Alfred Mele on free will

Alfred R. Mele’s recent article on free will in The Chronicle of Higher Education is short and to the point. First, Mele argues that most people do not have a conception of free will that involves anything obviously spooky, such as a non-physical soul. Second, he argues that arguments against free will that rely on Libet-style experiments are inconclusive. He may be right on both points, but this falls well short of vindicating free will (to be fair to him, all he concludes is that it’s too early to bet the farm on free will’s non-existence … a rather weak claim).

As to the first point, Mele refers to a simple study that he conducted (he suggests that there were more, but only offers us the one):

In one, I invited participants to imagine a scenario in which scientists had proved that everything in the universe is physical and that what we refer to as a “mind” is actually a brain at work. In this scenario, a man sees a $20 bill fall from a stranger’s pocket, considers returning it, and decides to keep it. Asked whether he had free will when he made that decision, 73 percent answer yes. This study suggests that a majority of people do not see having a nonphysical mind or soul as a requirement for free will.

Just a few points about this. First, even as reported only 73 per cent of participants in the study thought that free will is compatible with physicalism – that seems to be the essence of what they were asked. That’s an overwhelming majority, no doubt, but the fact remains that 27 per cent of people in the study apparently thought that free will is not compatible with physicalism. Furthermore, we are told nothing to assure us that the participants were a random sample of the population (any population you might consider relevant, whether it be the population of the local community where the study was conducted, the population of, say, the US, or the population of the English-speaking world). Was the study biased, for example, towards highly educated people, or more secular people, who might be more accepting of physicalism than the general population?

Thus, the study is suggestive, but it does not, at least as it’s reported, prove a great deal. It suggests that many people, perhaps a majority, think that free will is compatible with physicalism, and perhaps that it’s compatible with causal determinism (if we make an assumption that the two ideas tend to be closely linked in people’s minds). That’s important, in that the study at least provides data casting doubt on the view that “free will” just means, in the English language, something … well, something incompatible with physicalism and determinism. It provides one set of data, albeit not necessarily a very reliable one, that many people actually have a much more mundane conception of free will.

However, we don’t know enough to draw strong, specific conclusions about how the participants in the study approached the question they were asked. Did they answer quickly and intuitively, or did they think about it in a more theoretical way? Perhaps some thought it was sufficient that the person not handing back the $20 was able to deliberate, was not externally impeded or coerced, etc. Perhaps some thought it enough that nothing described in the scenario implied the truth of anything like fatalism (it is arguable that this is the issue that really bugs the folk … both now and historically). Perhaps, for all we know, the participants, or many of them, had rather confused conceptions of free will, involving a jumble of ideas, but at least the majority of them did not have a clear conception that “free will” just means (in part) the actions or deliberations of some kind of spooky non-material thing.

Furthermore, we should ask about the arguments of writers who insist that the meaning of “free will” – what the expression actually conveys in ordinary use in the English language – is the deliberation of some sort of non-material thing whose activity transcends any natural order of causation. How is this semantic claim grounded? What evidence do we have for it? Perhaps Mele’s little experiment is not decisive against the people who define “free will” in such ways, but their semantic claim doesn’t seem to have any more solid grounding. On the contrary, it generally seems to be an intuition based on the life experience of those making the claim, but that is hardly scientific and others may have very different life experiences. At the most it relies on highly indirect evidence, such as evidence that the folk tend to have a dualist theory of mind.

As for Mele’s point about how to interpret Libet-type experiments, this is an area where I hesitate to get involved. I don’t want to pretend to expertise that I lack. Still, I’ve never been convinced that the experiments prove much at all, and not for the sorts of reasons that Mele gives. While Mele argues about how to interpret the data, I wonder whether he isn’t chasing a mirage here. Even if he is correct about everything else, his arguments only seem to be relevant if he is trying to defend a position that we make decisions, or perhaps final decisions, entirely consciously – or at least that this happens in important cases. But, Libet aside, is that even remotely like the experience we actually have? While agonising conscious deliberations may play some kind of important role in reaching some of our decisions on difficult issues, we never seem to reach these decisions in an entirely conscious way.

Not surprisingly, therefore, courtroom advocates (barristers and trial lawyers) are trained to immerse themselves in the detailed evidence relating to a case, but not to think they can work out the case theory consciously. At least that’s my experience: they are told, in effect, to let the unconscious mind do the creative work, or much of it. I doubt that it’s different for other professions, irrespective of what practitioners are explicitly taught.

The way we reach factual conclusions, arrive at theories and understandings, make decisions about what to do next, and do so on, will generally, at some stage, involve an Aha! point (or more than one) where the answer (or some aspect of it) seems to “come to us” from the unconscious parts of our minds. Not only is this the actual phenomenology of decision-making, there are reasons to question whether we can coherently imagine or describe a situation in which every aspect of our decision-making is fully conscious – just try to do so! I defy you to. (Neil Levy has a 2005 article questioning whether the idea is even conceptually coherent, though I don’t know whether it represents his current view.)

A suitably deflated concept of free will may match up with the ideas of the folk, at least in a rough way, and may also be instantiated in the real world – e.g. it may be enough that fatalism is not true and/or that we are often able to make uncoerced decisions after an adequate time for deliberation (which may, in part, be conscious). If these are the sorts of things that the folk, or many of them, primarily have in mind when they think they have free will, then free will is not only compatible with physicalism and determinism. It also seems compatible with the role that is played by the unconscious mind. On the other hand, it may not be compatible with the role played by the unconscious mind if we always (or even typically) have unconscious desires, fears, etc., that are at odds with our consciously held values, and which we’d experience as alien if we knew about them.

In any event, no matter how much we fence and skirmish over how to interpret the Libet data, we are not going to be able to defend free will unless we have a conception of free will that allows a large role for the unconscious mind in our decision-making processes.

Leave a comment ?


  1. I’m surprised Blackford even considers polling data WRT whether “free will is compatible with physicalism.” I can understand why Mele does: he’s in Tallahassee (where I’m sure virtually any absurdity can get the assent of 27% of the yahoos–creationism certainly can), and he’s getting 4.4 M$ from Templeton. But, unless I’m missing something, the question is empirical: given a set S of data about the physical world, is some account A of free will justifiable? If so, how is polling relevant?

  2. All kinds of routine decisions tend to become autonomous and unconscious. It is the important ones that give an orientation to our lives which embody the question: do I wish to become the sort of person that would do x? These sorts of decisions are not amenable to laboratory testing. We have to make up our minds. It is this make that is significant here and this is where I suggest the discussion of free will is most fruitful. Can we live outside our era and achieve the godlike detachment that an absolute free will would require? No. Is such a detachment senseless? Yes because no particular consciousness has that ubiquity. It has a point of view. A will that is free is free within the bounds implied in the concept of having a will, i.e. being a person. A will that is free is one that uses the Criss Cross code at those important personal juncture. That road having been crossed a number of times consciously the process becomes automatic.

  3. As far as unconscious mind is concerned, I believe (from published medical studies)that the conscious mind is notified of the decisions of the unconscious mind about 1 second after the decision is reached. However it has no way of knowing about the delay and so there is usually the illusion that the conscious mind made the decision itself.

  4. The 2005 paper doesn’t quite represent my current view, because (oddly enough) Mele convinced me that I was mistaken. I was mistaken in thinking it was conceptually incoherent that consciousness (somehow) be the mechanism that makes decision. I was not mistaken, though, in thinking that it wouldn’t enhance freedom in any way were consciousness to be involved in that kind of way. I wrote this up for a popular audience for a writer’s festival presentation; it subsequently appeared in a book called *The Antipodean Philosopher*.

  5. swallerstein (amos)

    Free will may be a question of degrees of freedom.

    Many decisions are made unconsciously, but we have the ability to over-rule those unconscious decisions through increased self-awareness. However, some people are more aware than others of their unconscious tendencies, habits and ways of reacting and those people are “freer” than those who are less aware of the their unconscious tendencies, habits and ways of reacting, because their awareness of their unconscious mind permits them to correct unconscious “choices”.

    Even those who are more aware than average of the the way their unconscious mind operates still at times act or react habitually or guided by unconscious factors and thus, no one is ever wholly free, yet, as I said, some people at some moments are relatively free.

  6. Alfred Mele was recently interviewed by Richard Marshall over at 3:AM. Just thought it might interest some.

  7. On the one hand we have the semantic question of what the folk mean by “free will” and on the other the empirical question of what human agential capacities are, including the role of consciousness (phenomenal, access, Levy’s awareness and any other discriminable variety). It’s the latter question that should inform policy debates on our responsibility practices, for instance, should reasons-responsive but likely fully determined choosers be retributively (non-consequentially) punished for their transgressions? Arguing about what free will “really” means is a bit of a sideshow and a barrier to debating more substantive questions.

  8. gadflymagazine

    Oh No! What could be more substantive than the question of free will itself? It is because we do no want to address the fundamental issues that we find ourselves with seemingly insoluble ethical dilemmas. If you make a basic mistake no amount of wallpaper will cover the cracks.

  9. For a while – roughly since I started thinking rigorously about fMRI technology due to Sam Harris – I’ve been pondering how neurofeedback will effect the crude idea that “we can know if you’ve been lying”. If that’s true then I can now at the same time as you that I have been exposed, assuming I’m exposed to the same information as you. So it’s not that you can know it before me, it’s that I can now it “immediately” (i.e. at the same time as you) that I’ve been caught with my pants down. Ouch, my head is spinning…

    I understand that what is happening is that the fMRI is leaving an electronic imprint on some medium that we evaluate later. But, importantly, I am one of you. So none of you will know it before me, assuming I’m exposed to the same information as the rest of you (i.e I’m robbed of my illusion that I can deceive and that I thought would protect me from your vicious predation).

    If neurofeedback works at least to some extent, then will I alter by evolutionary necessity my underlying methodology of deceiving so that none of you and only I understand that I’ve been lying? Otherwise, won’t your predatory instincts extinguish me?

  10. Free will discussions are actually discussions about consciousness. Libertarian free will aside, everyone* agrees that physical systems drive decisions.

    If we take the above experiment** and replace a person with a chess computer making a move no-one* says the computer has free will. If the computer requires a login and stores previous games against each user and uses that information to make its next move, does it have free will? Suddenly not everyone says no, it is certainly deterministic, but it uses prior information to alter it’s decision tree and isn’t that the basis (albeit an incredibly simplistic version) of the compatibalist argument? What, then, differentiates the chess computer form a human? Surely it can only be argued the conscious mind which evaluates prior information with the current situation to decide rather than an ‘unthinking?’ computer.

    Thus, the argument between determinists and compatibalists comes down to the existence/role of consciousness.

    Or did I miss something important?

    *who has something worth contributing

    **another flaw in the experiment was having humans as actors. No matter how you frame it, many people will grant humans characteristics (soul, free will, empathy etc.) that they would not grant an alien or a chimpanzee given the same framing.

  11. Dennis Sceviour

    It is to the advantage of the dictatorial to deceive the people into believing that there is no free will. It could be argued that if Free Will cannot be scientifically verified, then it should at least be invented. This explains much about how Free Will appears in the subconscious mind, and not as an overt realism. Even assuming Free Will is a contrivance, it does not help to define it. As everyone has access to Free Will, then everyone is also free to define and interpret it, unless bounded by relative Compatibilism.

    Eastern philosophy is divided on the existence of Free Will. Japanese Zen Buddhism appears to deny its existence, but Hindu philosophy shows a compatibilist view:

    Is everything pre-ordained or are we what we are because of exercise of free will? A cow is tethered to a pole with a long rope. The cow feels she is free to roam anywhere but the perimeter of the area in which she can move is fixed. Similarly, every human being has a free will but the length of rope is governed by God (Ramakrishna).

  12. “Do you believe in Free will”?

    “Yes, I had no choice: It was preordained…”.

    Oh..and when did you stop beating your wife?

  13. Stephen Lawrence

    “”Do you believe in Free will”?

    “Yes, I had no choice: It was preordained…”.

    I think Leo’s reaction is a good example of what just about everybody thinks before reflection.

    What they think is if they couldn’t have done otherwise in the actual situation they had no choice.

    I believe I’ve seen enough to know that.

    Coyne and Harris are aware of their incompatibilist intuitions which is why they disbelieve in free will. So was Einstein. So was Darwin which is why he disbelieved in ultimate responsibility and so is Richard Darwkins which prompted him to write his edge piece on retribution.

    I see little room for doubt over the intuitions people do have but still it’s good to gather the empiricale evidence.

    I think to gather it the question needs to very clearly point to the problem. I don’t think using the word physicalism is good, nor even determinism. Pre-determinism would be better but better still is: Would you have free will if there were only one future you could get to from the actual circumstances of your birth?

  14. @ Tom Roche – the first question is “What does the expression ‘free will’ actually mean in ordinary conversation (as opposed to some stipulated definition)?” The research is meant to work out what the expression actually conveys to people. Whether it conveys something compatibilist or something incompatibilist is a fact about what meanings people assign to the words as a composite expression.

    We can then ask whether such expressions as “You have free will” or “You don’t have free will”, as understood by ordinary people, are true. The polling (well, it’s not quite polling) is to try to establish what we can about the first question.

  15. I do agree with the other Tom – Tom Clark – that any policy responses need to be based on the facts, whatever we think they are, rather than on whether we do or do not have free will as understood by the folk. Once again, free will talk may not be the best talk to use in all this.

  16. Dennis Sceviour

    The Libet experiments have glaring shortfalls. First, there is the disregard for the null set in the experiments. This is often a shortfall of scientific experimentation. The Libet tests was based on a random choice, but no measurements were made for a person who could (or would) choose not to choose.

    Second, the choices themselves were isolated to measurements based on supposed random selections. However, I suspect the same results could have been achieved if a person made choices based on an inspiration from God. Thus, no information about the relationship between Free Will and Determinism has been revealed. In my opinion, the Libet type experiments will never reveal anything. However, this does not mean that scientific verification of Free Will has failed. It may only mean that a scientific experiment on Free Will has not been discovered yet.

  17. If you parse the “experiment” it actually talks about the choices involved in keeping the money. Few would worry about free will if no actions where involved, and no ethical questions concerning those actions needed to be answered. Therefore the question becomes: what influences our choices – consciously, preconsciously, and unconsciously? Starting with the later, defining “unconsciously” as that which never reaches our awareness, free will makes no sense. That leaves the first two – with the preconcious being that process that precedes, and then becomes available, to our awareness i.e. consciousness. What happens in the automatic preconscious process, has been and will be further studied with methods such as the fMRI. It is only at the level of consciousness where freedom of choice becomes possible. Furthermore, and this is my main point, the choices can only be be made between all the imagined outcomes (by definition coming into awareness in a serial manner) and then compared. That is our only freedom – to choose between outcomes that we have imagined – and that is of course limited by all the factors in our brains that are not under our control.
    From a social point of view then, we can only judge a persons actions if we know what options where imagined before those actions took place. How we deal with persons who seem to have imagination powers that are relatively incomplete (for that social group)and therefore lead to “bad” choices, is another question. Saying a person had free will – i.e. was able to access in their imagination all the possible outcomes is just a shortcut that relieves us(the judges)of the need to consider the whole picture. Furthermore the full data, to consider all possible outcomes, is not available to any but a god-like being. That should humble all those that judge another human being by invoking “free will”.

Leave a Comment

NOTE - You can use these HTML tags and attributes:
<a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>