Hayek on neural networks and free will

Share on FacebookShare on Google+Email this to someoneTweet about this on Twitter

A while ago, a commentator on one of my posts asked me to elaborate on my view on free will which I claimed was a useless concept. I suppose the next question would have been how my agnosticism about the existence of free will is reconcilable with classical liberalism. Well on this issue I get my bearings from Hayek and this paper by Gary Dempsey of Cato on Hayek’s views on free will and his anticipation of the ‘neural networks’ idea should answer these questions, as well being a treat for enthusiasts of both evolutionary theory and Hayek. Here is the abstract:

This paper examines the evolutionary epistemology of the Austrian economist Friedrich Hayek. I argue that Hayek embraces a connectionist theory of mind that exhibits the trial-and-error strategy increasingly employed by many artificial intelligence researchers. I also maintain that Hayek recognizes that his epistemology undermines the idea of free will because it implies that the mind’s operation is determined by the evolutionary interaction of the matter that comprises ourselves and the world around us. I point out, however, that Hayek responds to this implied determinism by explaining that it can have no practical impact on our day-to-day lives because, as he demonstrates, the complexity of the mind’s evolution prevents us from ever knowing how we are determined to behave. Instead, we can only know our mind at the instant we experience it.

Regarding the intellectual genealogy of the neural networks concept, here is a key quote:

Once a ‘thick’ net of ordering connections is established in the mind, says Hayek, a range of possible neural routing patterns is engendered. Simultaneous classification, in other words, results in “a process of channeling, or switching, or ‘gating’ of the nervous impulses” (1967, p51). Yet Hayek is emphatic that this ‘lock-and-dam’ system of neural connections does not in and of itself specify the neural routing patterns that will be employed by the mind. Instead, neural connections constitute “dispositions” (1978c, 40) and only through competition among many different neural dispositions and combinations of dispositions will distinctly functional patterns be discovered. Hayek thus embraces the view that the physiological apparatus that enables us to know the world is itself subject to the pressures of the natural selection process. This view should not sound unusual to readers aquainted with the writings of nueroscientist William H. Calvin (1987, 1996b) and Nobel-laureate neurobiologist Gerald M. Edelman (1982, 1987), and it bears special noting that Edelman and Hayek were familiar with each other’s work. In fact, Edelman cites Hayek in his book Neural Darwinism, [3] and Hayek cites Edelman in his book The Fatal Conceit.

Also note the anticipation of the memetics concept:

Hayek makes it clear that the discovery of neural rules is not going on in only one mind, but in everyone’s mind, and that the discoveries made in one mind can “infect” (1967b, p47) other minds through speech and example. As such, he argues that humans are intelligent, in part, because neural rules can be accumulated and transmitted from person to person, generation to generation. “What we call the mind,” says Hayek, “is not something that the individual was born with…but something his genetic equipment helps him acquire, as he grows up…by absorbing the results of a tradition that is not genetically transmitted” (1988, p22). [5] In other words, language, morals, law, etc., are not discovered ex nihilo by each mind, but simply constitute an epidemic of “imitation” (ibid., 24), of successful neural rules combining and spreading through populations. Under this view, “learning how to behave is more the source than the result of insight, reason, and understanding” (ibid., p21) and “it may well be asked whether an individual who did not have the opportunity to tap such a cultural tradition could be said even to have a mind” (ibid., p24).

Finally a direct quote from Hayek himself on the normative implications of his brand of materialism:

we may…well be able to establish that every single action of a human being is the necessary result of the inherited structure of his body (particularly of its nervous system) and of all the external influences which have acted upon it since birth. We might be able to go further and assert that if the most important of these factors were in a particular case very much the same as with most other individuals, a particular class of influences will have a certain kind of effect. But this would be an empirical generalization based on a ceteris paribus assumption which we could not verify in the particular instance. The chief fact would continue to be, in spite of our knowledge of the principle on which the human mind works, that we should not be able to state the full set of particular facts which brought it about that the individual did a particular thing at a particular time (1989, pp86-87).

Update
Martin in Comments asks if the ideas above have any implications for the criminal justice system. Well, the short answer is no. Hayek was prominent in the resurgence of classical liberal thinking so it obviously didn’t affect his *normative* views. As noted, Hayek’s resolution of the dilemma is that we are unable to behave as if we have no free will so it doesn’t matter. When I say the concept of free will is ‘useless’ perhaps I should have been more exact – it is useless or meaningless for positive theory. I always thought that as a materialist this conclusion should not be very surprising and that all materialists would be comfortable with this implication. However free will may be a useful metaphysical fiction for public policy purposes.

That is , for normative purposes and everyday purposes we treat people *as if* they have free wil (and in fact the latter is unavoidable except for autistics)l. When we say that X is of sound mind and we hold him responsible for a crime and we punish him, how can we distinguish that from cases where we don’t punish X because he isn’t of sound mind?

Well, a punishment is basically a price signal we send into the environment of people contemplating the costs and benefits of the act similar to acts commited by X. The main aim of the justice system from a utilitarian perspective is deterrence. Let’s say the objective of the system is to minimise the incidence of particular acts. Absent the punishment more of those acts would be committed, with the punishment a sting is introduced which all people otherwise inclined to act like X in future will then be forced to take into account – none of this requires free will in principle. X may decide not to rob a bank because it’s too much hassle – whether this hassle comes from the fear of the law or that he can’t afford to buy the right mask or doesn’t know how to handle a gun properly is irrelevant from the perspective of positive explanation. Similarly we don’t need to believe that plants have free will when they grow a certain way because of the direction of the sun.

What do we mean when we conclude that X can be held responsible for his crime and Y cannot because Y is schizophrenic? Well it means this extra price of robbing a bank we introduce into the system isn’t going to be inputted by people of the category of Y into the calculus of costs and benefits in a way that would deter them from robbing banks because the signal receiver is too jammed. Again, no real need to *explain* the differences in doctrine in terms of free will.

15 Comments

  1. Very interesting stuff. I’m not sure free will is a useless concept, however-i.e, how do we reorganize the criminal justice system to comply with our new paradigm that murderers are simply the “result of the inherited nervous system… and external influences.” Did the World Trade Center really fall simply because Mohammed Atta’s genotype interacted with some zealous religious instruction? Of course, I realize you believe we have no choice in how we reorganize the criminal justice system, but I’d be interested to hear your preordained comments anyway.

  2. Once again, one has to distinguish between the causative and the descriptive. Description is not cause. To describe all neurological processes that are involved in an human act or thought is not the same as identifying was causes the act or thought.

    This is a case where science needs the guidence of some basic philosophy.

    And even if free-will is relative, or illusory, it is something that, unless we have a sense of having it, we become dysfuntional and depressed more than when we don’t have a sense of it.

  3. Even some purely physical processes are unpredictable. The predictive-determinist paradigm is pretty much dead everywhere. As you go higher up the great chain of being, predictability becomes increasingly less. Prediction of human behavior is really quite limited, usually dealing with averages of groups or prediction of broad categories of individual behavior (e.g. schizophrenia, knowledge of which does not allow you to predict most behavior of individual schizophrenics).

  4. Martin
    I thought the answer is pretty obvious from Hayek’s resolution of the dilemma. We are unable to behave as if we have no free will so it doesn’t matter. When I say the concept of free will is ‘useless’ perhaps I should have been more exact – it is useless or meaningless for positive theory. I always thought that as a materialist this conclusion should not be very surprising and that all materialists would be comfortable with this implication but I shouldn’t presume you are one. However free will may be a useful metaphysical fiction for public policy purposes though even this is doubtful but it does make exposition easier. That is , for normative purposes and everyday purposes we treat people *as if* they have free will. When we say that X is of sound mind and we hold him responsible for a crime and we punish him, how can we distinguish that from cases where we don’t punish X because he isn’t of sound mind. Well, a punishment is basically a price signal we send into the environment of people contemplating the costs and benefits of the act similar to acts commited by X. The main aim of the justice system from a utilitarian perspective is deterrence. Let’s say the objective of the system is to minimise the incidence of such acts. Absent the punishment more of those acts would be committed, with the punishment a sting is introduced which all people otherwise inclined to act like X in future will then be forced to take into account – none of this requires free will in principle. X may decide not to rob a bank because it’s too much hassle – whether this hassle comes from the fear of the law or that he can’t afford to buy the right mask is irrelevant. What do we mean when we conclude that X can be held responsible for his crime and Y cannot because Y is schizophrenic? Well it means this extra price of robbing a bank we introduce into the system isn’t going to be inputted by people of the category of Y into the calculus of costs and benefits in a way that would deter them from robbing banks because the signal receiver is too jammed. Again, no real need to *explain* the differences in doctrine in terms of free will.

  5. Hayek was prominent in the resurgence of classical liberal thinking

    pardon my ignorance.. but what exactly is classical liberal thinking?? is it the opposite of liberal thinking?? i though Hayek was a pure capitalist.. and i’m sure that Hayek is celebrated by neo-cons…

  6. if he was a pure capitalist he would not be celebrated by neo-cons ;0) in fact hayek was far less capitalistic than say mises-he accepted the insertion of government into the works of the economy to some extent-but it was an ordo-liberal sort of view-taxation is OK but nationalization is bad.

  7. “We are unable to behave as if we have no free will so it doesn’t matter.”

    Yes. I agree. Conversely-this also renders determinism equally useless. As William James would say, the resolution of this debate lacks “cash value”-but really, you have to talk about something.

  8. Ms Gumbo – classical liberalism is the appropriate term for philosophies favouring limited government. The only reason we have to use the appendage ‘classical’ is because you silly Americans got things topsy turvy to the point where people who call themselves conservative are really classical liberal and people who call themselves liberal are socialist.

  9. >Conversely-this also renders determinism equally useless.

    Martin – read the article, it depends on what you mean by determinism. Hayek was a materialist who did not find the free will concept helpful but he wasn’t a determinist. However the relevant issue here is that there is no logical basis for free will – this is closer to the truth than the opposite picture of reality. Make of that what you will, or are you saying ‘truth’ for normative purposes trumps truth for positive ones?

  10. this is like the movie The Matrix, with many levels…we act and feel as if we have free will, but only because through sheer determinism of the evolutionary process it is forced on us. It’s like thinking about the famous two-slit experiment in physics, if it makes sense you really don’t understand it.

  11. By useless, I mean what Godless said-no predictive power, i.e. whether my next actions are determined, random, or subject to my free will, either “truth” is equally useless in predicting that action. No cash value in the knowledge.
    I’m not sure what you mean by “positive truth”-rather than base firm conclusions on soft definitions, please define that term.
    Meanwhile, assuming arguendo free will is an illusion-why did the concept of free will evolve and seemingly invade almost every human brain?

  12. “why did the concept of free will evolve and seemingly invade almost every human brain?”

    Daniel Dennett and others have said that the “illusion” of free will has evolutionary advantage. If that’s so, wouldn’t actual free will have an even greater evolutionary advantage?

    The problem of free will from a materialist perspective is that materialism cannot explain how free will may arise, so it a priori discards the notion. I think it would be useful to devise a sort of Turing Test for free will and say that if a person, animal or thing exhibits such and such behavior – acts as though it has free will, then it has free will.

    Because we can’t explain it from the materialist paradigm and may need to resort to “sky hooks” doesn’t discredit the idea or make it useless.

  13. right justapolak-of course materialism is also faced with the problem of defining “matter,” though most just slide right over that little foundational problem in their rush to construct their edifice of “no nonsense” truth.

  14. When thinking about free-will we have to consider some things: a) there is no “homunculus” or center-stage-of-being where decisions are made. That is just a feeling. Split brain experiments can easily divide the mind and dupe the conscious part of you into inventing stories to explain why you followed a hidden instruction. (in other words the “central stage” is an easily discovered illusion) …so ‘freedom’ is zero in that sense. b) cause and effect: if all brain (machine) and environmental (stimulus) variables could be accounted for (and perhaps random quantum ones as well) predictive power would raise to %100. In other words, it is assumed that the brain may be very complex, but still runs on a set of evolutionary defined instructions par stimuli. …So ‘freedom’ is zero in that sense as well. (and I reject the McKibben claim that freedom is simply what we are ignorant about)

    So in what sense is man free? I think, fundamentally, life itself* and freedom are synonymous; more specifically intelligence (beginning at its most rudimentary form) and freedom. Man is more ‘free’ than a chimp which is more ‘free’ than a cat, which is ‘more free’ than a fish or cricket. But all are “free” in some sense that they all have individual personality. Even fish and bugs develop unique behavior or ‘individuality’ to a certain extent. In a somewhat ascending order each level of organism has more complex creative behavior and goal formation and more variables are needed to predict its behavior. Now among men ‘freedom’ has a different meaning. This sense of the word is the most important one to me. Those with better methods for thinking are more free than those who have worse. Someone who understands the scientific method is going to be more in command of himself and his environment than someone who doesn’t. A skeptic has more genuine freedom than a rube.

    why did the concept of free will evolve and seemingly invade almost every human brain?

    So if free will is so useless, then why did we evolve the sensation of it? Well there are many unknowns here but an important thing to remember is that just b/c something doesn’t do what we thought it did that doesn’t make it useless. The best explanation I’ve seen for why there is a conscious sense of self is that it is a by-product from the important survival need to understand other living things. “The theory of mind” is the innate understanding of other living things as autonomous actors with intention, goals, motivation- complex internal thought (and I’m talking about all animals). It is the basis for not only social behavior but learning behavior (humans are born imitators). So the brain formulated a tool for trying to figure out what other living organisms are thinking (for prediction and learning), and that “sense of mind” just naturally gets applied to our own minds.

    *and by ‘life’, I’m not necessarily speaking of organic life. I think man has already created AI at least as sophisticated as much lower order life, and probably more so.

  15. Jason M writes “ there is no “homunculus” or center-stage-of-being where decisions are made. That is just a feeling.”

    Well then where is the “homunculus” which experiences feelings and the “conscious part” which can be duped?

    Discussions about free will often strike me as confused, impractical, quasi-metaphysical ramblings that go nowhere. I accept free will simply from self-inspection and don’t get caught up in its implications for whatever metaphysical notions I may adhere to, and materialism is a metaphysical notion – an extrapolation far beyond what’s justified by the evidence.

    “if all brain (machine) and environmental (stimulus) variables could be accounted for (and perhaps random quantum ones as well) predictive power would raise to %100.”

    That’s a big IF and only possible from a god’s eye perspective. Who even knows if it’s possible in theory. As I understand it, quantum phenomenon are inherently unpredictable and can only be predicted probabilistically. It’s only at the macro level that phenomena appear to be deterministic. This allows science to make accurate predictions and technology to manipulate matter and energy, but it does not justify extrapolating determinism to all scales.

    “A skeptic has more genuine freedom than a rube.” What should we be more skeptical of, free will or the materialistic determinism that has us denying our most basic intuitions?

a