Coyne’s latest defense of determinism: why it fails

Edge‘s big question for 2017 is: “What scientific term or concept should be more widely known?” The compilation of answers (205 in all) makes for fascinating reading. For his part, Professor Jerry Coyne has nominated physical determinism as “a concept that everyone should understand and appreciate.” Unfortunately, Coyne’s defense of this concept leaves a lot to be desired. As I’ll argue below, even if you reject interactionist dualism (as most scientists do), you can still believe in libertarian free will.

Professor Coyne begins by mis-defining determinism as the notion that “all matter and energy in the universe, including what’s in our brain, obey the laws of physics.” I know of no philosopher who defines determinism in this way. The Stanford Encyclopedia of Philosophy, for instance, roughly defines causal determinism as “the idea that every event is necessitated by antecedent events and conditions together with the laws of nature.” Coyne says nothing about antecedent conditions, and fails to even mention the notion of necessitation.

To illustrate what’s wrong with Coyne’s definition, I’ll use an analogy which is often cited by philosophers: the game of chess. All the pieces obey the rules (or laws) of the game, but those laws don’t tell the player where to move the pieces. Even if the pieces were capable of moving themselves without the help of an outside agent, nothing in the rules of the game would determine the moves that followed. That’s because the rules of chess merely constrain the set of moves which are allowed, without determining the movement of any of the pieces. What Coyne needs to show is that the laws of physics are more than mere constraints, and that for any given collection of molecules, they narrow down the set of possible outcomes to just one, and no more.

Coyne continues:

At a given moment, all living creatures, including ourselves, are constrained by their genes and environment to behave in only one way — and could not have behaved differently. We feel like we make choices, but we don’t. In that sense, “dualistic” free will is an illusion.

Here, again, Coyne is making an unwarranted leap of logic. I would happily grant that I am constrained by my genes and environment. What Coyne needs to show, however, is that I am constrained to behave in only one way.

Coyne seems to believe that a basic knowledge of physics is all you need to establish the truth of determinism:

This must be true from the first principles of physics. Our brain, after all, is simply a collection of molecules that follow the laws of physics; it’s simply a computer made of meat. That in turn means that given the brain’s constitution and inputs, its output — our thoughts, behaviors and “choices” — must obey those laws. There’s no way we can step outside our mind to tinker with those outputs. And even molecular quantum effects, which probably don’t even affect our acts, can’t possibly give us conscious control over our behavior.

Coyne’s appeal to physics is a misguided one. The claim that physics rules out free will is often heard. I find it curious, however, that proponents of determinism seldom tell us exactly which laws of physics imply the truth of determinism. The law of the conservation of mass-energy certainly doesn’t; and neither does the law of the conservation of momentum. Newtonian mechanics is popularly believed to imply determinism, but this belief was exploded over two decades ago by John Earman (A Primer on Determinism, 1986, Dordrecht: Reidel, chapter III). Sometimes the Principle of Least Action is said to imply determinism. But since the principle does not apply to a system which is non-holonomic (i.e. a system whose geometrical constraints involve not only the coordinates but also the velocities), and usually doesn’t apply if the system is dissipative (i.e. if it includes frictional forces), then I fail to see how it could be realistically applied to the human brain – which we have absolutely no reason to regard as holonomic, and which is subject to frictional forces (both wet and dry). So much for Coyne’s theatrical appeals to “the first principles of physics.” However, there is a more fundamental problem with his case against free will.

Let us grant (for the sake of argument) Professor Coyne’s materialist premise that it is my brain that thinks and chooses, rather than:

(i) a Cartesian soul interacting with my body, or
(ii) an embodied person (me) performing a non-bodily action (which is what Aristotelian-Thomists believe happens when we make a rational decision), or even
(iii) a person (me) performing an action (choosing) which has some non-physical properties, in addition to its physical properties (as property dualists maintain).

Even if we accept that thoughts and decisions are actions performed by our brains, and that these actions can be fully described in physical terms, the truth of determinism does not follow. To rule out the possibility of free will, we need to import an additional premise: that any movement of a body can always be fully explained in terms of the interactions between its physical constituents (i.e. the particles of which its is composed) and the other bodies in their surroundings. This premise might sound trivially obvious, but what it denies is that bodies have any holistic physical properties which affect the way in which their constituent particles move. When you put it that way, it’s not at all obvious.

Holism and the failure of supervenience

Suppose now that brains possess holistic properties (which are physical but not attached to any particular part of the brain), and that thinking and choosing are simply holistic properties of conscious, normal human brains. Then since physics makes no reference to holistic properties in its mathematical descriptions of bodily motion, it follows that physics will be unable to fully account for the movement of any body possessing these holistic properties – which in turn means that physics can provide us with no guarantee that a set of inputs acting on the brain when its constituent particles are configured in physical state S will always generate the same output, since the brain’s holistic properties may turn out to vary, even when its constituent parts (i.e. its neurons) are configured in the same state.

The notion which I am proposing here, that a body whose physical constituents have exactly the same properties on two successive occasions may yet turn out to have holistic properties which are quite different on both occasions may strike some readers as outrageous, as it runs afoul of the widely accepted principle of supervenience, which the philosopher Donald Davidson pithily expressed in a now-famous quote:

“[M]ental characteristics are in some sense dependent, or supervenient, on physical characteristics. Such supervenience might be taken to mean that there cannot be two events alike in all physical respects but differing in some mental respect, or that an object cannot alter in some mental respect without altering in some physical respect.”

To someone who accepts the principle of supervenience, the contrary notion, which I am defending here, might seem to smack of occultism. In reality, it is nothing of the sort. For if holistic properties of bodies are not confined to a particular point in space, it should hardly be surprising that these same properties are not confined to a particular point in time, either – which means that we cannot infer what values these properties take, merely by examining the behavior of a body, here and now. The problem with the principle of supervenience is that it rests upon a “snapshot” view of bodies, which is biologically implausible.

An advocate of supervenience might reply that the principle can be reformulated to accommodate the passage of time: “there cannot be two events which are alike in all physical respects over an extended period of time, but differing in some mental respect.” But what period of time are we talking about here? One second? One minute? One hour? Any answer we give will be an arbitrary one. And in the case of human beings, we can certainly envisage an individual performing the same behavior on two separate occasions, but having entirely different reasons and intentions for doing so, on each occasion. You may tell me that subtle differences would still show up in their brains, on the two occasions, which a skilled neurologist could detect. That may well be so, but I see no reason why it must – even when the behavior being repeated occurs over a very, very short interval of time.

Perhaps the supervenience proponent is envisaging two (living, conscious) bodies, whose physical states are identical over their entire lifetimes, and asserting that it is impossible for there to be any mental differences between these bodies. But in that case, the assertion is simply wrong. For it will never be the case that the two bodies are receiving identical input from the outside world. The mere fact that they are situated in different locations prevents that. Hence their physical states will never be the same – which renders moot the question of whether their mental states will be the same. Thus the plausibility of supervenience turns out to rest on an absurd counterfactual.

But even if the claim that two living bodies whose physical states are identical over their lifetimes will have the same mental states were true, we could not use it to infer that if a particular living body is in the same physical state S over an extended period of time, on two successive occasions, its mental states are the same on both occasions. (And if we allow that a body’s mental states can affect its behavior, it follows that we cannot be sure that it will behave in the same way on both occasions.)

We have seen that attempts to argue for the principle of supervenience end in ignominious failure. At this point, a defender of determinism may argue that the idea that the mental states of bodies are holistic is wildly implausible, and not worthy of serious consideration – but I would ask: why? Attempts to localize semantic propositions in the brain have, to date, been wholly unsuccessful (although scientists have had limited success in localizing words that are stored in the brain). That being the case, we cannot speak of decisions based on propositional reasoning as having a location in the brain, either. And even if some decision-making region of the brain could be identified – the frontal lobe is often suggested, although other areas appear to be involved as well – it still would not follow that decision-making is an event which could be reduced to the neuronal level. In other words, the question is not whether the holism of the mental is true, but to what extent it is true.

Coyne’s dismissal of randomness as irrelevant to the question of free will

At one point in his article, Coyne briefly alludes to the possibility that at the quantum level, events might not be determined, after all – but he dismisses this possibility, on the grounds that it has no relevance to our decision-making: “molecular quantum effects, which probably don’t even affect our acts, can’t possibly give us conscious control over our behavior.” If Coyne is asserting that quantum randomness in no way implies free will, then I agree; but if he is claiming that randomness is incompatible with free will, then he is profoundly mistaken. Consider the following two rows of digits:

1 0 0 0 1 1 1 1 0 0 0 1 0 1 0 0 1 1
0 0 1 0 0 0 0 1 1 0 1 1 0 1 1 1 0 1

The above two rows of digits were created by a random number generator. The digits in some of these columns add up to 0; some add up to 1; and some add up to 2.

Now suppose that I impose the non-random macro requirement: keep the columns whose sum equals 1, and discard the rest. I now have:

1 0 1 1 1 0 0 0 0 0 1
0 1 0 0 0 1 1 0 1 1 0

Each row is still random (at the micro level), but I have now imposed a non-random macro-level constraint on the system as a whole (at the macro level). That, I would suggest, what happens when I make a choice. Choices are holistic events in the brain, which constrain not only the spatial pattern but also the temporal pattern of neuronal firings in the brain, leaving them random at the micro-level, but imposing a distinctive pattern at the macro-level, which varies with the choice being made. A choice, I would suggest, is not typically made at a point in time; rather, it usually occurs over a segment of time – for instance, the length of time taken to perform a deliberate bodily movement (which may be seconds, minutes or even hours in some cases).

Finally, it may be asked how the brain could possess the ability to “discard” (or veto) an ensemble of micro-level states in the brain which does not correspond to the desired macro-level pattern – just as I did when I kept the columns of digits whose sum was 1, and eliminated the rest. The scientific answer is that a combination of feedback and feed-forward processes is known to regulate our voluntary movements, and that the brain continually makes minor adjustments to the motor impulses associated with these movements, even as it executes them. Exactly how it aggregates signals in different regions and rules out the ones it doesn’t want is a question I’ll leave to scientists who study the brain. All that concerns us here is that the aggregation of signals is known to occur in the brain, in connection with motor tasks, so the question of mechanism is a purely academic one.

We are beginning to see how a combination of holistic macro-level properties and micro-level randomness makes it possible for human brains to manifest the elusive phenomenon that we call free will. All that remains is to dispose Professor Coyne’s objection that experimental data have already ruled out free will.

Has free will been experimentally disproven?

In his latest article on determinism, Professor Coyne cites a range of experimental findings which appear to militate against the idea of free will:

Physical determinism of behavior is also supported by experiments that trick people into thinking they’re exercising choice when they’re really being manipulated. Brain stimulation, for instance, can produce involuntary movements, like arm-waving, that patients claim are really willed gestures. Or we can feel we’re not being agents when we are, as with Ouija boards. Further, one can use fMRI brain scans to predict, with substantial accuracy, people’s binary decisions up to ten seconds before they’re even conscious of having made them.

Coyne’s first two objections merely establish that my belief that I freely chose to execute a bodily movement is neither infallible nor incorrigible (and why should it be?). What they do not establish is that my bodily movements are never the result of my free choices – which is an entirely different question. I would suggest that the real reason why people tend to claim that involuntary arm movements caused by stimulating the brain are really willed gestures is that they have an overwhelming psychological need to maintain internal consistency, when giving a narrative of their actions and choices, as brain scientist Nicholas Humphrey points out in his Edge article on Referential Opacity. And the movement of Ouija boards is explained by the mechanism of the ideomotor effect, where the movement of the board is caused by subconscious signals from the brain, which cause a person’s hands and arms to move the board in a certain way. It is misleading of Coyne to cite signals that we are not even conscious of making as an example of voluntary agency.

What about Coyne’s claim that fMRI brain scans can predict people’s decisions up to ten seconds before they’re aware of having made them? A video of the famous “No free will” experiment by John-Dylan Haynes (Professor at the Bernstein Center for Computational Neuroscience in Berlin), has previously been cited by Coyne as experimental evidence debunking the notion of free will. According to the video, an outside observer, monitoring my brain, can tell which of two buttons I’m going to push, several seconds before I consciously decide to do so. I have discussed this experiment in a 2012 article, titled, Is free will dead? It turns out that there are several problems with Coyne’s claim that it refutes the idea of free will.

First, as Coyne acknowledges in a 2011 post on his Web site, Why Evolution Is True, entitled, The no-free-will experiment, avec video, “the ‘predictability’ of the results is not perfect: it seems to be around 60%, better than random prediction but nevertheless statistically significant.” Sorry, but I don’t think that’s very impressive.

Second, the experiment was deliberately designed to exclude the possibility of reflection. In the experiment, as narrator Marcus du Sautoy (Professor of Mathematics at the University of Oxford) puts it, “I have to randomly decide, and then immediately press, one of these left or right buttons.” Now, most people would say that a reflective weighing up of options is integral to the notion of free will.

Third, the experiment relates to an artificial choice which is stripped of several key features which normally characterize our free choices:

(a) it’s completely arbitrary. It doesn’t matter which button the subject decides to press. Typically, our choices are about things that really do matter – e.g. who the next President of the United States will be. And we make choices for a reason, rather than for no reason at all.

(b) it’s binary: left or right. In real life, however, we usually choose between multiple options – often, between an indefinitely large number of options – for example, when we ask ourselves, “What career shall I pursue after I graduate?”

(c) it’s zero-dimensional. Normally, when we make choices, there are multiple axes along which we can evaluate the desirability of the various options. In the experiment described above, there were no axes along which we could weigh up the desirability of the two options (left or right button), as there was literally nothing to compare.

(d) it’s impersonal. We are social animals, and most of our choices relate to other people – e.g. “Whom shall I marry?” Pressing a button, on the other hand, is a solitary act.

(e) it contains no reference to second-order mental states. Typically, when we choose, we give careful consideration to what other people will think of our choice, and how they’ll feel about it. To entertain these thoughts, we have to be capable of second-order mental states: thoughts about other people’s thoughts.

(f) it’s future-blind. The choice of whether to press the left button or the right button is a here-and-now choice, with no reference to future consequences.

(g) it has no feedback mechanism. Not only do choices typically have consequences, but the results of our choices are usually communicated back to us in a way that influences our future behavior. In the button-pressing experiment, nothing is learned by the subject.

Fourth, the experiment described by Coyne made no attempt to evaluate Benjamin Libet’s hypothesis of “free won’t”: “while we may not be able to choose our actions, we can choose to veto our actions.” What happens if the subject is permitted to decide in advance which button they will press, but is also free to change their mind at the last minute? Can a trained outside observer, who is monitoring an MRI scanner, pick up this sudden change of mind on the subject’s part? Coyne does not tell us.

A fifth criticism that can be made of Haynes’ experiment is that the time scale involved makes it meaningless to speak of free will or its absence, just as it would be meaningless to ask what color a hydrogen atom is. Typically, our free choices are preceded by an extended period of deliberation, followed by the brain’s preparation for the execution of a bodily movement, followed by activation of specialized areas of the brain which are responsible for the contraction of specific muscles in the body. It could therefore be argued that freedom is a property which does not attach to the decision to act here and now, but to the entire process leading up to the decision.

So, what can the predictability (60% of the time) of an arbitrary, binary, impersonal choice, which involves no weighing up, no worries about what other people might think, no thought of the future and no feedback, possibly tell us about the existence of free will in human beings? Absolutely nothing.

Adina Roskies, a neuroscientist and philosopher who works on free will at Dartmouth College in Hanover, New Hampshire, wasn’t too impressed with Professor Haynes’ experiment either. “All it suggests,” she declared, “is that there are some physical factors that influence decision-making”, which shouldn’t be surprising.

Summary

The arguments I have discussed above are the only arguments against free will which Professor Coyne puts forward in his latest article. It appears to me that Coyne’s case against free will is based on the flimsiest of scientific and philosophical foundations: he fails to adequately define the notion of determinism; his theoretical arguments against free will rest on a misunderstanding of physics and illicitly assume the truth of reductionism and the falsity of holism; he overlooks the possibility that randomness in Nature, while not implying the existence of free will, may help make it possible to realize; and finally, the experimental data which he cites to discredit the notion of free will fail to do so. I conclude that the possibility of free will remains open, and that it can be defended without appealing to dualism.

What do readers think?

44 thoughts on “Coyne’s latest defense of determinism: why it fails

  1. Fantastic cake and eat it too. A long gripe about controlling for confounding variables as a bad thing!

  2. I agree that holistic macro-level properties are enough to salvage what we ordinary think of as volitional agency, but I’d like to hear more about why this is sufficient for libertarian freedom rather than just garden-variety compatibilism.

    I’d also like to know why you don’t think libertarian freedom doesn’t entail interactionist dualism. That’s always been my objection to it. If it’s because you think of the mind/brain-body-environment as a hylomorphic (structure/stuff) relation, is that enough to give you libertarian free will?

    I guess it’s because I think of libertarian freedom as a Cartesian innovation (and in fact the driving force behind his mind/body dualism) that I’m not sure you can really have one without the other.

  3. Jerry Coyne is a deeply religious man. And physical determinism is his religion.

    One of the creationist ideas is “front loading”. I find it quite amusing that Coyne is an evolutionist, and a critic of creationism. Yet his own views on determinism amount to a front loading theory of everything. Coyne fails to see the contradiction, and I suppose that failure could be seen as weak support for determinism.

    Quoting Coyne:

    Why is it important that people grasp determinism? Because realizing that we can’t “choose otherwise” has profound implications for how we punish and reward people, especially criminals.

    Coyne seems to see that as an important part of his argument. But it is nonsense. If physical determinism is true, as Coyne asserts, then the way that we treat criminals was already determined at the time of the big bang.

  4. To see the incoherence of either position, simply assume either side and try to test it against reality. Try to define such a test.

  5. I think we ought to have a government agency to match up criminals who cannot “choose otherwise” with judges and juries who cannot “choose otherwise.”

    Seems fair.

  6. Mung:
    I think we ought to have a government agency to match up criminals who cannot “choose otherwise” with judges and juries who cannot “choose otherwise.”

    Seems fair.

    They tried that once. But the people they hired had no choice but to not show up for work.

  7. Neil Rickert:
    One of the creationist ideas is “front loading”.I find it quite amusing that Coyne is an evolutionist, and a critic of creationism.Yet his own views on determinism amount to a front loading theory of everything.

    Exactly right! Know the precise details of every particle/wave/field in the Universe at a particular moment and you have the entire history and future too, set in stone! Yet some events such as nuclear decay are not deterministic, so phew, we don’t need to worry!

  8. I’m not sure physics says nuclear decay is undetermined. Just not by local causes.

    I prefer to think something is amiss with the concepts. We keep wanting to analogise everything with billiard balls, but that leads to conclusions that don’t work.

  9. Mung:
    I think we ought to have a government agency to match up criminals who cannot “choose otherwise” with judges and juries who cannot “choose otherwise.”

    Seems fair.

    Who chooses?

  10. Hi Kantian Naturalist,

    You write:

    I agree that holistic macro-level properties are enough to salvage what we ordinary think of as volitional agency, but I’d like to hear more about why this is sufficient for libertarian freedom rather than just garden-variety compatibilism.

    I’d also like to know why you don’t think libertarian freedom doesn’t entail interactionist dualism. That’s always been my objection to it. If it’s because you think of the mind/brain-body-environment as a hylomorphic (structure/stuff) relation, is that enough to give you libertarian free will?

    Good question. Personally I think that human beings are capable of performing both bodily and non-bodily actions, and that thinking and choosing fall into the latter category. I don’t think of a human being as two things (a soul and a body) as Descartes did, but as one thing. So in that respect, I’m in the hylomorphic camp. But when you ask the hylomorphists how I manage to pick up my pen, most of them (Ed Feser included) reject the idea that my thoughts or choices make my arm move. They prefer to say that they’re the reason why my arm moves when I pick up my pen. The problem I have with that is that it leaves unanswered the question of what does make my arm go up, and that it seems to epiphenomenalize thought and/or choice, by denying that it makes anything happen: to the extent that it makes a difference, it is as a goal or final cause, not as an efficient cause. So I incline towards the view that my immaterial thoughts can actually interact (non-locally) with the brain – not by pushing and pulling, but by excluding or vetoing aggregations of signals which are incompatible with the movement that the agent intends. In that respect, I’m closer to the Cartesians than the Aristotelian-Thomists.

    However, I readily admit that the question of how I can perform a non-bodily action (thinking or intending) which controls my brain is a difficult one. It sounds strange to many people. I’m also aware that there are some Christians who subscribe to a materialist view of man on Biblical grounds – e.g. Jehovah’s Witnesses and (I think) Seventh Day Adventists. They believe that when I think, it is my brain that does the thinking, and that when I die, nothing is left – until Judgement Day, when God resurrects us. I think there is a slight chance (maybe 5 to 10%) that they are right on this point, and for that reason, I think it is unwise for religious believers to defend free will on the assumption that dualism is true. I’m much more certain of free will than I am of dualism. That was why I wrote this post: in order to demonstrate that even if you’re a materialist – indeed, even if you’re an atheist – you can still believe in libertarian free will.

    You argue that holistic macro-level properties can only salvage the notion of volitional agency, rather than libertarian free will. The answer lies in my paragraph:

    Suppose now that brains possess holistic properties (which are physical but not attached to any particular part of the brain), and that thinking and choosing are simply holistic properties of conscious, normal human brains. Then since physics makes no reference to holistic properties in its mathematical descriptions of bodily motion, it follows that physics will be unable to fully account for the movement of any body possessing these holistic properties – which in turn means that physics can provide us with no guarantee that a set of inputs acting on the brain when its constituent particles are configured in physical state S will always generate the same output, since the brain’s holistic properties may turn out to vary, even when its constituent parts (i.e. its neurons) are configured in the same state.

    The upshot of the preceding paragraph is that if we are capable of performing holistic activities, then from the standpoint of physics, they are unpredictable. Now that doesn’t make them free in the libertarian sense, but it does make them non-deterministic. What would make them free, then? If these holistic activities embody my will, then they are not only unpredictable, but volitional acts. Put that together and I would argue that you have full libertarian free will. In other words, what I’m saying is that bodies with a sufficient degree of neural complexity may turn out to have wills of their own, as an emergent property. Now that does sound mysterious. But it’s perfectly compatible with materialism, albeit a rather peculiar variety. Cheers.

  11. “Exactly right! Know the precise details of every particle/wave/field in the Universe at a particular moment and you have the entire history and future too, set in stone! ”

    But you can’t do that, even in principle.

  12. Alan Fox: Exactly right! Know the precise details of every particle/wave/field in the Universe at a particular moment and you have the entire history and future too, set in stone! Yet some events such as nuclear decay are not deterministic, so phew, we don’t need to worry!

    More evidence that we’re living in a simulation.

    Grad Student 1: “It’s going to take me weeks to implement this nuclear decay algorithm the prof came up with, but she wants it tomorrow.”

    Grad Student 2: “Ah, just hack in a random number generator — she’ll never check.”

  13. AhmedKiaan:
    “Exactly right! Know the precise details of every particle/wave/field in the Universe at a particular moment and you have the entire history and future too, set in stone! ”

    But you can’t do that, even in principle.

    Hell, no!

  14. petrushka:
    I’m not sure physics says nuclear decay is undetermined. Just not by local causes.

    I prefer to think something is amiss with the concepts. We keep wanting to analogise everything with billiard balls, but that leads to conclusions that don’t work.

    OK, what about cosmic rays? Get a cosmic ray hit your DNA and you could be in for a SNP, which if it happens to catch you in the gonad, might contribute to a new beneficial allele. Are comic ray emissions, and especially the precise direction they take predictable from the prior state of the Universe?

  15. Alan Fox: OK, what about cosmic rays? Get a cosmic ray hit your DNA and you could be in for a SNP, which if it happens to catch you in the gonad, might contribute to a new beneficial allele. Are comic ray emissions, and especially the precise direction they take predictable from the prior state of the Universe?

    Might need to predict where that gonad is too.

  16. Alan Fox: Are comic ray emissions, and especially the precise direction they take predictable from the prior state of the Universe?

    Magic eight-ball says the future is cloudy.

    My reading on this says it is quite possible to have a system that is unpredictable from within the system, yet fully determined. I think there’s a difference between determined and predicable.

    But I think Jerry drops the ball here. We, as sentient beings, are evolved to attempt to predict the future (the consequences of our actions) and to act according to our expectations. That is what “free will” means in a social and legal sense.

    We are not likely to become “fatalists” and start acting as if putting our hand on a hot stove is unimportant, because an omniscient observer knows we are going to do it.

    In the ordinary, garden variety sense of the term, we have free will. We are the result of learning, and consequences matter. I do not believe capital punishment is socially constructive, but if the world were such that it were a useful deterrent, I might not oppose it. I certainly wouldn’t oppose it on the grounds that everything is determined.

  17. vjtorley,

    Very interesting! From what you’ve said here, it looks like your thesis that thought and choice are “non-bodily” is functioning as a premise to get you from volitional agency to genuine libertarian freedom. Am I interpreting you correctly? Without that additional premise, one has actions that are under-determined by physics but not genuinely free (in the sense that you think we need).

    I can easily agree that thinking about where to have lunch is not overly bodily, whereas as actually walking to the cafe is. But that doesn’t mean that reasoning of the sort you envisage doesn’t supervene on brain-body-environment transactions. For it could be that when I decide not to go out, because I need to save money, that could be explained in terms of how prefrontal activity inhibits subcortical urges and prompts.

    In short, I think that defenders of agent causation need to worry much more about Spinoza’s criticism that we believe in free will to the extent that we are ignorant of the causes of our actions.

  18. Neil Rickert:
    Jerry Coyne is a deeply religious man.And physical determinism is his religion.

    One of the creationist ideas is “front loading”.I find it quite amusing that Coyne is an evolutionist, and a critic of creationism.Yet his own views on determinism amount to a front loading theory of everything.Coyne fails to see the contradiction, and I suppose that failure could be seen as weak support for determinism.

    Quoting Coyne:

    Coyne seems to see that as an important part of his argument.But it is nonsense.If physical determinism is true, as Coyne asserts, then the way that we treat criminals was already determined at the time of the big bang.

    I agree with your assessment. And I think Vince’s excellent critique makes the shortcomings of Coyne’s “argument” quite clear. My only reservation about it is that there are many ways to define “supervenience” and the assumption of each version produces a different result viz. the mind/body problem. But Coyne’s defense of hard determinism (basically “It’s physics, man!”) is pretty primitive.

  19. walto: But Coyne’s defense of hard determinism (basically “It’s physics, man!”) is pretty primitive.

    I’d go a few step beyond ‘primitive’ here.

    I’m enough of a naturalist to think that nothing can happens within the history of the universe that entails a violation of any the laws of fundamental physics.

    Fundamental physics, as we currently understand it, consists of general relativity, quantum mechanics, and thermodynamics.

    But since we don’t have a comprehensive theory of all three, even at the level of fundamental physics, ‘promiscuous realism’ (John Dupre’s phrase) is unavoidable.

    From what I can see, there’s no way to get from promiscuous realism at the level of fundamental physics to metaphysical hard determinism without a lot of question-begging assumptions. Those assumptions should be viewed with suspicion by anyone who hopes to have their metaphysical views hew as closely as possible to what can be established by the methods of the empirical sciences!

  20. Hi Kantian Naturalist,

    Thank you for your response. You write:

    Very interesting! From what you’ve said here, it looks like your thesis that thought and choice are “non-bodily” is functioning as a premise to get you from volitional agency to genuine libertarian freedom. Am I interpreting you correctly? Without that additional premise, one has actions that are under-determined by physics but not genuinely free (in the sense that you think we need)…

    For it could be that when I decide not to go out, because I need to save money, that could be explained in terms of how prefrontal activity inhibits subcortical urges and prompts.

    Just to be clear: what I’m saying is that if a bodily activity were: (i) holistic, (ii) unpredictable and (iii) an embodiment of an act of will, then it would be a free choice, in the libertarian sense of “free.” Condition (ii) follows from (i) for reasons I have explained above: holistic properties don’t figure in the physics equations we use to predict motion. So the only questions remaining are whether there are any holistic activities occurring in the brain, and whether any physical activity can embody an act of will (dualists and hylomorphists say no, and I’m inclined to agree, but I’m willing to allow that maybe, the answer is yes).

    Your decision not to go out might well be because prefrontal activity in your brain inhibits subcortical urges and prompts telling you that you’ll be happier if you spend your money instead. That’s fine by me, provided that the prefrontal activity is itself an embodiment of an act of will.

    You may object that to the extent that the prefrontal activity in your brain successfully inhibits your urge to go on a spending spree, it does so by virtue of its physical properties rather than its volitional properties, which means that the causation involved is non-rational. Libertarian freedom, on the other hand, is essentially rational. I would reply that your argument proves too much, as the same objection could be deployed against any kind of activity (whether physical or incorporeal) causing you to fight and conquer your urges to go out and spend. For instance, if my Cartesian soul makes my body pick up a book and sit down to read it in the living room, rather than going out, it does so by virtue of its causal properties rather than its volitional properties. But the fallacy is the same in both arguments: namely, the assumption that causal properties (whether material or immaterial) cannot be volitional. While it’s clear that the local properties of particles (which move either deterministically or randomly) cannot be described as volitional, it’s by no means clear that the non-local, holistic properties of (say) the prefrontal cortex are incapable of being both causal and volitional at the same time.

    You add:

    In short, I think that defenders of agent causation need to worry much more about Spinoza’s criticism that we believe in free will to the extent that we are ignorant of the causes of our actions.

    If what I have proposed in my OP is correct, then Spinoza’s mind-body dual-aspect theory may be close to the mark, after all. Where Spinoza went wrong was in neglecting the possibility of holistic physical processes which are not determined, and in assuming the truth of determinism on a priori theological grounds: his God is wholly necessary and the only substance, so everything that happens in the world has to happen. But that’s a topic for another day.

  21. vjtorley,

    I can happily agree that there’s a need to respect and preserve the distinction between reasons and causes!

    However, I think that the distinction between personal-level descriptions and explanations and subpersonal-level descriptions and explanations is sufficient for us to do precisely this. Thus what is described at one level as my reasoning that it is better for me to have my lunch at home rather than go out can also be described at a different level as prefrontal cortical inhibition of limbic system activity.

    Put otherwise, the distinction between what is volitional and what is mechanistic (i.e. neuro-computational) is not a distinction in terms of ontologically basic properties — volitional properties and physical properties — but rather a distinction in terms of whether we are talking about the whole system (whether a human person as rational animal or a nonrational but minded animal) or its one of components (the brain).

  22. vjtorley: Just to be clear: what I’m saying is that if a bodily activity were: (i) holistic, (ii) unpredictable and (iii) an embodiment of an act of will, then it would be a free choice, in the libertarian sense of “free.” Condition (ii) follows from (i) for reasons I have explained above: holistic properties don’t figure in the physics equations we use to predict motion. So the only questions remaining are whether there are any holistic activities occurring in the brain, and whether any physical activity can embody an act of will (dualists and hylomorphists say no, and I’m inclined to agree, but I’m willing to allow that maybe, the answer is yes).

    I’m not clear on how that would distinguish libertarian free will from compatibilist free will. To be fair, I have never been clear on that distinction (if there is a real distinction).

    I’ll note that unpredictable need not contradict being physically determined. We generally use the term “chaotic” to describe behavior that is determined yet unpredictable.

    You seem to be taking “holistic” to be a natural kind. I don’t see that it is. Roughly speaking, to say that something is holistic is to say that there is a complete failure of attempts to analyze it into simpler components. But that is surely relative to our methods of analysis and to what we take to be the simpler components.

  23. petrushka:
    We, as sentient beings, are evolved to attempt to predict the future (the consequences of our actions) and to act according to our expectations. That is what “free will” means in a social and legal sense.

    We are not likely to become “fatalists” and start acting as if putting our hand on a hot stove is unimportant, because an omniscient observer knows we are going to do it.

    In the ordinary, garden variety sense of the term, we have free will. We are the result of learning, and consequences matter. I do not believe capital punishment is socially constructive, but if the world were such that it were a useful deterrent, I might not oppose it. I certainly wouldn’t oppose it on the grounds that everything is determined.

    No that is not free will in legal and social terms. Free will in that terms means that you know that the stove is hot and will burn your hand and knowing that you choose to put or not your hand according with your “will”.
    The problem for Coyne determinism is not only the “free will” but th “will” alone. If determinism is true we have not anly the illusion to choose freely, we have the illusion of will. We do not will, we act according the chemical reactions that are determined by the laws of chemistry.

  24. Blas: The problem for Coyne determinism is not only the “free will” but th “will” alone. If determinism is true we have not only the illusion to choose freely, we have the illusion of will.

    The ability to learn from experience is not an illusion. The capacity to learn and be modified by learning is the legal test. That is implicit in “knowing” the consequences of actions and “knowing” the laws governing behavior.

    From the legal standpoint, an agent is a black box that can be defective in any of several ways. It can be incapable of understanding the law, incapable of understanding consequences, or incapable of obeying the law. Simplified, those who have the mental capacity and who disobey are “punished”. Those who are mentally incapable of obeying the law or incapable of understanding it are “treated”.

    The legal system has been coping with this problem for centuries. There are no easy solutions, and so the law sometimes seems to behave stupidly. to me it is purely a problem of utility. What kind of law enforcement has the most social utility.

    I have no magic answers. no one does. that’s why we have politics.

  25. Blas: The problem for Coyne determinism is not only the “free will” but th “will” alone. If determinism is true we have not anly the illusion to choose freely, we have the illusion of will. We do not will, we act according the chemical reactions that are determined by the laws of chemistry.

    Certainly it’s right that if determinism is true, then the feeling of “will” is a kind of illusion. That was Spinoza’s position, and in a related way, also Nietzsche’s.

    But determinism of that metaphysically demanding sort does not follow from the sciences alone. Coyne is simply mistaken in thinking otherwise.

  26. petrushka: The legal system has been coping with this problem for centuries. There are no easy solutions, and so the law sometimes seems to behave stupidly. to me it is purely a problem of utility. What kind of law enforcement has the most social utility.

    I have no magic answers. no one does. that’s why we have politics.

    I’m with you here. ASSUMING we have libertarian free will, and organizing our entire legal and political systems accordingly, has worked since before written history. For me, determinism and solipsism are akin – we can’t show either of them is wrong, but so what?

  27. petrushka: What kind of law enforcement has the most social utility.

    I have no magic answers. no one does. that’s why we have politics.

    That´s the problem, who are going to enforce the law and based on what “social utility”. Unfortunatly for TSZ “science” cannot define “social utility”.
    Looking at the mankind history probably is far better keep not only the “will illusion” for maintaining the social order but keep also the “religion illusion”.
    And yes, french revolution was far more violent than spanish conquest of America, fascism was far more violent than cruzades and imposition of comunism was far more violent than the imposition of islam.

  28. Blas,

    Looking at the mankind history probably is far better keep not only the “will illusion” for maintaining the social order but keep also the “religion illusion”.

    What would you do with people who did not share the religion illusion?

  29. ALso, secularism has brought unprecedented levels of freedom and peace to society. I can’t believe there’s still people wanting to go back to the middle age.

  30. Blas,

    And yes, french revolution was far more violent than spanish conquest of America, fascism was far more violent than cruzades and imposition of comunism was far more violent than the imposition of islam.

    I’m impressed you can even lift a brush that broad.

  31. Flint: I’m with you here. ASSUMING we have libertarian free will, and organizing our entire legal and political systems accordingly, has worked since before written history. For me, determinism and solipsism are akin – we can’t show either of them is wrong, but so what?

    I see the point you’re making here, but I have some reservations about how you’re making it.

    I’m fine with the idea that choice and moral accountability are deeply held pragmatic presuppositions of human agency, though I’d be hesitant to say how far back in history that goes. (See the Iliad, for example.) But sure, the way that we assign praise and blame, and try to motivate people’s behavior (and justify, excuse, or apologize for our own) has a lot to do with the pragmatic presupposition of personal choice and moral accountability.

    I’m less sure about the idea that this presupposition amount to presupposing libertarian free will. I think of libertarian freedom as the idea that every free choice is a causa sui, a self-caused act. As far as I know, no one had that idea before Descartes. It’s how he tries to avoid the metaphysical consequences of the mechanistic worldview to which he is otherwise committed, and it’s also what motivates his mind-body dualism.

    (Also, I think determinism and solipsism can be shown to be false, but that’s a different issue!)

  32. Blas: And yes, french revolution was far more violent than spanish conquest of America

    Astonishing display of ignorance there. Nicely done.

  33. dazz:
    ALso, secularism has brought unprecedented levels of freedom and peace to society. I can’t believe there’s still people wanting to go back to the middle age.

    I’d give a lot to be middle-aged again!

  34. dazz: You have to be kidding me.

    I guess it depends whether you consider importing European diseases to the native population a violent act. It certainly resulted in a vast number of deaths, perhaps up to 90% of the indigenous population.

    ETA Executions for the period known as the Reign of Terror amounted to around 17,000 for the whole of France.

  35. Alan Fox: I’d give a lot to be middle-aged again!

    LOL, my english sucks balls, oh well

    Alan Fox: I guess it depends whether you consider importing European diseases to the native population a violent act. It certainly resulted in a vast number of deaths, perhaps up to 90% of the indigenous population.

    ETA Executions for the period known as the Reign of Terror amounted to around 17,000 for the whole of France.

    Pretty sure many, many more were killed and enslaved by the colonizers in America

Leave a Reply