Get That Chip Out of My Brain!

There has of late been some discussion of free will and determinism, and particularly the relative merits of compatibilism versus incompatibilism, at various blogs. (See, for example, here, here and here.)

I must confess that I’ve not followed these discussions closely, despite having a longstanding interest in this issue (see here and here, for instance), so I don’t really have anything substantive to say about the debate, except, I guess, that I’m inclined towards the sort of incompatibilism espoused by Jerry Coyne (my hands were strangely reluctant to type that).

However, this does seem like an opportune moment to ask the readers of Talking Philosophy for their advice and opinions about an interactive activity that I put together at Philosophy Experiments, which explored some of these issues through a look at a Frankfurt Case and some other stuff. It’s here:

Get That Chip Out of My Brain!

Thing is, I programmed the activity about six months ago now, but I was never happy with it, and haven’t added it to the front page of the site (it’s been played quite a lot because of traffic that comes in via Google, etc).

Basically, my view is that most people will find the stuff about “Transfer NR” (John Martin Fischer & Mark Ravizza) confusing and philosophically suspect – it seems tricksy – and I tend to think that I ought to rewrite the whole activity, focussing on the Harry Frankfurt stuff, which I think works much better.

If anybody felt inclined to play through the activity (it’ll only take a few minutes), and let me know if they agree, disagree, or have any other thoughts, that would be really helpful. If it turns out that even a few people think it doesn’t work, then I’ll almost certainly rewrite the thing (because I think there is a good interactive exercise in there somewhere, but I’m not sure this is it).

Leave a comment ?


  1. I don’t see the chip contradiction. There are two possible definitions of ‘could have done otherwise’. (i.e., was actually coerced/unable to deviate or merely potentially coerced). You conflate the two.

  2. Dennis Sceviour

    “Yes, it’s true…”
    This is one of the toughest questions in philosophy, and my opinion waxes and wanes as to whether free will or determinism is best. Some philosophers (Michael Ruse for example) hold that moral claims are not true or false, moral claims are right or wrong. Would the nature of the question change if it was re-worded:

    “Yes, it’s right that a person can only be morally responsible for what they have done if they could have done otherwise.”

    “…if they could have done otherwise.”
    This is a big if. Once an action has been done, how does one prove that it could have been done otherwise? The proof is speculative and involves interpretations of intent, not action. That is, while the free will theory is entertaining, the deterministic view has the stronger theory.

    Mad Professor Coine with the microchip is science fiction, and thus in my opinion may lead to erroneous results. A more realistic example would be appropriate. For example, the assassin had been threatened with extortion if he did not carry out the assassin. Or, the assassin had been enlisted by a commando group and trained (brainwashed) to be a mercenary killer. Or, the professor had trained attack dogs to kill. Would the dogs be morally responsible?

  3. As always in these discussions, there are (at least) two senses of moral responsibility we need to distinguish.

    The first is to say that X “is morally responsible” (in a vague, but perspective-independent, universal sense). Determinism seems to logically preclude this sense of moral responsibility.

    The second is captured by “I choose to hold X morally responsible”. It seems that the compatibilist can embrace this sense of morally responsible despite determinism being true. In this second sense, “morally responsible” is simply a description of my attitude toward X’s actions. If my attitude will alter X’s future behavior (or Y’s, or Z’s), then it’s no contradiction for me to hold her “morally responsible” even though she could not have acted otherwise.

  4. s. wallerstein (amos)

    I agree with Dennis that the microchip narrative is science fiction and hard to think about.

    As he says, brainwashing on an impressionable, weak person, perhaps in the military or in a terrorist group or in a fanatical religious sect, might be easier to reflect upon.

  5. Guys – The alternatives don’t work (at least not in a straightforward way), because the whole point of the Frankfurt Case is that the chip is never needed. In other words, the chip ensures that it isn’t possible for Huntington to act in any way other than he does act. But, as it happens, the chip isn’t required, because Huntington forms his own intention to kill Bunny Amore.

    So what it seems to show – though I think it doesn’t – is that the fact that somebody cannot act in any way other than they act doesn’t preclude them being morally responsible.

    Harry Frankfurt does discuss things such as threats, hypnosis, special poisons, and directly manipulating a person’s brain, but any of these would have to occur with entirely certain effect, and only at the point at which the person we’re considering deviates from any intention to commit the act in question.

  6. Dennis Sceviour

    “Huntington forms his own intention.”
    How do we know this to be true? This is the key element to answering the question, and just an assumption as saying he acted of his own volition is not sufficient to make a decision between free will and determinism. The question assumes what it is trying to prove; that is, forms his own intention, acting on his own volition, and free will are all synonymous phrases.

  7. @Dennis

    Sorry, I’m not sure quite what you’re saying there:

    a) I’m not sure what you take to be “the question” (the question Frankfurt is interested in is whether you can have moral responsibility even in the absence of the ability to do otherwise)?

    b) It’s only an assumption on one understanding of “forms his own intention”. I simply meant that he wasn’t brainwashed, coerced, poisoned, the chip didn’t fire, etc.

    c) One can still quite happily think he formed his own intention in that sense (i.e., wasn’t brainwashed, etc); and yet be both a determinist and incompatibilist (determinist, because it was always the case that he was going to form the particular intention he formed; and incompatibilist, because that fact means that we lose all the stuff we want in place – particularly to do with moral responsibility (obviously Frankfurt is denying that this is true) – when we talk about free will);

    d) So, in that sense, the statement (I’m guessing that’s what you’re talking about when you say “the question”), doesn’t assume what it’s trying to prove (because it leaves open both compatibilism and incompatibilism).

  8. Dennis Sceviour

    a) Yes, that is the question.

    b) We have to agree on the meaning of “forms his own intention.” If it is not free will, then how else could it be formed? I cannot see any other assumption possible, as you have excluded external influences. If there are external influences, as a realistic situation would probably entail, then it would form part of the scenario, and leave open the question of compatibilism.

    c) I am not sure what is meant by “it was always the case that he was going to form the particular intention.” I suppose it is vague enough to leave open both compatibilism and incompatibilism. We must agree to disagree on the Frankfurt microchip examples since I hold that a fictitious example will probably lead to erroneous results.

    d) I agree the question as given leaves open incompatibilism. However, something extra (besides microchips) is necessary to leave open the question of compatibilism.

    I hope this brief critique is of some use to you.

  9. Hey Dennis

    b) I can’t even make sense of contra causal free will, let alone agree that “forms his own intention” has to be understood in those terms. But I’m not sure there’s anything at stake here (at least not in terms of our discussion, since I think we basically agree with each other). I was just using the term to distinguish the situation where the chip doesn’t fire, Huntington isn’t poisoned, etc., from the situation where it does, he is, etc. If you want to argue that it’s not appropriate to say that he “forms his own intention” in the former cases (where the chip doesn’t fire, etc), then…. well, I’m not sure you’re right (I think one might reasonably use that expression even if determinism and incompatibilism are true, because the entity that “forms the intention” just is the physiological organism that functions deterministically, etc), but it doesn’t make any difference to any substantive position I hold (because given the truth of determinism, it results in an incompatibilist position, to which I’m committed anyway). Also, in terms of the activity, if you’re positively claiming that determinism is true, and that because it’s true it makes no sense to think that people form intentions, then presumably you would just respond that Huntington was not morally responsible for what transpired (which is how I would respond).

    c) Okay, well I think that’s just a more general rejection of thought experiments then.

    d) Yes, I agree. I don’t think the Frankfurt cases work. But… a lot of people would disagree with us.

  10. I got zero contradictions. Where’s my prize?

  11. @Ben – Your prize is that at some point in the future you’ll get to talk to me about meta-ethics again. Now won’t that be a treat for you…! 😉

  12. I had no contradictions. I make no claim to having anything like a great brain, just sort of normal I suppose. However it seems to me that from the first question onwards one sets oneself on a path. Answers to subsequent questions I check for compatibility with replies to the earlier questions. I do in this connection, have sincere viewpoint, and am not just trying to get it right, my replies were what I thought it best to do. The test took some time with much pondering and trying to diagram what was happening with some use of Euler’s diagrams.
    I can see no defects in the test, and enjoyed doing it, especially as ethics is not very high in my philosophical interests.

  13. Dennis Sceviour

    Correct me if I am mistaken, but you may be mixing continuity with philosophical compatibilism. (If I am not correct, my apologies)

    I dislike how some philosophical writers like to highjack terms and create different meanings. I stubbed my toe recently on two different meanings for cognitivism.

  14. Like Don and Ben, I get zero contradictions but I did feel, as Don wrote, that my response to the first question pointed out how I “should” answer the other questions. I don’t think I did this deliberately to avoid contradictions — in other words, I think my responses were honest — but subconsciously…I don’t know. I felt this more with this activity than with the others but I also found it very interesting. The discussion at the end seems to be very consistent with how I think about free will and responsibility.

  15. In setting out an experimental philosophical test for free-will you have a choice (!).

    This is in the respect of the question: are you seeking to find data to champion one theory or another, or to see if there are multiple modes present within us (and what factors determine the mode used)?

    Personally I suspect that out of the 4 modalities of free-will: false/true; and, determinism: false/true. You will find circumstances that ALL cases are found to be applicable on occasion.

    That is to say we have free-will and a non-deterministic CNS basis (metaphysical libertarianism) for some decisions (e.g., voting in a general election), and on occasion we behave without free will in an autonomous way utilizing CNS functionality that is best described as deterministic (e.g., pressing the brake of a car when a child runs into the road). Indeed there is room for two stage models and other variants perhaps.

    Saying that we can not conceive with our understanding of the physical processes a non-deterministic basis for the CNS (disregarding Quantum Mind arguments) is not sufficient reason to dismiss the possibility of a non-deterministic CNS. That would be like dismissing the possibility of a bee flying because we can not work out how it manages it.

    The analogy is like have cache, RAM and Disk based memory on an optimized performance computer. Having evolved a mixed strategy may be best.

  16. Don and Keith

    Thanks guys. I think it’s inevitable that there is a sense in which one’s early responses will prime one’s future responses. But, as you both suggest, pretty much all one can do is to strive to answer honestly, etc.

    Maybe the activity isn’t quite as flawed as I suspected (notwithstanding that some people just don’t think thought experiments are very useful).

    @Martin – Well, it really wouldn’t be like dismissing the possibility of a bee flying, since the problem with non-determinism (notwithstanding QM randomness, etc) is that it violates the principle of causal closure; in other words, it’s not merely that it violates particular physical laws.

    Also, it just isn’t true that we don’t have any idea how a bee manages to fly (although this doesn’t alter the point you’re making, obviously). For example:


    If my attitude will alter X’s future behavior (or Y’s, or Z’s), then it’s no contradiction for me to hold her “morally responsible” even though she could not have acted otherwise.

    That can’t be right (or at least not all there is to it). My attitude towards my cat might well alter its future behaviour (if I tell it off, for example, or don’t feed it), but it certainly isn’t in the least bit clear it makes sense for me to hold it morally responsible.

    Also, if moral responsibility is merely a matter of my attitude towards some action of X, and some attempt to alter their future behaviour, it would seem to follow that it would be entirely permissible for me to hold them morally responsible for actions that are coerced, occur under the influence of poison, etc, etc… which is, well… problematic, I’d say.

  17. @Jeremy,

    Thanks for the link about the bee’s 😎 . Of course it was true that bee’s flew before we worked it out is my point, like black swans existing before we discovered them.

    Alot of philosophers, perhaps not as up on their physics as some, fixate on quantum uncertainty as a means of achieving acausal non-determinism, I too think that is a red herring. My bet, hunch, is that the mechanism is more likely quantum entanglement that does provide the possibility of acausal connection/synchronicity across space-time. It is what Einstein thought not likely, saying it was spooky and violated his preferred worldview of locality (though Bell’s inequalities and subsequent experiment shown that reality is more of an idealism and non-local). People remember his quote about “God not playing with dice” and think this is the same point, it is not.

  18. For a recently revised view on Quantum Approaches to Consciousness, see…

    And for Free-will and the Quantum system behavior…

  19. PS: it is hardly the case that this point of view is modern afterall!

    “If the atoms never swerve so as to originate some new movement that will snap the bonds of fate, the everlasting sequence of cause and effect—what is the source of the free will possessed by living things throughout the earth?”

    ~ Titus Lucretius Carus, Roman philosopher and poet, 99–55 BC.

  20. Bad Robot: Huntington intends something he can’t not do; but he still has a choice and intent! He is sort of in a continuum between a robot and a free agent—to the extent that he is determined, he cannot choose (hence is not morally culpable—but neither is determinist society for prosecuting him—society is no more free than individuals in that case)—but the scenario says he does have a choice and intent alongside his powerlessness to do otherwise: His identity has been split into a real Huntington (the free agent) and an automaton (the determined one) that is not really him. I’d argue that it isn’t really Huntington who would kill if programmed to kill—even though the real Huntington did intend to kill. If, once Huntington chooses not to kill in this scenario, he is then forced to and he would no longer be a free agent, and hence is not really Huntington-the-free-agent. I.e. either Huntington kills or a robot kills—they are not the same “person.” The robot is NOT Huntington—and thus either (a) Huntington intends to kill (and is culpable) or (b) he intends not to kill, and is then another entity entirely (the bad robot) who kills. In reality, we are both determined and free, imo, often a bit to blame and a bit innocent too: grey-matter rather than black and white.

  21. To summarize/revise that rather convoluted paragraph: If Huntington chose not to kill, then it is not Huntington who kills, but someone else (the scientist who installed the chip) who does the killing. Huntington has a choice to kill or to cease to exist as a free agent. One body doing the killing, two free agents with intent. No contradiction here?

  22. @Martin – The idea of atoms “swerving” precedes Titus Lucretius Canus by about 300 years; it’s Epicurus’s – fairly arbitrary – attempt to escape the deterministic implications of Democritus’s atomism.

    @JD Casten – Yes, I think that’s a defensible position. A little while ago, somebody sent me an email making the same argument. They expressed it this way:

    Finding Huntington morally responsible, while also stating that people are not morally responsible if they couldn’t do otherwise, is not necessarily contradictory. The catch lies in the notion that, if the chip fired, then it is not Huntington who is acting because his will is no longer involved. Therefore, I would say that there is a sense where “he” can do otherwise, because when the chip fires “he” stops acting and it is the chip and its creator that become the responsible actors.

    I think the way to falsify the Frankfurt case is: (1) by pushing the idea that it’s the firing of the chip that constitutes the “can do otherwise”; and/or (2) by pushing the idea that de facto the chip is a red herring: it doesn’t fire, so all one is left with is somebody acting in a manner that was determined from the moment of their birth.

  23. Jeremy, you said (and oh how I wish I could make the blockquotes work):

    “That can’t be right (or at least not all there is to it). My attitude towards my cat might well alter its future behaviour (if I tell it off, for example, or don’t feed it), but it certainly isn’t in the least bit clear it makes sense for me to hold it morally responsible.”

    I have two responses to this. The first is that cats don’t respond to our holding them “morally responsible” whereas people do (though see also my second response, below). Cats do respond to being told off and having their food withheld, but that’s not the same thing as responding to my moral attitude toward it. A healthy human will know, or try to know, what my moral attitude toward her is, even if I don’t mete out an explicit punishment or reward, and then adjust her behavior. And as a healthy human, she’s quite good at picking up on subtle signs of my attitude toward her, so in order to influence her, it’s best if I actually feel the attitude (opprobrium, or approval, whatever…) This is all it takes to understand why we have the “moral emotions” that we have.

    The second response is that cats might actually respond to being held morally responsible, though in a very limited sense. That is, if I catch the cat doing something bad (and BTW how is it that I can catch it doing something bad if it’s not to some extent morally responsible? If a rock rolls over my car it’s bad, but I’m not “catching the rock doing something bad”) I think some cats might actually respond to my “moral attitude” toward them. I don’t know cats well enough, but I think many cat owners do act very much as if their cats respond to their attitude. (As my mother used to say to her cat: “Cat, you are the devil incarnate!”) Now maybe this qualifies as responding to being held morally responsible or maybe not, but it does seem that there’s a continuum of this capability. Dogs are better at it than cats are. Sociopaths are really bad at it. Rocks are horrible at it. And the more of this capability you have, the more sense it makes for me to hold you morally responsible.

  24. @Thos

    Congrats on the blockquote! 🙂

    I actually think the question of whether a cat responds to being held morally responsible is quite difficult. But we can probably agree that it doesn’t do so in any sort of conceptually sophisticated way (and likely to the extent that it does respond, it’s just behavioural).

    But the cat thing is still illustrative. The point is that if, as you hold, “morally responsible” is simply a description of my attitude towards somebody, and an expectation that I’ll change their behaviour as a consequence, then what we can legitimately hold people to be morally responsible for seems to be entirely arbitrary.

    Not least, certainly in the form you have it, it seems possible that I could hold somebody morally responsible for things that they haven’t actually done. It’s possible you’ll argue that it’d be hard to alter people’s future behaviour in this circumstance, but actually I don’t think it would be – it’s entirely possible you’d reduce the likelihood of some behaviour x by holding someone responsible for doing x when they haven’t actually done it. (Indeed, it’s possible you’d even make them feel guilty about x, etc).

    If you’re interested, Richard Double and Ted Honderich have played out a version of this argument. Honderich argues that how we respond to the truth of determinism is just a matter of attitudes. Double counters that this means that anything goes:

    If our responses to determinism are simply attitudes, and if attitudes are neither nor false, then determinism has no logical, moral or psychological consequences in the sense of correct conclusions that have to be accept by all rational persons. There are only reactions to determinism, none of which is objectively better than all the rest.

  25. Re Dennis Sceviour January 9, 2012 at 8:04 pm :
    I am wondering if you could perhaps point out in what I wrote how Compatibilism and Continuity are conflated. I used the word Compatibility in the sense that I may be contradicting something I had already said. I had no thought concerning the theory that Free will and Determinism can be viewed as somehow not mutually destructive. I do not see that I mentioned the word Continuity but it may be implied by my method of ascertaining if it be possible to reason from one stage to another, without encountering a hazard.
    In my opinion the frequent occurrence of different meanings for a term is one of the bugbears of philosophy, (as you say the Hijacking of terms) resulting in what are arguments merely about verbal differences. This should not happen if definitions of terms are initially stated by the disputing parties. For instance where arguments in identity are concerned it is not always clear if ‘The Same’ means The Same in all respects or The Same in some respects.

  26. Dennis Sceviour

    Re: Posted by Don Bird January 9, 2012 at 7:16 pm
    “Answers to subsequent questions I check for compatibility with replies to the earlier questions.”
    I interpret this to mean that you checked for continuity between questions. Thus, I conflated compatibility and continuity. Sorry for the misunderstanding.

    Re: Posted by Don Bird January 10, 2012 at 12:16 pm
    “…the frequent occurrence of different meanings for a term is one of the bugbears of philosophy”
    I agree. I have put it on my list to remember to avoid using the bugbear “same”.

  27. it seems possible that I could hold somebody morally responsible for things that they haven’t actually done.

    Yes, I suppose that’s what I would say, if it were possible to read minds. If, for example, I read your mind and ‘heard’ you plotting to torture a kitten for your own pleasure, I would, indeed, adopt a negative moral attitude toward you. What matters, when holding someone morally responsible, is knowing what their behaviors are likely to be (more commonly conceived as “what type of person they are”). Most often, we figure this out by observing what their behaviors have been, but in theory I would have no objection to holding someone morally responsible if I were absolutely certain that they were about to do something morally reprehensible. I could even imagine a future gameshow where we monitor peoples’ intentions, put them in artificial situations, and ring the bells and flash the lights just before they act, and reward them for having the right intentions or punish them for having the wrong intentions. Geez–I should go pitch that to a cable network!

    Honderich argues that how we respond to the truth of determinism is just a matter of attitudes. Double counters that this means that anything goes

    Thanks for introducing me to this, and your patience with my untutored musings. Without having read H&D, I suspect that I’ll side with Honderich. I certainly wouldn’t agree that relativism toward determinism means “anything goes”. Morality doesn’t have to be ‘objective’ in the sense that I think Double is implying in order for us to sensibly adopt moral stances and prefer some moral stances to others. This fits with my position as a compatibilist, so I claim +1 philosophy points for consistency.

  28. @Thos – This is going to get very complicated, very rapidly. I’m likely going to reply to you in a blog post (because I think the issues your stance raises go to the heart of why it is problematic to think ideas of moral responsibility, etc., survive determinism + incompatibilism).

  29. @Jeremy,

    Yes, indeed…

    “The idea of atoms “swerving” precedes Titus Lucretius Canus by about 300 years; it’s Epicurus’s – fairly arbitrary – attempt to escape the deterministic implications of Democritus’s atomism.”

    And at their time it was arbitrary. But with our current understanding of quantum entanglement (specifically that our conscious decision to observe a system at one point in space-time results in instantaneous changes to an entangled system remote – perhaps at the other end of the universe) we have means to “swerve their path”. What these classical philosophers noted was for free-will we need non-determinstic acuasal (in the sense of non-local) behavior, they pinpointed it. Subsequently we have found a mechanism to provide it. Conway et al (see my previous link) illustrate this is needed and viable. We have not concluded on a mechanism within the CNS that would support it, though candidates are proposed as hypotheses (see my previous link to the Stanford entry).

    What we can reasonably ask ourselves is, “would free-will be a evolutionary advantage?” and, “with the demonstrable means of quantum entanglement could a biological system evolve to utilize it in the CNS?” (see )

    If the answers to these questions are in the ballpark of, “possibly yes” then it’s a very promising area of research for consciousness studies and the philosophy of mind/free-will.

    It is simply (IMO) no longer viable to say there is no physical basis to argue for free-will in the CNS.

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>