If you see a difference, then please explain the difference between the ability to choose and free will.

I have noticed a tendency of many people to suggest that the ability to make a choice and the property or concept of free will are two different things.

I want to be clear: I am not implying that they are or are not. I am curious to understand how people view this. It seems that you could see:

A. No difference
B. A distinct difference
C. A distinction without a difference
D. Something else

I would love to see your explanation, no matter what your answer. I have no interest, in this context, in what any philosopher you can quote had to say. I am curious about your understanding and your ability to state it. 

Views: 361

Reply to This

Replies to This Discussion

"Could you cite your sources please?"

Can you cite yours? Burden of proof: You made the claim that quantum mechanics "proves" that the universe is fundamentally random. Where did you get that idea, other than from the popular (lay) interpretation of quantum theory? I don't think you'll find any quantum physicists making that claim. You might want to ask one.

(Hint: There's an important assumption that quantum indeterminacy is based on, and that assumption has not been proven true. This fact is well-known among physicists, and is not new or controversial.)
There's a few things I didn't understand. For instance, "post-determination" . . . I can't make heads or tails of that word. I'll have to wait for clarification on that one.

Have to get going - but that one was a response to something that has not been elegantly answered for me actually (except Wonderist ... in a way) - how is determinism not pre-determinism?

For example - since we know all the things that determine the moon's orbit, unless something drastically changes in the next five minutes (or five minutes from when you read this - I've already written it by then) the position of the moon is predetermined whether anyone predicts it or not. It's not whether we predict it or even know how to predict it. It's that all the factors that will place the moon where it will be in five minutes are determined in their own right. So, in fact, if something did change where the moon should be in five minutes, well, that was determined by other factors in advance of the the event as well, and the moon still ends up where it was going to end up no matter what.

So it is with us - even our thoughts, and which ones we act on, and which ones we don't, and precisely when we act on them, etc. Therefore, if everything is determined, it doesn't matter where you are on the timeline, the future is as set as the past.

Even no argument is made against that, there are determinists who don't want to say that anything is predetermined (maybe simply because it starts to sound like some god did the determining? - which isn't what I'm suggesting at all.) But, even if 'choice' is a more complex determinant equation than, say, the path a tornado takes - if free will is defined (as many do, by the way) as no different from free will - then choice is an illusion - or just another mechanism of determinism.

Some someone - and at this point I don't remember who - said that things are 'predetermined' but not 'post determined.' Frankly, I still don't get that - unless undetermined factors are coming in to play.
Right. All I'm saying is, just because Aristotle thought the heavenly bodies moved in crystal spheres doesn't mean they weren't moving in a different way - determined by things we didn't predict.

What this discussion has actually shown me is that some people think 'free will' and 'choice' mean something different and some don't. Frankly, deterministically speaking, whether you make a distinction or not, on some level, choice is an illusion as well. Because of uncertainty, chaos, Schrodinger's cat - or the 'simple' magnitude of complexity itself, choice may well remain an impenetrable illusion.

What I wonder is this: If one of the determining factors is my inability to predict (in other words - the fact that I am forced to guess) how deterministic is my guess. That is to say, is there any actual randomness in the guessing part of the equation?

You see, computer programs don't have random factors. They use pseudo-random strings (strings generated by a known algorithm that look so random - there is a 'test' for this - that the algorithm is nearly impossible to reverse engineer given the string.) We don't know how to generate true random. If a human mind does contain truly random determinants - then in that way alone, it currently differs from a computer program. If there is no such thing as a truly random event, then it is not a difference in and of itself.

So what I am putting out there for inspection is the idea that we could be right about something partly by an accident of an absolutely unpredictable event.

Wonderist, among others, suggests that 'chaos' can be seen as both deterministic and completely unpredictable - which, if true, would make choice something more than just an 'impenetrable illusion.'
"on some level, choice is an illusion as well"

Only if you insist that choice can only be 'real' choice if it is non-deterministic. If you accept a simpler version of choice such as simply selecting from options based on criteria, then purely deterministic choice is not an illusion. All you have to do is give up the unjustified assumption that deterministic choice is not 'real' choice. Again, you've not addressed this point, even though I mentioned it several times already.

"That is to say, is there any actual randomness in the guessing part of the equation?"

It could all be pseudo-random all the way down, including quantum, and you'd still be able to have conscious choice. This is only a problem if you are stuck on the idea that choice *must* be non-deterministic. If you let go of that idea, you lose nothing, and choice still has meaning.

"If a human mind does contain truly random determinants - then in that way alone, it currently differs from a computer program."

Again, not if the program accepts input. The input can be, for example, a truly random source of bits. This is sometimes known as a probabilistic Turing machine, but really any computer can accept random input. Accepting input is another point you've not really acknowledged.

"So what I am putting out there for inspection is the idea that we could be right about something partly by an accident of an absolutely unpredictable event."

I would attempt to be more clear or cautious about mixing 'random' with 'unpredictable', and also whether something has to be 'absolutely' unpredictable or whether it's sufficient for it to only be 'somewhat' unpredictable. For example, I make a big deal of the fact that we can't *perfectly* predict the future ahead of time, but that is not to say that we can't make *any* useful predictions. We definitely can. We do it all the time, every day of our lives. They just aren't perfect predictions.

Whether things are truly random or merely unpredictable is an open question, and likely to remain unsolved forever. Even if we found a 'deterministic' explanation of quantum mechanics, there could still be a deeper layer of randomness below that, hidden away from us just as quantum indeterminacy was hidden from us during Newton's time. We might dig deeper and find it, or not. We'll probably never know with absolute certainty either way. My ultimate point is that it doesn't matter, either way. The universe still works, random or pseudo-random, either way.

Also, things don't need to be 'absolutely' unpredictable for unpredictability to play a fundamental role. Even quantum mechanics describes probabilities not as completely unpredictable, but as having certain wavelike distributions. In other words, we know the electron is probably here and probably not way over there, but we don't know exactly, and we can't predict perfectly.

Relating this to your question of the randomness of prediction, personally I don't think it matters whether the prediction process uses 'true' randomness or not. It may, or it may not, and it can still produce good predictions either way. An obvious example is statistical reasoning. A less obvious example are the links I gave earlier on stochastic computing.

All methods of using randomness to make predictions will work, regardless of whether you give them 'true' random input, or merely unpredictable pseudo-random input. In the latter case, the only important part is that the pseudo-random algorithm is not known by the prediction program itself (which can lead to wonky results, such as over-fitting the theory to the data, for some badly-designed programs) or known by someone trying to thwart the program.

All that said, I would almost always prefer restricting myself to questions of unpredictability rather than questions of true randomness, and I would probably not use the phrase 'absolutely unpredictable', since that to me implies true randomness, where 'practically unpredictable' would do just as well, and without the implication.

I think the final important point to remember is that unpredictability in the predictor and/or unpredictability in the universe does not automatically make prediction impossible. It just makes it imperfect. Even extremely chaotic, long term events can be predicted with great accuracy, if you're careful to qualify your predictions. A good example would be entropy and the heat death of the universe. Nobody knows what the final state of the universe is going to be, but we can make pretty damn good predictions as to how quickly heat death will approach, and how cold it will be at future times. Global warming is another good example.

We don't need perfect predictions to make 'pretty good' predictions. All we really need, for evolution's purposes, to evolve conscious choice in a deterministic universe, is to make slightly more accurate predictions than chance alone. If you can make slightly better predictions, you can make slightly better choices, and take slightly more effective action, out-competing the other guy. Eventually, natural selection amplifies that slight advantage into a big win. I predict that's what happened to Neanderthal humans. It didn't take coordinated effort like systematic genocide for us to wipe them out, just slightly more effective conscious choice. That could have been deeper meta-awareness to take into account more information, better raw predictions from brain intelligence, better choice-making judgment faculties, or better cultural/social structure to make a chosen course of action more effective.

This is one of the other reasons I don't think it makes sense to say 'choice' is illusion. It has obvious, practical consequences, both in our own lives, and also in the bigger picture. If you don't acknowledge 'choice' as meaningful, even if it can be explained purely deterministically, you're left with no conceptual tools to explain the consequential phenomena of choices.
Actually - if I accepted the mind isn't any less deterministic than anything else, hadn't used the words 'totally interdependent' then I guess you could say that I elided the idea of inputs. But I didn't.

Even to most atheists - 'choice' suggests that the chooser is 'author' of the choice. You are pointing out that, indeed, the choice is determined in a fixed 'web' (better term?) of causation. That's what I've been drilling down to.

The input can be, for example, a truly random source of bits.

So far, we don't know how to generate random using a computer. Also, are you suggesting there is such a thing as truly random? If something is purely determined - it can be argued it can't be truly random.

Now Heisenberg, chaos theory, and things so far unobserved or unobservable may 'put the lie' to pure determinism and allow for true random. And, if it exists, we may find a way to generate a random number in a computer. Or, like you suggest, the computer could find its random factors outside of itself (inputs). Sure.

But, in your mind, how does 'true random' fit into determinism?
"Even to most atheists - 'choice' suggests that the chooser is 'author' of the choice. You are pointing out that, indeed, the choice is determined in a fixed 'web' (better term?) of causation."

Part of the 'web' is the 'author'. I don't see any problem here. We went over this before also, in our discussion of fuzziness. Fuzziness does not cripple the idea of identity, and therefore 'author'. The author exists, the imagined options exist, the choice process exists, the final choice of option exists, the final course of action exists, it is all deterministic, and it still makes sense to call it 'choice'.

The author authored the choice (especially if you throw consciousness/meta-awareness into the mix), the choice influenced the actual future in a post-deterministic way, the choice was effective. Simple counting machines, dripping faucets, and tornadoes don't implement the same process since they don't evaluate options, select one based on some criterion, and act based on the selection. Simple computer programs can perform a choice process, but lack meta-awareness and consciousness. Humans have it, and so have more discretion on which choices they make. But even without consciousness, current computers can still make complex choices based on semi-intelligent assessment of varied inputs. They can even invent options of their own, producing non-human innovation.

"So far, we don't know how to generate random using a computer. "

Irrelevant to my point. You stated that humans and computers would be different if humans could access internal randomness. But computers can do the same things if they simply have an input stream that is random. In other words, you can interpret internal human randomness as merely deterministic human behaviour plus a random input stream, same as a computer with a random input stream.

"Also, are you suggesting there is such a thing as truly random?"

No, I was responding to your point about humans being different if they had internal randomness. You can equalize the situation by giving the computer a random input stream. *If* humans have true randomness available, *then* you could just give true randomness to a computer through its input. My hypothetical was in response to yours, where you already included true randomness into the scenario.

"But, in your mind, how does 'true random' fit into determinism?"

There is a special case of determinism called indeterminism which is like determinism-plus. Things act mostly deterministically, except there is an underlying true randomness. This is essentially equivalent to the popular interpretation of quantum mechanics. My views on conscious choice are compatible with both indeterminism and post-determinism. With indeterminism, you just need a layer of stochastic processes to tame the randomness. We already have such stochastic reasoning in science, both in quantum mechanics and in classical statistics. Again, whether the universe is truly indeterminate or unpredictably pseudo-randomly post-determinate (the less-popular view of quantum mechanics), it works basically the same. My point is, either way, it doesn't matter, conscious choice exists in both systems.
"there are determinists who don't want to say that anything is predetermined (maybe simply because it starts to sound like some god did the determining? - which isn't what I'm suggesting at all.)"

It's not necessarily about god, although it also argues against omniscient prophecy. The issue is whether the determination happens 'pre' or not. It's all determined, no one is arguing against that. At each step of the way, the future is determined entirely by conditions in the present, agreed. There is only one actual future that will happen, no dispute there. The question is how does the prefix 'pre' play into it.

Determinism alone (no prefix) covers the 'one actual determined future' scenario. So what can 'pre' add to the mix? Usually, it is used to add that the 'future is known ahead of time', in other words, the determination of events occurs *prior* (hence pre-) to the events actually happening, and can somehow be either *known* or *trusted*. This is qualitatively different than simply 'one actual determined future'. It has certain implications and consequences that simple (no prefix) determinism does not necessarily have. For example, it implies that the future is laid out like a pre-recorded video tape, or a perfectly detailed 'future history of the universe', and time is simply moving step by step through each frame of the video, or word by word through the book. At time A, the future B is already known, and person X can either know or trust that future B will occur exactly as 'pre-determined'.

Some common interpretations of pre-determinism which are not necessarily entailed by no-prefix determinism: a) new age types often say "there is a plan/reason/purpose for everything, so no matter what happens, everything is how it should be/how it was planned from the beginning" (complacency) b) some skeptics often say "the future is already set, so there's no point doing anything to change anything, as it will happen however it was pre-determined to happen anyway" (apathy, dread, or helplessness) c) woo woos often say "we've got the true prophecy, and it will necessarily come true, because it was pre-determined (by god, usually)" (omniscient prophecy; also hubris). Interpretations a) and c) are easier for skeptics to see through, but some determinist skeptics also fall into pre-determinism with claims like b). None of those are necessarily entailed by no-prefix determinism. Post-determinism rejects b) as well as a) and c).

Post-determinism is simply a word I came up with to replace the notion of non-pre-determinism, which I found clumsy and confusing. It specifically rejects pre-determinism. No-prefix determinism encompasses both pre- and non-pre-, but I found it tiresome to repeatedly say "But determinism doesn't necessarily mean pre-determinism," which was often interpreted as lacking a full understanding of just how pre-determined no-prefix determinism must be. To head-off that misunderstanding, I started to say, "I advocate non-pre-determinism", but that grew old fast, so I just flipped the prefix and now we can talk about post-determinism.

The key thing that post-determinism claims is that "under (no prefix) determinism, if we assume the future is pre-determined, then true knowledge of the pre-determined future can easily be used to thwart that knowledge and avoid that future, contradicting that supposedly 'true' knowledge, and thereby invalidating the assumption of pre-determinism." (Proof by contradiction.) It also claims, less strongly, that "if the future is actually pre-determined, then it is impossible for true knowledge of that future to be known *within* the universe, ahead of actual events," or more succinctly, "pre-determinism is always hypothetical, never practical."

As a consequence of these claims, post-determinism can be defined as "the future is determined by the present, after the fact," in contrast to pre-determinism's "the future is determined by the present, before the fact." Pre-determinism can be extended all the way back to the beginning of time as, "the future is completely determined by the past, with no regard for present conditions." Post-determinism cannot be so extended into the past. The present is *always* relevant to post-determinism. In fact, it's really the only thing that's relevant for determination of the next state of the future. In the end, after the fact, in *hindsight*, you could theoretically trace the deterministic causes and effects, from beginning of time to the present, and see that it was all perfectly determined. But you cannot extend that hindsight into *foresight* and see what the *true* perfectly determined future will be. At best, all you can do is make imperfect predictions with incomplete knowledge of the *true* actual future that will occur.

Proof of post-determinism's strong claim is simple: A sufficiently determined entity can use knowledge of its own pre-determined behaviour to do oppositely or differently from what was predicted. If you report to me that the true pre-determined future, which says I'm going to go right, I can go left instead.

Even a simple machine can thwart pre-determinism. The inversion machine (or maybe I should call it the anti-prophecy machine) takes a single input, 0 or 1, and outputs the opposite number. Input 0 outputs 1. Input 1 outputs 0. It is physically impossible to feed in *true* knowledge of the pre-determined output of the anti-prophecy machine to the machine's input, and *also* get that number as output. If the pre-determined output is 0, and you feed in the input 0, the output will be 1, contradicting the prophecy. Likewise, 1 produces 0, also contradicting the prophecy.

In fact, you don't even have to know what the pre-determined future is for the anti-prophecy machine to work. The simple fact is that it is *impossible* to give it an input which is the true pre-determined output. The only input you can give it is a *false* prophecy. It instantly and deterministically contradicts true prophecies.

The point of the anti-prophecy machine example is that it scales up to *any* entity which is given true knowledge of the pre-determined future. That knowledge can easily be thwarted to contradict it, which also contradicts pre-determinism. Humans can do it, little machines can do it, any deterministic system can do it.

So, this leads to the weaker claim that if the universe is actually pre-determined, then knowledge of its future cannot be known inside the universe before the events, ahead of time. If it were, it could be used with an anti-prophecy machine to thwart it, contradicting pre-determinism.

So, even if the future is pre-determined, we can never know it practically, because knowing it practically would defeat pre-determinism. Pre-determinism is always hypothetical, never practical.

Practically speaking, only post-determinism, or non-pre-determinism, makes sense. You cannot know or trust in a pre-determined future. The determination of events will always take the present moment into account. The present is not irrelevant to the outcome of the future, in fact it is the only relevant factor. The future is determined by the present, after the fact, not before the fact. We are not following a pre-recorded videotape or pre-written book. The book is being written at this very moment, based not just on the first word of the book, but also on the current word.

In fact, although the current word was determined by the first word in a strict chain, it is only the current word that is relevant to determining the next word. We can't skip chapters and read ahead, because doing so would give the current word extra knowledge, changing it. And since only the current word is relevant to determining the next word, the next word will also be changed, and the peeked-at future chapters will be over-written by different words, effectively proving that they were never the pre-written story in the first place.

It is this last analogy, which hopefully makes the whole thing clearer, which is relevant to the topic of free-will vs. choice.

Using our imaginations, we can make imperfect, but still useful, predictions of the future, or 'imagined future possibilities'. These imagined possibilities can act like prophecy. We can feed them into our anti-prophecy machine and determine a course of action which will avoid that imagined future. We can't read future chapters word-for-word, but we can spot the trends in the story and say, "I can see where the story's headed, and I don't like it. I would rather have a different plot." Using this imperfect but useful prediction, we can essentially 're-write' the imagined future chapter -- which is actually only 'writing' (not 're-writing', since it was never written ahead of time anyway) the next word, one word at a time, until we reach the next chapter. This chapter will almost certainly be different than the one we imagined, both because our prediction was imperfect in the first place, and also because we used the prediction to select from different courses of action, based on the criterion of which future we desired most strongly, and which actions would be most likely to achieve that future and avoid all the others. This is 'choice'.

Free will (contra causal), on the other hand, would be like magically taking the pen from the universe's hand and writing a whole new story, disconnected from the words that came before, uncaused/undetermined by present circumstances. Free will escapes determinism, choice does not.

The key point is that with pre-determinism, choice has no power at all. If I reported to you the true pre-determined future, which dictated you will go right, then you will necessarily go right. You cannot use that knowledge to go left. Simple no-prefix determinism does not entail that, and post-determinism rejects it explicitly. If you can use that knowledge, then that is not pre-determinism, and choice is meaningful. Choice is not just counting numbers or blowing in the wind. The future is determined by the present, not just the past, and the present includes my predictions of possible futures, as well as my currently-chosen course of action. My choices do affect what the actual future will be.

And just to prove it, I choose to press the Add Reply button, rather than closing the page and ditching what I just wrote. I predict this will bring about my more-favoured possible future of having this post appear on this thread, and avoid my less-favoured possible future of losing what I wrote. If my present choice is effective, then it will help to determine the actual future that occurs. Will my prediction come true? Let's see what happens....
Great. And I totally agree that that free will is a very unlikey proposition (etc.) The 'complete independence' suggested by 'free' can be dispensed with. We are pretty certain that, in fact, there is complete interdependence.

And, while 'will' or 'choice' may remain very similar in definition - (and also probably fully determined, whether we can see the whole picture or not) to the extent that 'choice' is an illusion (within the context of one fixed future) because we have zero chance of 100% accurate predication, it is an impenetrable illusion. Wherever we cannot predict the shape of all the causal relationships, the virtual is, de facto, actual.
"Nothing can be zero chance - only nearing it. "

Logical impossibilities such as logical contradictions are said to have zero probability. The anti-prophecy machine introduces a contradiction to having knowledge of 100% perfectly accurate predictions of the future. Whatever the prediction says, the machine does the opposite.
Asimov also extrapolated the 'law of big numbers' and actuarial math to create the concept of 'psycho-history' where it remained impossible to accurately predict the actions of an individual but large numbers of people could be predicted with great accuracy.
Okay - an asymptote approaching zero ...
"If the pure determinism of LaPlace were available, all things past and future would be theoretically knowable."

Even with LaPlace, you can never know the true output X of an anti-prophecy machine fed that prediction as input X. Either your prediction will be false, or you will be unable to perform the actual experiment as described (you will have to feed it a false prediction of its output, rather than the true prediction).

In any case, you wouldn't be able to predict your own behaviour, as you would have to recursively feed in your own knowledge of the predictions in an endless loop. A sufficiently determined scientist could always prove LaPlace wrong by doing oppositely or differently than whatever the equations predict.


© 2018   Atheist Nexus. All rights reserved. Admin: The Nexus Group.   Powered by

Badges  |  Report an Issue  |  Terms of Service