Definition of Free Will

For the serious discussion of weighty matters and worldly issues. No off-topic posts allowed.

Moderators: Azrael, Moderators General, Prelates

User avatar
jules.LT
Posts: 1539
Joined: Sun Jul 19, 2009 8:20 pm UTC
Location: Paris, France, Europe

Re: Definition of Free Will

Postby jules.LT » Mon Jan 14, 2013 3:56 pm UTC

Cease asking questions and making claims about a concept that you haven't even tried to define: freedom.
You'll be less wrong.

And determinism is not on a scale. The universe either is or isn't.
The slightest bit of non-determinism makes it non-deterministic, especially considering the laws of chaos.
Bertrand Russell wrote:Not to be absolutely certain is, I think, one of the essential things in rationality.
Richard Feynman & many others wrote:Keep an open mind – but not so open that your brain falls out

User avatar
ucim
Posts: 6896
Joined: Fri Sep 28, 2012 3:23 pm UTC
Location: The One True Thread

Re: Definition of Free Will

Postby ucim » Tue Jan 15, 2013 6:50 pm UTC

jules.LT wrote:Free will has to be decided at the "person" level, that (almost) goes without saying.
Yes, and it is (almost) never said. However, the question "do you have free will" is often decided at the mechanism level.

Consider the cases where you (the reader) might have decided that somebody is not acting of their own free will, such as when they are under the influence of alcohol, drugs, failure to take drugs, stress, lack of sleep, chemical imbalance, psychotic episode, coersion (implied or otherwise), or habit.

In some cases (e.g. implied or overt coercion) you can stay at the person level...
Spoiler:
I stay late every day at work for free, because my boss has indicated in various ways that I won't need to come in any more if I don't. My boss smugly claims that I am acting of my own free will, and he has no role in it (while he benefits from my free labor). I think otherwise - that I am being coerced into supplying more than I had bargained for. A similar argument would apply to sposal abuse. Either way we can remain at the "person" level
and in other cases we must enter the mechanism level.
Spoiler:
Under the influence of alcohol or hormones, and viewed at the person level, I can still imagine future courses of action, I can still deliberate, I can still plan, and I can still control my actions. But because I am under the influence, I am likely to end up making a different choice. This would also apply to being under the influence (or not under enough influence of) drugs, electrolytes, blood sugar, etc. In all cases, we must enter the mechanism level to answer the question in a manner consistent with our predispositions.
In the first sense, the idea of free will is tied in with the idea of coercion and restriction (externally shaping the option space from which a person chooses), while in the second sense, the idea of free will has become a proxy for the idea of intellectual vs emotional decisionmaking. They are very different ideas and perhaps merit different terminology.

In the first sense, the question of an answering machine or an AI or a dog having free will doesn't really apply very well. We are looking at the question: given an entity, under what circumstances does it have free will?

In the second sense, the aforementioned question very much applies, and is at the root of sentience, consciousness, and sapience. It is the question we will have when the flying saucers land, and when the networked computers decide we are superfluous. (Of course, by that time, the answer may no longer matter!)

Jose
Order of the Sillies, Honoris Causam - bestowed by charlie_grumbles on NP 859 * OTTscar winner: Wordsmith - bestowed by yappobiscuts and the OTT on NP 1832 * Ecclesiastical Calendar of the Order of the Holy Contradiction * Heartfelt thanks from addams and from me - you really made a difference.

Technical Ben
Posts: 2986
Joined: Tue May 27, 2008 10:42 pm UTC

Re: Definition of Free Will

Postby Technical Ben » Tue Jan 15, 2013 10:28 pm UTC

jules.LT wrote:Cease asking questions and making claims about a concept that you haven't even tried to define: freedom.
You'll be less wrong.

And determinism is not on a scale. The universe either is or isn't.
The slightest bit of non-determinism makes it non-deterministic, especially considering the laws of chaos.

I would say I see it to be a scale. I can predict part of it to 100% certainty, other parts to 0% and everything in between. So I can say I know for certain (for sufficient definitions of certain) some parts are 100% determined, and others 100% undetermined. What do we call a system that has both? Say I have a machine, that always flashes a red light every 5 seconds. It also flashes a green light randomly. Do we say that the machine is 100% random? That in the "system" or "universe" of the machines construction it is only possible to be 1 complete state? Either the machine is "all random" or "all determined"? I'd say neither of those statements are correct. Would you?
It's all physics and stamp collecting.
It's not a particle or a wave. It's just an exchange.

User avatar
jules.LT
Posts: 1539
Joined: Sun Jul 19, 2009 8:20 pm UTC
Location: Paris, France, Europe

Re: Definition of Free Will

Postby jules.LT » Tue Jan 15, 2013 11:38 pm UTC

That's not determinism, it's predictability. Look up determinism on Wikipedia.
In a deterministic universe, "random" merely means that you have no way to determine what the result will be.
Bertrand Russell wrote:Not to be absolutely certain is, I think, one of the essential things in rationality.
Richard Feynman & many others wrote:Keep an open mind – but not so open that your brain falls out

Greyarcher
Posts: 708
Joined: Thu Oct 18, 2007 3:03 am UTC

Re: Definition of Free Will

Postby Greyarcher » Wed Jan 16, 2013 3:39 am UTC

Ha! It's pretty neat to see some of this vocalized, since I recognize some of these variants as things I've never managed to drag to a satisfactory conclusion. Models of the self, freedom, identity, mechanisms, and competing desires are particularly recurring themes.
elasto wrote:
ucim wrote:The idea of "being a slave to your own mental processes" separates "you" from "your mental processes", but in fact the separation is invalid. Your mental processes are an integral part of "you", unless you mean something else by "you".

Ding ding! This is the key to the dilemma for me. To interpret 'free will' to mean 'you' can make decisions apart from what 'your deterministic mental processes' would themselves make is a meaningless definition. Your conscious mind is part of the decision making mechanisms - and a very important part - so you both do and do not have free will depending on how you choose to look at it.

As to what the higher functions are for if they mostly just echo what the subconscious has already concluded, I'd suggest they are for complicated prediction/planning - such as 'if tonight I try to steal the food that Bob has hidden, will he realise it was me and, if he does, what will he do about it?'

And, having arisen, higher functions are sort of involved in mundane decisions like 'I've got meat and veg on my plate, which do I want to eat first?' but I think those kind of decisions are mainly made by the subconscious and then rationalised and rubber-stamped by the conscious.
I like that last paragraph, as I've been wondering about things that arise from the sub-conscious and their relation to a "you". Maybe because I do not subscribe to a particular model of "myself" I am unable to rubber-stamp things as "me". For instance, although I may seem to set a direction for my thoughts, I cannot say that any particular thought I have is something I decided upon. Rather, it is just a thought that arises from my sub-conscious. "Decisions" are the same.

Perhaps Ucim's interesting application of the concept "wholistic" would work here. A decision is something I might consider "mine" if I am wholly in agreement with the decision and reflected on it beforehand; but even then, I'm not sure. Setting, in this manner, criteria for "owning decisions" somehow seems...arbitrary and artifical. Perhaps it is superfluous from the start to add a concept like "decision" or "me" to this body and mind; the mind thinks, the body acts, but I'm not sure about adding a "me who makes decisions" in there.

Naturally, I use "I" and "me" in common parlance. But when I think more deeply on the question of the self, identity, causality, desires, freedom, and decision...the "me" certainly dissolves away.

It may be that no model of the self that covers these subjects satisfies any particular goal I have; therefore, I feel no need to subscribe to a particular model for framing "me" or these elements of "me" in a deeper manner. But I do seem to understand humans in general in a sort of mechanistic way without adding "freedom" or "choice" concepts........

Ah! I see. Adding certain concepts of "freedom" or "choice" seems unnatural and unnecessary to my understanding of humans (and myself!). The confusion I feel when dealing with these concepts is because I'm trying to shoehorn in things that don't really fit and have peculiar interactions or nonsensical elaborations with regards to possible concepts of "self", "mind", and "causality". Somehow, it feels like I've had a minor epiphany. All along, I've been puzzled because of concepts that I've heard various forms of but don't seem to fit with each other properly? But, rather, there was no need to have all the concepts and make them fit in the first place? Interesting.

Thank you all. It's an odd conclusion to reach, since it's not a conclusion about a way for the concepts to all fit together in my head (which is apparently what I think I expected); rather, it's a conclusion that it was unnecessary to have all these concepts present and fitting together in a model of my understanding of humanity.

(Well, I may reflect again and reconsider later on the conclusiveness of my thought. But for now, at least, I am content that I have settled myself with the subject.)
In serious discussion, I usually strive to post with clarity, thoroughness, and precision so that others will not misunderstand; I strive for dispassion and an open mind, the better to avoid error.

Technical Ben
Posts: 2986
Joined: Tue May 27, 2008 10:42 pm UTC

Re: Definition of Free Will

Postby Technical Ben » Wed Jan 16, 2013 10:58 am UTC

jules.LT wrote:That's not determinism, it's predictability. Look up determinism on Wikipedia.
In a deterministic universe, "random" merely means that you have no way to determine what the result will be.

I did not state the mechanism of the machine. It could be a deterministic machine, it could be a random machine, it could be anything. The example shows that we cannot argue that it has to be one or the other or that only the 2 options are available.

If we cannot state what my machine is restricted by, through external observation, we cannot state what my decision mechanisms are restricted by through external observations (or by extension what mechanism of cause is in the rest of physics, if still making external observations). Can we? Thus we cannot say anything about the restrictions applied on my decisions. Except for those we do observe, but only as far as we observe them. So we can say I'm restricted to deciding to eat things that exist, but unrestricted in what type of food I eat.

If I am unrestricted, the term used is "free", right? I have "free" ability to choose in at least some (so a restricted amount) of things. Unless we can state how my machine with 1 red light blinking every second and 1 green light randomly blinking is deterministically made or random. I'm happy to go back and look over my ideas if I can get a better definition of my machine. If it's a "this or that" machine, I'll go read up on this and that. :)
It's all physics and stamp collecting.
It's not a particle or a wave. It's just an exchange.

Greyarcher
Posts: 708
Joined: Thu Oct 18, 2007 3:03 am UTC

Re: Definition of Free Will

Postby Greyarcher » Wed Jan 16, 2013 8:58 pm UTC

Technical Ben wrote:If we cannot state what my machine is restricted by, through external observation, we cannot state what my decision mechanisms are restricted by through external observations (or by extension what mechanism of cause is in the rest of physics, if still making external observations). Can we? Thus we cannot say anything about the restrictions applied on my decisions. Except for those we do observe, but only as far as we observe them. So we can say I'm restricted to deciding to eat things that exist, but unrestricted in what type of food I eat.
Pardon, Ben, but it looks like you jumped from "the machine's restrictions cannot be observed or ascertained" in the first sentence to "so the machine is unrestricted" in the last sentence. The most relevant thing inbetween seems to be "we cannot say anything about the restrictions on the machine" which itself seems to directly contradict the last sentence.
In serious discussion, I usually strive to post with clarity, thoroughness, and precision so that others will not misunderstand; I strive for dispassion and an open mind, the better to avoid error.

Technical Ben
Posts: 2986
Joined: Tue May 27, 2008 10:42 pm UTC

Re: Definition of Free Will

Postby Technical Ben » Sat Jan 19, 2013 2:16 pm UTC

Only if the insistence is that the machine is unrestricted. My example is only that we cannot say if it is or is not. Is that fair?
It's all physics and stamp collecting.
It's not a particle or a wave. It's just an exchange.

User avatar
ucim
Posts: 6896
Joined: Fri Sep 28, 2012 3:23 pm UTC
Location: The One True Thread

Re: Definition of Free Will

Postby ucim » Sat Jan 19, 2013 3:40 pm UTC

I'm not sure that "restricted", in an internal sense, even makes sense. All machines are restricted in that they can do only what they can do. They are unrestricted in that they can do everything they can do. That is the nature of a mechanism. I don't really see that there is any meaning here.

My car is restricted in the sense that it cannot microwave a chicken. It is unrestricted in that it can be driven on any strong flat-ish surface it is put on.

If it blows a gasket and no longer functions, is this what you are calling a restriction? I would just call it "broken". But either way, there is an implied concept of how the machine "is supposed to operate", which I'll call a Platonic Ideal (admittedly abusing overloading the term). It is no longer operating in its Platonic Ideal way, ergo it is broken.

This is fine for cars and answering machines (which are designed to certain specifications, so in these cases there does exist a Platonic Ideal). It is, however, problematic for evolved and emergent entities like animals, people, and sufficiently advanced AI such as the gaianet.
Spoiler:
Gaianet: (n) the emergent entity composed of a massive number of networked computers, the people who operate and are dependent upon them, and the relationships they form with each other by dint of this dependence. As the banking system, medical system, social networking system, employment system, and other basic infrastructures of society become tightly intertwined due to networked computers, the richness of the rapid and massive data interchange they foster, and the control upon individual options that it entails, they form a gaianet, and we become a part of this new organism, which can reasonably said to have its own emergent will.
These emergent entities do not have a Platonic Ideal in this sense; there is no easy way to say when they are "broken". We can sometimes say that one person (for example) behaves differently from a fuzzy idea of "normal" that we devised after observing other people (who all behave somewhat differently from each other), but that's not the same as "broken". This is one of the biggest problems with mental health issues.

If this is what you mean by "restricted" (in the internal sense), then it is a very different concept from "restricted" in the external sense. I would go so far as to say that the lexicographical baggage that the word carries makes the word inappropriate for the ("internal sense") uses it is being applied to here in this discussion of "free will", even though it has significant bearing on the issue of whether one has free will when they are (for example) under the influence of (naturally or artificially introduced) substances.

Jose
Order of the Sillies, Honoris Causam - bestowed by charlie_grumbles on NP 859 * OTTscar winner: Wordsmith - bestowed by yappobiscuts and the OTT on NP 1832 * Ecclesiastical Calendar of the Order of the Holy Contradiction * Heartfelt thanks from addams and from me - you really made a difference.

Technical Ben
Posts: 2986
Joined: Tue May 27, 2008 10:42 pm UTC

Re: Definition of Free Will

Postby Technical Ben » Sat Jan 19, 2013 5:24 pm UTC

Both. Internal or external. Prove that my machine is restricted to be either random or deterministic. Or that we can say it is defiantly one or the other.

If we cannot confirm it is random and/or deterministic (deterministic or not, random or not etc), then how can we confirm those are the only two options?

Does our example of stating "choice can only be random or not/deterministic or not" rule out free will any more than stating "Ice cream can only be strawberry flavor or not/raspberry flavor or not" rules out chocolate flavor? Are we missing other possible options?
It's all physics and stamp collecting.
It's not a particle or a wave. It's just an exchange.

User avatar
ucim
Posts: 6896
Joined: Fri Sep 28, 2012 3:23 pm UTC
Location: The One True Thread

Re: Definition of Free Will

Postby ucim » Sat Jan 19, 2013 6:18 pm UTC

Technical Ben wrote:Prove that my machine is restricted to be either random or deterministic.
I'm not sure what you are asking here, and I'm pretty sure the request is not well-formed. It also uses "restricted" in yet a third sense - that of assigning an element to a set. This is not what is meant by "restricted" in the context of this discussion, and so can be confusing.

Also, the word "deterministic" contains a lot of baggage. What exactly do you mean, in this case, by that word?

I suspect you are talking about something akin to "speed of divergence" - if you run your machine ten times from identical starting conditions, a machine whose "random component" is relatively inconsequential will take longer for the ten instances to diverge from each other's paths. One whose "random component" is more critical will diverge more quickly. But in both cases, there will be divergence.

If there is no random component (in the machine or its inputs), then there will be no divergence.

Either there is no divergence, or there isn't no divergence. We don't have to know which category your machine falls into in order to ascertain that those are the only two possibilities. But I'm not sure that this has any bearing on free will.

Jose
Order of the Sillies, Honoris Causam - bestowed by charlie_grumbles on NP 859 * OTTscar winner: Wordsmith - bestowed by yappobiscuts and the OTT on NP 1832 * Ecclesiastical Calendar of the Order of the Holy Contradiction * Heartfelt thanks from addams and from me - you really made a difference.

Technical Ben
Posts: 2986
Joined: Tue May 27, 2008 10:42 pm UTC

Re: Definition of Free Will

Postby Technical Ben » Sat Jan 19, 2013 7:25 pm UTC

How can we confirm our stating conditions are identical? I know nothing of the inside of the box. I only know of it's green and red flashing light.

Confirm if the box, via observation only, is divergent or not. If it is random or not. If it is deterministic or not.

If we cannot confirm, describe or know my box (in this simple thought experiment, or could we even set it up as a real experiment?), then no one can tell me "TechyBen's decisions are based of deterministic inputs and are thus predictable in essence" or "TechyBen's decisions are based on random effects, and thus random in essence". Is that true? If we cannot describe my choice mechanism as random or not (or deterministic or not), then we cannot state "free will cannot exist because we only have the options random or not", can we?

The words are not the problem, or their definitions. We can go back and choose better words. I'm happy to consider better terms here. What mechanism do we choose by? Random or deterministic mechanisms. But before you answer, what mechanism does my box in the illustration act by? Random or deterministic?

If there is no random component (in the machine or its inputs), then there will be no divergence.

I can only observe the output. It's a box giving signals. Can I state anything about it's divergence, random components etc? If I can't about a simple box, how can I about a persons decision mechanism?

Either there is no divergence, or there isn't no divergence.

I could rephrase that. "Either there is strawberry ice cream, o0r there isn't strawberry ice cream" is a true statement also. But neither are defined as required for a decision or choice mechanism are they? If I say I see choice as "free from determinism" or "an ability to make a decision free from forceful effect", is there a requirement for determinism, randomness or divergence that would remove my ability to claim for free will? As the flavor of ice cream does not prevent me from considering free will, why would observation of some things being random (or not), prevent me from considering free will?
It's all physics and stamp collecting.
It's not a particle or a wave. It's just an exchange.

User avatar
ucim
Posts: 6896
Joined: Fri Sep 28, 2012 3:23 pm UTC
Location: The One True Thread

Re: Definition of Free Will

Postby ucim » Sat Jan 19, 2013 8:25 pm UTC

Technical Ben wrote:If we cannot describe my choice mechanism as random or not (or deterministic or not), then we cannot state "free will cannot exist because we only have the options random or not", can we?
Free will does not reside in randomness - at least not the way I see free will, nor the way I use it in casual parlance. The idea of "will" is more a statement about the state of the mechanism than about its function - it can be thought of as a shorthand for a set of states which would tend to favor a certain outcome. Whether or not the mechanism involves randomness is not relevant.

Spoilered for tangency:
Spoiler:
... and what is randomness? The digits of pi are random, but they are also exactly the digits of pi. If the intervals between green light flashes correspond to the digits of pi, are they random?
Free will, at its essence, involves the ability to choose. Considering your machine, what, exactly, is doing the choosing, and what is it choosing between?

Jose
Order of the Sillies, Honoris Causam - bestowed by charlie_grumbles on NP 859 * OTTscar winner: Wordsmith - bestowed by yappobiscuts and the OTT on NP 1832 * Ecclesiastical Calendar of the Order of the Holy Contradiction * Heartfelt thanks from addams and from me - you really made a difference.

Greyarcher
Posts: 708
Joined: Thu Oct 18, 2007 3:03 am UTC

Re: Definition of Free Will

Postby Greyarcher » Sun Jan 20, 2013 12:48 am UTC

Technical Ben wrote:If we cannot confirm, describe or know my box (in this simple thought experiment, or could we even set it up as a real experiment?), then no one can tell me "TechyBen's decisions are based of deterministic inputs and are thus predictable in essence" or "TechyBen's decisions are based on random effects, and thus random in essence". Is that true? If we cannot describe my choice mechanism as random or not (or deterministic or not), then we cannot state "free will cannot exist because we only have the options random or not", can we?
By putting aside everything that is constituted by random and deterministic processes, would any remaining element be "more truly" TechyBen? Even if we insist that the human mind is a bit of a mystery machine, and so we can hide a little god of the gaps inside, I'm not sure that little hidden bit could more meaningfully be called "us". And I can't see how it would arrive at its decision nor how its operation could more meaningfully be called "me" or "free".

If we do not have full knowledge of the human machine, it seems natural that we extend our model of how the world generally works to that machine and use that as a baseline unless we learn otherwise. Adding some sort of "freedom" to make humans special seems unnecessary and inappropriate. We are already extraordinary in our superior intelligence. We are also animals, mammals, primates--and we should be understood in light of these things.
Technical Ben wrote:Only if the insistence is that the machine is unrestricted. My example is only that we cannot say if it is or is not. Is that fair?
I admit, I like leaving these sorts of questions shelved if they cannot be properly answered one way or another. But, as you can see in my previous paragraph, practically speaking I think we should understand humans as we understand the rest of the world. And we should be cautious not to let the differences between other animal machines and human machines cause us to jump to various self-indulgent conjectures.
In serious discussion, I usually strive to post with clarity, thoroughness, and precision so that others will not misunderstand; I strive for dispassion and an open mind, the better to avoid error.

Technical Ben
Posts: 2986
Joined: Tue May 27, 2008 10:42 pm UTC

Re: Definition of Free Will

Postby Technical Ben » Sun Jan 20, 2013 12:27 pm UTC

ucim wrote:
Technical Ben wrote:If we cannot describe my choice mechanism as random or not (or deterministic or not), then we cannot state "free will cannot exist because we only have the options random or not", can we?
Free will does not reside in randomness - at least not the way I see free will, nor the way I use it in casual parlance. The idea of "will" is more a statement about the state of the mechanism than about its function - it can be thought of as a shorthand for a set of states which would tend to favor a certain outcome. Whether or not the mechanism involves randomness is not relevant.

Spoilered for tangency:
Spoiler:
... and what is randomness? The digits of pi are random, but they are also exactly the digits of pi. If the intervals between green light flashes correspond to the digits of pi, are they random?
Free will, at its essence, involves the ability to choose. Considering your machine, what, exactly, is doing the choosing, and what is it choosing between?

Jose

My machine does not. It only proves you/me/anyone cannot state "Free will cannot exist, because the only 2 options we have are random or not (or by opposite description, determined or not)". As our model seems incomplete, we cannot make such sweeping statements about it. From what I can tell.

What chooses inside the machine? Well, that's the question of the riddle. If the box is sealed, how can I know? Can I make assumptions? If I cannot, what do I call things I cannot, nor ever, know?

Will seems easy to contemplate. It's what we want. Either naturally or just by internal free willed choice. We can choose to choose strawberry ice cream. Right? We wish (or will for), strawberry ice cream over chocolate. Or will may not be actionable (there may only be chocolate), but it's something we can define and try to act on. Is it limited to predetermined or random events though? I'd say we cannot say either, or confirm that is the complete set of options. Perhaps free will acts with both, and is a separate set all together

As strawberry, being a fruit, and "fruit or not" is still separate from chocolate, being a bean and "bean or not", and neither preventing other flavors of ice cream. Likewise, being told the only flavor of choice is "random or not" by some and "determined or not" by others and "determinism and randomness are the opposites of each other" are all true statements, but do they effect choice? I don't see them being the exhaustive options of flavors of types of "choice" mechanism. :)

Greyarcher, my argument is not that the human mind is a bit of a mystery, but that the entire system or universe or physical laws are. So how can we state our current models or observations are so complete and exhaustive so as to rule out any type of "free" will?

Adding some sort of "freedom" to make humans special seems unnecessary and inappropriate

Only to make them special in the same way we invoke temperature to define them different from "stars" or "ice". Or we invoke some form of difference we can observe. I observe people act differently to rocks, animals and computers.

I admit, I like leaving these sorts of questions shelved if they cannot be properly answered one way or another. But, as you can see in my previous paragraph, practically speaking I think we should understand humans as we understand the rest of the world. And we should be cautious not to let the differences between other animal machines and human machines cause us to jump to various self-indulgent conjectures.

I'm not sure what you mean by this. If I observe a bird can fly, it is "special" in that it's not a fish, that swims and cannot fly. Likewise, if I instead of attributing "flying ability" to birds but "decision making ability" to humans, what is "special" about it? Why shelve the fact and honest reflection that we cannot properly answer questions about "closed boxes"? Which is better, assuming the box has elves in it (unnatural and imaginary), is powered by quasars (natural but improbable), or to simply state "we do not know"?
It's all physics and stamp collecting.
It's not a particle or a wave. It's just an exchange.

User avatar
ucim
Posts: 6896
Joined: Fri Sep 28, 2012 3:23 pm UTC
Location: The One True Thread

Re: Definition of Free Will

Postby ucim » Sun Jan 20, 2013 3:28 pm UTC

Technical Ben wrote:My machine does not.
Your model does not what? I presume you mean it does not choose. But it does. Every second it chooses whether or not to flash a red light. (As it turns out, it always chooses to do so). It also chooses whether or not to flash a green light (though to this day nobody has been able to figure out how it decides!). Granted, this is not much of a choice, but it's also not much of a machine.

Technical Ben wrote:What chooses inside the machine? Well, that's the question of the riddle.
Actually, that's not what I asked. I asked "What chooses?". The point of my question is to ascertain whether you are thinking of the machine as a whole (Fred decided to go to the grocery store) or as a collection of interacting components (Fred's lack of nourishment caused a rise in certain enzymes which depressed certain brain functions and allowed other bits of cellular circuitry to gain dominance, causing his legs to propel him in the direction of the nearest grocery).

Your answer tells me that you are thinking in the latter way. You are "opening the box", so to speak. When you do so, the idea of "free will" stops being applicable, because you can see the mechanism. It doesn't matter whether there is randomness or not - the mechanism has no choice but to do what it does.

However, overall, the mechanism makes a choice.

In making this statement, we must now think of the machine as a whole. When I do this, stepping back so that I am not looking at the mechanism itself, the idea of free will starts to become applicable. It applies to a whole ("greater than the sum of its parts"), not to the simple collection of parts itself.

It's a way of looking at something.

Technical Ben wrote:...It only proves you/me/anyone cannot state "Free will cannot exist, because the only 2 options we have are random or not (or by opposite description, determined or not)".
The reason given does not support the statement made. In my mind, free will has nothing to do with whether or not the mechanism embodies randomness. So, what is it that the phrase "free will" means to you? This thread is about defining the term (or exploring the limitations of a given definition). Why do you tie randomness to the idea of free will? And while we're at it, what does "choose" mean to you?

When you say...
Technical Ben wrote:Will seems easy to contemplate. It's what we want.
... you are now looking at "we" (a person) wholeisticly, and not examining the mechanism. So when you compare a person to your machine (for the purposes of examining the concept of free will) you are not contemplating them in the same way. This is crucial.

Technical Ben wrote:As strawberry, being a fruit, and "fruit or not" is still separate from chocolate, being a bean and "bean or not"
Chocolate is in the "not fruit" category. It doesn't matter if it is a bean or not. Similarly strawberry is in the "not bean" category.

Jose
Order of the Sillies, Honoris Causam - bestowed by charlie_grumbles on NP 859 * OTTscar winner: Wordsmith - bestowed by yappobiscuts and the OTT on NP 1832 * Ecclesiastical Calendar of the Order of the Holy Contradiction * Heartfelt thanks from addams and from me - you really made a difference.

Technical Ben
Posts: 2986
Joined: Tue May 27, 2008 10:42 pm UTC

Re: Definition of Free Will

Postby Technical Ben » Sun Jan 20, 2013 4:17 pm UTC

How do you know it chooses? We can define choice, but how can we check my machine is choosing for any type of choice or mechanism? It's is a "closed box". We can only observe the outputs. We might be able to observe inputs, but for simplicity, for now we will only consider the outputs.

I might have missed some point, but how can we state my machine chooses to flash a light (green or red, randomly or not)?

I'd say in this context Fred and Fred's need for nourishment are the same thing, but by different names. Our ability to give the same things different names (I can call a Cat a Feline etc), does not make the similar things different does it? I'd concede if the two things are different (tiger V lion). But here, to me, Fred and his need to eat are the same thing, can we separate the two?

When you do so, the idea of "free will" stops being applicable, because you can see the mechanism.

It would not stop free will being applicable. Free will (my ability to choose which I observe so far to be removed from force or external requirement/determinism) may be randomly constructed, it may be deterministically constructed, it may be constructed by whatever means it is constructed by. I would say I do not know. To judge I have an exclusive list to construct decision mechanisms out of would be a folly on my part. So I'd reserve judgement and leave it as "free will".

the mechanism has no choice but to do what it does.
However, overall, the mechanism makes a choice.

This I find contradictory. Can you explain?

In my mind, free will has nothing to do with whether or not the mechanism embodies randomness.
Same here. :)
I've not tied it to free will, others have argued it was. To choose is to make a preference, decision or will. I can choose chocolate flavor or strawberry. I can choose via pre-existing preference, randomly or choose not to choose. As I observe the amount of choices seems unexhaustible and the level of influence, while being a factor, can always be vetoed by whatever choice mechanism I have internally, it is "free".

As to definitions of "we" and "mechanism", they are all perfunctory to me. That's why I used the example of the box. We cannot know if it's a whole box or a construct of smaller mechanisms. Can we? We just state "it is what it is". Or even "it is what we observe". My decision mechanism is, observed to me to be "free". :)

Chocolate is in the "not fruit" category. It doesn't matter if it is a bean or not. Similarly strawberry is in the "not bean" category.

By that though, "Not fruit and not bean" are not exhaustive in their exclusion, are they? Likewise, how can I be told that we have exhausted the possibility of my decisions being free from force/determinism and out of my ability to effect?
It's all physics and stamp collecting.
It's not a particle or a wave. It's just an exchange.

Greyarcher
Posts: 708
Joined: Thu Oct 18, 2007 3:03 am UTC

Re: Definition of Free Will

Postby Greyarcher » Sun Jan 20, 2013 8:04 pm UTC

Technical Ben wrote:Greyarcher, my argument is not that the human mind is a bit of a mystery, but that the entire system or universe or physical laws are. So how can we state our current models or observations are so complete and exhaustive so as to rule out any type of "free" will?
Our understanding of the world is incomplete. However, I don't think that avoids the problems I mentioned, some of which I shall discuss again below.

Technical Ben wrote:
Adding some sort of "freedom" to make humans special seems unnecessary and inappropriate

Only to make them special in the same way we invoke temperature to define them different from "stars" or "ice". Or we invoke some form of difference we can observe. I observe people act differently to rocks, animals and computers.
I disagree; if I'm not mistaken, you want humans to be "free"--I think you mentioned so in an earlier post. You see positive meanings or implications associated with the term, and wish to apply that label. It's not a necessary element for understanding humans; rather, it adds no understanding because you've been working the "ignorance" angle of argument to add a mystery "freedom" mechanism in that area of ignorance.
(I will also add, since I think I need to be careful, that an explanation that allows us to make sense of the world is not the same as something that deepens our understanding of the natural world. Gods were a popular explanations in the old days for various natural phenomena, but using them as an explanation surely did not deepen our understanding of the natural world. So please do not claim that "freedom" improves our understanding of humans by explaining something like unexpected of unpredictable behaviors; it would be essentially no different from utilizing gods.)

If there are positive consequences or implications to being "free" then you should simply try to figure out how to keep those positive elements without the need to add this vague trait.

Technical Ben wrote:
I admit, I like leaving these sorts of questions shelved if they cannot be properly answered one way or another. But, as you can see in my previous paragraph, practically speaking I think we should understand humans as we understand the rest of the world. And we should be cautious not to let the differences between other animal machines and human machines cause us to jump to various self-indulgent conjectures.
I'm not sure what you mean by this. If I observe a bird can fly, it is "special" in that it's not a fish, that swims and cannot fly. Likewise, if I instead of attributing "flying ability" to birds but "decision making ability" to humans, what is "special" about it? Why shelve the fact and honest reflection that we cannot properly answer questions about "closed boxes"? Which is better, assuming the box has elves in it (unnatural and imaginary), is powered by quasars (natural but improbable), or to simply state "we do not know"?
If a question cannot be answered through study of the world yet, I postpone making a conclusion on the question--normally. So if the question is, "Are human decisions 'free'?" then I would normally "shelve" the question because our studies of the brain and genetics have yet to reach a depth where we understand the many influences and interactions that go into the process.

However, I cannot simply leave it at that in my discussion with you, because you seem to shift from that non-conclusion to other dubious lines of thought. So I explained further, that practically speaking we shouldn't add extra concepts until they became necessary (especially when the way you use 'freedom' has some of the problems I explained above/below). Rather, we just use our usual methods of understanding the natural world to understand humans as well.

Also, I was talking about "freedom" and not decision-making ability. I think it is careless to equate the two.

As for "freedom" and treating humans as "special"...don't worry too much about the word choice; that's mostly a reference to various historical tendencies that I was reminded of.
-----but on that note, don't you think it is fundamentally unusual to argue for an unknown mechanism that is neither random nor deterministic but allows you to retain the claim that humans are "free" and all the intuitions you associate with that? Maybe you are more impartial and objective than I think and I'm reading too much into your word choice, but it doesn't seem like you're merely trying to deepen your understanding of humans here. Rather, you are trying to arrive at the conclusion that humans are free, and so your methodology is biased at its root. Am I wrong?
In serious discussion, I usually strive to post with clarity, thoroughness, and precision so that others will not misunderstand; I strive for dispassion and an open mind, the better to avoid error.

Technical Ben
Posts: 2986
Joined: Tue May 27, 2008 10:42 pm UTC

Re: Definition of Free Will

Postby Technical Ben » Sun Jan 20, 2013 9:40 pm UTC

disagree; if I'm not mistaken, you want humans to be "free"

I would say I observe them to be. Not that I may wish (any more than I do actually make decisions, so naturally am bias to there being a decision making ability I term "me making a decision") One not observed to fit the description pre-determined or random.

This is not ignorance or mystery any more than us not knowing how 2 black holes combine is mystery or ignorance. It would just be something we have yet to understand (I do believe we may do now, but it's still a good example). Just because some people have a difficulty understanding how humans make decisions, is no reason to say humans cannot make decisions, and in fact some other force or method or mechanism that we do understand makes the decision instead, is it?

I offer no explanation or label of the decision making mechanism other than "it is as I observe it to be". I'll give it the name "free will", but only in that I observe choice to be based on will, and to act freely. That makes little comment on the mechanism, right? Likewise, if you can tell me what my illustrative Box is made of, if it is allowed of disallowed from having free will, we can likewise say if humans are allowed or not too?

If there are positive consequences or implications to being "free" then you should simply try to figure out how to keep those positive elements without the need to add this vague trait.

What is vague about the trait? If I tie a rope, the other end is termed "free". "freedom of degrees of movement" appear to be valid terms. So, is "freedom to choose" a valid term?

So if the question is, "Are human decisions 'free'?" then I would normally "shelve" the question because our studies of the brain and genetics have yet to reach a depth where we understand the many influences and interactions that go into the process.
Fair enough. I've no comments if that is what you wish to do. Are you saying we cannot make any meaning full decision based on what we can observe?

Rather, we just use our usual methods of understanding the natural world to understand humans as well.

Well, that's fine up to the point humans break that observation or understanding. If we rule humans out as "natural" (which is the problem I see with all of these discussions) then a human does not act as a rock does. So if we know about rocks, it tells us nothing about humans. Likewise, how can this natural world (rocks) tell us about humans? We first define humans as natural also, right? Now why is any observation I make of humans, that they decide freely, not valid as a natural observation? Or is it valid to say humans can, and do, make decisions all by themselves and against all other influence at times?

Rather, you are trying to arrive at the conclusion that humans are free, and so your methodology is biased at its root. Am I wrong?

It would be a surprise to arrive at an opposite conclusion to observations, yes. :P I observe for instance that a rock has no ability to choose to go with the laws of gravity or not, it always falls. A bird has the choice to fly and defy gravity (only in that it can work against it, the law still stands), but is limited to it's needs. Where as a human can both defy gravity and decide when they wish to, at a whim even. Would you agree these are differences? If we can only observe the third with humans, then we may only observe the third with humans, not in any other system (rocks, birds etc).
It's all physics and stamp collecting.
It's not a particle or a wave. It's just an exchange.

User avatar
ucim
Posts: 6896
Joined: Fri Sep 28, 2012 3:23 pm UTC
Location: The One True Thread

Re: Definition of Free Will

Postby ucim » Mon Jan 21, 2013 1:47 am UTC

Technical Ben wrote:How do you know it chooses? We can define choice, but how can we check my machine is choosing for any type of choice or mechanism? It's is a "closed box".
Well, I would define making a choice as executing a single action from a set of more than one possible actions. In the case of your box, (viewed as a black box), the possible actions are to flash a light, or to not flash a light. At any given point, the machine executes one of these possible actions. Ergo, it chooses.
Spoiler:
The sticking point here of course is what it is that you consider "possible". This depends on how you are looking at the machine - either as a black box (where both choices are seen to be possible outcomes) or as a specific mechanism (where the only possible outcome will be the result of the laws of physics applied to that mechanism and that state at that moment). Note that this case also includes the case where the laws of physics include randomness.
Technical Ben wrote:I'd say in this context Fred and Fred's need for nourishment are the same thing, but by different names.
They are different ways of looking at the same thing. That makes them different in this context.

Technical Ben wrote:
ucim wrote:the mechanism has no choice but to do what it does. However, overall, the mechanism makes a choice.
This I find contradictory. Can you explain?
Look under the spoiler, above, and give it a good think. Then consider: when viewed as a black box, the mechanism makes a choice. It selects an action from a set of possible actions. This is "making a choice" (by my definition above). However, when we actually look at its internals, the set of possible choices is reduced, because we can now see how the choice is made, and in the case of a deterministic mechanism, can figure out what that choice will be. So, when we look at it this way, it is executing one action from a set that only contains one action. This is not a choice.

So, at least in the case of a deterministic mechanism, the idea of making a choice rests on not examining the mechanism too closely. The idea of making a choice is an expression of the solution set we are using (the range of possible outcomes), which is itself a result of our knowledge or ignorance (willful or otherwise) of the internals of the mechanism.

In the case of a mechanism that has a significant random component to it, the set of possible actions may be reduced, but not to one. So, I suppose you could say that the random part "chooses" (from whatever its possibilities are), but then the rest of the mechanism works by rote, as it were. But in any case, when you simply look at the entire mechanism as a black box (which you do when you say "It's is a 'closed box'"), the machine still does make a choice.

Technical Ben wrote:
ucim wrote:In my mind, free will has nothing to do with whether or not the mechanism embodies randomness.
Same here.
Then don't bring it into the discussion. It sets off too many irrelevant tangents.

Technical Ben wrote:As to definitions of "we" and "mechanism", they are all perfunctory to me.
That is your mistake. You need to be clear (to yourself and to us) what you mean by these words (such as when you say "Will seems easy to contemplate. It's what we want.") because when you are, you will see the non-parallelism in your reasoning.

Technical Ben wrote:By that though, "Not fruit and not bean" are not exhaustive in their exclusion, are they?
No, that is just one category. That category, plus its negation ("fruit or bean") is exhaustive. Even the set of all sets that are not members of themselves belongs to one of these two sets (not fruit and not bean).

Technical Ben wrote:Likewise, how can I be told that we have exhausted the possibility of my decisions being free from force/determinism and out of my ability to effect?
The word "likewise" makes no sense to me in this context. And I think it would be enlightening to split the question into two parts... one dealing with force, and the other dealing with determinism. They are disparate ideas.
Technical Ben, responding to Greyarcher wrote:I would say I observe them to be [free].
That is not an observation. That is a conclusion. It is a conclusion which I claim results from invalid reasoning. What, exactly, is the observation that leads you to this conclusion?

More generally, from that same post, you seem to think of humans as somehow "different in a different way" which has implications for free will. I don't see it that way. Humans are natural, just like rocks, birds, and cats. But I think you are confusing "natural" with "possessing a mechanism". What is it that you mean by "natural" and why is it important? The fact that both humans and rocks are natural does not make them identical.

Jose
Order of the Sillies, Honoris Causam - bestowed by charlie_grumbles on NP 859 * OTTscar winner: Wordsmith - bestowed by yappobiscuts and the OTT on NP 1832 * Ecclesiastical Calendar of the Order of the Holy Contradiction * Heartfelt thanks from addams and from me - you really made a difference.

Technical Ben
Posts: 2986
Joined: Tue May 27, 2008 10:42 pm UTC

Re: Definition of Free Will

Postby Technical Ben » Mon Jan 21, 2013 1:55 pm UTC

The set of actions observable by us on the box are different to the set of actions available to the box. If we are external observers, how can we prove the set of actions available? Can we prove or is it safe to assume we have and exhaustive list?

the possible actions are to flash a light
Or is it we observe this to be so, that a third option, to flash a blue light, may be available? How can we determine this?

It includes all cases. It's a box, and we only see it flashing lights. What do we state it's "state" is? Can we say it's a random box or a predetermined box? If I see the choice mechanism from humans as closed to observation currently, what statements am I allowed to make about human choice?

They are different ways of looking at the same thing. That makes them different in this context.

Is a coin different if we look at the heads over the tails? So, is the problem in viewing choice mechanisms in the mechanisms it's self, of the inability of heads to be tails? Perhaps we need to stop trying to force heads to be tails (getting confused over if human choice is random or deterministic) and agree to call both the same thing under the term "coin" (or here "free will/choice"). Would you agree it would be silly to argue over which a coin is, a head or a tail? :)

However, when we actually look at its internals...

First look at the internals of my box before we can make that statement. Although, as it's an illustrative black box, it may be rather difficult. Likewise, look into the makings of the laws of physics or the decisions of people. What statements can we say for sure? What statements can we not conclude (that is, which statements do we have insufficient information for)?

In the case of a mechanism that has a significant random component to it, the set of possible actions may be reduced, but not to one. So, I suppose you could say that the random part "chooses" (from whatever its possibilities are), but then the rest of the mechanism works by rote, as it were. But in any case, when you simply look at the entire mechanism as a black box (which you do when you say "It's is a 'closed box'"), the machine still does make a choice.

Have we confirmed we can reduce the mechanics down to "random or not" as you mention above? Or have we reduced the mechanics down to "deterministic or not" as you mention in the previous paragraph? I'd not see that we have yet. First we need to prove such states for my illustrative black box, don't we? Else how can I know I can reduce such decisions down for any other example, humans or otherwise?

As to "not bean", if something is described as "not random" and observed as "not deterministic", then is it contradictory? Or have I just ruled out 2 possible sets of how ever many sets there could be?

That is not an observation. That is a conclusion. It is a conclusion which I claim results from invalid reasoning. What, exactly, is the observation that leads you to this conclusion?

Is observing one end of a rope not tied down a false conclusion? Is observing a persons ability to choose opposite or freely from influence a false observation?

Humans are natural, just like rocks, birds, and cats.

Yes. I agree. Humans are not "different" from being "natural" and by are by definition are "different" than a "cat (etc)". Why is it ok to say "A human is different than a cat because humans do not have claws" but not ok in your mind to say "A human is different than a cat because humans have free will"? What part of that sentence makes the "difference" too different for your comfort?

Would you say it's correct to state rocks do not have free will, but humans do? Why would it be concluded that if rocks do not have free will (there is no observed mechanism in rocks for free will) that there cannot be one observed with humans?
It's all physics and stamp collecting.
It's not a particle or a wave. It's just an exchange.

User avatar
ucim
Posts: 6896
Joined: Fri Sep 28, 2012 3:23 pm UTC
Location: The One True Thread

Re: Definition of Free Will

Postby ucim » Mon Jan 21, 2013 3:48 pm UTC

Technical Ben wrote:The set of actions observable by us on the box are different to the set of actions available to the box. If we are external observers, how can we prove the set of actions available?
In math, there are things that cannot be proven. Some of those things are false (which is why they cannot be proven) and some of those things are true (which is odd, but true).

In science, nothing can be proven. Theories gain or lose credibility based on experiment, observation and analysis. New theories are suggested and are subjected to the same process. It is, I suppose, an act of faith that this process leads to a better understanding of our world. It certainly seems to lead to one that is more consistant with observation, which is as good as it gets.

But we are not even doing science here. This is more like philosophy, or even lexicography (the point of this thread is to attempt to define a philosophical concept).

Technical Ben wrote:
ucim wrote:the possible actions are to flash a light
Or is it we observe this to be so, that a third option, to flash a blue light, may be available? How can we determine this?
A blue light is a light. In any case, you have presented your box and told us what it does. I took that at face value. If we are presented with a box whose function we do not know, and we observe it to flash one light in a pattern, and another light in a seemingly random fashion, and we never see it flash a third light (do we see the unlit bulb?), then we are left with a situation that could in the future surprise us. I don't see how any of this is relevant.

Technical Ben wrote:If I see the choice mechanism from humans as closed to observation currently, what statements am I allowed to make about human choice?
You are allowed to make any statements you want, but the only sensible ones would involve seeing the human as a whole. We are familiar with humans and know what their capabilities tend to be. Therefore we have a good idea of what the behavioral choice set is, and know that it is pretty vast. Nonetheless, a given human at any given time behaves in one manner. This is by my earlier definition, a choice. It is a choice even if the choice is random (as you have said earlier).

Technical Ben wrote:Is observing one end of a rope not tied down a false conclusion? Is observing a persons ability to choose opposite or freely from influence a false observation?
To the first question, it is not a conclusion, it is an observation (of the condition of the end of the rope). You cannot conclude from this observation alone that the end of the rope is free. It could be frozen, it could be glued, it could have a price tag on it. :)

To the second question, it is not an observation, it is a conclusion. This conclusion can reasonably be made only after many observations of a person's choices under different circumstances.

Free as in beer, free as in speech, and free as in not-tied-down are three different concepts which share the same word. Free as in will is yet a fourth sense of the word. Also, observations and conclusions are vastly different animals. Don't confuse them if you want sensible results.

Technical Ben wrote:Why is it ok to say "A human is different than a cat because humans do not have claws" but not ok in your mind to say "A human is different than a cat because humans have free will"? What part of that sentence makes the "difference" too different for your comfort?
The undefined term "free will", which is being bandied about as if we all know what it means. The point of this thread is that we don't all agree on what it means. Your comments illustrate this pretty effectively.

But before discussing free will, we need to agree on what "choice" means, since it is central to the concept of free will.
proposing a definition of 'making a choice', ucim wrote:Making a choice can be defined as executing a single action from a set of more than one possible actions.
Does this makes sense to you?

Jose
Order of the Sillies, Honoris Causam - bestowed by charlie_grumbles on NP 859 * OTTscar winner: Wordsmith - bestowed by yappobiscuts and the OTT on NP 1832 * Ecclesiastical Calendar of the Order of the Holy Contradiction * Heartfelt thanks from addams and from me - you really made a difference.

shvedsky
Posts: 13
Joined: Mon Jan 21, 2013 6:39 am UTC

Re: Definition of Free Will

Postby shvedsky » Tue Jan 22, 2013 7:15 am UTC

ucim wrote:Free will, at its essence, involves the ability to choose.


Could you clarify why?

To me, free will doesn't necessary involve the ability to choose. It only requires—and is equivalent to—the ability to understand what "you" might and might not do, to realize how "your" actions affect the world, and to reflect upon that. "You" don't necessarily have to be able to choose what to do in order to realize that what "you" have not done could have had different impact.

User avatar
jules.LT
Posts: 1539
Joined: Sun Jul 19, 2009 8:20 pm UTC
Location: Paris, France, Europe

Re: Definition of Free Will

Postby jules.LT » Tue Jan 22, 2013 1:37 pm UTC

If you take actions, then you have made choices relative to those actions. Unless they were the result of a simple automatic reflex, in which case they are not an expression of free will.
I can't see how you could express free will without choice.

Also, a posteriori rationalization has nothing to do with free will. You could very well be killed the instant you take a free decision, and that decision would still be an expression of free will.

proposing a definition of 'making a choice', ucim wrote:Making a choice can be defined as executing a single action from a set of more than one possible actions.

Absolutely not. A pseudo-random function does not make choices.
In order for the selection of an action to be a "choice", there has to be analysis of the situation, simulation of the effect of different possible actions and a decision based on those deliberations.
Bertrand Russell wrote:Not to be absolutely certain is, I think, one of the essential things in rationality.
Richard Feynman & many others wrote:Keep an open mind – but not so open that your brain falls out

User avatar
ucim
Posts: 6896
Joined: Fri Sep 28, 2012 3:23 pm UTC
Location: The One True Thread

Re: Definition of Free Will

Postby ucim » Tue Jan 22, 2013 2:04 pm UTC

jules.LT wrote:
proposing a definition of 'making a choice', ucim wrote:Making a choice can be defined as executing a single action from a set of more than one possible actions.
Absolutely not. A pseudo-random function does not make choices.
In order for the selection of an action to be a "choice", there has to be analysis of the situation, simulation of the effect of different possible actions and a decision based on those deliberations.
This sounds more like your proposed definition of "free will".

I think it is useful to have a single common word for "executing a single action from a set of more than one possible actions". Paralleling math, we could call this "making a weak choice", and use your definition for "making a strong choice". Both concepts are useful, and separating the two out could be the source of some enlightenment.

Jose
Order of the Sillies, Honoris Causam - bestowed by charlie_grumbles on NP 859 * OTTscar winner: Wordsmith - bestowed by yappobiscuts and the OTT on NP 1832 * Ecclesiastical Calendar of the Order of the Holy Contradiction * Heartfelt thanks from addams and from me - you really made a difference.

User avatar
jules.LT
Posts: 1539
Joined: Sun Jul 19, 2009 8:20 pm UTC
Location: Paris, France, Europe

Re: Definition of Free Will

Postby jules.LT » Tue Jan 22, 2013 2:34 pm UTC

For a start, wikipedia defines choice as "the mental process of judging the merits of multiple options and selecting one or more of them". Other sources give similar results.
It looks a lot like my proposed definition of free will because the two are intricately intertwined: free will requires choices.

Find a word that doesn't have such an important additional meaning in this discussion, or use the whole expression...
ucim wrote:
Technical Ben wrote:What chooses inside the machine? Well, that's the question of the riddle.
Actually, that's not what I asked. I asked "What chooses?". The point of my question is to ascertain whether you are thinking of the machine as a whole (Fred decided to go to the grocery store) or as a collection of interacting components (Fred's lack of nourishment caused a rise in certain enzymes which depressed certain brain functions and allowed other bits of cellular circuitry to gain dominance, causing his legs to propel him in the direction of the nearest grocery).

Your answer tells me that you are thinking in the latter way. You are "opening the box", so to speak. When you do so, the idea of "free will" stops being applicable, because you can see the mechanism. It doesn't matter whether there is randomness or not - the mechanism has no choice but to do what it does.

However, overall, the mechanism makes a choice.

In making this statement, we must now think of the machine as a whole. When I do this, stepping back so that I am not looking at the mechanism itself, the idea of free will starts to become applicable. It applies to a whole ("greater than the sum of its parts"), not to the simple collection of parts itself.

I kinda get what you say, and I agree with the general idea, but for the nth time:
The idea of free will doesn't stop being applicable "when you see the mechanism". It's just not applicable TO the mechanism.
If you're talking about the whole, but keeping in mind how the mechanism works, it doesn't change anything about the free will status of the whole.

Seriously, can't you just say "only a complex system can make choices; the bits in it contribute to the choice, but they don't make choices themselves". Just like no part of your feet "walks": they merely form part of the process.
Bertrand Russell wrote:Not to be absolutely certain is, I think, one of the essential things in rationality.
Richard Feynman & many others wrote:Keep an open mind – but not so open that your brain falls out

User avatar
ucim
Posts: 6896
Joined: Fri Sep 28, 2012 3:23 pm UTC
Location: The One True Thread

Re: Definition of Free Will

Postby ucim » Tue Jan 22, 2013 3:57 pm UTC

jules.LT wrote:For a start, wikipedia defines choice as "the mental process of judging the merits of multiple options and selecting one or more of them". Other sources give similar results.
If we are arguing from authority, the first dictionary I grabbed (American Heritage) has, as its first definition of "choose", "to select from a number of possible alternatives". The second dictionary I opened (Concise Oxford) has, as its first definition, "Select out of greater number".

Other definitions are also shown, but the point is that the word is commonly used to simply mean "select", without any additional baggage, and is a perfectly good word to use in the weak sense, especially when dealing with entities whose internals we do not know well enough. (The dog in dog food commercials chooses one bowl over the other, my computer algorithm chooses which folder to drop an incoming message into.)

jules.LT wrote:for the nth time: The idea of free will doesn't stop being applicable "when you see the mechanism". It's just not applicable TO the mechanism.
Consider the two phrases synonymous if you like. Generally in this discussion, opening the box (looking at the mechanism) is followed by discussion of the mechanism. The idea of free will does not apply TO the mechanism, and thus, does not apply to the discussion OF the mechanism. This is what I mean by "stops being applicable".

jules.LT wrote:If you're talking about the whole, but keeping in mind how the mechanism works, it doesn't change anything about the free will status of the whole.
Agreed. Fully.

jules.LT wrote:Seriously, can't you just say "only a complex system can make choices; the bits in it contribute to the choice, but they don't make choices themselves". Just like no part of your feet "walks": they merely form part of the process.
That's pretty much what I am saying (with the quibble that we haven't decided what is "complex enough")

I think we are saying the same thing in different ways (each for our own reasons). I say it my way because there are some that would argue that in the case where the mechanism is deterministic, the whole does not have free will (because the parts don't). I maintain that even if the parts don't, the whole may. This is my way of drawing the picture.

Jose
Order of the Sillies, Honoris Causam - bestowed by charlie_grumbles on NP 859 * OTTscar winner: Wordsmith - bestowed by yappobiscuts and the OTT on NP 1832 * Ecclesiastical Calendar of the Order of the Holy Contradiction * Heartfelt thanks from addams and from me - you really made a difference.

Trebla
Posts: 387
Joined: Fri Apr 02, 2010 1:51 pm UTC

Re: Definition of Free Will

Postby Trebla » Tue Jan 22, 2013 4:27 pm UTC

jules.LT wrote:Seriously, can't you just say "only a complex system can make choices; the bits in it contribute to the choice, but they don't make choices themselves". Just like no part of your feet "walks": they merely form part of the process.


I'm not convinced this is a good analogy. Walking is a process which is made up of lots of littler processes. Making a choice is an atomic piece of that process that's the result of a decision making process... the decision making process may be complex and have contributors, but the flip between "this and that" is atomic and the mechanism that actually "flips that switch" (for lack of a better spur of the moment term) is something different from the process. There is a moment where the complex system's decision making process hits an event where some reduced piece of the complex system actually makes what we're calling a "choice." I agree that the bits contribute to the choice, but there is a single point of failure in the process where a reduced portion of the system is the actual cause of the "choice." I'm not nearly as eloquent a speaker as most here, so I'm probably not being as clear as I'd like.

"The person makes the choice, but not a part of the person" seems incorrect... can it be reduced to the point where "That synapse firing is the actual mechanism that singularly defines the individual's actions (starts the process, at least)?"

User avatar
jules.LT
Posts: 1539
Joined: Sun Jul 19, 2009 8:20 pm UTC
Location: Paris, France, Europe

Re: Definition of Free Will

Postby jules.LT » Tue Jan 22, 2013 4:30 pm UTC

ucim wrote:If we are arguing from authority, the first dictionary I grabbed (American Heritage) has, as its first definition of "choose", "to select from a number of possible alternatives". The second dictionary I opened (Concise Oxford) has, as its first definition, "Select out of greater number".

Other definitions are also shown, but the point is that the word is commonly used to simply mean "select", without any additional baggage, and is a perfectly good word to use in the weak sense, especially when dealing with entities whose internals we do not know well enough. (The dog in dog food commercials chooses one bowl over the other, my computer algorithm chooses which folder to drop an incoming message into.)

I'm afraid "to select" has the exact same additional baggage, which is why a dictionary does not need to mention anything further once it's mentioned that word.
Laughably, google definitions define "choose" as "select" and "select" as "choose" :-P
Arguably, dogs do make choices, even if they're simpler than some of ours.
Current computers definitely don't. People don't really think that computer make "real choices", it's just a turn of phrase.

ucim wrote:I think we are saying the same thing in different ways (each for our own reasons). I say it my way because there are some that would argue that in the case where the mechanism is deterministic, the whole does not have free will (because the parts don't). I maintain that even if the parts don't, the whole may. This is my way of drawing the picture.
My way of saying it doesn't have a problem with whatever characteristics one may want to assign to the inner mechanisms, and it doesn't introduce mysterious phenomenons like concepts "disappearing".
Bertrand Russell wrote:Not to be absolutely certain is, I think, one of the essential things in rationality.
Richard Feynman & many others wrote:Keep an open mind – but not so open that your brain falls out

User avatar
jules.LT
Posts: 1539
Joined: Sun Jul 19, 2009 8:20 pm UTC
Location: Paris, France, Europe

Re: Definition of Free Will

Postby jules.LT » Tue Jan 22, 2013 4:42 pm UTC

Trebla wrote:I'm not convinced this is a good analogy. Walking is a process which is made up of lots of littler processes. Making a choice is an atomic piece of that process that's the result of a decision making process... the decision making process may be complex and have contributors, but the flip between "this and that" is atomic and the mechanism that actually "flips that switch" (for lack of a better spur of the moment term) is something different from the process. There is a moment where the complex system's decision making process hits an event where some reduced piece of the complex system actually makes what we're calling a "choice." I agree that the bits contribute to the choice, but there is a single point of failure in the process where a reduced portion of the system is the actual cause of the "choice." I'm not nearly as eloquent a speaker as most here, so I'm probably not being as clear as I'd like.

"The person makes the choice, but not a part of the person" seems incorrect... can it be reduced to the point where "That synapse firing is the actual mechanism that singularly defines the individual's actions (starts the process, at least)?"

It sure feels instantaneous, but then again our feeling about it is a post hoc rationalization of an unconscious process...
Same as "there is a moment when the foot leaves the ground": at the molecular level there really isn't. How many molecular bindings count for being "in contact"? (no two atoms are in contact anyway)

As for a single synapse shot being the moment of choice, I don't see what would distinguish it from the one before or the one after. Or the many ones all around it, for that matter, because I don't see how anything significant can happen in our brain through just one synapse. The further back you go in the origins of a brain output, the more elements will be involved, just like a person has innumerable grand-grand-grand-[...]-parents. Until you reach the first brain cell that differentiated in your mother's womb (and that's if you only take brain cells into account).
Bertrand Russell wrote:Not to be absolutely certain is, I think, one of the essential things in rationality.
Richard Feynman & many others wrote:Keep an open mind – but not so open that your brain falls out

Trebla
Posts: 387
Joined: Fri Apr 02, 2010 1:51 pm UTC

Re: Definition of Free Will

Postby Trebla » Tue Jan 22, 2013 5:09 pm UTC

jules.LT wrote:It sure feels instantaneous, but then again our feeling about it is a post hoc rationalization of an unconscious process...
Same as "there is a moment when the foot leaves the ground": at the molecular level there really isn't. How many molecular bindings count for being "in contact"? (no two atoms are in contact anyway)

As for a single synapse shot being the moment of choice, I don't see what would distinguish it from the one before or the one after. Or the many ones all around it, for that matter, because I don't see how anything significant can happen in our brain through just one synapse.


I really don't know, intuitively it seems like there must be a horizon point at which the process reduces to an atomic decision before branching back out into myriad results... I don't know if there is a single synapse firing, or a cluster, or if that concept of such a horizon isn't even valid in this regard. Going with the family tree analogy, my many progenitors (decision making process) lead to a single me (the choice) which leads to many heirs (resulting action). The "atomic piece" where my progenitors stop being my progenitors and become my heirs is the single switch that is me.

Again, maybe that's not how the brain works.

Presumably there's one planck moment where there is a force pushing down at least one molecule compressing the space between some molecule of the foot and ground (that we colloquially call "contact") and the next planck moment where that's not correct. Is that not true? I really don't know molecular mechanics well enough to say, but my layman understanding would lead me to believe that. Maybe there are two or more molecules that break "contact" in the same planck moment transition (what's that called? a planck second?) but that's still vastly reduced from "the foot leaving the ground" which is an entirely different thing than the process of walking.

Edit: corrected plank to planck... that's how little I know!

User avatar
jules.LT
Posts: 1539
Joined: Sun Jul 19, 2009 8:20 pm UTC
Location: Paris, France, Europe

Re: Definition of Free Will

Postby jules.LT » Tue Jan 22, 2013 5:30 pm UTC

That's where the analogies break.
Many factors come into a choice. For each, numerous synapses are active. And afterwards countless synapses are active again.
I find it absurd that there could be one central synapse in the process where the "choice" happens, but I can't prove it isn't the case.
Let's consider the case anyway: how could this single jolt of electricity reasonably be called a "choice"?
I contend that the choice can only be seen as an attribute of the whole.
Bertrand Russell wrote:Not to be absolutely certain is, I think, one of the essential things in rationality.
Richard Feynman & many others wrote:Keep an open mind – but not so open that your brain falls out

User avatar
ucim
Posts: 6896
Joined: Fri Sep 28, 2012 3:23 pm UTC
Location: The One True Thread

Re: Definition of Free Will

Postby ucim » Tue Jan 22, 2013 5:36 pm UTC

jules.LT wrote:Laughably, google definitions define "choose" as "select" and "select" as "choose" :-P
My dictionaries do the same, and I suspect most do. After all, English words are being defined by other English words. However, there is no indication that your "additional baggage" is a required part of the definition, and it is not the way I and most of the people I work with use the word in casual parlance. One can choose after deliberation, or one can choose without deliberation. And in casual parlance, computers are said to choose between alternatives. The word does not necessarily carry baggage. And given that there is no common word for "choosing without contemplation being necessary", and given that both "choose" and "select" are used to fill that void, I find them perfectly good words for the concept.

Given that a distinction is helpful here, I propose "weakly choose" and "strongly choose", but am open to alternatives.

jules.LT wrote:People don't really think that computer make "real choices", it's just a turn of phrase.
"Turn of phrase" is the same as "usage of the word". But now you are distinguishing between "choices" and "real choices"... that seems like a value judgment to me. You are attaching the baggage right there. And in any case, why could a computer not be said to imagine future courses of action, deliberate about reasons, plan actions, and control actions in the face of competing desires. Isn't this exactly what a chess computer does, in its own way, when it chooses to move the bishop instead of the rook?

Trebla wrote:Presumably there's one planck moment where there is a force pushing down at least one molecule compressing the space between some molecule of the foot and ground (that we colloquially call "contact") and the next planck moment where that's not correct. Is that not true?
That is probably not true. Molecules are mostly empty space anyway, and are far bigger than the Planck distance.

Trebla wrote: I agree that the bits contribute to the choice, but there is a single point of failure in the process where a reduced portion of the system is the actual cause of the "choice."
I don't see why that should be so.

Jose
Order of the Sillies, Honoris Causam - bestowed by charlie_grumbles on NP 859 * OTTscar winner: Wordsmith - bestowed by yappobiscuts and the OTT on NP 1832 * Ecclesiastical Calendar of the Order of the Holy Contradiction * Heartfelt thanks from addams and from me - you really made a difference.

Trebla
Posts: 387
Joined: Fri Apr 02, 2010 1:51 pm UTC

Re: Definition of Free Will

Postby Trebla » Tue Jan 22, 2013 5:40 pm UTC

jules.LT wrote:That's where the analogies break.
Many factors come into a choice. For each, numerous synapses are active. And afterwards countless synapses are active again.
I find it absurd that there could be one central synapse in the process where the "choice" happens, but I can't prove it isn't the case.
Let's consider the case anyway: how could this single jolt of electricity reasonably be called a "choice"?
I contend that the choice can only be seen as an attribute of the whole.


Then we're talking about the entire decision making process as being the choice? I will admit to being less interested in the "many factors" than I am in the fundamental underlying... I don't even know what to call it, so I'll stick with "flip of the switch" and the unique driver of that flip as contrasted with the earlier example of a black box computer (which is odd since I'm far less interested in individual fat molecules than I am in female, how do I stay PC... mammary glands, as a counterexample).

I will drop back out now and allow the thread to continue at that level. I'm with you on the "can a single jolt be called a choice?" and really don't know how to answer it.

Jose wrote:I don't see why that should be so.


It may not be so, I can't back it up, just that it seems like the complexity would be reducible. That's a limit of my understanding of reducible complexity.

User avatar
jules.LT
Posts: 1539
Joined: Sun Jul 19, 2009 8:20 pm UTC
Location: Paris, France, Europe

Re: Definition of Free Will

Postby jules.LT » Tue Jan 22, 2013 7:14 pm UTC

ucim wrote:One can choose after deliberation, or one can choose without deliberation.
One can choose with very little deliberation, but with no deliberation at all it isn't a choice, it's an automatism.
ucim wrote:But now you are distinguishing between "choices" and "real choices"... that seems like a value judgment to me. You are attaching the baggage right there.
I'm merely asking what people "really" mean by that word. When the casual use and use of a word in a more serious context are different, I think we can dismiss the casual use.

ucim wrote:why could a computer not be said to imagine future courses of action, deliberate about reasons, plan actions, and control actions in the face of competing desires. Isn't this exactly what a chess computer does, in its own way, when it chooses to move the bishop instead of the rook?
Most computers use radically different structures in their information processing, but chess computers seem rather close to what we do. I don't know enough about the phases of their "decision" processes to know for sure if I'd call it "choice". There is also the possibility that there is another criteria to add to the list for something to be "choice".

Trebla wrote:I will drop back out now and allow the thread to continue at that level. I'm with you on the "can a single jolt be called a choice?" and really don't know how to answer it.
My answer would be that there is no "unique driver" :-P

Thanks for your contribution, anyway. It was interesting ^_^
Bertrand Russell wrote:Not to be absolutely certain is, I think, one of the essential things in rationality.
Richard Feynman & many others wrote:Keep an open mind – but not so open that your brain falls out

User avatar
ucim
Posts: 6896
Joined: Fri Sep 28, 2012 3:23 pm UTC
Location: The One True Thread

Re: Definition of Free Will

Postby ucim » Tue Jan 22, 2013 8:38 pm UTC

Jules.LT wrote:There is also the possibility that there is another criteria to add to the list for something to be "choice".
This sounds suspiciously like moving the goalposts; ensuring that "choice" is defined in such a way as to support a pre-conceived notion of free will.

Jose
Order of the Sillies, Honoris Causam - bestowed by charlie_grumbles on NP 859 * OTTscar winner: Wordsmith - bestowed by yappobiscuts and the OTT on NP 1832 * Ecclesiastical Calendar of the Order of the Holy Contradiction * Heartfelt thanks from addams and from me - you really made a difference.

User avatar
jules.LT
Posts: 1539
Joined: Sun Jul 19, 2009 8:20 pm UTC
Location: Paris, France, Europe

Re: Definition of Free Will

Postby jules.LT » Tue Jan 22, 2013 9:16 pm UTC

Free will is what we describe as free will. We just need to find out what exactly that is, solve the inconsistencies and draw the conclusions.
Bertrand Russell wrote:Not to be absolutely certain is, I think, one of the essential things in rationality.
Richard Feynman & many others wrote:Keep an open mind – but not so open that your brain falls out

shvedsky
Posts: 13
Joined: Mon Jan 21, 2013 6:39 am UTC

Re: Definition of Free Will

Postby shvedsky » Wed Jan 23, 2013 6:52 am UTC

Maybe the "metrological" approach could work? For instance, what experiment could I carry out to verify that I have free will?

User avatar
jules.LT
Posts: 1539
Joined: Sun Jul 19, 2009 8:20 pm UTC
Location: Paris, France, Europe

Re: Definition of Free Will

Postby jules.LT » Wed Jan 23, 2013 10:08 am UTC

I can't remember if you accepted the proposed list of capacities as necessary for free will, but we might want to add other criteria:
- Dynamic goal/sub-goal adjustment (but not so for choice).
- Maybe something related to how we understand concepts within networks of related ideas, and make correspondences between concepts according to those links. Although I don't think so.

About goals:
Humans originally share with all self-perpetuating processes the ultimate "goal" of self-perpetuation. That is our nature.
Our main subgoals are avoiding harm and seeking gratification.
The relationship between those (and other, further removed subgoals) and the original ultimate goal can get clouded. Because we are so complex and because of the random "errors" introduced by evolutionary processes, we can have goals that are irrelevant or even disadvantageous for self-perpetuation. Those goals are still sometimes recognized by the consciousness as important or even prevailing.

I think that there is something about how those goals and their subgoals are set that is essential to free will, although I'm not yet sure what.
Bertrand Russell wrote:Not to be absolutely certain is, I think, one of the essential things in rationality.
Richard Feynman & many others wrote:Keep an open mind – but not so open that your brain falls out

User avatar
XTCamus
Posts: 176
Joined: Fri Sep 02, 2011 11:59 pm UTC
Location: "...on that dizzying crest"

Re: Definition of Free Will

Postby XTCamus » Wed Jan 23, 2013 7:57 pm UTC

shvedsky wrote:Maybe the "metrological" approach could work? For instance, what experiment could I carry out to verify that I have free will?

You should be able to prove that you yourself have free will, or at least a degree of free will, as most here do not consider this a binary question, by first thinking of a simple act that you are capable of (or even recording your intent somewhere if you want "evidence"), and then performing that act. I.e. I intend to raise my right arm. Repeat as needed until you have sufficient data to convince yourself. Most people already strongly believe in this degree of free will so they never bother actually performing experiments.

The harder question is what evidence would you need from someone else, (or from an ape, or from an AI program) that would convince you that they have a certain degree of free will? This has been asked here before, but has not led to anything close to consensus. Many seem to view it as practically impossible. I would like to see this thread work harder on this question though. If an AI program was to explain to me why it was convinced it did have free will, and had performed one of the above experiment itself several times, I would readily admit the possibility, or even grant for the time being, until shown otherwise that the AI program did have a degree of free will.

Likewise if an ape learned enough sign language to make similar claims, I would grant the same. However in my case, I already tend to think that apes and even dogs, already have a rudimentary degree of free will (but I am probably in the minority here on this). But I would need to see more from an AI before graning them even instinctive "animal" equivalency.


Return to “Serious Business”

Who is online

Users browsing this forum: No registered users and 10 guests