Free Will In A Non-Magical Universe

What might free will look like? Suppose you had a intelligent black box and wondered if it had free will, or in the horribly imprecise terms of recent discussion was “deterministic or random?” Again suppose you can copy this black box. You pose an question for your black box(es) and repeat for many iterations, posing the same question/problem each time. What then would you expect if this box had free will.

It seems to me the answer would be that the box had free will, then the box would give a distribution of answers, all of which “make sense” from a logical, creative, or emotionally reactive point of view. It also seems, given our understanding of the human brain and of what the possibilities of our universe regarding setting limitations both from a complexity (and if need be quantum) view point of the “same” initial conditions” this result is exactly what we expect from brains like ours.

What other sort of “Turning-like” tests might you pose to an array of black boxes to determine if they have free will or not that might better determine the question of whether free will might or might not make sense? Sugggestions?

8 Responses to Free Will In A Non-Magical Universe

  1. It seems to me the answer would be that the box had free will, then the box would give a distribution of answers, all of which “make sense” from a logical, creative, or emotionally reactive point of view.

    I think the ‘distribution of answers’ aspect is a problem. A Magic-8 ball will give you a ‘distribution of answers’. As to requiring the answers to ‘make sense’ from a logical, creative, or emotionally reactive POV….well you seem to be limiting yourself to human intelligence aren’t you? Might not a sophisticated enough computer system achieve free will? If it did, would it look like human free will? I suspect not. How we think is bound up with what we are. We think like humans because we are humans. I don’t see any particular reason to assert, though, that non-humans could never think or that if they did their thinking would look like human thinking. Quite the opposite I’d suspect.

  2. Also why a ‘distribution of answers’? Does not free will include the ability of someone to be stubbornly consistent? Always saying ‘no thank you’ when offered coffee maybe free will as much as someone who says yes 50% of the time!

  3. Boonton,

    Also why a ‘distribution of answers’?

    That depends on the question I assumed that a good test would have more than one reasonable answer.

    Seems to me

    Do you want coffee?

    Is not a good problem/question to pose. Except to rule out your 8-ball, e.g., “cannot predict now” is not a very good indications of an intelligent response (actually quite a few of the standard responses don’t do well with “do you want coffee?”.

    I think the ‘distribution of answers’ aspect is a problem. A Magic-8 ball will give you a ‘distribution of answers’.

    Let’s see. If the box always gives the same answer, then it doesn’t have free will because it’s mechanically giving the exact same result. If it gives a distribution its “random”. Again, suppose it has free will, how would that look? What sort of test might you do that distinguishes between free will and not-free will.

  4. Let’s see. If the box always gives the same answer, then it doesn’t have free will because it’s mechanically giving the exact same result. If it gives a distribution its “random”. Again, suppose it has free will, how would that look? What sort of test might you do that distinguishes between free will and not-free will.

    The box may always give the same answer beause it always chooses to give the same answer. Likewise if it changes its mind a lot, it may appear to give a random answer when in fact it may just have a very erratic free will.

    We’re assuming cause and effect here. Ask thousands of people some question and you’ll get a random set of answers. Yet if anyone has free will we’d say it’s them.

    I’m starting to think that maybe the free will issue is something of an illusion. The desire to assert that the ego is ‘not empty’ to use Zen Buddhist terminology. We have a desire to assert that we are something other than what we are….hence we reject the notion that the brain controls behavior, we want it to be a disembodied soul. But then what controls the soul as it controls the brain? Well it’s ‘free will’ But what is that exactly? We end up running in circles.

  5. Boonton,

    Ask thousands of people some question and you’ll get a random set of answers.

    Not exactly. What you won’t get it “I cannot predict that yet” in response to “Do you want some coffee?” That is you shouldn’t get (from people) answers which last contextual connection to the question. They aren’t completely random.

    I’m starting to think that maybe the free will issue is something of an illusion. The desire to assert that the ego is ‘not empty’ to use Zen Buddhist terminology. We have a desire to assert that we are something other than what we are….hence we reject the notion that the brain controls behavior, we want it to be a disembodied soul.

    Huh?

    Seems like where you’re going with this is that a turning like test is not (for you) sufficient to determine whether a subject possesses free will, even if you can make copies of it and retry. Absent any reasonable test that you can think of, you’ve decided that consciousness is an illusion. My reply to the last is that we’re not talking about the ego or consciousness being an illusion or not. I don’t see consciousness as we perceive it as a requirement for free will. My point is irrespective of the existence of the soul (which a fellow like JA is going to reject) man has free will by virtue of his construction. As I noted, I fail to have the imagination to figure out how a creature (BB) can have creative genius and not possess also free will. JA (and I thought you) failed to understand how deterministic physics can coexist with free will. I don’t see the problem with the two coexisting. I hold that’s his (and your?) failure in imagination. His (JAs) categories of that things are deterministic or random is a bad category, which I thought I had demonstrated by pointing out deterministic randomness and determinism within randomness (PV=nRT).

    It seems to me if a thing (black box in this case) can make decisions and those decisions are not mechanically determined trivially and it is clear that the BB’s decisions are not externally controlled (a puppet doesn’t have free will, its actions (and will) belong to the puppeteer) … then it has free will.

    As I noted the other week, however, I don’t think it follows that free will is required for moral (social) responsibility. Some people do, but in response to them, I think men have free will.

  6. As I noted the other week, however, I don’t think it follows that free will is required for moral (social) responsibility.

  7. Seems like where you’re going with this is that a turning like test is not (for you) sufficient to determine whether a subject possesses free will, even if you can make copies of it and retry.

    Well Turing doesn’t really address the problem of the ‘philosophical zombie’. It basically is a double blinded test that tells us if some ‘black box’ appears to be intelligent or have free will. Maybe that’s the best we can ever do but it does mean that there’s a very real possibility that it doesn’t. A highly sophisticated video game may have an artificial character that appears intelligent and has free will. Assuming it consistently passed your Turing test would you feel confident that it really had it, or would it just be video game designers figuring out how to take advantage of our mental biases to fool us? Turing does seem to lead us to a road of a type of relativism where we ditch the idea that there’s a real reality ‘out there’ and assume reality is just whatever our heads seem to tell us it is.

    JA (and I thought you) failed to understand how deterministic physics can coexist with free will. I don’t see the problem with the two coexisting.

    I think the problem is that free will is being left poorly defined. If you simply mean cannot be predicted, sure we have it. If you mean cannot be predicted at all, even in the form of probability distributions….ok. But things can still be deterministic and not able to be predicted. I think your test of giving ‘right answers’ in ‘context’ would suffer from cultural bias. For example, a committed post-modernist who feels that language is a form of control and oppression that must be broken by purposefully breaking its rules and throwing out non-sequitors may fail your test by providing ‘wrong answers’ that either sound like gibberish or seem devoid of understanding context. Yet he would have as much free will as you.

  8. Boonton,
    Does the “copy” answer (in which copies either answer the same or different) make a difference to the gedanken-free-will-Turinglike test? It seems to me there are those who would insist that if the answer was the same, then the subject is mechanical and has no free will or if it is different then it is random and has no free will. The problem there is that it seems to me there is no possible answer to the test that yields free will. If you pose a test with no correct answer, your test is flawed, not necessarily the subject.

    I thought philosophical zombies lacked consciousness . Is consciousness required for free will? Why?

    Free will seems simple enough to me. Free -> unconstrained by outside influences. Will -> can make meaningful choices.

Leave a Reply

Your email address will not be published. Required fields are marked *

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