Compatibilism: What's that About?

[Jarhyn]

Choosing is something talking organisms say...

Then you don't understand in even the least bit what we are talking about. Maybe see Elixer's latest post in the proof of God thread.

We are talking about properties that pertain to systems specifically "the properties of deterministic systems".

You have posed that choosing operations in the definition of my model (which happens to be the same model physicists use to describe the universe as deterministic) are not choosing operations in the definition of my model.

I can point absolutely to this thing that I call choice as a set of numbers and their relationship on a graph of the system state through time.

So, there are still arguments being made, I guess, from ignorance.

P1 Deterministic systems do not allow local free will
P2 The Universe is a deterministic system
C The universe does not allow local free will

You will need to prove P1.

P1 is a mathematical hypothesis.

To prove 1 you will need to provide a mathematical proof.

To disprove 1 we just need to find ANY deterministic system with a well defined property of "local free will" existing on its graph.

Then you have to also prove that "free will" is not a property of stochastic systems because deterministic systems may contain locally stochastic systems.

So really, your goal is to prove that no property within all of math and graph and systems theory is describable by the phrase "free will."

I can point absolutely to this thing that I call choice as a set of numbers and their relationship on a graph of the system state through time.

So if I have free, and I have will, as related concepts (such that a will can be described as having freedom) on a graph description of a deterministic system, even if it happens in a stochastic locality of that deterministic system, I have disproved that determinism requires the absence of free will.

 

[fromderinside]

Choosing is something talking organisms say...

...
We are talking about properties that pertain to systems specifically "the properties of deterministic systems".

You have posed that choosing operations in the definition of my model (which happens to be the same model physicists use to describe the universe as deterministic) are not choosing operations in the definition of my model.

I can point absolutely to this thing that I call choice as a set of numbers and their relationship on a graph of the system state through time.

OK. A set of numbers is, at most, an abstraction. And it is so only if the numbers arise from empirical material operations.

So if I have free, and I have will, as related concepts (such that a will can be described as having freedom) on a graph description of a deterministic system, even if it happens in a stochastic locality of that deterministic system, I have disproved that determinism requires the absence of free will.

Seems to me your argument is based on logical arguments. I argue that logic without necessary tie-ins to material foundations is subjective in nature rather than objective. Below is the article to which I refer in making this inference.

Logical mechanism design: https://www.mit.edu/~irahwan/docs/KER2011a.pdf

Game theory is becoming central to the design and analysis of computational mechanisms in which multiple entities interact strategically. The tools of mechanism design are used extensively to engineer incentives for truth revelation into resource allocation (e.g. combinatorial auctions) and preference aggregation protocols (e.g. voting). We argue that mechanism design can also be useful in the design of logical inference procedures. In particular, it can help us understand and engineer inference procedures when knowledge is distributed among self-interested agents. We set a research agenda for this emerging area, and point to some early research efforts.

Specifically, I want you to examine the Social choice function which is what I believe to what you are referring to when you advocate choice.

Your generalization associating logical choice to determinism is very limited. You are using a logical construction rather than an empirical material construction. I find the logic interesting but not definitive re materiality. I think a reasonable interpretation would be you are using logical positivism techniques.

Now I use such when I'm designing an experiment but my experiment is against the real world rather than a logical world. I examine material outcomes rather than logical ones.

I say an observer responds by building a word model of what her effectors are doing rather than is she making a conscious choice. And as Pavlov proved salivation is not choice.

I spent some time studying signal detection theory, the ideal versus the human. Similar models, similar differences. Humans are not ideal observers. However one can make inferences for future experiments. It is from that kind of thinking I came to the realization that the basis for human and most mammal hearing is driven by sound in motion, detecting where things are going.

 

[DBT]

You seem to want to change the rules

No, you ARE changing the rules.

A deterministic system is a system in which no randomness is involved in the development of future states of the system.

Yes. That essentially is determinism. Your wording doesn't contradict anything I have said, nor the definition given in Stanford.

A stochastic system has a random probability distribution or pattern that may be analysed statistically but may not be predicted precisely.

Again, we are not talking about a stochastic system. The subject is the compatibility of free will in relation to determinism....not random or stochastic systems.

This is then applied to our world as a statement.

If you do not understand what this means and why determinism must be approached from the level of state machines and math, you are hopelessly lost.

You miss the point entirely.

 

[DBT]

And, the forum lost most of my reply, whatever.

You don't even understand that the Virtual Particle activity of our universe means that to declare our universe deterministic is a declaration of pure faith.

So, there are still arguments being made, I guess, from ignorance.

P1 Deterministic systems do not allow local free will
P2 The Universe is a deterministic system
C The universe does not allow local free will

You will need to prove P1.

P1 is a mathematical hypothesis.

To prove 1 you will need to provide a mathematical proof.

To disprove 1 we just need to find ANY deterministic system with a well defined property of "local free will" existing on its graph.

An interesting thing here is that this still doesn't answer the fact that P2 is also plainly being taken on your faith and only tolerated here so we can have a debate at all: the universe is possibly stochastic (not deterministic).

Then you have to also prove that "free will" is not a property of stochastic systems because deterministic systems may contain locally stochastic systems.

So really, your goal is to prove that no property within all of math and graph and systems theory is describable by the phrase "free will."

Whether the universe is stochastic has bearing on this debate, which is about free will in a deterministic universe.

If you want to argue about free will in a stochastic universe, start your own thread on the subject.

 

[DBT]

P1: A freely chosen will is when someone chooses for themselves what they will do, while free of coercion and other forms of undue influence.
P2: A world is deterministic if every event is reliably caused by prior events.
P3: A freely chosen will is reliably caused by the person's own goals, reasons, or interests (with their prior causes).
P4: An unfree choice is reliably caused by coercion or undue influence (with their prior causes).
C: Therefore, the notion of a freely chosen will (and its opposite) is still meaningful within a fully deterministic world.

In a determined world, the possibility of doing otherwise is an illusion.

A "possibility" exists solely within the imagination. We cannot drive a car across the possibility of a bridge. We can only drive across an actual bridge.

But a possibility is not an illusion, because we cannot build an actual bridge without first imagining a possible bridge. The notion of a possibility serves an essential mental function, enabling us to plan our deliberate actions to accomplish some deliberate intent. The thought of a possibility has empirical functionality, that is, it causes things to happen in the real world.

The thought of a possibility is an empirical event, and just like every other event, it is causally necessary from any prior point in time.

If determinism allows any action to be taken, it's not determinism. The word 'determined' means final. Final as in fixed. Finalized or fixed by prior states of the world.

Imagining building a bridge doesn't appear in a vacuum. Conditions in the world in relation to human needs, wants and engineering skills brings forth the idea of building a bridge, and if the need is strong enough, the impetus to put imagination into action.

Countless elements and events went into imagining building a bridge and carrying it out.

Each and every diner is restricted to the option that was determined.

Each and every diner will necessarily consider multiple possibilities from the menu as they decide for themselves what they will order for dinner. The multiple possibilities are just as inevitable as the single option that was finally chosen.

Determinism changes nothing. Everything happens in precisely the same way.

The brain of each and every diner acquires information from the menu, which is processed (brain state, proclivities, etc), conscious deliberation generated and related action initiated, you choose Lobster on that occasion. The only possible result on that occasion.

Information processing, not free will.

''If Determinism is true, human beings lack the ability to think in a manner that is not 100% caused by prior activity that is outside of their control, and thereby lack Free Will. By the same token, if human beings have Free-Will, they are capable of thinking in a manner that is not 100% caused by prior activity that is outside of their control, which rules out Determinism." ,,, - Bruce Silverstein, B.A. Philosophy

Nope. If you start out with false assumptions you will end with false conclusions, as Mr. Silverstein does. The assumption that our thoughts are 100% caused by things that are not our thoughts is false. Thinking, like walking, is something we do. It is not done for us by some external entity. It is actually us doing the walking. It is actually us doing the thinking.

It's a brief summary with a bit rhetoric thrown in, but essentially sound.

Sensory information acts upon the system. As described in prior posts, the brain has evolved to acquire information from the environment and respond to it in adaptive ways (if functional)

We respond to objects and events in the external world, which our brain represents in conscious form. We don't choose our genetic makeup or neural architecture, how our brain functions or responds.

The illusion of conscious control is revealed when things go wrong with the system.

This is a simple empirical observation that cannot be dismissed by metaphysical abstractions.

Given a determined world, the 'choice' you made was determined before you entered the room.

The choice was determined by the choosing. The choosing did not occur until I entered the restaurant, sat at the table, and scanned the menu to see what possibilities were available. Determinism does not mean that any events happen before they happen.

The choice was determined by the interaction of information within the system prior to conscious awareness, narrator function, etc...

As you said yourself;
''Determinism means that events will proceed naturally (as if "fixed as a matter of natural law") and reliably ("without deviation").
''All of these events, including my choices, were causally necessary from any prior point in time. And they all proceeded without deviation from the Big Bang to this moment.'' - Marvin Edwards Post #887

Yes, but the Big Bang happened when it happened, and not a moment before. And my choosing happened when it happened, and not a moment before.

Yes, events unfolding as determined, state by state, in each and every moment in time with 'the way things going thereafter fixed as a matter of natural law.'

The illusion you're experiencing is another result of figurative speaking. Because my choice was inevitable, it is AS IF it had been already made at the time of the Big Bang. But all figurative statements are literally false. Empirically and objectively, the choice was made by me in the restaurant. The Big Bang played no meaningful role in my decision. That's why the waiter brought me the bill, rather than bringing the bill to the Big Bang.

You don't exist in isolation. The world brought you forth. The world acts upon you, and you respond according to the state and condition of the brain that processes the information it acquires and responds to.

But you were not free to choose.

And yet I did choose. So, you are mistaken. In fact, deterministically speaking, it was causally necessary that I, and no one else, would be doing the choosing.

Choice entails the possibility of an alternative. Necessitation ensures that that no alternative is possible.

Repeatedly wrong. Causal necessitation ensured that there would be a menu of alternatives that I would choose from!

There is a menu. The menu has a range of options. Given determinism, the option you take on this occasion is the only possible action. Determinism doesn't allow alternate actions in any instance in time, only what is determined by the state of the system in any given instance in time.

It is not you who choice, but the system, the world, that brings you to that location and precisely that action in relation to that menu.

All of these events, including the menu of alternate possibilities, were causally necessary from any prior point in time. And they all proceeded without deviation from the Big Bang to this moment.

The menu has a list of options, but each diner has only one possible action in any moment in time.

''In neuroscientific circles, ... "Even though there’s no one in charge of its operations, the brain generates a strong intuition of personal agency, borne out by the obvious fact that persons accomplish all sorts of things in all manner of ways.''

No surprises here either. The last sentence though is the key. The brain's sense of self is "borne out by the obvious fact that persons accomplish all sorts of things in all manner of ways." Our sense of self simply comes from empirically observing ourselves choosing to do all sorts of things, like ordering a dinner from the restaurant menu.

We see ourselves doing it, and oddly enough, we conclude that we did it.

But conscious self is not doing it. The distinction being that conscious self in not in control, that it's specifically the brain that is generating the experience of self-awareness and conscious control, an illusion that is exposed whenever things go wrong with the brain.

We as conscious people have no access to the means of our experience.

Tenet: Consciousness is associated with only a subset of nervous function

''Based on developments of the past four decades, there is a growing subset consensus – that consciousness is associated with only a subset of all of the processes and regions of the nervous system''

''Consistent with the subset consensus, many aspects of nervous function are unconscious.Footnote 3 Complex processes of an unconscious nature can be found at all stages of processing (Velmans Reference Velmans1991), including low-level perceptual analysis (e.g., motion detection, color detection, auditory analysis; Zeki & Bartels Reference Zeki and Bartels1999), semantic-conceptual processing (Harley Reference Harley1993; Lucas Reference Lucas2000), and motor programming (discussed in sect. 3.1). Evidence for the complexity of unconscious processing is found in cases in which the entire stimulus-response arc is mediated unconsciously, as in the case of unconsciously mediated actions (e.g., automatisms).

There is a plethora of evidence that action plans can be activated, selected, and even expressed unconsciously.Footnote 4 In summary, it seems that much in the nervous system is achieved unconsciously. This insight from the subset consensus leads one to the following question: What does consciousness contribute to nervous function?''

''Regarding the skeletomotor output system, one must consider that all processes trying to influence skeletomotor behavior must, in a sense, “go through it.” Each system giving rise to inclinations has its peculiar operating principles and phylogenetic origins (Allman Reference Allman2000): One system “protests” an exploratory act while another system reinforces that act (Morsella Reference Morsella2005). Because each skeletomotor effector can usually perform only one act at a time (e.g., one can utter only one word at a time; Lashley Reference Lashley and Jeffress1951; Wundt 1900), there must be a way in which the inclinations from the many heterogeneous systems can be “understood” and processed collectively by the skeletomotor output system. To yield adaptive action, this process must also integrate information about other things (e.g., the physical environment).''

 

[Jarhyn]

A set of numbers is, at most, an abstraction

You are the one who decided to come in here saying that "the universe is describable as a set of numbers and that ser of numbers has no local property describable as "free will", when you say 'the universe is deterministic and this means there is no free will'

The key word is "deterministic", and saying physics (an application of the language of math) is perfectly descriptive of it.

You can either accept the model of "the universe is describable with math", in which case you can say "the universe is deterministic" and then use properties of deterministic systems to make declarations about it, or you cannot utter "the universe is deterministic" sensibly because you don't know what the fuck you are talking about.

You are using an abstraction and the necessary mechanics of that abstraction to describe the universe and most importantly,

Abstractions, if they are descriptive, still describe real properties.

A compiled program is still described by the source code, it is still the shape of it's logic, and still absolutely "it's geometry" even if there is much more that is being abstracted

 

[Jarhyn]

Your wording doesn't contradict anything I have said, nor the definition given in Stanford.

Yes it does. The universe is apparently stochastic.

Whether the universe is stochastic has bearing on this debate, which is about free will in a deterministic universe

No, it really doesn't because deterministic systems may contain locally stochastic systems.

If stochastic systems may have local free will then a deterministic system with a locally stochastic system may have the describable property of free will, if a stochastic system may have free will.

YOU want to discuss this on the level of an academic, you have to approach the topic academically.

 

[Marvin Edwards]

P1: A freely chosen will is when someone chooses for themselves what they will do, while free of coercion and other forms of undue influence.
P2: A world is deterministic if every event is reliably caused by prior events.
P3: A freely chosen will is reliably caused by the person's own goals, reasons, or interests (with their prior causes).
P4: An unfree choice is reliably caused by coercion or undue influence (with their prior causes).
C: Therefore, the notion of a freely chosen will (and its opposite) is still meaningful within a fully deterministic world.

If determinism allows any action to be taken, it's not determinism. The word 'determined' means final. Final as in fixed. Finalized or fixed by prior states of the world.

That would seem to be a logical fact, but it is not a meaningful nor a relevant fact.

To say that the Big Bang finalized all future events suggests that it is was the Big Bang that pointed the gun at the bank clerk and insisted that she hand over the money. Or that it was the Big Bang that landed the Apollo rockets on the Moon. Or that It is was the Big Bang that bombed our naval fleet at Pearl Harbor. Or that the Big Bang chose my breakfast for me.

And yet, the Big Bang played no meaningful or relevant role in any of those events.

Meaning. What's that about?
Relevance. What's that about?

Imagining building a bridge doesn't appear in a vacuum. Conditions in the world in relation to human needs, wants and engineering skills brings forth the idea of building a bridge, and if the need is strong enough, the impetus to put imagination into action.

Countless elements and events went into imagining building a bridge and carrying it out.

Really? I thought you said that the Big Bang already finalized the bridge. If it were finalized, by the Big Bang, then nothing more needed to be done. Right?

The brain of each and every diner acquires information from the menu, which is processed (brain state, proclivities, etc), conscious deliberation generated and related action initiated, you choose Lobster on that occasion. The only possible result on that occasion.
Information processing, not free will.

Why would information processing be meaningful or relevant if the choice were fixed and finalized by the Big Bang? "Final" does mean final, right? Nothing else left to be done, right?

Once you realize that the Big Bang was just the start of things, and not the final end of things, we find all of the other causal mechanisms showing up, including ourselves and our choices, and the meaningful and relevant role that we and our choices play in fixing and finalizing future events.

Sensory information acts upon the system. As described in prior posts, the brain has evolved to acquire information from the environment and respond to it in adaptive ways (if functional)

Exactly. There is the menu, in the environment of the restaurant, and before we can eat we must choose what we will order for dinner.

We respond to objects and events in the external world, which our brain represents in conscious form. We don't choose our genetic makeup or neural architecture, how our brain functions or responds.

Fortunately, it is not necessary to choose our genetic makeup, or our neural architecture, or how our brain functions. It is only necessary that we choose our dinner from a menu of the possible meals.

The illusion of conscious control is revealed when things go wrong with the system.

Whether the choice is made consciously or made unconsciously, it is still our own brains that are doing the choosing. And it is our own choosing that controls our chosen actions.

Yes, events unfolding as determined, state by state, in each and every moment in time with 'the way things going thereafter fixed as a matter of natural law.'

Then I should expect natural law to pay for my dinner. What? That doesn't work?
And how about the bank robbery caused by natural law. Why aren't the police arresting and jailing natural law?

Oh. So we don't hold the past and the laws of nature responsible for their works? Hardly seems fair.

It is not you who choice, but the system, the world, that brings you to that location and precisely that action in relation to that menu.

Then tell the waiter to bill the system or bill the world. If they are responsible for my ordering the lobster dinner then they should pay the bill.

But conscious self is not doing it. The distinction being that conscious self in not in control, that it's specifically the brain that is generating the experience of self-awareness and conscious control, an illusion that is exposed whenever things go wrong with the brain.

Again. It makes no difference whether conscious self is an illusion or not. It is empirically the brain, specifically my brain, that chose to order the lobster dinner. And it must now pay the bill.

How the brain functions, how it manages to make choices from the menu, is irrelevant to the issue of whether or not it actually makes choices. The fact is that the brain actually does make choices, whether or not to order the lobster, and whether or not to rob a bank. In both cases it will be held responsible for its deliberate actions.

Neither the past nor the laws of nature will be held responsible, because they cannot be corrected. They are as they are, and there's nothing we can do about it.

But the guy who ordered the lobster can be required to pay for the lobster. And the guy who robbed the bank can be placed in a correctional facility.

 

[The AntiChris]

Whether the universe is stochastic has bearing on this debate, which is about free will in a deterministic universe

No, it really doesn't because deterministic systems may contain locally stochastic systems.

I find this confusing. Are these "stochastic systems" deterministic?

 

[pood]

I haven’t had time yet to finish the Hoefer article, but it’s interesting, sophisticated and all-encompassing.

In connection with the article, I want to comment on one point Marvin made earlier. He raised the issue of the Minkowski/Einstein block universe (it was first mooted by Minkowski in 1908 and only later accepted by Einstein). He seemed to say, or imply, that a block universe would threaten free will.

As Hoefer notes, determinism might well be bi-directional. That is, while we assume the past determines the future, it might well be equally true that the future determines the past. This comports well the logic of compatibilism, which holds that I freely do what I do because of antecedent causes; and had those causes or past events been different, I might have done differently. But by equal logic we could say that what I do now, determines past events.

The reason is that all physical processes are fundamentally time asymmetric. It is only at the macro level, or more specifically in the context of human observation, that an arrow of time emerges. This is a bit of a mystery, but is thought to have to do with entropy, or it may in part be due to Kant’s claim that time and space are human intuitions.


Temporal asymmetry comports well with the block universe, and the main observational evidence for the block universe comes from special and general relativity, with their relativity of simultaneity. If different observers disagree on what set of events “now” quantifies over, and they do under relativity theory, then presentism, the thesis that only “now” exists and must be the same for everyone, is refuted. This leaves eternalism, the thesis that all moments in time exist in the same way that all locations in space exist. This is the block universe.


In such as universe, the future is as fixed as the past, and like the past, cannot be changed. But for the purposes of free will, this fact does not matter. We know the past is fixed, and that we cannot change it, but from this it obviously does not follow that the past was necessary. It’s a fixed fact of the past that I had eggs for breakfast yesterday, but from this it can’t be inferred that I had to have eggs. In exact fashion, just because it is true today (a fixed fact of the future) that tomorrow I will have eggs for breakfast, it does not follow that tomorrow I must have eggs. The block universe is just a modern iteration of Aristotle’s problem of future contingents, and the answer to that problem, with respect to free will, is modally the same in the block: true propositions about the past, present, and future take their truths from the way that the world was, is, and will be, and not the other way around. Such propositions do not impose their truths upon the world, any more than the “laws” of nature impose their truths upon the world.

If the block universe is true, we can’t change the future any more than can change the past. The worry is that if we can’t change the future — if is “already” so to say, set in stone, then we have no free will. The worry is misguided. No one who thinks we have free will believes that free will means we ought to be able to change the past. Why, then, should free will require that we be able to change the future?

The aforementioned Norman Swartz has put the point cogently: Free will does not require us to be able to change anything. It only requires that we be able, in part, to make the past, present, and future, be, what they were, are, and will be. We can do this just as easily in a block universe and as in a non-block universe. My past selves (temporal parts) freely make certain things to be true, and my present and future temporal parts also freely make certain things to be true. The block world is a tenseless view of time that does not threaten free will in the slightest but comports well with relativity theory and with the temporal fundamental symmetry of all physical processes, including the functioning of brains.

 

[Jarhyn]

Your wording doesn't contradict anything I have said, nor the definition given in Stanford.

Yes it does. The universe is apparently stochastic.

Whether the universe is stochastic has bearing on this debate, which is about free will in a deterministic universe

No, it really doesn't because deterministic systems may contain locally stochastic systems.

If stochastic systems may have local free will then a deterministic system with a locally stochastic system may have the describable property of free will, if a stochastic system may have free will.

So it's very much part of the topic.

This is, ultimately, why I find hard determinists to be exercising on obtuseness, I think.

Admittedly I didn't see it until this thread, and I've been tossing around these ideas (essentially having the same conversation as you lot have been having with each other and me, but with myself) for the last 6 years.

Whether the universe is stochastic has bearing on this debate, which is about free will in a deterministic universe

No, it really doesn't because deterministic systems may contain locally stochastic systems.

I find this confusing. Are these "stochastic systems" deterministic?

Stochastic and deterministic as "descriptors of systems" are a thing of math.

Stochastic systems are ones which contain "probabilistic" elements which make their modeling require statistical analysis rather than mechanical analysis.

A deterministic system which contains a locally stochastic system would, for example, be a system where I supply a mathematical seed, and a number of AI initialized with iterations of that seed for their network weights, and then every iteration select as winners for reproductions and mutations those which most successfully navigate a maze, where their behavior may intersect physically. The losers stop existing in future frames as such, and get rewritten by the winners.

These things will have stochastic elements to their strategy (which is a measure of their degree of freedom) because they cannot derive the RNG, and even if they could, they cannot model the entire contents of another network on the basis of that RNG. I only give them so many neurons! It is globally deterministic, and has locally stochastic elements according to their mathematical definitions.

There are probably less complicated examples of this fact, using more simplified graphs, It's just easiest in the fields of systems theory that I understand fairly solidly.

 

[The AntiChris]

Whether the universe is stochastic has bearing on this debate, which is about free will in a deterministic universe

No, it really doesn't because deterministic systems may contain locally stochastic systems.

I find this confusing. Are these "stochastic systems" deterministic?

Stochastic and deterministic as "descriptors of systems" are a thing of math.

Stochastic systems are ones which contain "probabilistic" elements which make their modeling require statistical analysis rather than mechanical analysis.

A deterministic system which contains a locally stochastic system would, for example, be a system where I supply a mathematical seed, and a number of AI initialized with iterations of that seed for their network weights, and then every iteration select as winners for reproductions and mutations those which most successfully navigate a maze, where their behavior may intersect physically. The losers stop existing in future frames as such, and get rewritten by the winners.

These things will have stochastic elements to their strategy (which is a measure of their degree of freedom) because they cannot derive the RNG, and even if they could, they cannot model the entire contents of another network on the basis of that RNG. I only give them so many neurons! It is globally deterministic, and has locally stochastic elements according to their mathematical definitions.

There are probably less complicated examples of this fact, using more simplified graphs, It's just easiest in the fields of systems theory that I understand fairly solidly.

Thanks for responding but I'm still not clear if the local "stochastic systems" you talk of are essentially deterministic.

This is important because if they're not essentially deterministic then we're not talking about compatibilist free will (I think this may be what DBT is getting at).

 

[Jarhyn]

Thanks for responding but I'm still not clear if the local "stochastic systems" you talk of are essentially deterministic.

This is important because if they're not essentially deterministic then we're not talking about compatibilist free will (I think this may be what DBT is getting at).

No, the locally stochastic systems are stochastic within their locality, which here is confined specifically to the behavioral success property.

I localize (contextualize) my observation, and within that locality (context) of the overall graph behavior and now I have observed a real property of that location (context). It doesn't matter that I can zoom out and see that this is not a property across the whole of time and space, though I expect it is. It has a real property in this location.

But moreover, saying things about whether a system is deterministic or stochastic does not mean that free will is an illusion, it just makes it a contextual property, particularly to the very absolute game theory that we must operate with.

 

[Marvin Edwards]

...
Admittedly I didn't see it until this thread, and I've been tossing around these ideas (essentially having the same conversation as you lot have been having with each other and me, but with myself) for the last 6 years.

It didn't take me that long.

After my father died, I spent time in the public library, browsing the philosophy section. I think I was reading something by Baruch Spinoza that introduced the issue of determinism as a threat to free will. I found this troublesome until I had this thought experiment (whether I read it in one of the books or just came up with it myself, I can’t recall).

The idea that my choices were inevitable bothered me, so I considered how I might escape what seemed like an external control. It struck me that all I needed to do was to wait till I had a decision to make, between A and B, and if I felt myself leaning heavily toward A, I would simply choose B instead. So easy! But then it occurred to me that my desire to thwart inevitability had caused B to become the inevitable choice, so I would have to switch back to A again, but then … it was an infinite loop!

No matter which I chose, inevitability would continue to switch to match my choice! Hmm. So, who was controlling the choice, me or inevitability?

Well, the concern that was driving my thought process was my own. Inevitability was not some entity driving this process for its own reasons. And I imagined that if inevitability were such an entity, it would be sitting there in the library laughing at me, because it made me go through these gyrations without doing anything at all, except for me thinking about it.

My choice may be a deterministic event, but it was an event where I was actually the one doing the choosing. And that is what free will is really about: is it me or is someone or something else making the decision. It was always really me.

And since the solution was so simple, I no longer gave it any thought. Then much later, just a few years ago, I ran into some on-line discussions about it, and I wondered why it was still a problem for everyone else, since I had seen through the paradox more than fifty years ago.

 

[Jarhyn]

...
Admittedly I didn't see it until this thread, and I've been tossing around these ideas (essentially having the same conversation as you lot have been having with each other and me, but with myself) for the last 6 years.

It didn't take me that long.

After my father died, I spent time in the public library, browsing the philosophy section. I think I was reading something by Baruch Spinoza that introduced the issue of determinism as a threat to free will. I found this troublesome until I had this thought experiment (whether I read it in one of the books or just came up with it myself, I can’t recall).

The idea that my choices were inevitable bothered me, so I considered how I might escape what seemed like an external control. It struck me that all I needed to do was to wait till I had a decision to make, between A and B, and if I felt myself leaning heavily toward A, I would simply choose B instead. So easy! But then it occurred to me that my desire to thwart inevitability had caused B to become the inevitable choice, so I would have to switch back to A again, but then … it was an infinite loop!

No matter which I chose, inevitability would continue to switch to match my choice! Hmm. So, who was controlling the choice, me or inevitability?

Well, the concern that was driving my thought process was my own. Inevitability was not some entity driving this process for its own reasons. And I imagined that if inevitability were such an entity, it would be sitting there in the library laughing at me, because it made me go through these gyrations without doing anything at all, except for me thinking about it.

My choice may be a deterministic event, but it was an event where I was actually the one doing the choosing. And that is what free will is really about: is it me or is someone or something else making the decision. It was always really me.

And since the solution was so simple, I no longer gave it any thought. Then much later, just a few years ago, I ran into some on-line discussions about it, and I wondered why it was still a problem for everyone else, since I had seen through the paradox more than fifty years ago.

It comes down to me to the idea that people wish to complain that because free will only exists as a relational property rather than a global property, it doesn't exist at all.

No.

It exists exactly as a real relational property, particular to the context of decision of event on model.

 

[The AntiChris]

Thanks for responding but I'm still not clear if the local "stochastic systems" you talk of are essentially deterministic.

This is important because if they're not essentially deterministic then we're not talking about compatibilist free will (I think this may be what DBT is getting at).

No, the locally stochastic systems are stochastic within their locality, which here is confined specifically to the behavioral success property.

Apologies for being dense but are you saying that locally stochastic systems are not deterministic?

What I'm getting at is that if, as you seem to imply in your reply to DBT (post #916), free will is reliant on local stochastic (non-deterministic?) systems then the free will you're talking about is not compatibilist free will.

 

[Jarhyn]

Thanks for responding but I'm still not clear if the local "stochastic systems" you talk of are essentially deterministic.

This is important because if they're not essentially deterministic then we're not talking about compatibilist free will (I think this may be what DBT is getting at).

No, the locally stochastic systems are stochastic within their locality, which here is confined specifically to the behavioral success property.

Apologies for being dense but are you saying that locally stochastic systems are not deterministic?

What I'm getting at is that if, as you seem to imply in your reply to DBT (post #916), free will is reliant on local stochastic (non-deterministic?) systems then the free will you're talking about is not compatibilist free will.

So, the wording I use around this is at least intended and supposed to be something very precise and sharp.

It may take a few readings of it to fully understand all the exact shapes in it, and perhaps I misshaped them and they are accidentally nonsense instead of what I wished to communicate!

Imagine a system, as I described before:

A system where I supply a mathematical seed, and a number of AI initialized with iterations of that seed for their network weights, and then every iteration select as winners for reproductions and mutations those which most successfully navigate a maze, where their behavior may intersect physically. The losers stop existing in future frames as such, and get rewritten by the winners.

This system contains no randomness in the process of determining next state. It satisfies the core property of "deterministic", because the RNG is a pseudorandom process, a thing of chaos.

I can run it a million times, and every frame will be rendered exactly the same, assuming sufficient shielding against interference of external junk on the field of it's memory.

But as demonstrated, this deterministic system contains locally stochastic elements: the individuals in it cannot organize to model the RNG or it's implications on their behavior despite the fact that we can provably see their universe is deterministic.

 

[The AntiChris]

What I'm getting at is that if, as you seem to imply in your reply to DBT (post #916), free will is reliant on local stochastic (non-deterministic?) systems then the free will you're talking about is not compatibilist free will.

So, the wording I use around this is at least intended and supposed to be something very precise and sharp.

I'm sure it was but I'm afraid it meant little to me!

Imagine a system, as I described before:

A system where I supply a mathematical seed, and a number of AI initialized with iterations of that seed for their network weights, and then every iteration select as winners for reproductions and mutations those which most successfully navigate a maze, where their behavior may intersect physically. The losers stop existing in future frames as such, and get rewritten by the winners.

This system contains no randomness in the process of determining next state.

This looks deterministic to me. I'll leave it there.

Thanks for your efforts.

 

[Jarhyn]

What I'm getting at is that if, as you seem to imply in your reply to DBT (post #916), free will is reliant on local stochastic (non-deterministic?) systems then the free will you're talking about is not compatibilist free will.

So, the wording I use around this is at least intended and supposed to be something very precise and sharp.

I'm sure it was but I'm afraid it meant little to me!

Imagine a system, as I described before:

A system where I supply a mathematical seed, and a number of AI initialized with iterations of that seed for their network weights, and then every iteration select as winners for reproductions and mutations those which most successfully navigate a maze, where their behavior may intersect physically. The losers stop existing in future frames as such, and get rewritten by the winners.

This system contains no randomness in the process of determining next state.

This looks deterministic to me. I'll leave it there.

Thanks for your efforts.

"It satisfies the core property of "deterministic", because the RNG is a pseudorandom process, a thing of chaos."

Of course it's deterministic. It's defined and developed as such, specifically to satisfy that definition.

The issue is that it contains locally stochastic elements.

This is the thing I say means something very specific: it means that deterministic systems are capable of having and representing, in their graph, all the properties of stochastic systems.

if I say "B has a meaningful property of P, and things with property A may contain objects which meaningfully have property B with all meaningful properties of B, then A may have a meaningful property P."

There is unknowable and unreconcilable chaos even that in the fully deterministic system, and because of the inability to so model recursively, it becomes stochastic from that context, the local one to the ones in the race.

Some things they will model, namely the laws of motion in their environment. Depending on how much I give them, they may even model the laws of their universe entirely, down to memory quanta!

They might debate whether the RNG is random or pseudorandom, on the ethics of finishing the race knowing half of them will die, and even if they discover that it is all deterministic they will still know that they are blind to what it would take for them to determine anything with that knowledge in anything but a chaotic efficient competition. It just makes them more capable of another layer of examining whose will is the freest, whose model draws them.

But they won't be able to escape their own inability to model the whole of their time and space with them in it, too. They are incapable of being a perfect systemic model and are themselves "locally stochastic".

 

[Marvin Edwards]

What I'm getting at is that if, as you seem to imply in your reply to DBT (post #916), free will is reliant on local stochastic (non-deterministic?) systems then the free will you're talking about is not compatibilist free will.

So, the wording I use around this is at least intended and supposed to be something very precise and sharp.

I'm sure it was but I'm afraid it meant little to me!

Imagine a system, as I described before:

A system where I supply a mathematical seed, and a number of AI initialized with iterations of that seed for their network weights, and then every iteration select as winners for reproductions and mutations those which most successfully navigate a maze, where their behavior may intersect physically. The losers stop existing in future frames as such, and get rewritten by the winners.

This system contains no randomness in the process of determining next state.

This looks deterministic to me. I'll leave it there.

Thanks for your efforts.

"It satisfies the core property of "deterministic", because the RNG is a pseudorandom process, a thing of chaos."

Of course it's deterministic. It's defined and developed as such, specifically to satisfy that definition.

The issue is that it contains locally stochastic elements.

This is the thing I say means something very specific: it means that deterministic systems are capable of having and representing, in their graph, all the properties of stochastic systems.

if I say "B has a meaningful property of P, and things with property A may contain objects which meaningfully have property B with all meaningful properties of B, then A may have a meaningful property P."

There is unknowable and unreconcilable chaos even that in the fully deterministic system, and because of the inability to so model recursively, it becomes stochastic from that context, the local one to the ones in the race.

Some things they will model, namely the laws of motion in their environment. Depending on how much I give them, they may even model the laws of their universe entirely, down to memory quanta!

They might debate whether the RNG is random or pseudorandom, on the ethics of finishing the race knowing half of them will die, and even if they discover that it is all deterministic they will still know that they are blind to what it would take for them to determine anything with that knowledge in anything but a chaotic efficient competition. It just makes them more capable of another layer of examining whose will is the freest, whose model draws them.

But they won't be able to escape their own inability to model the whole of their time and space with them in it, too. They are incapable of being a perfect systemic model and are themselves "locally stochastic".

From my perspective, all events are causally deterministic, but may be practically impossible to predict. That's why we have concepts like random and chaotic, to identify unpredictability except through probabilities. The causation is reliable, but the prediction is not.