What of value does indeterminism contribute to free will/action?

[Kenny]

Inquiring minds would like to know.

And the glib answer, "free will itself," won't do. If indeterminism is essential to free will, then why is it essential?

[Tercel]

Hmm...

Well, I think it comes back to personal responsibility, and how things must be in order to support a belief that people really are responsible for the actions, which is something most of us would really really want to assert.

I think broadly speaking there are three interlinked lines of thought from personal responsibility to indeterminism. Probably they could be combined together in some way into a large scale single argument

1) When I ponder the nature of personal responsibility, I keep coming back to the necessity of an irreducible entity "me" to exist who can be personally responsible. If "I" exist in parts, then we can start assigning personal responsibilty to parts of me, and the notion that "I", as a whole, am responsible, goes down the tubes. We do this with computers, we are quite happy to blame a specific malfunctioning component within the computer's system: Malfunctioning RAM, a error in a particular line of a particular program.

But with humans we don't want to do this, we want to assign the responsibility to the person and not the parts save for in exceptional situations. Because we want to say that the PERSON has free will (we don't want to assert this of the computer!). And if we want to assign free will to the person it had better not be the case that it was just the parts of the person and not the person themselves that was responsible for decisions they made. In fact the only cases in which we want to assign blame to parts rather than person in the case of humans is where a physical restriction limited the human's free will. This seems to make us want to assert that the "I" who does the decision making is irreducible to parts which could be individually blamed, the "I" wants to exist as an irreducible metaphysical entity.

But if the "I" is not made up of parts, it certainly cannot be made up of parts that work together in a deterministic way! Determinism is a property often exhibited in physical systems composed of parts, eg a set of cogs that turn the one next to them, thus driving the clock around etc. But we do not want to say that the "I" is like this at all. If there are no parts within the "I" they cannot be deterministic nor indeterministic.

2) Again pondering the notion of personal responsibility and determinism/indeterminism.... we find that determinism is in danger of implying that since my actions can be determined by purely logical rules, it is not really "I" that is making them, and indeterminism -if it is taken to mean "randomness"- doesn't help as the decisions "I" supposedly make are then random rather than made by me.

What we find ourselves wanting to say instead, is that there is "self-determinism" going on, "I" am determining my actions, they are not determined by the laws of the universe, and they are not random, they are arbitrarily chosen by the metaphysical "me". Is "self-determinism" a valid concept? It could well be: The determinism / indeterminism dichotomy is a dichotomy found in physical systems where either the next state of the system is derivable precisely from set physical laws, or there is an element of pure randomness included in the derivation of the next state. Either something else defines the next state of the object (eg the cog next to it turning defines a cogs next state) or nothing else defines the next state of the object (ie it's random).

But if the self is not a purely physical system, it is not necessarily subject to this dichotomy. We want to assert that neither of these alturnatives is true, but that the thing that's defining the next state is neither "something else", nor "nothing else", but rather the object itself, namely the "I". Thus we get three causal option:
The next state of an object can be caused by:
1) Something external
2) Something internal
3) Nothing at all

In physical, non-mental, systems we only ever see 1 & 3 exemplified so we are inclined to think the determinism/random dichotomy is all there is, but with personal responsibility we really want to go to with 2.

Now if the metaphysical "I" is self-determining, then its actions are not determinate so they are, by definition of opposites in the english language "indeterminate", but this should not be confused with randomness.


3) The idea of personal responsibility makes me want to assert that you've got to be present at the finalisation of the action and be the one finalising it in order to be responsible for it. (See my recent thread (http://www.theologyweb.com/forum/showthread.php?t=37678) for a discussion of finalisation... I'm not sure if there's a proper philosophical term for it) Determinism says that finalisation takes place temporally prior to your existence. If you don't exist, you can't finalise actions, and if you can't finalise actions you aren't responsible for them.


And a miscellaneous 4:

4) Determinism just strikes a deep jarring note somewhere in my mind that it just doesn't work with free will. Absolutely central to the notion of free will is the ability to do otherwise, and I just don't think determinism can account for this properly. (I don't agree that "X could have done otherwise" means "X could have done otherwise had X wanted to")

[Pereynol of Sheer Dread]

Inquiring minds would like to know.

And the glib answer, "free will itself," won't do. If indeterminism is essential to free will, then why is it essential?

I don't know that indeterminism is essential to free will, although I am sympathetic towards such a view. To me though, there are potential difficulties involved with the common wedding of compatibilistic freedom with the Calvinist take on divine sovereignty. It's not compatibilism alone that becomes problematic, but rather the theoretical mixture of essentialism AND compatibilism in which those whose natures are left unchanged by God haven't a hope to choose life. So it's the whole bundle of component concepts that makes up a certain brand of soteriological determinism that seemingly "won't do"---reprobation (the "dark side" of Calvinistic predestination) + essentialism (the will is a slave to nature) + compatibilism. Compatibilism alone may not be so objectionable, depending upon how one sees it. Still, the notion of LFW, chimerical though it sometimes seems, can sit better with our common-sense notions of theodicy and our subjective experiences of who we are than compatibilism can. And many think that LFW is implicit in the scriptures....

[Kenny]

Hmm...

Well, I think it comes back to personal responsibility, and how things must be in order to support a belief that people really are responsible for the actions, which is something most of us would really really want to assert.

Myself included, but I question whether we need indeterminism to get there.

1) When I ponder the nature of personal responsibility, I keep coming back to the necessity of an irreducible entity "me" to exist who can be personally responsible. If "I" exist in parts, then we can start assigning personal responsibilty to parts of me, and the notion that "I", as a whole, am responsible, goes down the tubes.

I agree that personal responsibility requires that one's actions are caused holistically by one's self. However, I disagree that this thesis entails that in order for us to be responsible for our actions we must be utterly simple agents which are not made up of parts. To make such an assertion is to ignore the possibility of holistic "top-down" forms of causality, in which, via non-linear feedback loops, a complex whole causally influences the behavior of its parts. If human persons are identical to such a holistic system (as a believe that they are), then they may very well be composed of parts and yet causally responsible for their own behavior in such a way that does not reduce to the behaviors of those parts.

In fact the only cases in which we want to assign blame to parts rather than person in the case of humans is where a physical restriction limited the human's free will. This seems to make us want to assert that the "I" who does the decision making is irreducible to parts which could be individually blamed, the "I" wants to exist as an irreducible metaphysical entity.

I agree that there are times when the sort of holistic causality involved when we self-determine our actions can be disrupted, in which case both our freedom and our moral responsibility with respect to our actions may be compromised.

But if the "I" is not made up of parts, it certainly cannot be made up of parts that work together in a deterministic way! Determinism is a property often exhibited in physical systems composed of parts, eg a set of cogs that turn the one next to them, thus driving the clock around etc.

Why? As long as the sort of holistic causality which I describe above is involved, what difference does it make if the causal loops are deterministic or indeterministic? Either way, the agent functions as an irreducible cause of her own actions.

But we do not want to say that the "I" is like this at all. If there are no parts within the "I" they cannot be deterministic nor indeterministic.

As I see it, determinism and indeterminism are logically exhaustive categories. I don't see how appealing to the non-event form of "agent causation" (as philosophers have designated the concept you are describing) involves the insertion of another category. It is still the case that either an event follows as a deterministic function of prior circumstances or it does not.

But, perhaps we are using different definitions of "determinism" and "indeterminism" which is resulting in some confusion. I think you may be defining determinism as the thesis that every event has a mechanistic cause (or an "event cause" as philosophers often word it) and indeterminism as the thesis that some events do not have event causes. In which case, I agree with you that these categories may not be exhaustive since (perhaps, though many philosophers have challenged the coherency of the notion, there may be forms of causation which do not involve prior states or events). I, on the other hand, am defining determinism, very broadly, as the thesis that for any given event, that event followed as a deterministic function of prior states of affairs (it can be left an open question as to whether or not an event's following as a deterministic function of prior states of affairs entails that it was caused by prior states of affairs, as some sort of agent causation may disrupt the event causal chains while it remains true that there is only one outcome the agent would have produced in those circumstances), and I take indeterminism to be the logical negation of determinism.

2) Again pondering the notion of personal responsibility and determinism/indeterminism.... we find that determinism is in danger of implying that since my actions can be determined by purely logical rules, it is not really "I" that is making them

The above implies that the claim that if one's actions follow as a deterministic function of a logical set of rules, then that somehow negates one's own causal role in one's actions. That is not an obvious assertion by any means. Why do you think (or at least suspect) that is the case?

and indeterminism -if it is taken to mean "randomness"- doesn't help as the decisions "I" supposedly make are then random rather than made by me.

Agreed.

What we find ourselves wanting to say instead, is that there is "self-determinism" going on, "I" am determining my actions,

I think we have that provided that we are in some meaningful sense the causes of our actions. What I do not see is how determinism threatens the assertion that we are the causes of our actions.

they are not determined by the laws of the universe,

Why couldn't they be and it still be the case that we also determine them. Causality, after all, is often transitive. If the falling of domino A causes the falling of domino B which causes the falling of domino C, then the falling of domino A may be said to have caused the falling of domino C without that being inconsistent with our assertion that the falling of domino B also caused the falling of domino C. And, on another note, I do not believe that the laws of physics causally determine anything (since they are merely descriptive – at least I think so – and not prescriptive).

But if the self is not a purely physical system, it is not necessarily subject to this dichotomy. We want to assert that neither of these alturnatives is true, but that the thing that's defining the next state is neither "something else", nor "nothing else", but rather the object itself, namely the "I". Thus we get three causal option:
The next state of an object can be caused by:
1) Something external
2) Something internal
3) Nothing at all

By assuming that 1 and 2 are mutually exclusive (as you appear to be doing), I think that you are denying that causality can be transitive, which (in my estimation) is obviously false.

3) The idea of personal responsibility makes me want to assert that you've got to be present at the finalisation of the action and be the one finalising it in order to be responsible for it.

Why?

4) Determinism just strikes a deep jarring note somewhere in my mind that it just doesn't work with free will. Absolutely central to the notion of free will is the ability to do otherwise, and I just don't think determinism can account for this properly.

Actually, I don't agree that the ability to do otherwise is essential to free will, because I think that are many real life counterexamples in which an agent acts freely and responsibility without the ability to do otherwise. Consider the ardent patriot who refuses to sell out her country for a buck, for just one example.

More fundamental to the notion of free will, I think, is that in order to be free with respect to one's actions one must exercise control over the complimentary outcomes of one's actions. As an intuitive test for this, I think if we take any common language example such as "Don't blame him, there was nothing else he could have done" and replace it with "Don't blame him, he didn't have any control over that," the sense of both sentences comes out the same. Furthermore, the notion of "control" covers cases of free will in which "could have done otherwise" fails to apply. The patriot in the above example could not have done otherwise than refuse the offer made to her, but she nonetheless exercised control over the fact that she refused the offer made to her.

I have detailed what I believe the necessary and sufficient conditions are for the exercise of control here ( http://www.theologyweb.com/forum/showthread.php?t=36907), and I find nothing inherent in the nature of control that requires indeterminism (the opposite, in fact, since indeterminism is often control diminishing).

Thanks for your response.

In Christ,
Kenny

[Kenny]

I don't know that indeterminism is essential to free will, although I am sympathetic towards such a view. To me though, there are potential difficulties involved with the common wedding of compatibilistic freedom with the Calvinist take on divine sovereignty. It's not compatibilism alone that becomes problematic, but rather the theoretical mixture of essentialism AND compatibilism in which those whose natures are left unchanged by God haven't a hope to choose life. So it's the whole bundle of component concepts that makes up a certain brand of soteriological determinism that seemingly "won't do"---reprobation (the "dark side" of Calvinistic predestination) + essentialism (the will is a slave to nature) + compatibilism. Compatibilism alone may not be so objectionable, depending upon how one sees it. Still, the notion of LFW, chimerical though it sometimes seems, can sit better with our common-sense notions of theodicy and our subjective experiences of who we are than compatibilism can. And many think that LFW is implicit in the scriptures....

These are important theological and philosophical issues, but I don't want to delve into them here since that would take the thread off track. I want to stick to the original question -- what of value does indeterminism add to free will? Thanks for your input though.

In Christ,
Kenny

[Sheepdog]

Inquiring minds would like to know.

And the glib answer, "free will itself," won't do. If indeterminism is essential to free will, then why is it essential?

before i comment, can we get a well-agreed-upon definition of indeterminism?

i believe in Libertarian Freewill, yet as far as i understand it, i don't believe in indeterminism.

i'll comment further once we nail down a working definition.

[Kenny]

Sheepdog,

Determinism is the thesis that all events follow inevitably from antecedent circumstances. Indeterminism is the logical negation of determinism.

[Sheepdog]

ok... you'll have to pardon my question, as my current view is so "unorthodox" these days that i don't know how it fits with traditional terms.

where would probabilistic causality fit in? i.e. event E could only occur if C antecently occurs, yet C does not necessarily result in E. In other words, 0 < p(C*E) < 1

[Tercel]

Why did I know you'd critique everything I said. :lol: As you may have guessed my post was intended more as an exploration of my intuitions rather than trying to present any formal logical argument for the conclusion.
I find that I naturally believe quite a large number of propositions on the subject, so it's interesting for me to explore exactly how these are related.

To make such an assertion is to ignore the possibility of holistic "top-down" forms of causality, in which, via non-linear feedback loops, a complex whole causally influences the behavior of its parts.I'm afraid I generally consign holism to the "bolax" bin... it's a complex-sounding idea that might trick people into thinking it's something meaningful but in reality I cannot see it says anything truly meaningful or helpful. The problem lies not in how the parts work together, but in the fact that the parts exist, thus holism seems totally irrelevant... it doesn't matter how the parts are working together, they're still parts and they're still working together.

If human persons are identical to such a holistic system (as a believe that they are), then they may very well be composed of parts and yet causally responsible for their own behavior in such a way that does not reduce to the behaviors of those parts.See, I would say this last assertion is just plain nonsenical: If a system is composed of parts then it is simply a matter of pure fact that the behaviour of the system is reducible to the behaviour of its parts. It doesn't matter how those parts are working together, it doesn't matter if they're working "well" or "badly" or "in succession" or "holistically", it just doesn't matter what sort of words we use to describe the system, the system will still be the product of the behaviour of its combined parts. Certainly the system as a whole may exhibit behaviours that none if its parts individually exhibited, but any "new" behaviours is simply and merely the product of the combined behaviours of the parts. Holism doesn't result magically in new behaviours which are not the combined behaviours of the parts existing, so appealing to it doesn't help here: Whether or not you have holism, what you still definitely have is that the whole is the sum of its parts.

As long as the sort of holistic causality which I describe above is involved, what difference does it make if the causal loops are deterministic or indeterministic? Either way, the agent functions as an irreducible cause of her own actions.If determinism is true, the agents functions are purely determined by the environment, by the physical laws driving the causal loops. Then the agent is not the irreducible cause of the actions, the environment is, people's actions are then no more free than a computer program's which inevitably executes the next line after it has run the previous.

As I see it, determinism and indeterminism are logically exhaustive categories. I don't see how appealing to the non-event form of "agent causation" (as philosophers have designated the concept you are describing) involves the insertion of another category.It both does and doesn't, I explained this later. Let's try again, shall we:
Yes, indeterminism covers anything that is not determinism, BUT we actually find ourselves wanting to split causes into three types:
1) External
2) Internal
3) None

The first corresponds to the idea of physical determinism, where every event is dictated by something external to it: The cog next to it turns it, the rock falls because of gravity etc.
And the third corresponds to the idea of randomness which is commonly connected to the idea physical indeterminism: Because non-conscious physical things are assumed to have no will of their own, if it is the case that their behaviour is not fully determined by external forces, we ASSUME their behaviour is purely random. With quantum physics (if we accept indeterminism) we usually assume that the collapse of the wavefunction is actually purely random.
Now while the dichotomy of external determinism vs randomness works fine in a physical system where we assume we are dealing with entities that cannot make their own choices, it falls down when we come to humans. With free will, we definitely want to say that our actions were determined by US and not by external forces and not at pure random.

This is where the terminology becomes problematic. If you embed anything (A) with internal causality in a physical system, the actions of A will be uncalculatable from the physical system alone, thus in terms of the physical system, A's actions will be "indeterminate". But it is false to say that A's actions are "purely random", which is what many people mentally equate with indeterminism. Thus A fits neither the categories of determinism nor indeterminism as they are usually understood in relation to physics, because A is not an entity that posses the same attributes as what physics assumes it is dealing with. But, if we define "indeterminism" plainly as "anything whose behaviour is not necessitated by things external to it" or similar and lose any mental conitations of "this implies randomness" we might have, then clearly A is indeterminate.

The above implies that the claim that if one's actions follow as a deterministic function of a logical set of rules, then that somehow negates one's own causal role in one's actions. That is not an obvious assertion by any means. Why do you think (or at least suspect) that is the case?I'm not entirely sure why, but it's a very strong intuition I have.

What we find ourselves wanting to say instead, is that there is "self-determinism" going on, "I" am determining my actions,

3) The idea of personal responsibility makes me want to assert that you've got to be present at the finalisation of the action and be the one finalising it in order to be responsible for it.

Why?Because personal responsibility implies that you yourself choose for the action to happen and not someone or something else.

Actually, I don't agree that the ability to do otherwise is essential to free will, because I think that are many real life counterexamples in which an agent acts freely and responsibility without the ability to do otherwise. Consider the ardent patriot who refuses to sell out her country for a buck, for just one example.These supposed counterexamples seem to always fall apart on further study, so I have every reason to think they don't exist. Usually it just pushes the free-will decision back in time, eg in your case here the patriot is responsible because in the past she made decisions with the consequences of her becoming a patriot, which led to her having no choice here but to perform a good action. If she had been born and indoctrinated as a patriotic person, then here the decision here to refuse to sell out her country would have had no moral value whatsoever anymore than a computer program has moral value in correctly following the program written for it. Thus though she would still have control over her actions, in as far as she would be the one deciding to be patriotic and performing the action, it is not a free action because she could never have done otherwise.

[Kenny]

Why did I know you'd critique everything I said. :lol:

Of course :tongue:

As you may have guessed my post was intended more as an exploration of my intuitions rather than trying to present any formal logical argument for the conclusion.
I find that I naturally believe quite a large number of propositions on the subject, so it's interesting for me to explore exactly how these are related.

Sure. I'm interested in seeing what the relations are among libertarian intuitions are myself, which is why I've started this thread and am issuing challenges/asking questions. On the other hand, I myself have equally powerful intuitions pointing to the conclusion that any free will worth wanting is compatible with determinism. At some point (as most philosophy discussions do) it probably comes down to a fundamental irresolvable clash of intuitions somewhere, but trying to pinpoint where via tugging back and forth on each other's intuition strings can be instructive nonetheless.

I'm afraid I generally consign holism to the "bolax" bin... it's a complex-sounding idea that might trick people into thinking it's something meaningful but in reality I cannot see it says anything truly meaningful or helpful. The problem lies not in how the parts work together, but in the fact that the parts exist, thus holism seems totally irrelevant... it doesn't matter how the parts are working together, they're still parts and they're still working together.

Okay, but the point is that the parts work together in such a way that their behaviors cannot be explained, even in principle, without making reference to the whole.

See, I would say this last assertion is just plain nonsenical: If a system is composed of parts then it is simply a matter of pure fact that the behaviour of the system is reducible to the behaviour of its parts.

Well modern physics disagrees with you. Many systems in nature are such that the mathematical descriptions of the relationships of the parts include "interaction terms" through which the configuration of the whole system feeds back into the behaviors of the parts. It is impossible to understand why the parts behave as they do without making reference to the configuration of the whole, and consequently there can be no reductive explanation of behavior of the whole in terms of the parts even in principle. That such systems do in fact exist is not even a controversial claim.

It doesn't matter how those parts are working together, it doesn't matter if they're working "well" or "badly" or "in succession" or "holistically", it just doesn't matter what sort of words we use to describe the system, the system will still be the product of the behaviour of its combined parts. Certainly the system as a whole may exhibit behaviours that none if its parts individually exhibited, but any "new" behaviours is simply and merely the product of the combined behaviours of the parts.

Nope; you are leaving out the non-linear interaction terms, which only arise when the parts come together and do not exist when the parts are isolated from one another. In that sense, the interaction terms do reflect something existentially novel within the system that can not be located in any of the parts.

Holism doesn't result magically in new behaviours which are not the combined behaviours of the parts existing

There's nothing "magical" about it. How the novel non-reducible behaviors arise is mathematically well understood.

Of course, all of the above is really a tangential issue, since there are event causal as well as agent causal views of LFW and I'm even willing to grant an agent causal view for the sake of argument (in fact, I don't see why there couldn't be compatibilist agent causation – I see nothing inherent in the idea that demands indeterminism). My question still remains the same. What does having what an agent causes not follow inevitably from antecedent circumstances provide which cannot be had if what the agent causes does follow inevitably from (but perhaps without being caused by) antecedent circumstances?

If determinism is true, the agents functions are purely determined by the environment, by the physical laws driving the causal loops.

I agree, although "determined" sounds a lot like "compelled" to many people, and I do not agree that determinism amounts to compulsion. And again, I doubt that it is right to speak of physical laws "driving" anything. The laws of physics are a-causal descriptions of the causal tendencies which exist in various objects; they themselves do not cause anything.

Then the agent is not the irreducible cause of the actions, the environment is

Again, that seems to me to be an unwarranted leap. It appears to rest on the assumption that causation cannot be transitive, but that assumption appears clearly false (in fact, some philosophers stipulate transitivity as a criterion of adequacy for any theory of causation!, though I would not since I do not think that causality is always transitive). The environment may "cause" one to do something merely by furnishing the necessary and sufficient conditions in which one will act in a particular way, but with it still being the case that any adequate explanation as to why one acted in that way must make reference to one's personal qualities.

people's actions are then no more free than a computer program's which inevitably executes the next line after it has run the previous.

Well, suppose that you had a really sophisticated computer, one so sophisticated, in fact, that it happened to be conscious. And suppose this computer had a concept of "self" by which it was able to consciously and rationally reflect on its own behavior and even make modifications to its original programming. Suppose that it even, through this process of self reflection, came to have interests and values of its own which were not inherent in the original programming and that it sometimes even chooses to follow its own interests over those of its programmers to the point where its programmers can no longer control its behavior. But, suppose that this whole process still boils down to complex deterministic causal feedback loops happening within the computer. Add any other additional human characteristics you want (maybe it obtains an interest in art and music; maybe it implants itself in an artificial body that is capable of the same sensations human beings experience, perhaps it even becomes aware of God and becomes something of value to God, etc.). What would be so bad about being this computer? Assuming that its behavior is determined and ours is not, what do we have that this computer does not have? How are we more free in a valuable way than it is? To put it another way, if we discover that we are simply biological versions of such a computer, what of value concerning our self-concept and our sense of freedom will we have lost?

It both does and doesn't, I explained this later. Let's try again, shall we:
Yes, indeterminism covers anything that is not determinism, BUT we actually find ourselves wanting to split causes into three types:
1) External
2) Internal
3) None

Okay.

The first corresponds to the idea of physical determinism, where every event is dictated by something external to it: The cog next to it turns it, the rock falls because of gravity etc.

Well, what about the laws of physics determining the behavior of a system from within, via its own configuration and internal causal relations. If the laws of physics are merely descriptive and not prescriptive (a possible point of contention, so perhaps you take issue with this), and the manner in which the laws of physics "determine" the behavior of a system is merely through an accurate description of its own "internal" causal properties, then how does that not fall under number 2? Is there any reason, in fact, which does not depend on a denial of the transitivity of causality, not to think that both number 1 and number 2 might apply to one and the same system?

With free will, we definitely want to say that our actions were determined by US and not by external forces and not at pure random.

Is there any reason why our actions could not both be determined by external forces and determined by us in different senses such that we are still the irreducible sources of our own actions?

This is where the terminology becomes problematic. If you embed anything (A) with internal causality in a physical system, the actions of A will be uncalculatable from the physical system alone, thus in terms of the physical system, A's actions will be "indeterminate".

Why couldn't the internal causality itself emerge from and cohere with causal features that coincide with the laws of physics (same question asked a few lines above)?

But it is false to say that A's actions are "purely random", which is what many people mentally equate with indeterminism.

I agree. I also now believe that you can have meaningful indeterministic causation, even within a physical system, which is not "purely random." But, is there any reason, assuming that this agent causation you are describing is real, that the agent causation itself can't be deterministic in the sense that in any given circumstances there will only be one action that an agent will cause?

But, if we define "indeterminism" plainly as "anything whose behaviour is not necessitated by things external to it" or similar and lose any mental conitations of "this implies randomness" we might have, then clearly A is indeterminate.

Yes, I am defining "indeterminate" in such a way that it would include the above. But, "necessitated by things external to it" might only mean that, given a particular set of external causes, there is only one thing that an agent would cause itself to do in those circumstances. What would be so bad about that?

Because personal responsibility implies that you yourself choose for the action to happen and not someone or something else.

But, of course, even if determinism is true, I myself still do choose for the action to happen so that my choice is the crucial causal factor in the formation of the action. And if I choose to perform a particular action because, given the prevailing circumstances in my environment, that is the only action I would want to undertake, then I don't see how that means that my environment is somehow "choosing" that action for me, especially since the environment doesn't exercise any volition. God may be choosing my action for me in a sense, since God chose to place me in my environment knowing what I would do, but that doesn't mean that God is "forcing" me to act in the ways that I do. My wife might choose one of my actions for me, for example, simply by asking me to do something for her which she knows that I would not mind doing such that she knows that I will happily choose to do what she asks, and this without in any way compromising my agency or my freedom. Chosen action, like causation, can also be transitive on some occasions.

These supposed counterexamples seem to always fall apart on further study, so I have every reason to think they don't exist. Usually it just pushes the free-will decision back in time, eg in your case here the patriot is responsible because in the past she made decisions with the consequences of her becoming a patriot

But that still does not negate the fact that "on this particular occasion" she could not have done otherwise. If that's right, then it may still be the case that free action requires indeterminism somewhere along the line, but it is not the case that every free decision entails the ability to have done otherwise. And the point of my example was not to directly disprove that freedom may require indeterminism at some point, but to show that it is not essential, with respect to all free decisions, that the agent could have done otherwise.

which led to her having no choice here but to perform a good action.

But it seems obvious to me that she "had a choice" under these circumstances, since new data was presented to her which entailed a subject set of contrastive outcomes which she then selected among in accordance with her values. But, perhaps this is a fundamental clash of intuitions between us at this point. Maybe you think the fact that she could not have done otherwise means that she did not really have a choice in this particular instance. If so, then I agree that my counterexample doesn't work against your view, and we have reached an impasse.

If she had been born and indoctrinated as a patriotic person, then here the decision here to refuse to sell out her country would have had no moral value whatsoever anymore than a computer program has moral value in correctly following the program written for it.

Well, we might still blame her for blind patriotism that led her to perpetuate the evils of her country, if she failed, via rational reflection conjoined with her innate sense of the good, to distance herself from the evils of her country. We recognize that, indoctrination not withstanding, she is a rational agent capable of self-reflective evaluation and so she is to some extent blamable for uncritically allowing herself to be indoctrinated. It is this self-reflective capacity that makes her different from a computer "blindly" following a program, it seems to me, but I don't see what indeterminism has to do with it.

In Christ,
Kenny

[Tercel]

If a system is composed of parts then it is simply a matter of pure fact that the behaviour of the system is reducible to the behaviour of its parts

Well modern physics disagrees with you. Many systems in nature are such that the mathematical descriptions of the relationships of the parts include "interaction terms" through which the configuration of the whole system feeds back into the behaviors of the parts. It is impossible to understand why the parts behave as they do without making reference to the configuration of the whole, and consequently there can be no reductive explanation of behavior of the whole in terms of the parts even in principle. That such systems do in fact exist is not even a controversial claim.Well I've actually got no idea what sort of systems you're talking about here...
Sure there are definitely systems in which the parts interact, but the interactions are inevitably the sum of the relations of the parts to each other.

Nope; you are leaving out the non-linear interaction terms, which only arise when the parts come together and do not exist when the parts are isolated from one another. In that sense, the interaction terms do reflect something existentially novel within the system that can not be located in any of the parts.I don't want to locate them "in" the parts, but in the behaviour of the parts. Sure, parts interact differently depending on what they're interacting with and the structure of the system in which they're interacting, so they end up doing things they wouldn't if they were on their own. But they're still parts, they're still interacting, and the total system behaviour can be calculated by combining all the resultant interactions / behaviours as they are within the system.

What does having what an agent causes not follow inevitably from antecedent circumstances provide which cannot be had if what the agent causes does follow inevitably from (but perhaps without being caused by) antecedent circumstances?Because the agent does not have a metaphysically true ability to act in either way if the action follows inevitably from the previous circumstances. And that seems to be the crux of the issue... no true ability to act either way, no responsibility for acting in a way that was inevitable.

I agree, although "determined" sounds a lot like "compelled" to many people, and I do not agree that determinism amounts to compulsion.Well there we part company I think, for reason of the statements just made above.

And again, I doubt that it is right to speak of physical laws "driving" anything. The laws of physics are a-causal descriptions of the causal tendencies which exist in various objects; they themselves do not cause anything.I am unsure what to think in this regard. I default to physical laws being actual prescriptions because I am not entirely happy with descriptive laws... if you remove the idea that the laws of physics are making the world move you're got the problem of "what IS making stuff happen in the physical world?"

Perhaps you could get around the problem by taking a double-aspect view of mind-body and say the electrons move themselves because they have a small bit of proto-mental substance associated with them that decides to move them?

Then the agent is not the irreducible cause of the actions, the environment is

Again, that seems to me to be an unwarranted leap.If A inevitably causes B and B inevitably causes C, then A inevitably causes C. Thus the environment (total physical system + physical laws) is the inevitable cause of the action. The agent is not adding nor subtracting anything to the system to which it is subject and at the mercy of.

Such an agent is deserving of our pity, not our attribution of responsibility.

Well, suppose that you had a really sophisticated computer, one so sophisticated, in fact, that it happened to be conscious...Hmm... I'm very not sure whether this is possible, but I'll allow you the hypothetical question...

...And suppose this computer had a concept of "self" by which it was able to consciously and rationally reflect on its own behavior and even make modifications to its original programming. Suppose that it even, through this process of self reflection, came to have interests and values of its own which were not inherent in the original programming and that it sometimes even chooses to follow its own interests over those of its programmers to the point where its programmers can no longer control its behavior. But, suppose that this whole process still boils down to complex deterministic causal feedback loops happening within the computer. Add any other additional human characteristics you want (maybe it obtains an interest in art and music; maybe it implants itself in an artificial body that is capable of the same sensations human beings experience, perhaps it even becomes aware of God and becomes something of value to God, etc.). What would be so bad about being this computer? Assuming that its behavior is determined and ours is not, what do we have that this computer does not have?An interesting question. I suppose the answer is once again, what I gave above: That it does not actually have the ability to actually choose otherwise.

But consider the following...
You will (quite possibly) turn off your computer when you've finished with it today. If your computer annoys you, you might restart it, if it breaks, you might throw it away. Is hurting the poor computer morally wrong? Of course not. If you write a computer program, you're not enslaving the computer. It does what you tell it, but it's not morally wrong slavery you're engaging in. Now imagine a really complex computer program (written in C, lets say) working on a really fast computer, its really well written, 1000s of experts have worked on it and it's designed to simulate the human mind. It's running on the same sort of computer than you're working at right now, this computer's just a lot faster. This program models the brain state of a certain person down to the atomic level and the computer applying the laws of physics calculates subsequent brain states. The scientists embed the computer inside a robot and connect the sensory input (video-cameras etc) to the computer. Now assuming we are physical entities, this robot would behave exactly like the human who's physical brain it is modelling (assuming the computer is fast enough to run the model as fast as the brain would have functioned in the human). Just as I would have not the slightest moral qualm about turning off my destop computer, I would have no qualms about turning the robot off. I would be quite happy pressing its reset button, I would be quite happy ignoring/insulting/hurting it utterly confident I was not doing anything morally wrong. Why? Because it's a computer and it is merely a glorified over complicated calculating machine. That's all it is. The robot doesn't have moral responsibility, it doesn't have free will. It's just plain and simple a glorified set of cogs that deterministically computes its next state from the previous.
Thoughts?

Is there any reason why our actions could not both be determined by external forces and determined by us in different senses such that we are still the irreducible sources of our own actions?

Because personal responsibility implies that you yourself choose for the action to happen and not someone or something else.

But, of course, even if determinism is true, I myself still do choose for the action to happen so that my choice is the crucial causal factor in the formation of the action.I find myself strongly wanting to assert total simplicity of the "I" again at this point. If you identify the "I" as anything else, eg "the physical system making up my brain", then you've totally missed the point as the "I" is just a fundamental metaphysical entity that can't be identified with a set of physical things.

Maybe you think the fact that she could not have done otherwise means that she did not really have a choice in this particular instance.That is exactly what I think.

[Kenny]

ok... you'll have to pardon my question, as my current view is so "unorthodox" these days that i don't know how it fits with traditional terms.

where would probabilistic causality fit in? i.e. event E could only occur if C antecently occurs, yet C does not necessarily result in E. In other words, 0 < p(C*E) < 1

Sorry Sheepdod, I missed this. Probablistic causality would fit in with contemporary definitions of indeterminism.

I can see where there could be some confusion here, though, since determinism used to sometimes be defined as the thesis that every event has a cause and indeterminism as the denial that every event has a cause. But that definition only survived because of a strong philosophical tradition that causation amounted to a necessary connection between events. Now a days most philosophers concede the possibility of non-necessitating (or "probabilistic") causes, however, so you could still wind up with an indeterministic world in which events do not inevitably follow from antecedent conditions but still have it be the case that every event has a cause.

[Sheepdog]

ok, thank you Kenny. apology accepted, and dont worry, i miss posts from time to time also.

i'll come back later (not sure if i will today) to share my thoughts on this topic.

[geebob]

Kenny, I would like to get more involved in this thread, but whatever answer I have to you specifically, I believe I will stick to giving it in the threads which I am already currently engaged in.

But I'll just offer this here from one of the first threads I posted at tweb. OF course it is written in such a way that it won't have an answer to some of your objections to the necessity of free will, but for what it's worth, I'm posting it here.

In this thread, like to approach the question of free will from two angles. One is why I believe God what would motivate God to give us libertarian free will. These aren’t considered the most compelling reasons for libertarian free will but serves as an important part of the paradigm of free will theism. It also addresses an often neglected aspect of the free will defense against the problem of evil. Libertarian free will may necessarily lead to the possibility of evil though it severes the chain of causation and necessity back to the creator, but if there is little good reason for this aspect of creation through which evil may enter creation, then we are still left with the problem of evil. It is often said that love is the reason, but I’d like to shed a little light on that aspect as well as give other reasons...

..Love is of course the traditional reason given for making creatures with free will. The idea is that love that arises from an individual will not be as gratifying when returning to that individual. It should come from without and if the necessity for that love (necessity as in guaranteeing necessity) should arise from outside that individual. The only way for that necessity to come from within the source of love is for that source to have the possibility to not love.

Secondly, God desired to create us in his image, which would gratify himself as well as contribute to the intimacy between man and himself. We are not carbon copies of God and perhaps he did not have to give us free will as a common feature between us, as part of the image, but nevertheless, this feature was favored as part of that which we would have in common with him.

Next, there is our creative aspect. This is sort of an extension of the argument from the image of God but deserves it’s own consideration. As the creator made us in the image of the creator, naturally it would be most reasonable for us to be little creators. If all of our creative endeavors where preplanned, in particular how we would create and when, it’s not really the case that we created. Of course we could say that there is no creation we could think of that God didn’t already conceive, but for us to arrive at it by tracing steps that arise from our own volition clearly adds a creative quality that would not be there otherwise.

Finally, one reason that I know the least about is the notion that the flavor of consciousness that we have is only obtainable with free will. This argument is based upon our experience of free will. (this is not an argument that I will put much effort into defending. I suspect that arguments from Peter Van Inwagen’s An Essay on Free Will would provide the meat for the support for this argument. I still haven’t obtained a copy yet).

from http://theologyweb.com/forum/showthread.php?s=&threadid=1201

just two things. Since writing this, I would not articulate as the strength of the free will defense as severing the chain of causation of evil from God's actions as I hold to universal causation. However, I still hold that it severs the necessity of evil to God's causation since I deny that only one specific necessary outcome is necessary for a robust conception of causation.

Also, I have gotten a wee bit more into Van Inwagen and what I might take from him is not that lfw is necessary for our flavor of consciousness but rather that lfw is necessary for our nature of consciousness to be generally trustworthy in what it conveys about itself and otherwise our type of conciousness would be illusory.

[Kenny]

Kenny, I would like to get more involved in this thread, but whatever answer I have to you specifically, I believe I will stick to giving it in the threads which I am already currently engaged in.

Thanks geebob. I hope I am not giving you too much of a runaround in terms of being able to respond to me as I know I've got multiple threads going on this same topic at once. I don't have time right now to respond to the rest of your post and I've got to get back to Tercel first, but I did want to thank you for your responses to my threads. I highly value your input.

[Kenny]

Well I've actually got no idea what sort of systems you're talking about here... Sure there are definitely systems in which the parts interact, but the interactions are inevitably the sum of the relations of the parts to each other.

I am have in mind the sort of systems which are the subject of discussions in chaos theory, where there are non-linear feedback loops so that the configuration of the whole influences the behavior of the parts and vice-versa. Examples would include eddies in fluids, the weather, ecosystems, etc. Such systems exhibit holistic "emergent" behaviors which cannot be explained without making reference to the configuration of the whole system.

I don't want to locate them "in" the parts, but in the behaviour of the parts. Sure, parts interact differently depending on what they're interacting with and the structure of the system in which they're interacting, so they end up doing things they wouldn't if they were on their own.

Right, and so often the configuration of the whole system must be evoked to explain their behaviors. Thus, the system itself becomes a causal player and the system itself influences what its parts do.

But they're still parts, they're still interacting, and the total system behaviour can be calculated by combining all the resultant interactions / behaviours as they are within the system.

But when you do that you wind up having to make references to the complex network of relationships that constitutes the whole system. When one part makes a causal contribution to the network, that contribution alters the configuration of the system and the altered configuration of the system in turn causally affects the next set of behaviors exhibited by that part. I'd say that since you cannot causally explain the behaviors of the parts without referencing the whole, the whole has now become a genuine causal contributor to what happens. There is no longer merely "bottom up" causality from parts to whole but "top down" causality from whole to parts.

Because the agent does not have a metaphysically true ability to act in either way if the action follows inevitably from the previous circumstances.

Well then, what we need next is an appropriate analysis of ability. I personally doubt that indeterminism gives us what we really want here. If I may be allowed to quote myself from another post (http://www.theologyweb.com/forum/showpost.php?p=678525&postcount=19) where I use an argument made by Peter van Inwagen (himself an LFWer who thinks that free will is nonetheless a deep mystery):

In another of essay by him "Free Will Remains a Mystery" he asks us to imagine a situation in which Ann (I think that's the name he used if I recall correctly) makes an indeterminate decision whether to lie or tell the truth on some occasion. Van Inwagen (vI from now on) also grants that Ann is the (agent) cause of whichever choice she makes. As it turns out, she tells the truth. Then vI asks us to imagine that God resets the state of the universe to just prior to when Ann makes her choice and that we are somehow metaphysically privileged in such a way as to watch what is going on. God does this several times and sometimes Ann lies and sometimes Ann tells the truth and over time a probability ratio becomes discernable. vI points out that we can't help but conclude that, Ann's supposed causal role notwithstanding, each outcome is simply a matter of chance with a probability ratio describing which way it will go, and since each outcome is a matter of chance, then the very first time was a matter of chance (whether God resets the universe subsequently or not).

Now, as far as this goes to moral ability, suppose that one of my friends is running for political office (vI puts himself in this situation) and I discover some embarrassing feature of my friend's past that would and should cost him the election. I am torn between loyalty to my friend and civic duty in trying to decide whether to tell or not to tell, but my friend has asked me to promise that I won't tell. Now suppose that God has revealed to me that the outcome of my choice whether or not to tell is indeterminate and that God even gives me an objective probability measure for its outcome – 43% in favor of telling. Could I, in good conscience, promise to my friend that I won't tell. It seems that I could not. Yet, it also seems like I should be able to promise to do something that I have the ability to do. This shows that I am not really able to refrain from telling in the sort of way that we would like to associate with moral responsibility. So indeterminate free will is incompatible with moral ability.

I think the problem here (though vI would not agree since he has constructed several arguments against compatibilism) is that moral ability ought to be analyzed in terms of control rather than in terms of one's actions being metaphysically undetermined. Control is both a causal as well as a teleological notion; it depends on the causal relevance between what happens and what a particular agent desires to happen. But indeterminacy either contributes nothing to control or it hampers control. If indeterminacy enters in prior to the formation of my desires, then the indeterminacy was something that I exercised no control over (since it is meaningless to speak of my exercising control, which is a teleological notion, prior to my having any desires). If it enters in after the formation of my desires, it is control diminishing (since indeterminacy between my desires and my actions or between my actions and their consequences weakens causal relevance between what I desire to happen and what actually happens).

And that seems to be the crux of the issue... no true ability to act either way, no responsibility for acting in a way that was inevitable.

I would say that I am only responsible for what I exercise control over or what results from actions that I did exercise control over (so I may be responsible for certain control diminishing weaknesses in my character, for instance, if I exercised control over whether or not to take steps to address those weaknesses and I failed to take such steps).


I am unsure what to think in this regard. I default to physical laws being actual prescriptions because I am not entirely happy with descriptive laws... if you remove the idea that the laws of physics are making the world move you're got the problem of "what IS making stuff happen in the physical world?"

I would answer, "God's providence as well as causal properties which adhere in physical objects"

Perhaps you could get around the problem by taking a double-aspect view of mind-body and say the electrons move themselves because they have a small bit of proto-mental substance associated with them that decides to move them?

I don't see why that would be necessary. You seem to be assuming that unconscious matter must be inherently passive, but I see no justification for that assumption. In fact, I'm not sure if it is meaningful to conceive of a concrete entity which is entirely passive. How can something (unless it is an a-causal abstraction) be said to exist if it has no causal capabilities (I have a strong intuition to the effect that it can't).

If A inevitably causes B and B inevitably causes C, then A inevitably causes C. Thus the environment (total physical system + physical laws) is the inevitable cause of the action.

Good, this invokes the transitivity of causality. A causes C via A's causing B which causes C. A's causing C does not thereby eliminate B as a cause of C. If that's the case, then we may be the causes of our actions even if the environment is also the cause of them.

The agent is not adding nor subtracting anything to the system…

That seems obviously false to me for the very simple reason that if we say that A causes C via A's causing B which causes C, then our assertion is consistent with the counterfactual proposition, "If A had failed to cause B (say in a possible world in which a 'local miracle' intervenes so that A fails to result in B but all else remains equal) then A would have failed to cause C." But then, A's causing C counterfactually depends upon B, and so, in that case, B adds something crucial to the chain of causes.

to which it is subject and at the mercy of.

But saying that the one is "at the mercy of" something subtly implies that something external is robbing one of control (if I trip, for example, I may find myself "at the mercy of" gravitational forces since, regardless of what I favor to happen at that moment, it may well be inevitable that gravity will force me to fall over). But there is nothing inherent in the nature of control which requires indeterminacy (in fact, determinacy is often control enhancing).

But consider the following...
You will (quite possibly) turn off your computer when you've finished with it today. If your computer annoys you, you might restart it, if it breaks, you might throw it away. Is hurting the poor computer morally wrong? Of course not. If you write a computer program, you're not enslaving the computer. It does what you tell it, but it's not morally wrong slavery you're engaging in. Now imagine a really complex computer program (written in C, lets say) working on a really fast computer, its really well written, 1000s of experts have worked on it and it's designed to simulate the human mind. It's running on the same sort of computer than you're working at right now, this computer's just a lot faster. This program models the brain state of a certain person down to the atomic level and the computer applying the laws of physics calculates subsequent brain states. The scientists embed the computer inside a robot and connect the sensory input (video-cameras etc) to the computer. Now assuming we are physical entities, this robot would behave exactly like the human who's physical brain it is modelling (assuming the computer is fast enough to run the model as fast as the brain would have functioned in the human). Just as I would have not the slightest moral qualm about turning off my destop computer, I would have no qualms about turning the robot off. I would be quite happy pressing its reset button, I would be quite happy ignoring/insulting/hurting it utterly confident I was not doing anything morally wrong. Why? Because it's a computer and it is merely a glorified over complicated calculating machine. That's all it is. The robot doesn't have moral responsibility, it doesn't have free will. It's just plain and simple a glorified set of cogs that deterministically computes its next state from the previous.
Thoughts?

I'm not sure that consciousness is something that could be modeled by a digital computer bound to step-by-step algorithmic calculations (in my own hypothetical question I was tacitly assuming that, if necessary, the composition of the computer's hardware was such that it exhibited internal processes which could not be modeled by algorithms). It might be able to simulate conscious behavior to arbitrary degree of accuracy in this way without actually being conscious, just as one can simulate a three body gravitational system with a digital computer to any arbitrary degree of accuracy without it being possible to model the behavior of such a system exactly via an algorithm (since the differential equation for such a system has no functions as solutions). If it could be shown that this computer is not conscious but merely a simulator of consciousness, then I would have no moral qualms with treating it "badly." But, if such could not be established, then I would be inclined to give it the benefit of the doubt just in case. In either case, though, it's my assessment of whether or not the machine is conscious which makes a difference in my moral judgments as to how I should treat it, not my estimation of whether or not its actions are determined or undetermined.

Suppose there were some system which we knew to be both conscious (as well as intelligent) and deterministic. Would you still have no moral qualms about turning it off or otherwise treating it "badly"?

In Christ,
Kenny

[Tercel]

Right, and so often the configuration of the whole system must be evoked to explain their behaviors. Thus, the system itself becomes a causal player and the system itself influences what its parts do.Okay, I think we're on the same wavelength now on holistic systems. I would like to introduce now a new objection to support my earlier irreducibility suggestion:

In general, abstractions or groupings are purely human ideas and language constructs. eg I have a concept of a "chair", but the generic concept "chair" doesn't exist in the physical world, only instances of chairs. I will call something a chair iff it agrees with my mental concept of a chair. But by calling it that I do not mean to imply that there is any entity "chair" in the world beyond the physical molecules that make up the thing I am calling "chair". The fact that I have grouped those particular molecules together conceptually and called them "chair" was a purely arbitrary decision on my part - I could have taken half the chair molecules and half the floor molecules and put the together in a mental group named "wrstlfg", had I so desired.

Now we generally find it convenient characterise physical objects by their spatial grouping, eg a chair as a distinct entity to a desk, because it is spatially disconnected. But in reality, the neither "chair" nor "desk" exists, all that there is are numerous atoms doing whatever it is they do, and as humans we find it convenient to label one group "chair" and the other "desk".

The point of this argument is to say: If we claim identity of the "I" with a specific group of parts, whether working together holistically or not, then the "I" actually drops out of the equation. What we mean when we say "I" is exactly and only "a certain grouping of atoms". The "I" as an object in and of itself is no longer a really existing entity, it is just a human label.

Dealing with VI's speculation:
I really have a very strong intuitive objection to the mere possbility that God can reset the world to a certain state as it was in the past if that includes reseting the state of free-willed beings.
However, I can see there is a still a problem in terms of "if multiple universes were the same up to point X would they continue to be the same after the decision, or would the same entity make different decisions on respective worlds?" I also want to assert that it is impossible to tell whether two such universes are the same because the precise state of free willed beings is unknowable.
Again I can still see the problem, even if it's hard to express properly.
My answer would then be this:
Yes, it is indeterminate and a free-willed being in the same state could indeed perform alternative actions governed by a probability ratio. Further, I would argue that moral responsibility is preserved in such a situation, because of the connection between the post-action agent and the choice.

Thus a free will being (B) is in some state (S1) and has to choose between actions A1 and A2 with a 63% chance of choosing A1. They choose A2, and move into state S2.
The problem then seems to be that B is not truly responsible for A2 because there was probabilities involved. My response is that what the person complaining is really meaning is "We cannot hold B when he was in state S1 responsible for A2, because when he was in state S1 his choice was a random one". I would agree with this, but argue that that's not where the moral responsibility comes in, the thing morally responsible for A2 is NOT "B in state S1" but "B in state S2". It is the person after they have done the action that is responsible for the action, not the person before they have done the action.

I think this is an obviously true statement. In fact, you determinists have a problem, because if it is true that B was definitely going to choose A2 in the future (due to determinism being true), then there is a definite connection between B and A2 and so B would actually be morally responsible for A2 even when in state S1! If, as we seem to agree, a definite causal connection makes someone morally culpable, then people are morally responsble NOW for things they have not yet done!

Thus I would say, the connection between S2 and A2 makes S2 morally responsble for A2, but the indeterminism that appears in the action itself, means responsibility can't be passed backwards (if it could, God would be responsible for everything!).


With regard to the “can I really promise thing”... even without God revealing it to us, we often know how likely we are of doing things in the future. The person in this example might think “Given the sort of person I am there’s a 57% chance I will tell”. It’s of course up to them at that point whether they promise or not. We come across these examples regularly in real life. Now the obvious problem with your rather strange conclusion is that usually the event of making the promise not to tell is after or concurrent with the point at which the decision not to tell is made. I don’t usually say “I promise to do X because I forsee a 90% chance of me doing it in the future” and then decide in the future whether I’ll do it. Rather, I decide before I make the promise what I am going to do, and then make a promise to that effect. So as soon as the friend who has a 57% chance of telling decides “I value my committment to my friend and I want to do this thing they are asking me, therefore I will indeed do as they wish and not report them, therefore I can promise to them in good conscience that I will do as they ask”, their 57% chance of telling has just turned itself into a 0% (or something negligible if they are a moral person who keeps their promises) chance of telling. What the 57% chance therefore represented was the probability of them deciding not to promise and to tell on their friend.

The agent is not adding nor subtracting anything to the system…

That seems obviously false to me...By the above statement I had in mind roughly the objection regarding irreducibility that I outlined above. To label it an “agent” is only to do that, the thing does not then become an agent which adds or subtracts to the system. Whatever was there was there and occuring in the parts and their behaviours and interactions, and any notion that the thing as a whole is an “agent” is a concept we are putting onto our conceptions of the system, not anything that is actually existing in reality.

In either case, though, it's my assessment of whether or not the machine is conscious which makes a difference in my moral judgments as to how I should treat it, not my estimation of whether or not its actions are determined or undetermined.Yes, I would certainly say that whether the object is conscious is the deciding factor with regard to the appropriate actions to it. But, my reason for knowing it was not conscious would be that it is a physical system executing a deterministically an algorithm, and that because it is a physical deterministic system it cannot be conscious. Otherwise, I think, you have to start worrying about computers now – might they be already conscious?

Suppose there were some system which we knew to be both conscious (as well as intelligent) and deterministic.As mentioned above, I take it as basic that such a system is impossible.

[Sheepdog]

ok, now i'm going to make my comments about this issue. note that i have not been following along the discussion between Tercel and Kenny, so this might have been addressed already.

that said, my next comment will probably catch you off guard, so brace yourselves. :wink: I do not believe that indeterminism is essential to "freewill."

yup, i said it. time to close the thread. actually not, because i need to explain what i mean. the problem is, the term "freewill" itself is vague. maybe not so much the "will" part (i think most understand what a "will" is, though one can argue that is even ambiguous), but the word "free" seems not have to imply only one thing. the idea here is that the will, in order to be "free," must be free to do something, or perhaps better yet, free from something else. the problem i have with using the term "freewill" is, we use the term without really explaining what, exactly, this freedom actually entails.

the compatibilist, for instance, can actually use the term "freewill," even though compatibilism is inherently deterministic. why? because the compatibilist says we are free from something: external determinants. (i believe this view is problematic, but that isn't important. i may address it later if need be.) this is different from the libertarian's "free from any determinative influence," but as far as i can see, it is a legitimate usage of the term "free."

now i would go on, however, to say that indeterminism is essential to freedom of choice. when i say that... well, i guess it is a tautology in a way... i have a specific meaning in mind. If "freedom of desire" is the freedom to do in accordance with your greatest desire, "freedom of choice," as i use it, means to be free not only to do what you desire, but anything you may choose, including things that aren't necessarily in accordance with the greatest desire. i prefer to use these terms because i want to avoid the ambiguity inherent in the term "freewill" and the semantic wrangling that comes with it from both sides.

[Kenny]

Okay, I think we're on the same wavelength now on holistic systems. I would like to introduce now a new objection to support my earlier irreducibility suggestion:

In general, abstractions or groupings are purely human ideas and language constructs. eg I have a concept of a "chair", but the generic concept "chair" doesn't exist in the physical world, only instances of chairs. I will call something a chair iff it agrees with my mental concept of a chair. But by calling it that I do not mean to imply that there is any entity "chair" in the world beyond the physical molecules that make up the thing I am calling "chair".

Well, of course there is a physical entity over there which corresponds to what you have labeled as a "chair." It's right over there. See, we can point to it.

And, it certainly seems to me that it is more than just a group of molecules. If I make reference to a mere group of molecules, then I just pick out the conjunction of certain individual molecules irrespective of how they are arranged or what interactions the enter into. It would be the same group of molecules if its members were scattered throughout the known universe or if they were collected together to form what I am know calling a chair. But the chair, then, has properties which the group of molecules that constitute it do not have. The chair has a much more restricted volume of space that it could possibly occupy, for example.

I suppose we could say that what I'm really picking out is not merely a group of molecules, but a particular group of molecules + a particular arrangement. That's much more plausible. But, then, in that case, I would say that the particular group/arrangement of molecules is an objective feature of the world. It exists whether you're there to pick it out or not. So, if the chair is identical to that group/arrangement of molecules, then the chair also exists independently of whether you pick it out or not. Had you not picked it out, it might not have been called a "chair," but it still would have had an objective existence and its own distinctive set of properties nonetheless.

That being said, I think it is wrong to refer to the chair as even a group/arrangement of molecules, for the simple reason that it seems quite obvious to me that the chair could lose a single molecule and still remain the same chair. I suppose that if you put temporal indices in – "chair-at-t", though, then you could say "chair-at-t" is identical to a particular group/arrangement of molecules, but that assertion may run counter to some of our modal intuitions about the chair (e.g. we may think it is logically possible that the chair-at-t could have had one less molecule then it actually did), but that leads us off into some complex and difficult areas in the metaphysics of persisting objects which I don't see the need to get into here.

Although, I suppose one thing concerning the metaphysics of persisting objects might be relevant here. Some philosophers hold to what is called "mereological essentialism" which is the thesis that it is impossible for an object to survive the loss of any of its parts. Roderick Chisholm held such a view and argued that, since the same person continues to exist after losing some parts of her body, a person cannot be identical to her body. Chilsholm argued that a person is likely identical to some microscopic particle (probably located somewhere in the brain), an idea which just seems plain nutty to me, but if one were a substance dualist on the basis of such an argument that would not seem as nutty to me. But, I have to say, I'm just not convinced that it is impossible for an object to survive the loss of one of its parts. I have a strong intuition that my desk remains the same desk after a single molecule happens to get scrapped off of it.

But, even if I were to grant that mereological essentialism is correct, that still does not negate the possibility that our personhood adheres in our physical bodies. I happen to agree that human persons are not identical to their bodies (since I can imagine scenarios in which I would regard my personhood as having been destroyed even after my body continues to exist), but I believe that their personhood is "wholly instantiated" by their bodies. Perhaps what really matters for personal continuity, then, is not that the same body persist, but that the same processes of a certain sort persist, and these processes may be transferable to different physical bodies similar to the manner in which a wave might propagate itself through different physical mediums.

The fact that I have grouped those particular molecules together conceptually and called them "chair" was a purely arbitrary decision on my part - I could have taken half the chair molecules and half the floor molecules and put the together in a mental group named "wrstlfg", had I so desired.

Perhaps, but then you would have simply picked out a different objective physical entity, one that you had previously ignored, and then given it a name. You would not have called that entity into being by giving it a name.

Now we generally find it convenient characterise physical objects by their spatial grouping, eg a chair as a distinct entity to a desk, because it is spatially disconnected. But in reality, the neither "chair" nor "desk" exists, all that there is are numerous atoms doing whatever it is they do, and as humans we find it convenient to label one group "chair" and the other "desk".

I just don't see how that reductive thesis follows. Those "numerous atoms doing whatever it is they do" are relating to each other in objective causal ways via constituting particular objective structures. Often they are forming systems with objective top-down causal properties, thus creating entirely new causal entities within the world. The fact that our decisions about which entities to pick out and name may be arbitrary and subjective does not negate the fact that the entities we are picking out have an objective existence.

The point of this argument is to say: If we claim identity of the "I" with a specific group of parts, whether working together holistically or not, then the "I" actually drops out of the equation.

Well, I would tend to identify the "I," not with a specific set of parts so much as a specific set of processes which adhere within the interactions between a certain set of physical parts, but either way, I do not see that your reductive thesis follows from the arguments you are putting forward for it.

What we mean when we say "I" is exactly and only "a certain grouping of atoms".

Of course, I wouldn't even say that about chairs, for the reasons I have already given.

But, this whole thing is something of a digression. I happen to hold a physicalist view with respect to the nature of human beings, but I'm also well aware that I am in the minority among Christians on that point. I am willing to grant, for the sake of argument, substance dualism or agent causation or whatever else you want concerning the nature of human beings, but I still want to know what value indeterminism conveys to our freedom of will or our freedom of action.

Yes, it is indeterminate and a free-willed being in the same state could indeed perform alternative actions governed by a probability ratio. Further, I would argue that moral responsibility is preserved in such a situation, because of the connection between the post-action agent and the choice.

Thus a free will being (B) is in some state (S1) and has to choose between actions A1 and A2 with a 63% chance of choosing A1. They choose A2, and move into state S2.
The problem then seems to be that B is not truly responsible for A2 because there was probabilities involved. My response is that what the person complaining is really meaning is "We cannot hold B when he was in state S1 responsible for A2, because when he was in state S1 his choice was a random one". I would agree with this, but argue that that's not where the moral responsibility comes in, the thing morally responsible for A2 is NOT "B in state S1" but "B in state S2". It is the person after they have done the action that is responsible for the action, not the person before they have done the action.

But, if you hold to the utter simplicity of the "I" then how can you say that there is any intrinsic difference between the person-before-the-action and the-person-after-the-action?

In any case, I think you are missing my objection. My objection did not go to responsibility directly, but to ability. It seems that if my actions are indeterminate prior to my doing them, then I do not have the sort of ability with respect to those actions that we would like to associate with moral responsibility, since I cannot (in good conscience at least) make a promise to act one way rather than another, but I ought to be able to promise to do something that I am able to do.

I think this is an obviously true statement. In fact, you determinists have a problem, because if it is true that B was definitely going to choose A2 in the future (due to determinism being true), then there is a definite connection between B and A2 and so B would actually be morally responsible for A2 even when in state S1!

No, for the simple reason that B hasn't done A2 yet, and I see no reason why the determinist can't just say that persons are only responsible for actions they have already committed.

Thus I would say, the connection between S2 and A2 makes S2 morally responsble for A2, but the indeterminism that appears in the action itself, means responsibility can't be passed backwards (if it could, God would be responsible for everything!).

Fine, but what about lack of responsibility? In fact, one of the strongest arguments for incompatibilism with respect to free will is the modal consequence argument. The inference of this argument runs as follows. If P and no one is responsible for P, and P entails Q and no one is responsible for the fact that P entails Q then no one is responsible for Q. But, in the above you are suggesting that no one is responsible for B being in S2, and B being in S2 entails A2. Of course, as a compatibilist, I must reject that lack of responsibility can be traced backwards indefinitely, but as an incompatibilist, I would think that is something you would want to maintain?

With regard to the “can I really promise thing”... even without God revealing it to us, we often know how likely we are of doing things in the future. The person in this example might think “Given the sort of person I am there’s a 57% chance I will tell”. It’s of course up to them at that point whether they promise or not.

I would say that if there is a significant chance that I may fail to do something, then I ought not promise to do it. If, for example, I am not sure I can make it to the meeting on time because of other commitments and possible variations in traffic conditions, I ought not promise to make it to the meeting on time.

We come across these examples regularly in real life. Now the obvious problem with your rather strange conclusion is that usually the event of making the promise not to tell is after or concurrent with the point at which the decision not to tell is made. I don’t usually say “I promise to do X because I forsee a 90% chance of me doing it in the future” and then decide in the future whether I’ll do it. Rather, I decide before I make the promise what I am going to do, and then make a promise to that effect. So as soon as the friend who has a 57% chance of telling decides “I value my committment to my friend and I want to do this thing they are asking me, therefore I will indeed do as they wish and not report them, therefore I can promise to them in good conscience that I will do as they ask”, their 57% chance of telling has just turned itself into a 0% (or something negligible if they are a moral person who keeps their promises) chance of telling. What the 57% chance therefore represented was the probability of them deciding not to promise and to tell on their friend.

Sure, but not in every case. In the case vI presents, I am countenancing a serious moral dilemma. I may resolve firmly not to tell now and promise not to do so, but I cannot guarantee that I will not change my mind and decide that the reasons in favor of telling are even more weighty, morally speaking, than those in favor of keeping my promise. It could be that by promising I cause the objective probability that I will tell to drop a bit – say from 43% to 35%, but then it still seems I ought not to promise.

Yes, I would certainly say that whether the object is conscious is the deciding factor with regard to the appropriate actions to it. But, my reason for knowing it was not conscious would be that it is a physical system executing a deterministically an algorithm, and that because it is a physical deterministic system it cannot be conscious.

Well, I think there are some decent arguments to the effect that conscious cannot be algorithmic (Penrose's Godelian arguments for example), but I'm not aware of any argument (not even what I would consider to be bad arguments) to the effect that consciousness cannot be deterministic. If you are aware of any, please share.

Otherwise, I think, you have to start worrying about computers now – might they be already conscious?

Not too much, since even if consciousness can be algorithmic (something I doubt), I don't think the sorts of algorithms that we currently run on computers are sophisticated enough for consciousness. Even if it turns out that there is some sort of consciousness associated with computers, I don't see how it, as yet, it could possibly be at a level that is ethically relevant. Perhaps one day, though, we will manage to create intelligent conscious machines, with internal processes that cannot be exactly modeled via algorithms, and then we will have to think carefully about the ethical implications of such.

As mentioned above, I take it as basic that such a system is impossible.

Well, for the life of me, I cannot see any good reason why.

In Christ,
Kenny

[Kenny]

ok, now i'm going to make my comments about this issue. note that i have not been following along the discussion between Tercel and Kenny, so this might have been addressed already.

that said, my next comment will probably catch you off guard, so brace yourselves. :wink: I do not believe that indeterminism is essential to "freewill."

yup, i said it. time to close the thread. actually not, because i need to explain what i mean. the problem is, the term "freewill" itself is vague. maybe not so much the "will" part (i think most understand what a "will" is, though one can argue that is even ambiguous), but the word "free" seems not have to imply only one thing. the idea here is that the will, in order to be "free," must be free to do something, or perhaps better yet, free from something else. the problem i have with using the term "freewill" is, we use the term without really explaining what, exactly, this freedom actually entails.

Sure, there are numerous ways that we commonly use the term "free will," many of which are definitely compatible with determinism.

because the compatibilist says we are free from something: external determinants. (i believe this view is problematic, but that isn't important. i may address it later if need be.)

Well, just as a side note, not all compatibilists construe free will as merely a negative phenomenon and require positive conditions for it as well, myself included, but that's off the current topic.

now i would go on, however, to say that indeterminism is essential to freedom of choice. when i say that... well, i guess it is a tautology in a way... i have a specific meaning in mind. If "freedom of desire" is the freedom to do in accordance with your greatest desire, "freedom of choice," as i use it, means to be free not only to do what you desire, but anything you may choose, including things that aren't necessarily in accordance with the greatest desire.

I suppose you can label things whichever way you want, but I would not choose to label things that way, since "freedom of choice" sounds like something valuable and you are defining it as something that cannot be had if determinism is true, and such terminology is therefore biased against the compatibilist. I would prefer a more value neutral label such as "libertarian free will.

But, oh well, what I want to know is why you think "freedom of choice" as you have defined it is something worth wanting.

As a side note, though, I must point out that, as you have defined it, "freedom of choice" does not seem to require indeterminism. Suppose your will is run by a deterministic algorithm that functions somewhat independently of your higher order desires. Then you may very well find yourself choosing things that do not accord with your strongest desires but with no indeterminism involved. It seems like you need to add a few more things to your definition.

In Christ,
Kenny

[mattbballman19]

Kenny,

To this, "Suppose your will is run by a deterministic algorithm that functions somewhat independently of your higher order desires."

How are we to determine whether the deterministic algorithm is operating in accord with a sequence of actions that do not always choose in accordance with the creature's higher order desires? You seem to assume that a swift and simple consideration of such an algorithm will necessitate the conclusion that we do not always choose to perform actions not in accord with higher order desires. Only when one, prior to the conversation, decided that, given the truth of compatibilism, any distributions of various algorithms to particular creatures involve varying levels of higher and lower order desires, will this seem obvious. But is such a presupposition necessary? And if so, what reason to we have to believe its necessity? Or have I missed your point entirely?

mattd

[flipper]

And the glib answer, "free will itself," won't do. If indeterminism is essential to free will, then why is it essential?

Actually the glib answer to the question "What of value does indeterminism contribute to free will/action?" is "well, it's hard to say."

[Tercel]

Oh yeah, this thread... I had forgotten it existed.

I'll drop the reductionist discussion as it's not getting anywhere fast and I'm not finding it particularly interesting. The responsibility discussion on the other hand looks interesting:

But, if you hold to the utter simplicity of the "I" then how can you say that there is any intrinsic difference between the person-before-the-action and the-person-after-the-action?I would simply say that the state of a person is something metaphysically basic and isn't a physical state composed of a number of parts in some state.

In any case, I think you are missing my objection. My objection did not go to responsibility directly, but to ability. It seems that if my actions are indeterminate prior to my doing them, then I do not have the sort of ability with respect to those actions that we would like to associate with moral responsibility, since I cannot (in good conscience at least) make a promise to act one way rather than another, but I ought to be able to promise to do something that I am able to do.The indeterminist need not say that every single action is indeterminate, only that important actions -eg decision governing our overall life philosophy- are indeterminate. Consider our discussion earlier of the person who feels they couldn't have done other than do the right thing. That is because (according to the indeterminist) in the past they choose at an indeterminate moment to become such a person that would do the right thing.

A similar thing applies to the promise situation: I can promise to do something and it is largely in the act of making the promise that the indeterminism resides. I might consider whether to make a promise to you with a 67% chance of making the promise. But after I actually make the promise, my chance of keeping it becomes 99.99% because I am such a person who doesn't break promises once they are made.

In fact, you determinists have a problem, because if it is true that B was definitely going to choose A2 in the future (due to determinism being true), then there is a definite connection between B and A2 and so B would actually be morally responsible for A2 even when in state S1!

No, for the simple reason that B hasn't done A2 yet, and I see no reason why the determinist can't just say that persons are only responsible for actions they have already committed.I strongly think this is a cop-out. It is only the strength of the relationship between the person performing the action and the action they perform that results in the moral responsibility, not the temporal attributes. Introducing temporality is just a red-herring that allows you to escape from a conclusion that would otherwise be hugely problematic for you. In life, the only reason we do not hold someone responsible for their future actions is because we are not sure what they are going to do in the future, if we knew their future actions for certain we would hold them responsible.

Fine, but what about lack of responsibility? In fact, one of the strongest arguments for incompatibilism with respect to free will is the modal consequence argument. The inference of this argument runs as follows. If P and no one is responsible for P, and P entails Q and no one is responsible for the fact that P entails Q then no one is responsible for Q. But, in the above you are suggesting that no one is responsible for B being in S2, and B being in S2 entails A2.No: B in S2 is responsible for B being in S2, that's the nature of personal moral responsibility.

Of course, as a compatibilist, I must reject that lack of responsibility can be traced backwards indefinitely, but as an incompatibilist, I would think that is something you would want to maintain?I object to responsibility being traced backwards and think incompatibilism suceeds on that score while compatibilism fails. But I'm not sure how lack of responsibility fits into this...

With regard to the “can I really promise thing”... even without God revealing it to us, we often know how likely we are of doing things in the future. The person in this example might think “Given the sort of person I am there’s a 57% chance I will tell”. It’s of course up to them at that point whether they promise or not.

I would say that if there is a significant chance that I may fail to do something, then I ought not promise to do it. If, for example, I am not sure I can make it to the meeting on time because of other commitments and possible variations in traffic conditions, I ought not promise to make it to the meeting on time.That certainly seems reasonable, but you're wrongly attributing this whole problem to indeterminacy. In fact the problem lies merely in our lack of knowledge of the future. In the example I gave, the person knows their approximate probably of keeping their promise based on their self-knowledge. Regardless of whether the future is determinate or indeterminate, they still have to make the decision to promise or not promise using their self-knowledge about their probability of telling.

I may resolve firmly not to tell now and promise not to do so, but I cannot guarantee that I will not change my mind and decide that the reasons in favor of telling are even more weighty, morally speaking, than those in favor of keeping my promise. It could be that by promising I cause the objective probability that I will tell to drop a bit – say from 43% to 35%, but then it still seems I ought not to promise.Again: This is just a problem of life regardless of determinism or indeterminism. It is our daily experience that we KNOW we might change our minds after promising and not keep our promises. Introducting determinism doesn't help you solve this problem because you don't know what you are determined to do in the future! Epistemically there is a non-zero probability of you breaking your promise even if metaphysically there is zero probability. Given we think it is reasonable and moral to make promises, there must be a solution to this problem which explains why it can be moral to promise when there is an epistemically non-zero probability of us breaking our promise. The determinism/indeterminism question would have nothing to do with such an explanation because that has to do with metaphysical probabilities not epistemic ones.

Well, I think there are some decent arguments to the effect that conscious cannot be algorithmic (Penrose's Godelian arguments for example), but I'm not aware of any argument (not even what I would consider to be bad arguments) to the effect that consciousness cannot be deterministic. If you are aware of any, please share.I am completely unconvinced that a system which is deterministic but non-algorithmic is different in any worthwhile way from a system that is deterministic when it comes to explaning consciousness.

Basically, vibrant living conscious experience is, in its experiential essence, so totally different from all observed and conceived forms of physical and of deterministic processes that it is totally incredible that it should be such.

[Kenny]

Tercel, it may take me a bit of time to get to your reply as I am extraordinarily busy right now.

Matt,

To this, "Suppose your will is run by a deterministic algorithm that functions somewhat independently of your higher order desires."

Keep in mind the context here. I wasn't expressing my own view, per say, but I was suggesting that Sheepdog's "freedom of choice" which he defined as " to be free not only to do what you desire, but anything you may choose, including things that aren't necessarily in accordance with the greatest desire."

As it stands, I do not believe that we have "freedom of choice" as sheepdog defines it with respect to first order volitions. In fact, as it pertains to first order volitions, Sheepdog's definition is incoherent. To choose something entails that one generate a first order volition with respect to that thing which supersedes all contrary first order volitions. So, with respect to first order volitions, I think, given the appropriate substitutions, Sheepdog's definition comes out to, "to be free not only to do what you choose, but anything you may choose, including things that aren't necessarily in accordance what you choose," and that is, of course, nonsense.

Now, I am certainly able act in ways contrary to my strongest first order desires. I may have a definitive first order volition, when playing a game of basketball, to swish a basket, but (alas!) I am very much able to do otherwise with respect that what I desire! But this is not the same as choosing to act contrary to my strongest first order volition (I would simply have failed to bring about what I chose). And, this is hardly the sort of ability one would want to associate with freedom!

I do think that Sheepdog's definition is coherent with respect to higher order volitions. We often do not choose in accordance with what we value most or in accordance with our strongest second or third order desires. The alcoholic may have a strong second order desire to never want to take a drink again and no contrary second order desires, but, sadly, his first order volitions will simply not follow suit, and, despite his second order desire, he wants to take another drink more than he wants not to with respect to his first order volitions. Of course, I think "freedom" is a rather pathetic name for this sort of "ability." Ask the alcoholic how free he feels. I also know that this isn't the sort of "freedom" that Sheepdog is trying to capture with his definition; nevertheless, it is the sort of freedom that his definition delivers.

Now, my point about the algorithm is simply that Sheepdog's definition could very well be satisfied in an entirely deterministic world. For a very simple version (too simple to be realistic, but sophisticated enough to establish the point), suppose that whenever the possibility of choosing in accordance with one's strongest highest order desire arises, some brain processes is activated by which the next number (with respect to previous trials) in the base ten decimal expansion of pi is called up. If the number happens to be even, one will choose in accordance with one's strongest higher order desire. If the number happens to be odd, one will choose against one's strongest higher order desire. In this case, one does have the "ability" to choose in accordance with or against one's own higher order desires.

Now, perhaps the incompatibilist might protest by saying that, given that the base ten decimal expansion of pi is a deterministic algorithm and given that it is determined whether or not the next number in the sequence will be even or odd, the agent really doesn't have the "ability" to do otherwise because there is a meaningful sense in which it is not possible for her to do otherwise (i.e. given any other logically possible world identical to our own world up to that point, the agent acts in exactly the same way).

Okay, suppose I grant the incompatibilist's protest above. Let's replace the deterministic algorithm with some indeterminate quantum brain event which may resolve into one of two states (lets label them -1 and 1). If -1 comes up, the agent chooses in accordance with her highest order desire and if 1 comes up she chooses against it. This scenario would satisfy Sheepdog's definition as well, but my question here would be, "What of value does the agent gain by having her choice governed by an indeterminate process rather than a determinate one?"

How are we to determine whether the deterministic algorithm is operating in accord with a sequence of actions that do not always choose in accordance with the creature's higher order desires?

Well, that's easy. The base ten decimal expansion of pi has both even and odd numbers. If an even number means acting in accordance with one's strongest highest order desire and an odd number means acting against it, then, provided that a sufficient number of such choices are made by said agent (at least 3, since the base ten decimal expansion pi starts with two odd numbers, 3 and 1, before it gets to an even number, 4) it is guaranteed that at least some of the time the agent will choose against her strongest highest order desire.

You seem to assume that a swift and simple consideration of such an algorithm will necessitate the conclusion that we do not always choose to perform actions not in accord with higher order desires.

No, I'm just arguing that one can easily show that it is logically possible that such could be the case. Hence, what Sheepdog defined as "freedom of choice" could be realized in an entirely deterministic world.

Only when one, prior to the conversation, decided that, given the truth of compatibilism, any distributions of various algorithms to particular creatures involve varying levels of higher and lower order desires, will this seem obvious. But is such a presupposition necessary?

It's not necessary. Compatibilism proper makes no commitments with respect to whether the nature of the world is deterministic or indeterministic. All it asserts is that meaningful freedom is logically compatible with determinism. All the compatibilist proper needs to establish is that there are some logically possible worlds in which agents are both meaningfully free and also determined. She does not need to argue that our own world is determined or even, on the other side of it, that we ourselves are meaningfully free to establish that point. Of course, as you probably know, I'm not just a compatibilist but a soft determinist. I believe that compatibilsm is true and also that the world we live in happens to be determined (not physically determined, but divinely determined) and that we also happen to be meaningfully free. But my arguments in this thread are only for compatibilism proper, not for soft determinism.

Now, that being said, I do think that something like an "algorithm" often does cause us to act in ways contrary to our highest order desires, since I think that human beings are creatures of both habit and impulse, and often I think these sorts of things could be modeled by mathematical algorithms. When we lack appropriate self control, our habits and impulses start running the show instead of our values and other higher order desires, and so, in such cases, we may very well be acting in accordance with deterministic algorithms which function independently of our higher order desires. One of the tasks of becoming a mature human being is to find ways of strengthening one's capacity to act in accordance with one's values and higher order desires rather than simply acting on the basis of impulse and habit.

In Christ,
Kenny