroscience tell us about whether we have these powers and capacities people associate with free will?

4. What implications will these neuroscientific discoveries have for our moral and legal practices?  Are those implications justified by the evidence and arguments?

40 Responses

  1. almele says:

    Excellent article, Eddy. I’m curious to see what people will agree & disagree with you about.

  2. Caesar says:

    I think Nietzsche summarized free will correctly and I will paraphrase: free will is like the ability to pull oneself out of a swamp by the hair. I would argue that it is not only wrong, it is also incoherent. If this is true, neurobiological arguments are beside the point. They are irrelevant. The self, whatever it may be, is not arise ex nihilo; it comes from some substratum or another. Whether this is substratum is spiritual or neurological is immaterial. The “I” is a creation, in other words, of forces of which the “I” is utterly unaware. If the “I” wants one thing or another, it is not because the “I” instantaneously created that desire; it is because of some configuration of the underlying substratum. Alter the configuration and one alters the desire and, ceteris paribus, the proceeding decision.

    Of course, most modern thinkers assume that his substratum is material, is composed of some combination of nervous tissue. This sometimes leads to an unnecessary focus on the findings of neurology. “Neurobiologists have shown that X, Y, or Z,” is a standard starting point for a debate on free will. But that is an unnecessary move. Suppose the self were entirely spiritual and that we could not, even in principle, relate cognitions to neural processes. What follows? A surfeit of interesting things, no doubt, but nothing about free will. Where does a desire come from? If the self is purely spiritual, one might say that it comes from the essence of the self; one cannot, however, argue that it arises because the self chose it. And even if the self chose it–why? What impelled the self to choose it? 

    My guess is that most debates about free will boil down to semantics. If free will is defined in a certain way–as the author noted–it is real. Although some have argued that the lay concept of free will is relatively free of metaphysics, I suspect it smuggles many metaphysical concepts through the door. In other words, the lay concept is not a concept that could survive sedulous scrutiny. There are more sophisticated variants around, and I suspect that they could. However, the debate about free will becomes empty at that point because almost everyone agrees that humans have the capacity to respond to reason (one criterion) or that consciousness does something (another oft-cited criterion); how much it does is very much debated). If one calls these attenuated capacities free will, I have no debate; I would, however, argue that the exercise is similar to arguing that god exists and then defining god as the universe. Who would argue otherwise? But what does the term add to our scientific/philosophical lexicon? 

    At bottom, my guess is that most of our free will fetishism arises from a desire to blame others for their wrongdoings. For some reason or another, this desire has been inextricably linked to the concept of responsibility. This connection, however, is utterly mistaken and free will qua metaphysical concept is utterly superfluous. We make very pragmatic distinctions between intentions and consequences and between behaviors that we could alter and those that we could not, and I don’t think the legal system or our moral system requires more. So, we would punish the person who intended to shoot another more than we would Dick Cheney. We would also punish the person who had a healthy brain more than one who had a diseased brain (diseased brain is impervious to reason?–at least somewhat). Free will is not necessary for these distinctions.

    What we should investigate, then, is not free will qua free will, but rather the feeling of free will. Why do we feel that we are the authors of our own lives? Is such a feeling advantageous? Is it inevitable? Et cetera. In the Baumeister lab, for example, researchers have studied the effects of belief in free will. The work is intriguing, but I take a different lesson from it than do most. If it is true that belief in free will leads to prosocial behavior, then that is a fault in our thinking. We need to inform people that free will is not a necessary for moral behavior or judgment. We should be practical about punishment and reward, not metaphysical. 


  3. Daedelus says:

    The signifiers ‘contemporary neuroscience’ and ‘free will’ appear odd in juxtaposition, like a !Kung Bushman seated next to a Roman Cardinal at a cafe table.  The scientist wonders if free will is the sort of hypothesis that is at all amenable to verification or refutation by experiment.  This is another way of asking whether or not the concept of free will is subject to an operational definition.  How would one approach the problem of determining whether or not a system under investigation is exhibiting the property of ‘free will’?  One might start with the idea of ‘unpredictability’.  One possible defiinition of ‘unpredictability’ would be the finding of behavior that is directly contrary to behaviors that are known to be invariably exhibited by members of the group under study; the finding of contrary behaviors in a ‘normal’ (not defective or diseased) member of the group would, under this operational definition, lend credence to the free will hypothesis.  For biological systems, we might say that self-preservation and propagation are known imperatives; the finding of willful suicide or sexual abstinence in otherwise healthy members of a species would suggest free will; under this definition, then, human beings would appear to possess free will. This is essentially Daniel Dennett’s position in “Freedom Evolves”.

    However, the capacity for self-destruction or remaining a virgin for a lifetime is not a very satisfying definition of free will.  We might point to extraordinary productions that seem to arrive ‘out of nowhere’, such as the Beethoven late string quartets, Shakespeare’s Hamlet or Bohr’s model of the atom.  Shakespeare’s productions are so extraordinary that some still dispute his authorship on the grounds of improbability.  Acts of inspired creation are unpredictable, lend credence to the notion of free will, but are not practical operational defiinitions for the poor scientist trying to prove or disprove a theory in the laboratory; she might be waiting a long time for another Cervantes to walk into her office.  This approach might still be feasible, when and if technology advances to the point that high-resolution functional mapping of the cerebral cortex becomes miniaturized and non-invasive to the point of portability, with remote monitoring.  Application of such technology to subjects with a highly developed sense of self-awareness, the Montaignes of the world, could at least tell us where the fount of creativity lies within the brain.  But it would not elucidate whether or not these productions were the result of a non-deterministic process or not.

    My own current view is that free will is the name we give to a chaotic system when that system happens to be a human being; I agree with the position expressed by Leonard Mlodinow and Stephen Hawking in their book “The Grand Design”.  The human brain is so complex that it would not be possible to construct a virtual state machine that could unerringly predict outcomes – given arbitrarily exact and extensive data about the system’s current state, and an exhaustive atlas of state transition rules – within a ‘reasonable’ period of time (say, before the end of the universe).  A virtual state model of the brain would  treat each dendritic receptor (about n*10^3/synaptic cleft) as a binary bit.  There are k*10^9 neurons/brain (the value of k has been reported between 80 and 200; for the purposes of this argument, these differences are trivial), and m*10^3 synaptic connections/neuron.  These figures were obtained from this reference: http://www.sciencedaily.com/releases/2010/11/101117121803.htm.  There are, therefore, S = 2^(k*n*m*10^15) possible states for one human brain and S-2 state transition rules in the atlas.  If every quark in our universe contained a universe, and if every quark in that universe contained another universe, and every quark represented a state or a transition rule, we would still not have enough information to construct our virtual state machine.  But the complexity doesn’t end there.   Learning is possible because some states lead to new connections being formed and old connections severed.  The virtual state machine transforms itself dynamically.  Furthermore, environmental factors such as metabolic state and ambient temperature, which are not predictable by the machine’s state at a given moment, have unpredictable effects. Finally, the state-transition rules themselves are non-deterministic because of stochastic processes such as diffusion of transmitter molecules across the synaptic cleft (it is possible for a chaotic system to contain stochastic elements without, itself, being stochastic). 

    The appearance of order out of a chaotic system is a wondrous phenomenon – wondrous, and unexpected.  Things needn’t have been so, but if they hadn’t been, we wouldn’t be here to think about how wondrous and unexpected our very existence is.  I abjure the project of trying to predict human behavior while at the same time admitting that it is theoretically possible, given a big enough virtual state machine (that is one possible definition of God, I suppose: a virtual state machine for the entire universe).  I follow Hans Vaihinger, in ‘Der Philosophie Als Ob’, and live my life morally, ethically, and socially ‘as-if’ I have free will.

  4. James Laird says:

    What do I think free will is? The ability for one human thought to interact with, affect, or influence another thought inside a physical brain, in a manner that’s not controlled solely by the four fundamental forces of physics (and isn’t random in nature either).

    Eddy, I believe contemporary neuroscience challenges the reality of free will, and it’s because they’re missing out on a fundamental concept.

    Neuroscience (NS) believes that all events are controlled solely by the four fundamental forces of physics in a predeterministic manner from the bottom-up. I’m thinking that the truth is more likely: “living forces” are an emergent property of all living systems, and those forces add into the net sum in real time, thereby helping to *determine* the path of reality.

    Due to a fundamental human reference issue, humans aren’t able to sense “living forces” exerted by other system levels. Humans only perceive of the *result* of the net sum of forces after the sum has already occurred for each moment of time. Therefore when scientists take apart a living system to better understand how it works (subdividing it repeatedly until they cannot divide it any further), they observe that all of the activity occurring in the subcomponents abides perfectly by the fundamental laws of physics. The fact remains however, that each of the system levels that they just took apart (i.e., each living entity that contributed toward higher-level life) is no longer alive and exerting forces from its previously existing level. Okay, in a nutshell that’s one concept that I honestly believe NS is missing, and it’s causing them to challenge the existence of FW.

    Here’s another fundamental concept that I believe is incorrect within contemporary neuroscience theory: NS effectively *assumes* that forces A1 and A2 located in field A are capable of summing together and *determining* how force B1 emerges in field B. That’s never been proven to be true, because forces located in different fields don’t add directly with one another. Here’s how that concept applies to free will: Electro-chemical (EC) forces located in your physical brain may cause thoughts to emerge, but those EC forces don’t determine the forces exerted *by* your thoughts. Therefore, the EC forces don’t determine the interaction of your thoughts. What I’m trying to say, is that NS believes that causality is one and the same as determinism, and I don’t believe that’s true. An event may be caused without being determined – that’s a fundamental concept that applies to all living things.

    I’ve created a website named tmsolf.org (The Missing Science of Living Forces) that presents carefully worded arguments supporting the existence of living forces. I believe the recognition of living forces is the next evolutionary step forward for mankind, and I welcome constructive criticism regarding the ideas on the website.

    Sincerely, James Laird

  5. bagua says:

    Contemporary neuroscience  supports the reality of free will. Everytime WE (physically,emotionally and spiritualy) are at some level of frequency .We do the things which are attached to that frquency at any particular time and place. The research in this field also started because we think of it. I was searching the net for studying about  life and death and during this ,i searched your site and entered it

    For past , Destiny was fixed . But for future ,we are all Architects and by our free will we can make ourselves what we want to make and do . The choice of making ourselves depends upon our past experiences and present awareness and wisdom. But only scientific basis can explain this to public. Our past experieces,awareness and emotions are stored in DNA’s and neurotransmitters plays an important roles in our decisions. By knowing all this ,we can think better and can make decisions by our will , which can be for our betterment and will help to get what we want

  6. kathleenvohs says:

    Eddy’s essay is a cogent, clear, and compelling view of free will and how neuroscience can support it. And also the hand-ringing that follows from some interpretations of neuroscientific findings. Neuroscience as a tool is not “the answer” to whether free will exists, nor should we expect it to be.

    I wanted to address the moral dimension in my comment: my research and others clearly shows that  when people believe that there is no such thing as free will, they behave in ways that (largely) are morally and practically detrimental to themselves and society. What boundaries there are to this, we do not know – but surely there are some. Nonetheless, the main effect stands, that free will beliefs support moral behavior and a disbelief in free will has some untoward societal costs. 

  7. Eddy Nahmias says:

    Thanks for these initial responses, Caesar, Daedelus, and James (I get the sense only one of you is using your real name!).  You’ve got the discussion off to a great start.  I won’t be able to respond to every issue raised but here goes:

    Caesar, I agree with you on two of your main three points.  The traditional free will debate may indeed be in large part a verbal (or “semantic”) dispute, with compatibilists and incompatibilists agreeing on the metaphysical facts about what sort of powers are possible in a deterministic world and simply disagreeing about which of those powers should be labeled “free will” (my collaborator Dylan Murray and I use this point to motivate experimental philosophy studies on how most people understand “free will” in a forthcoming paper, “Explaining Away Incompatibilist Intuitions”). 

    However, in that paper and elsewhere, I argue that most people understand free will, and the powers required to deserve praise and blame, in a way that does not commit them to the impossible abilities you describe—the ability to ‘pull oneself out of the swamp by one’s own hair’ or to create oneself ex nihilo.  So, on this point I disagree with you, and it is my main disagreement with Sam Harris and other willusionists who simply assume from the armchair that the ordinary definition of free will demands impossible powers.  You may find my critical review of Harris’ book Free Will useful:http://www.thephilosophersmagazine.com/TPM/article/view/15359/12081

    Finally, I agree with you that if one begins with a definition of free will as impossible self-creation, then neuroscience is irrelevant.  We don’t need science to help us learn that we can’t have impossible powers.  But as I say in the main blog post, some willusionists use neuroscience and psychology to challenge the ‘natural powers’ of free will by challenging the causal relevance of conscious and rational deliberation and self-control.  So, here we must look to the actual results to see if, once we pick out a relevant sense of free will, they challenge it. 

    And it is precisely this sense of free will that I think is useful in helping us consider the degrees to which people are morally responsible for their behavior.  After all, if we use “free will” to label powers no one could have, then it will be perfectly useless in helping us discern the degree to which people have free will or have the opportunity to exercise it in such a way that they should be held accountable for their actions.  There is indeed a difference between someone whose action is caused by her brain and someone whose action is caused by her brain tumor (one that knocks out her capacities for rational self-control).

    Daedelus, thank you for reminding us to consider the connection between free will and unpredictability.  I’m not sure what that connection should be.  I think most people don’t like the idea that their actions could be predicted by other people (even in principle, much less in practice)—though I have some recent experiments that suggest people might not be that averse to this possibility, at least as long as the prediction is not a step towards control by other people.  But presumably we also don’t like the idea that we are so unpredictable that we have no idea what we ourselves will do.  Randomness is not good for the sort of free will tied to self-control.

    But in any case, chaos theory (or nonlinear dynamics) suggests that even if determinism is true, complex systems—and the brain is more complex than anything—may be unpredictable in practice and perhaps even in principle.  I also think that in at least some cases, one’s first-person access to one’s own conscious mental states provides information about one’s mental states that could not be gained (in real time) by an outside observer even with complete information about the brain states that are the basis of those very mental states.  But I won’t try to defend this claim here!

    James, that last comment seems to jibe with your ideas.  And I also agree that the free will debate should focus more on the question of mental causation (e.g., whether and how conscious mental states can play a causal role in behavior) than on its obsession with determinism (which is neither necessary nor sufficient for epiphenomenalism about conscious mental states).  I tend to think that whatever sort of explanatory or causal non-reductionism applies to most high-level scientific entities, such as genes or neurons or tectonic plates, will be sufficient to get us what we need or want when it comes to (conscious) mental states.  So, we may disagree here about what sorts of “emergence” or “downward causation” will be required for free will.  But I may be wrong.

  8. ntadepalli says:

    Every individual is burdened by the unconscious , namely his desires,dispositions,beliefs,opinions,fears and ambitions.The unconscious usually plays a very weighty role in our decision-making process.

    Even in this scenario my argument is that the individual is free enough.

    On occasions , we recognise brain`s abilities to self-develop feedback functional circuits to regulate its earlier decisions to improve efficiency and in cases of extreme necessity even to adjust the unconscious to achieve purposes.

    These abilities ,even though they operate deterministically and automatically , reveal a freedom.

    Obviously we are not talking about any particular singled out decision-making,but how the individual conducts his life freely.

    All brains` activities referred are in auto-mode and there is no role for interventions from consciousness.

    The above may give a picture of freewill at variance with the generally accepted one.

  9. Roy Baumeister says:

    One of the biggest mistakes people make today is to assume that the brain is the sole or the original cause of behavior. There is a common assumption that once psychology understands the brain, it can dispense with other forms of explanation. I think this is a huge mistake that may not become apparent for a couple decades.

                I would make the analogy to assuming that we can understand human communication by analyzing the electronic technology inside a telephone. No matter how well you understand the technology, you won’t understand the conversation unless you also understand what the speakers are saying, which means knowing the language they use and knowing the context in which their remarks are made. The electronic technology works the same regardless of whether the person is lying or telling the truth, reasoning brilliantly or speaking a series of non sequiturs, and so forth. In the same way, understanding the mechanics of brain function will tell us only a little about human behavior, because the brain is just processing events in the social environment and helping the person understand them and react to them.

                I think neuroscience research is generally irrelevant to free will. Some neuroscientists think it is relevant, but as Nahmias says, they are mostly talking about obsolete notions of free will in terms of non-physical souls causing behavior. What ordinary people mean by free will, namely making choices with conscious reflection and without external coersion, is quite real. Of course the brain enables that to happen, but a brain alone couldn’t do it. The human brain is designed to learn the information and systems of the social group, starting with language. It learns to reason, mentally imagine and evaluate possible alternatives, and alter the body’s actions so as to bring about the chosen outcome.

                Moral action, for example, means tempering one’s desires and impulses with an understanding of abstract ideas of fairness, respect for rules, justice, and unselfishness. Those are not stamped into the brain — rather, the brain learns them from the social environment, including local variations in the rules. Hence although the brain is one contributing cause to moral action, it is only part of a wider causal network.

    • James Laird says:

      Roy, I enjoyed reading your comments about learning.

      Here’s an idea I’ve been thinking about regarding human learning: When a person is learning something new, the neural wiring of their brain is physically changing on the fly. In order for their neural wiring to change, forces must be exerted inside their physical brain that causes those changes. In addition, there must be *intelligence* associated with those forces; otherwise the changes to the neural wiring would be random in nature and the person wouldn’t learn anything. So here’s my big question: Where do the intelligent forces come from that change the neural wiring? When neuroscience (NS) looks at the learning process, they claim that everything is being controlled by the four fundamental forces of physics (4FFOP); but by doing that, NS advocates that human intelligence is somehow innate to the 4FFOP (e.g., gravity). Does that really make sense? What I’m trying to say, is that the forces that cause your neural wiring to change when you’re learning something new, *must* be exerted by your thoughts. If we believe that new forces (i.e., living forces) emerge at the thought level, and we believe those forces are what cause our neural wiring to change while we’re learning, then we can also believe that the processes in our brains aren’t controlled solely by the 4FFOP. If the processes in our physical brains aren’t controlled solely by the 4FFOP, then we can believe we have free will.

      The argument that eventually *proves* that humans have FW, is going to be based upon the fact that human intelligence isn’t innate to the 4FFOP.


  10. Will Truth says:

    I disagree that the “willusionist” definition of “free will” is dubious. Belief in free will really does require belief in some sort of supernatural power. The question is “what determines my so called voluntary actions?”. Is it the laws of physics/random quantum fluctuations? Or is it something else (i.e. “me”) ? It can’t be both.  You have to choose between: (i) the scientific materialist explanation that it is the laws of physics/random quantum fluctuations that are responsible; or (ii) the supernatural explanation that there is something outside the laws of physics/random quantum fluctuations that is at least partially responsible for our actions.

    I believe all our actions are technical “involuntary” from a strict scientific viewpoint, but that doesn’t mean that we shouldn’t be held responsible for all of them. The illusion of free will is useful because it is rational to reward or punish actions that we customarily regard as “voluntary”. We all know that there are punishments for deliberate murder, so that acts as an “input” to our brains and hopefully makes it less likely that we will murder someone. By the same reasoning, it is irrational to punish actions that we customarily regard as “involuntary” (e.g. striking someone while having an epileptic fit) because such punishments will have zero influence on whether people perform such actions or not.

  11. SilverAce says:

    I think what most people mean by free will, quite simply, is: at any particular moment, could someone have chosen to do a different action than they in fact did? 

    It seems like the neuroscientific/materialist view of free will gives a fairly unambiguous no to that.

  12. lingeringmind says:

    The author is clearly unable to give up the “big daddy” of the ancient & archaic idea of free will. He says that neuroscience must explain free will, not explain it away. Well, Kopernicus was able to expain away the idea of a geo-centric solar system. Frankly, I think its a good thing that mermaids, centaurs and unicorns can easily be explained away. Now its free will’s turn.

  13. Eddy Nahmias says:

    Thanks to Kathleen, Roy, Ntadepalli, and others for continuing the discussion.  Rather than address their specific points, I will try to tie them together.

    I agree that, in many ways, neuroscience studies the wrong level to tell us about the capacities required for free will, such as rational deliberation and conscious self-control (an excellent libertarian theorist on free will, Robert Kane, looks to neuroscience for evidence that quantum indetermism influences some neural interactions.)  Psychology studies the right level.  But of course, I’m optimistic that we will eventually be able to unify neuroscience and psychology on the model of other inter-level scientific endeavors, such as organic chemistry, which do not cease to refer to higher-level entities and their causal powers.  Genes and neurons exist and play an integral role in causal explanations, even though they are composed of molecules (which are composed of quarks, etc.).  Thoughts, consciousness, and language exist and play an integral role in causal explanations, even though they are composed of neurons (which are composed of molecules, etc.).  This is overly simplistic, but it does not seem to require a type of non-reductionism or downward causation that goes beyond what we see in other areas of science.  On the other hand, the brain and mind are the most complex things in the universe, so the “shape” of the theories may look quite different from anything we’ve seen before.

    Because neuroscience is, in general, not the right level to look at to understand the capacities required for free will, it also is not the right place to look for threats to free will (unless it’s just being used to reprise the standard threats of determinism or lack of self-creation, or unless it is giving us information about the mechanisms underlying the capacities for rational self-control, etc., and their potential limits or breakdowns). 

    Instead, psychology (and cognitive science) is more likely to suggest threats to free will.  As I briefly mentioned at the end of the post, psychology may inform us that we possess the capacities required for free will to a lesser degree than we tend to think or that our opportunities to exercise these capacities are limited more than we think.  We are often influenced by factors we are unaware of (that in itself may be OK), but also that we would not want to influence us were we to know about them (and that’s not OK–it suggests we are not able to control our decisions and actions in accord with reasons we endorse).

    Just one example: implicit bias.  There is a growing literature showing that even people who repudiate sexism and racism are susceptible to biases against women and minorities that influence their decision-making and behavior (indeed, even members of these groups are susceptible to such biases).  If I am on a hiring committee and explicitly aim to be unbiased in my assessments of candidates, then it seems my rational control is limited to the extent my assessments are influenced by implicit biases in ways I do not know about and hence cannot directly control (and to the extent that I was not responsible for developing these biases and that I cannot counteract their influence–e.g., before I even know they exist–I also seem not to be blameworthy for their effects on my decisions).  Now, once I know about the potential influence of these biases, I may also be able to learn methods to counteract them and gain control and responsibility.  Knowledge is power.  (But it won’t be easy.  Free will is an accomplishment.)

    • James Laird says:

      Eddy, I believe we can all agree that free will in the “weak sense” exists (i.e., given a choice, I can make a decision). However, in order for mankind to prove that free will in the “strong sense” exists (i.e., I could have taken a different path in life – I could have done otherwise), mankind *must* prove that the path of reality isn’t controlled solely by the four fundamental forces of physics. Until we develop that proof, the theory of predeterminism will prevail, and science will continue to acknowledge free will only in the weak sense.

      In order to prove that free will exists in the strong sense, we need to prove that new forces are an emergent property of living things.

      If it’s reasonable to believe the premise “the source of human intelligence isn’t directly from the four fundamental forces of physics”, then I believe a proof of FW in the strong sense is at hand. 

      Respectfully, James Laird

  14. trehub says:

    Those who believe that neuroscience challenges free will must have an implicit belief that our self is somehow separate from the natural workings of the brain. If one accepts the fact that the self is a real biological part of the brain, our decisions must necessarily be of our own free will. For more about the self, see  Where Am I? Redux, here: 


  15. twomeyw2 says:

    Defining free will as requiring a God, new forces, etc., seems absurd.  You are essentially defining it purely to win an argument in a way that does not align with the typical usage of the words.

    “The kidnapper held the victim against her will”  Oh wait a second, there is no free will, therefore the kidnapper really did nothing wrong, or we have to invent new language to appease those who adamantly cling to the belief that free will doesn’t exist based on their odd definition.

    If you really believe there is no free will, why bother trying to align inputs (e.g. punish kidnappers) with good / bad behavior at all?  Who are you (or anyone else) to claim that any behavior is good / bad.  Based on your reductionist view, you can’t claim that the subatomic reactions happening in your brain (and likeminded individuals) are superior to those happening in another person’s brain.  Physical phenomena is physical phenomena, there is no good or bad…. Right!?

    I agree with the author that a higher level consciousness with the ability to exert free will can exist on lower level physical phenomena, but whether my thoughts can be reduced to the interaction of subatomic particles (or anything else) is irrelevant.  When I make a free choice (without someone holding a gun to my head) I’m exercising free will.  I might have some biases influncing my decisions, but I’m certainly not a total slave to them.

    What if tomorrow we discovered that a God exists, or that there are “life forces”, etc., would that have any impact on the prior choices we made?  Would everyone have to agree that we were free all along?  Why not just stick with the common understanding of the words and recognize that (regardless of the ‘how’ of our brains), it is clear that they have the capability to rationally process information and take action on that information – we have free will.  A definition of free will that hinges on the mechanics of the brain is missing the point in my opinion.

  16. bugsey malone says:

    It is curious that the title of Mr Nahmias’s essay refers to the “reality” of free will, given that this treacly medieval concept continues to attract interminable philosophical dispute. The occasional problem neuroscience creates for itself lies in its tendency to somewhat hubristically overreach the boundaries of its own field, one that is otherwise perfectly positioned to test philosophical and (neuro)psychological hypotheses relating human motivation and action. The colourful Nietzsche quote above comes very close to the scientific psychological position; that we we are driven organisms, and that “free will”, in all its conceptual vagueness and metaphysical origins, refers to a psychological capacity to inhibit a range of primary and secondary biological drives and psychological impulses –  to breathe, eat, drink, have sex, murder, cheat, lie, etc – “the ability to pull ourselves out of the swamp by the hair”!

    Theoretical psychology and neuropsychology/neuroscience are finally able, after years of observation and research (with the aid of PET and other non-invasive brain investigation technologies), to collaborate with neuroscience in testing longstanding hypotheses with respect to localised brain functions associated with behaviour inhibition (such as the frontal lobes); that is to say, to investigate whether these specific functions are genetically or developmentally determined, and their degree of interaction. The conceptual and theoretical frameworks for these  neuroanatomical investigations,, however, are reliant upon a rigorous ab initio requirement for dispensing with the notion of ‘free will’ (to which we seem to have both a metaphysical and sentimental attachment) as being no more than a crude linguistic signifier for a capacity to sense or feel an urge or impulse, to inhibit it sufficiently long enough to think about it (i.e. become conscious of it, activating other localised neural processes)  before acting, or “choosing” not to act.  This parsimonious formulation accommodates our sentimental notions of freedom and fully embraces theories of morality.

    Theoretical psychology has been positing formulations of motivation and behaviour for over a hundred years by means of various testable models (the ‘superego’ in psychoanalysis,  ‘cognitive restructuring’ in behaviourism, ‘frontal lobes’ in neuroanatomy and neuropsycholog, etc). What they have in common is the acknowledgment that one cannot appeal to metaphysical concepts in the doing of science. It seems to me that the problem for neuroscience is that, rather like indulging a scruffy dog that should be kept outside at all times,  it continues to permit a distracting metaphysical concept to remain in the house while simultaneously ignoring one hundred years of theory and research in psychology/neuropsychology. Only by dragging the concept of ‘free will’ out of the lab by the scruff of the neck ONCE AND FOR ALL will there be enough room for other scientific disciplines dedicated to similar purposes to collaborate with neuroscience in determining and understanding what  makes us go and what makes us stop.

  17. roy turner says:

    do those who design and intepret the studies that show free will is non-existent  possess free will? Do they choose to do these studies?  If not, then there is no issue of the results being truthful. They are just what the neuroscientists are determined to say, and we can ignore them.

    • James Laird says:

      I think Roy Turner’s comment on 8/17/12 is right on:

      “do those who design and interpret the studies that show free will is non-existent  possess free will? Do they choose to do these studies?  If not, then there is no issue of the results being truthful. They are just what the neuroscientists are determined to say, and we can ignore them.”

      Roy’s comment illustrates that human intelligence must exert new emergent forces.


  18. Musing Choosing says:

    Thank you Caesar for the insights. Thank you SilverAce for the clear simplicity; I needed that.

    vagueness – Sometimes it is better to discard a vague word and impose a discipline of using more precise synonyms for the word’s various meanings. I get along just fine without the term “free will”—using “choice” and “cause” instead. Arguments vanish in the clarity.

    non-humans – Whatever terms you use, always test your definitions to see if they generalize beyond humans to other animals and machines.

    relativity – Beauty is relative to the observer. Randomness is relative to the observer. (Are the digits of π random? Depends on the observer’s familiarity with π.) Likewise, freedom and responsibility (a.k.a. “free will”) are relative to the observer. A prisoner in an unlocked cell who believes the cell to be locked is unfree from the prisoner’s point of view, free from the warden’s point of view. When parent A badly raises child B who grows up and secretly drugs unstable person C who stabs somebody with knife D, responsibility for the wound could be placed on A, B, C, or D by four different judges. The responsibility is relative to the observer.

  19. Aaron says:

    I know this is only a side note in the essay (which I thought was great) but it seems we assume a Greek like  notion when we conceive of the soul (I realise this is a big generalisation). While I know the idea of a sould is highly contested (and that’s fine) the way it seems to be conceived is as a detached, other wordly thing that is loosely and briefly associated with the material. From my limited knowledge of history this is a Greek idea and quie different to the Jewish view which envisaged a much more integrated reality when discussing the soul. The soul or ‘true person’ seems intimitely associated with the physical (hence a physical ressurection) and not so much a ‘ghost in the machine’ that seems to vanish before the search light. I’d be interested in any thoughts (or corrections.) Thanks. 

  20. Benson says:

    The common understanding of the concept of free will is that, faced with a certain situation, the protagonist could freely select which path to follow. It is understood that in this situation there could have been several choices and free will means that the selection was unconstrained by external considerations. How then could this assertion be tested? It would be necessary to arrive at the same decision point with the same external circumstances and demonstrate that more than one choice could be made. But this is impossible. One could not reproduce the circumstances because the second time would always have to include the fact that a particular choice had been made the first time. There is no way to actually demonstrate the existence or absence of free will. That is, though it takes the form of a question, it is not actually meaningful. It is like the question: What would the color of an elephant’s eggs be if an elephant laid eggs?

    Free will is often championed although there is no way to know exactly what is meant in each case since we do not have direct access to each other’s feelings. The fact that such feelings are reported to be widespread should tell us something about the nature of human beings but not about the concept of free will itself.

  21. seligman says:

    Here is an excerpt from a forthcoming paper (Navigating into the Future or Driven by the Past:Prospection as an Organizing Principle of Mind. Martin Seligman, Peter Railton, Roy Baumeister, and Chandra Sripada. Perspective in press)

    Freedom of the Will

    Philosophers have traditionally posed the question of free will in highly abstract metaphysical terms: If a universe obeys deterministic laws, can agents in this universe be free? Compatibilists argue that if one is careful not to confuse causation with constraint, then there is no opposition between free will and determinism, while incompatibilists deny this claim. By all accounts, this metaphysical debate has reached a stalemate. There are a variety of compelling arguments on either side, but nothing decisively tips the balance in one or the other way.

    We are not going to entering the metaphysical fray about determinism in this paper. Rather we see a need to develop additional perspectives on free will that might move us past the deadlock. One approach that has become increasingly popular within philosophical and psychological circles locates a more modest, distinctively anti-metaphysical, point of entry into the free will debate (Baumeister et al., 2011; Mele, 2001). The basic idea behind this approach is, very roughly, to put questions of psychology ahead of questions of metaphysics, and here is how it works. Start with the question of what component psychological mechanisms or capacities a creature needs to have in order to be free and autonomous. Build a catalog that encompasses the full assortment of ‘design features’ that make an agent free, using psychological and neuroscientific data from humans and other animals as critical guides to theory construction. Later, after the catalog is complete and the functional specifications of the items on the ‘free will inventory’ are fully fleshed out, then, and only then, are abstract metaphysical questions broached. In particular, questions can then be posed in a more structured and informed fashion about what kinds of physical laws, properties, or substances must exist in the universe (e.g., irreducible chanciness, certain forms of agent causation) in order for the capacities on the list to be realized. We call this the capacities-first approach to free will and we believe that prospection science provides a theoretical framework for organizing this kind of inquiry.

    What kinds of psychological processes appear to be implicated, when we take ourselves to be acting freely? Stillman et al. (2011) sought to discover how ordinary people understood free will. They asked participants to narrate an event from their lives in which they acted of their own free will or, in another condition, not of their own free will. Actions reflecting free will were more likely than the non-free-will actions to emphasize pursuit of long-term future goals. Free actions were more likely than the non-free ones to be about conscious deliberation and reflection. The free actions were also more likely to be consistent with the person’s moral values, and the free actions were also more likely than unfree ones to bring about positive outcomes. Taken together, these findings show that everyday understandings of free will are about the long-term, beneficial outcomes, aided by conscious reflection and principled commitments.


    These features point to the centrality to the experience of freedom of guidance via prospection. Acting freely involves the absence of constraint, but it is also fundamentally a matter of generating and evaluating multiple possible future courses of action, and electing an act in light of them. It follows, perhaps, that enhancing freedom will involve enhancing the power of generating and evaluating options.  Three distinctive design features of human prospection expand the complexity, time horizon, and accuracy of prospection.

    Complexity. Plans consist of sequences of actions linked in a coordinated way to achieve a goal. Plans are usually decomposable into parts, each of which achieves some proximal subgoal (Miller et al., 1960). In order to get to Boston, I need to fill the tire with air. Once that is done, I can drive to the gas station. With the tires filled and the gas tank full, I can drive all the way to Boston. Often parts of plans are themselves decomposable into further parts.

    Given their decomposable structure, efficient construction of plans requires a distinctive kind of prospective ability. It is not enough to simply prospect the outcomes of single actions. Rather, one must be able to perform sequentially linked prospections and the attendant ability to rearrange and build more flexible plans, not only expands the size of option sets, it makes available a vast array of new options that are dramatically more likely to achieve one’s goals.

    Time horizon. Some actions unfold over a short-duration: the rat in a T-maze selects going either right or left and receives a reward in a matter of moments. Humans, unlike other animals, can project years ahead and adjust current behavior accordingly. This contrast between humans and rats is set out not to deride the achievements of animal minds. When at a fork in a maze, expecting that danger lurks down one path and not another is an impressive feat. How much more impressive is it then that when at a fork in life, let’s say choosing a major in college, we have a the ability to prospect not just the immediate consequences of each option, but we can mentally ‘see’ in rich and vivid detail the various ways one’s life might unfold.

    Accuracy about the future. Improvements in accuracy about the future include being able to imagine oneself in different subjective states that one will experience in the future. Human prospection includes representations of states and motivations which differ from one’s current condition. If the anticipated motivation of its future self is one that the individual does not want to have, he can instead formulate options which prevent these anticipated future desires from arising. In one of the most famous works on free will in the modern philosophical literature, Harry Frankfurt (1988) argued that freedom of the will consists in having the will that one wants to have. According to Frankfurt, both humans and animals have the ability to have ‘first-order desires’, i.e., desires directed at doing this or that action. But humans also have the ability to step back and form second-order desires, desires about which first-orders desires one wishes to be effective.

    So specific design features of prospection expand freedom in at least three ways. Sequential prospection enables complex, flexible planning, expanding the number and quality of options. Prospection with a long time horizon enables options that unfold not just over days, but rather over years and decades. And prospection with meta-representation enables better accuracy about the future.


    We now turn from “freedom” to “willing” in order to explore the mechanisms by which a particular option is selected. There is one sense of “will’ that comes into play when we engage in the spontaneous or deliberate prospection of future possibilities. This feels “free” because the mind freely explores possibilities, and it feels like “freely willing” because what precipitates our action is the making up of our mind among these alternatives. Nothing more, no additional act of will, is required to act “as I see fit”.

    So The experience of “freely willing” is running through these prospections until one feels that one’s mind is made up, and then taking the course of action one has settled upon, and nothing more.

    The “settled outcome” is in an obvious sense one’s own idea, since it came about through one’s own unimpeded mental activity, without internal compulsion (which is insensitive to what one prefers) or external coercion (which prevents one from weighing  options without interference) or overpowering temptation (in which case the agent does not have the will he wants). No transcendental will is needed for the act to be “of one’s own accord”, no rational homunculus must “freely endorse” it–for when the agent settles his mind after freely exploring options by following what “seems best” , then if the agent also wants to have this be the ground of his choice, that is the agent freely endorsing it in every relevant sense, and performing the act because of his endorsement. Indeed, were we to describe the activity of such a homunculus, it would be no different from this.

    This description fits the phenomenology of freely willing. Had the agent thought more highly of another possibility, or had he been attracted to it whimsically and wanted his whimsy to guide him, he would have settled upon that one. This is what “I could have done otherwise if I’d wanted to” means. This, moreover, is a notion of free agency worth wanting—since it enables us to pursue what we want. And one need only reflect upon what one’s life would be like were one beset by compulsion, serious addiction, or coercion to see what a difference free agency in our sense makes.

    Our formulation is far from a complete account of process of willing, and we are especially aware of two gaps in our account. The first is “how is one’s mind made up” in leading to a “settled” evaluation among the options. In the section above on consciousness and commensuration, we postulated that a complex evaluative process among alternative options must exist, but we made few claims about the details of this process. In particular, we have assumed nothing about the role indeterminism or stochasticity might play (Glimcher, 2011; Grabenhorst & Rolls, 2011; Neuringer, 2002). But once prospection is put at the core of free will, this now becomes a tractable, empirical issue amenable to the traditional methods used in the fields such as judgment and decision-making.

    The second gap is the role that “controlled” versus “automatic” processes play in prospecting. Under some conditions, external circumstances frame the question which is prospected. “If you had ten million dollars to spend over the next year, what would you do?” sets off a panoply of automatic prospections without need for any voluntary action (“a clinic in North Philadelphia, no, in South Sudan.”). Under some circumstances, an effortful, controlled process frames the question which is prospected. “I have two deadlines this week, how should I meet them?” Under some circumstances, an internal state such as thirst cues the process. It is clear that a mixture of automatic and controlled processes are involved in the initiation, maintenance, and conclusion of the prospecting process, and the best we can say is that this now becomes a tractable and empirical issue.

    The central point of our analysis of willing is, however, that there need be posited no such thing as a “will.” It is worth remembering that the modern notion of a “will” as a thing is in fact a modern reification. Aristotle’s term boulesis, used for the desire which combines with an idea of an act to yield rational action, comes from the notion of “taking counsel” or “thinking over”, a form of prospection, and for Aristotle, decision (prohairesis) is “deliberative desire”, that is, a desire to do something here and now activated through deliberative assessment of available acts, not an inner act of willing (330 BC/1999, pp. 1112b26, 1139a21-b5). In ancient Greek thought and law, the equivalent of “willing”, ἑκών, meant being favorably disposed to seeing one’s idea of an act brought into being, and murder, φόνος, was a matter of foreseeing the death of another, wishing for it, and doing it as a result. In Old English and in many contemporary Germanic languages, “I will it” is synonymous with “I like it” or “I want it”, and in many Latinate languages, “willfulness”, (e.g., volonté in French) has the same root as “desire”, (vouloir).  “Free will” for the French is libre arbitre, roughly, “free weighing and judging”, with no reference to a special volitional faculty.

    Ultimately, our capabilities approach to the free will problem will face the question of what metaphysical assumptions, if any, are needed to fulfill the “job description” of free agency. It could turn out that our universe does not meet these requirements. But as far as we can see, the phenomenology of free agency can be understood, the psychological processes that underlie it described, the exceptions to it diagnosed, and the considerations that make it valuable to us illuminated, without challenging anything in current physics.

  22. Eddy Nahmias says:

    Thanks for all the recent comments.  Now we’re seeing some good resistance to my compatibilist and naturalist definition of free will and my claim that most ordinary people are not committed to incompatibilist (or dualist) conceptions of free will.  Though trehub and twonetw2 support my arguments, Will Truth, SilverAce, lingeringmind, and James are unconvinced about some parts of it.  Others (perhaps Bugsey, Benson, and Musing) wonder if these questions can be answered or if we should avoid getting too “metaphysical.” 

    There is a reason the debate has gone on thousands of years and continues (e.g., here on this blog!).  We have intuitions that support both sides of the debate.  So, one of my goals is to try to diagnose our intuitions so that we might better understand their source and whether some of them might lead us astray.  Let me try to diagnose some of the intuitions (comments) we’re seeing here:

    Try to give up dualism without losing your mind.  In my view, and supported by some of my experimental philosophy research, the main reason people find determinism (or, what is different, mechanism) threatening to free will is because they think that it means that our minds don’t matter—our mental states get bypassed by causes from the distant past (“it’s predetermined” or “fated” regardless of what I want or try), or by the neural processes in our heads (“my brain made me do it”).  As trehub suggests, assuming there is no non-physical soul or mind, then mental states are (or are functionally based in) neural states.  But that means we don’t lose our minds (or selves) when we give up souls.  This doesn’t mean, contra trehub, that “our decisions must necessarily be of our own free will,” because there are still cases where our decisions are limited by influences we do not know about or cannot counter (see my prior comment). As Aaron suggests, not all theories of the soul (or mind) separate it from the body (indeed, the Biblical conception is not so dualistic).  Once we understand how our embodied (“embrained”) minds work, including how consciousness and rationality work, we will be able to give up outdated theories that demand more than our experiences or intuitions do.

    But even if we think a material mind matters, we might still worry about whether we can choose otherwise (Benson and SilverAce define free will in these terms).  Are you sure you  want to be able to choose otherwise holding fixed everything about yourself, including your beliefs, desires, goals, reasons, etc. at the moment of choice?  I don’t.  I want to know what I should do by the time I make a choice.  Before I deliberate I don’t know what to do (that’s why I deliberate).  And at that point I think both alternatives are open to me (and they are).  But by the time I have to act, I want to know which choice makes more sense for me.  And if I succeed, then I do not want it to be possible that I choose the option that makes less sense.  I want the freedom to deliberate among the open alternatives I envision so as to reach a good choice and then to control my actions in light of it.  I don’t want the freedom to do otherwise than what I decide is best nor the freedom to act randomly because I can’t reach a good decision. 

    I appreciate the comments, and I’m sorry if I can’t address all of them in detail.

  23. SilverAce says:

    “Before I deliberate I don’t know what to do (that’s why I deliberate).  And at that point I think both alternatives are open to me (and they are).  But by the time I have to act, I want to know which choice makes more sense for me.  And if I succeed, then I do not want it to be possible that I choose the option that makes less sense.”

    Eddy — if you are not free to act differently, are you really free to deliberate differently? I don’t see how you are. Every thought, like every action, would also be a neural state determined by the previous state of the system (maybe with a little quantum indeterminacy tossed in), and so on ad infinitum. Never at any time is there any possibility that you could have thought differently. After all, each new thought is simply a decision to think that way–to think in the way that “makes more sense,” and as you say, you don’t want the option to be able to think in a way that makes less sense.

    If that’s the case, we couldn’t really have thought differently ever. At each point our thought is determined. It’s true that conscious thoughts and rational planning effect our decisions — but WE aren’t really in control of the content of those rational thoughts/planning because we could not have chosen for them to play out differently.

    How can this possibly be free will?

    • James Laird says:


      The way I interpreted Eddy’s comment, is that upon reaching a conclusion, he simply wants to follow through on that path.

      During deliberation, I believe Eddy was free to act differently. In other words, his thoughts exerted forces during deliberation, and those emergent forces were part of what determined his conclusion. Eddy could have deliberated differently and therefore come to a different conclusion, because his thoughts have “life” to them – they’re emergent and undetermined as is all life, and therein lies the essence of free will. That’s the “magic” that humans are seeking – life is emergent – it’s caused but not determined. (It’s not really magic; it’s simply reality, but it seems like magic is required because science has failed to recognize that the path of reality isn’t controlled solely by the four fundamental forces of physics.)

      I agree that each complex neural state in Eddy’s brain is associated with a thought (i.e., each wave of neural activity causes a thought), but the forces exerted from the neuron level don’t determine the forces exerted by his thoughts, because the forces exerted by his emergent thoughts are in a different field than the electro-chemical forces exerted by his neurons, and forces in different fields don’t add directly with one another.

      James Laird

  24. tedrey says:

    In questioning whether I have free will, it seems to me massively important to decide what one means by “I”.  If one recognizes a one-to-one correspondence between the objective material self, and the subjective mental self, then the I that is determined is exactly the same I who freely decides, the same I that one has always considered to be oneself.  So where’s the room for a problem?

    Curious that what seems so crystal clear to me will probably read like Jaberwocky to others.

  25. twomeyw2 says:

    Every thought, like every action, would also be a neural state determined by the previous state of the system (maybe with a little quantum indeterminacy tossed in)

    That ‘maybe’ pretty much invalidates this type of reasoning.  Isn’t a very popular interpretation of reality that in fact every possibility occurs (both at the quantum and macro level), though our particular existence only experiences one of those possibilities?  Therefore, if it is possible for you to choose red or white wine, you choose both – but only experience one particular reality.

    While certain issues related to free will may be debatable, determinism is not one of them.  I can prove 100% that our thoughts are not guaranteed to be predetermined by basing a decision (say to drink red or white wine) on the outcome of a quantum event.

    Even so, you might claim that we have no control over which reality (assuming an infinitely splitting universe) or probability (assuming some other driver) we exist in or experience.  In my above example, you could claim I was a slave to the quantum event (despite my choice to participate in the experiment). e.g.

    WE aren’t really in control of the content of those rational thoughts/planning because we could not have chosen for them to play out differently.

    But isn’t this a circular argument?  If I make a rational decision, why would I WANT it to play out differently?  If I wanted to think differently or make a different choice about something, I would have.

    This is the question you need to answer (at least to persuade me), because as I see it, you are claiming that to be free, I must be capable of producing undesirable outcomes (which seems closer to the definition of not being in control and not having free will).  Clearly, if free will requires there to be an absence of free will, free will cannot exist, period.

    — We say cruise control controls the speed of a car.  We don’t say that cruise control is not in control, the laws of physics are, or that for cruise control to be in control, the car should wildly change speeds in an undesirable manner.

    As I see it, I am both in control (at least to a certain degree) and unpredictable.

    • SilverAce says:

      Even so, you might claim that we have no control over which reality (assuming an infinitely splitting universe) or probability (assuming some other driver) we exist in or experience.


      If I make a rational decision, why would I WANT it to play out differently?

      I could give you many answers to this question. Here’s one: because in much of life, there is far more than just one right choice. There is a choice of values involved, not just a purely logical decision.

      Should you kill 1 person to save 2, for example? You could reason it out either way. And, upon deliberation, one choice will feel “righter.” You will then act on that righter-feeling choice. Does that mean that it actually had to be so, that no other choice could have felt rightest? That the puzzle could not have been seen in another way? That the path of the deliberations, its emotional content, the metaphors that were brought to mind, could not all have been different? Of course they could have–except that, according to the biological view, they couldn’t have.

      BTW, on a different note, to any moderators: ay I suggest that the commenting system be upgraded to something modern? It’s really unhelpful not to be able to even be able to preview your comments before posting them, for instance, or to have weird “Path: p.p.1 >> p.p1” statements at the bottom of the text-editing box.

  26. Eddy Nahmias says:

    SliverAce sggests that determinism (or perhaps physicalism?) means that ” Never at any time is there any possibility that you could have thought differently.”  I think this is the wrong way to think about what is possible even in deterministic universes.  The quoted statement sounds more like fatalism, the view that certain things (e.g., thoughts) will happen no matter what.  If a coin lands tails in a deterministic universe, it still could have landed heads.  It was not fated to land tails.  It would have landed heads had something been slightly different, and things could have been slightly different (or consider nearby possible worlds where something was slightly different such that the coin landed heads).  Similarly, our deliberations might have gone differently had something been slightly different.

    This may not satisfy you because it remains true that you cannot control whether things went differently such that things would have gone differently.  Nor could you control it if determinism were false (or if you had a soul).  But such lack of impossible control does not mean that nothing controls anything.

    • trehub says:

      Eddy wrote: “This doesn’t mean, contra trehub, that “our decisions must necessarily be of our own free will,” because there are still cases where our decisions are limited by influences we do not know about or cannot counter

      It seems to me that almost all of our decisions are influenced by the state of our pre-conscious/non-conscious brain and by constraints imposed by the world we live in. Who among us is free of bias? But our biases are an integral part of us, however they are formed. The bottom line is that we make our own personal decisions on the basis of who we are — rational and non-rational — in the context of the worldly constraints that we confront. 

  27. twomeyw2 says:

    I could give you many answers to this question. Here’s one: because in much of life, there is far more than just one right choice. There is a choice of values involved, not just a purely logical decision.

    But if my choice fulfills my values, I am exercising my free will.

    What I understand you as saying is that as my emotional drives are biologically ‘set’, my choices are in reality, enslaved to my emotional drives.

    But we are both emotional and rational beings. We have the capability (control) to modify and adopt our behavior and way of thinking by understanding and changing how we react to our emotional drives and even systematically changing the drives themselves.

    You may claim that my desire for self improvement may in fact be driven by my emotional drives. But so what?

    Without any drives, I would not be conscious. My brain would be like an idle Pentium III processor. I am not I without my drives. With drives, even drives that promote self modification, you’ll claim that I am a slave to those drives. So isn’t your argument not just that humans do not have free will, but that free will is impossible? If not, can you describe how free could exist, or what would be necessary?

    • Musing Choosing says:

      Nice work, twomeyw2. I like your proof above that choices can be unpredetermined by basing them on quantum events. Never thought of that.

      Second, the last sentence in your comment above neatly captures and forces the semantics issue. “…can you describe how free will could exist, or what would be necessary?” Before we can reason about free will we must agree on a meaning for the term. A test of whether we agree on a meaning is whether we can agree on a partition of a set of events into two classes: those events that are examples of free will and those that are examples of not free will.

      The wind causes a rock to tumble. (not free will) A cat chases a mouse. (not free will) A child chooses a piece of candy. (free will) To label these as “free will” or “not free will” is not an exercise in determining how the universe works. At this point it is simply an exercise in testing a definition.

      The Templeton Foundation seeks a way to move forward on the free will issue. I submit that twomeyw2 has found a way. Answer his question. How could free will exist? Characterize it. Just define it.

  28. ntadepalli says:

    progressive will towards freedom


    Further to my earlier comment I like to say :

    The activities of our brain are neither isolated nor immune from influences from within or beyond itself.The regulating and reforming influences are self-generated (to optimise behavioural responses)as determined by past experience and knowledge.

    I guess these influences are sensed as personal effort possessing freewill in conscious processes.

    ( The influences are no doubt present in unconscious processes also , unsensed )

    Though the resulting activities are automatic and deterministic,they reveal individual autonomy.

  29. Armadodecadron says:

    Suppose the total human personality contains and combines both an external, mental/spiritual self that possesses a relationship with (but is not completely responsible for directing), the creature of habit clearly demonstrated by recent neurosciences?  And suppose this relationship can at times be disagreeable between the two aspects, causing hesitation and friction in the directive processes of the conscious mind?

  30. Eddy Nahmias says:

    As the discussion comes to a close today, I want to emphasize a point that Musing brings up (and that has permeated this discussion; see also Seligman’s post).  Much of the debate about whether free will exists, or whether neuroscience (or determinism or naturalism or whatever) threatens our having free will, depends crucially on how we define free will.  There is an almost endless source of philosophical writing on free will for those who want to pursue it (Robert Kane’s Contemporary Introduction to Free Will offers a nice overview, or for a Lexicon of definitions, see Templeton’s Big Questions in Free Will project: http://www.freewillandscience.com/wp/?page_id=63).

    One reason I do empirical studies on people’s understanding of free will is that I worry that willusionists’ armchair claims about how people define free will are mistaken in such a way that it might have problematic implications.  Here’s a bit from my book in progress that explains what I mean:

    (1) Suppose that some people believe that free will involves both ‘supernatural powers’ (e.g., to be “soul” or “sole” cause) and the ‘natural powers’ of rational thinking and self-control (but most do not believe that the natural powers depend on the existence of supernatural powers, and some believe that free will requires only the natural powers). 

    (2) Suppose that scientists and the media inform people that free will is an illusion.

    (3) Suppose that neither they nor ordinary people are clear about exactly what is meant by ‘free will’ or exactly which powers are being questioned, such that people interpret the claims in terms of their own understanding of “free will”. 

    Then, (4) People are likely to doubt the efficacy of their natural powers of free will, such as rational thinking and self-control.

    (5) And if people doubt that their rational thinking or efforts of self-control make a difference, then they are less likely to exercise these capacities.

    (6) And typically, exercising capacities for rational thinking and self-control leads to more personally and socially advantageous behavior (i.e., wise choices often require us to exercise free will, understood in terms of these natural capacities).

    I think this argument helps to explain some of the results that Kathleen Vohs mentioned in an early comment—e.g., that telling people they lack free will leads them to cheat a bit more and help a bit less (she, Jonathan Schooler, Thomas Nadelhoffer and I are currently studying the scope and causes of these sorts of effects). 

    These points illustrate that even if debates about free will are largely “mere verbal disputes” or debates about definitions, they can have important practical consequences.  And these consequences will also arise in legal and political debates—for instance, to what extent are people in control of their behavior such that they deserve to be held legally responsible (and punished) or such that they deserve their social advantages or disadvantages.

    I don’t think it’s an over-statement to suggest that our overheated political debates derive in part from people’s different philosophical views about the degree to which people have free will and are responsible for their situation in life—the degree to which the outcomes of our lives derive from our own initiative or from good or bad luck and our family’s and society’s support or lack thereof. In my opionion, most Americans have an overinflated sense of free will and individual responsibility, but I think the best way to correct these views is not to tell people no one has any free will at all, but rather to tell them that people possess varying degrees of free will and have varying opportunities to exercise it, and to explore these degrees of freedom empirically.

  31. Roman Dawes says:

    Free will: The ability of an individual to know what neuroscience predicts a specific choice of action will be in a given instant and choose another action. Or, it’s the ability to know what we’re supposed to do and do something else.

    Although our will is not entirely free, scientific determinism as a possible explanation for human behavior is a chimera. Let’s theoretically journey 50 or 100 years into the future and assume there is a comprehensive theory of mind that the “willusionists” assume is possible. Wouldn’t this theory be able to calculate at a certain instance, given access to all variables of an individual’s experience, present and past, what a person’s response will be when given a certain, say, visual or auditory stimulus? And shouldn’t that response be the same even if the person conducting the experiment told the individual what that response will be?

    That second part is very important, because it shows a fundamental improbability for determinism even as a philosophical hypothesis, which is all that it is now. If our behavior is truly governed by physical laws, it should abide by scientific predictability in experiments that can be duplicated by every scientist who conducts it. If behavioral determinism is true, then, a future scientist will be able give a person a choice of buttons to press, tell that person which one their behavioral theory predicts we will press, and it must be impossible for him to press another, given that knowledge of the correct outcome is a variable factored into the behavioral equation.

    Asserting that physical laws determine the motion of inanimate objects not only asserts that a five-pound boulder, when dropped, will fall to the earth. It asserts that it is impossible for the boulder to instead shoot up into the sky or not fall to the earth at all. That’s what behavioral determinism claims of our behavior – not just that when going out for a walk in the neighborhood, you will turn right at the first intersection, but that it’s impossible that you will turn left. That’s an extraordinary implied assertion, and one that don’t think determinists or their opponents fully appreciate.

    The other extraordinary implication (and I think it’s extraordinary because it’s folly taken too seriously by bright people), is that the human ability to think in the abstract has no bearing on our behavior and the choices we make. In the thought experiment above, couldn’t a person, given the “correct” response calculated for a controlled stimulus, just contravene the theory and provide the “wrong” result. What’s amazing about behavioral determinism is how its adherents don’t believe that they have to take into account a human mind able to perceive in the abstract a behavioral theory that’s supposed to be operating on them. There is absolutely nothing like human sentience anywhere else in the physical universe governed by natural laws, and they don’t seem to think that matters.

    But isn’t our behavior nevertheless “caused” by something, no matter what? Well, yes, but that doesn’t mean a science of human behavioral determinism is possible. Scientific determinism requires the above-described predictability given the ability to control variables. Human behavior can be caused by chance, whimsy, an image in a dream, a forgotten image in a dream buried in the subconscious, or of course, the ability to perceive in the abstract the predicted outcome of a scientific experiment on their behavior.   

    Again, determinism doesn’t merely claim that we can predict that a person will become an artist. The hypothesis is that we will eventually be able to predict what that person will paint and which colors she will use for each image. That’s simply folly.

    The images we conjure in abstract thought are dimensionless – i.e., they are without length, breadth, depth or mass, and so will never be measurable. That’s why many of the variables that cause behavior will forever remain outside of the reach of the mathematical sciences. So,  although all behavior is caused, it is often caused by a subjective response to a mind operating independently and abstractly. And rarely was a choice not made actually impossible. That’s free enough will for me.

  32. Emcay says:

    I think that this article shows very clearly how contemporary neuroscience can act in favor of free will. Neuroscience will never be the measure by which free will is about to be (re)defined though, regardless of the progress in the scientific sphere. It will certainly remain a great tool to explain certain dimensions of our common free will.

  33. jsg says:

    The debate over free will assumes, incorrectly, that there is someone to be either free or not free. But since there is no individual apart from the processes that comprise him or her, there is no one to be either free or not free.