Kenny
September 7th 2004, 05:47 PM
I'm snipping a lot of this material from a post ( http://www.theologyweb.com/forum/showpost.php?p=699984&postcount=11) where I respond to geebob in my Preconditions for Control ( http://www.theologyweb.com/forum/showthread.php?t=36907) thread over in the Philosophy forum, since I think a discussion of this issue is significant in its own right, and I'm putting it here in Theology 201 because I want to put more of a theological spin on it. This OP should stand on its own, though the analysis I've done in the "Preconditions for Control" thread does provide much of the theoretical context for what I am saying here.
geebob, if you wish, you may respond to the parts from the original post that are reprinted here and then just focus on the parts left over in the other post (those dealing more directly with my analysis of control), or you can just respond to the whole thing over there (actually, you can do whatever you want, but I hope that by duplicating a part of our discussion, I'm not making responding inconvenient for you, if so, I apologize).
I have come to the conclusion that some versions of LFW are coherent (which is a shift in position for me, since I have argued to the contrary), though I still do not believe in LFW. These versions are what I refer to as "soft LFW." But, I don't know if most LFWers around here would consider these versions "enough."
The weakest form of soft LFW would involve indeterminacy in the deliberative process with respect to which reasons come to mind for or against a particular decision in the process of deliberation. It may be that if reason R occurs to agent S, then S will decide to do A, but if R does not occur to S, then S will decide not to do A, and whether or not R will occur to S is currently undetermined. This sort of soft indeterminist proposal has been suggested by the philosopher Alfred Mele (though Mele proposes it only as an epistemic possibility, neither endorsing it nor repudiating it as fact). As Mele points out, a compaibtilist can't say that such an indeterminist account of which reasons occur to us is anymore control diminishing than a determinist one, since on either view, which reasons come to mind is something the agent has no control over anyway (well, she may have some indirect control by choosing to attend to certain types of reasons more than others thereby raising the probability that similar types of reasons will come to mind, but the initial "coming to mind" of reasons at least is completely beyond her control). Of course, this weak "LFW" view does allow for a genuine indeterminacy in our actions, but it doesn't seem to be the sort that most LFWers want. It's difficult to see exactly how indeterminacy with respect to which reasons come to mind is an enhancement to an agent's freedom.
In fact, this indeterminacy seems more like a limitation to freedom than an enhancement to it (since, for one reason, it is the case that we are often not free to act in light of the best reasons, since those reasons did not occur to us). Also, an ideally rational self-controlled agent (which none of us are), we might imagine, would simply have all possible reasons for acting or refraining from acting in a certain way available to her, (as does God) and we might plausibly consider her more free with respect to her actions on that account (since the more reasons which are available for action or inaction, there becomes a wider variety of reason/intention states for an agent to enter into – i.e. S may perform A in order that R1 or S may perform A in order that R2…).
Personally, on a physical level, I think there likely is this sort of indeterminism in our deliberative process, via QM indeterminicy in our brains, but that this indeterminism opens up a natural causal window for divine (and possibly angelic/demonic) influence, so that, on a higher metaphysical level, all the causal gaps are closed (since, as Christians, we believe that these sorts of influences do occur, this seems to me like a very plausible place to locate it).
A stronger version of this "soft LFW" would be the likes of that proposed by Laura Ekstrom. Ekstrom locates the indeterminism in her account just prior to the formation of a definitive "preference" to act one way rather than another. Prior to the formation of such preferences, according to Ekstrom, an agent considers various other non-definitive preferences, acceptances (mental endorsements of propositions that have survived a process of critical reflection with a desire to obtain to the true and the good) etc., each of which raises or lowers the probability of a definitive preference coming about. Free acts are those acts which result from intentions deterministically caused by such indeterministically formed definitive preference states The formation of a definitive preference itself (which Ekstrom regards as also being an "act") need not be construed as a free act (since such a requirement would lead to an infinite regress).
I don't know if a lot of LFWer's would like this account, however, since it allows that LFW free acts can be (must be, at some point in the causal chain) determined by acts which themselves are not free. Ekstrom's account is an interesting twist on the compatibilist theme that an agent always acts in accordance with her strongest desires, but with indeterminism coming in before the formation of those desires.
As far as Ekstrom's account relates to control, I think a meaningful sort of control we can be said to have over our volitions is that of "value directed control," in which our set of values, conjoined to a set of well formed beliefs concerning which actions best serve those values, acts as a higher order selection mechanism directing our volitions. It seems that on Ekstrom's account, we may exercise maximal value directed control over which non-definitive preference/reason states we form (my value system may allow, as an option, for me to do A for R1, but disallow me to do A for R2 – e.g. maybe stealing the money to buy a fancy new car isn't an acceptable option, but, under the right conditions of desperation, stealing the money to supply a family member with a life-saving operation is), but then we still have disjunctive non-definitive preference/reasons states left over from this process. Consequently, after this maximal control process screens out certain possibilities, it still leaves open the possibility of doing A or not A for various disjunctive reasons. If the disjunctive reasons are such that either doing A or doing not A are both equally acceptable on my value system, then my value system does not definitively favor either outcome and so I am exercising no value directed control at that moment. If my value system does definitively favor one outcome over the other, and the outcome is still indeterminate, then my value directed control is diminished, since there is a chance that I could fail to do what I most value. Ekstrom concedes that there is diminished control on account of indeterminism, but she says that someone who desires not to have this indeterminism on that account shows that he values being in control more than he values being free.
Perhaps Ekstrom has me nailed there, but it seems to me that the ability to fail to do what one most values is a determent to freedom rather than an enhancement to it. I think that cases of addiction prove this point. The alcoholic may value being sober over being drunk, but he is not free to do what he most values. He is enslaved to his addiction (note: I also have in mind Paul's language in Romans chapter 7 here). I'm not denying that we do have the ability to act against what value the most (I think we sometimes do so), nor am I denying that we are morally responsible when we do so (often we are), but our doing so proves a detriment to our freedom nonetheless (which I think suggests that the relationship between freedom and moral responsibility is more complex than others, as well as myself previously, have presupposed).
As a compatibilist, though, I wonder what value such indeterminism conveys with respect to our freedom and responsibility. What would be the practical difference, for example, if (as Daniel Dennett has asked), such genuine indeterminism were replaced with something analogous to a deterministic "pseudo-random number generator." In either case, whether the outputs are genuinely indeterminate or determined, the degree of control exercised by the agent remains the same.
From the standpoint of the Calvinist/Arminian debate, what value would such soft versions of LFW add with respect to our choice to respond to God's grace? My initial answer would have to be "none," since if our response to God's grace were indeterminate, our control over our response would be just as diminished as it would if the final resolution of that choice were sovereignly determined by God.
Perhaps the soft LFWer might say that such indeterminism provides the agent with a certain causal "originality" regarding her actions and that such originality is valuable for reasons other than those associated with control or responsibility. I'll leave it as an open question as to whether that's the case (I would like to know why, exactly, it would be though) but even if it were so, would such a value be enough to justify God's leaving our eternal destinies up to chance when God could have just as easily (on these accounts) determined the outcome in a way that would no more diminish our control than if such chance were present?
In Christ,
Kenny
geebob, if you wish, you may respond to the parts from the original post that are reprinted here and then just focus on the parts left over in the other post (those dealing more directly with my analysis of control), or you can just respond to the whole thing over there (actually, you can do whatever you want, but I hope that by duplicating a part of our discussion, I'm not making responding inconvenient for you, if so, I apologize).
I have come to the conclusion that some versions of LFW are coherent (which is a shift in position for me, since I have argued to the contrary), though I still do not believe in LFW. These versions are what I refer to as "soft LFW." But, I don't know if most LFWers around here would consider these versions "enough."
The weakest form of soft LFW would involve indeterminacy in the deliberative process with respect to which reasons come to mind for or against a particular decision in the process of deliberation. It may be that if reason R occurs to agent S, then S will decide to do A, but if R does not occur to S, then S will decide not to do A, and whether or not R will occur to S is currently undetermined. This sort of soft indeterminist proposal has been suggested by the philosopher Alfred Mele (though Mele proposes it only as an epistemic possibility, neither endorsing it nor repudiating it as fact). As Mele points out, a compaibtilist can't say that such an indeterminist account of which reasons occur to us is anymore control diminishing than a determinist one, since on either view, which reasons come to mind is something the agent has no control over anyway (well, she may have some indirect control by choosing to attend to certain types of reasons more than others thereby raising the probability that similar types of reasons will come to mind, but the initial "coming to mind" of reasons at least is completely beyond her control). Of course, this weak "LFW" view does allow for a genuine indeterminacy in our actions, but it doesn't seem to be the sort that most LFWers want. It's difficult to see exactly how indeterminacy with respect to which reasons come to mind is an enhancement to an agent's freedom.
In fact, this indeterminacy seems more like a limitation to freedom than an enhancement to it (since, for one reason, it is the case that we are often not free to act in light of the best reasons, since those reasons did not occur to us). Also, an ideally rational self-controlled agent (which none of us are), we might imagine, would simply have all possible reasons for acting or refraining from acting in a certain way available to her, (as does God) and we might plausibly consider her more free with respect to her actions on that account (since the more reasons which are available for action or inaction, there becomes a wider variety of reason/intention states for an agent to enter into – i.e. S may perform A in order that R1 or S may perform A in order that R2…).
Personally, on a physical level, I think there likely is this sort of indeterminism in our deliberative process, via QM indeterminicy in our brains, but that this indeterminism opens up a natural causal window for divine (and possibly angelic/demonic) influence, so that, on a higher metaphysical level, all the causal gaps are closed (since, as Christians, we believe that these sorts of influences do occur, this seems to me like a very plausible place to locate it).
A stronger version of this "soft LFW" would be the likes of that proposed by Laura Ekstrom. Ekstrom locates the indeterminism in her account just prior to the formation of a definitive "preference" to act one way rather than another. Prior to the formation of such preferences, according to Ekstrom, an agent considers various other non-definitive preferences, acceptances (mental endorsements of propositions that have survived a process of critical reflection with a desire to obtain to the true and the good) etc., each of which raises or lowers the probability of a definitive preference coming about. Free acts are those acts which result from intentions deterministically caused by such indeterministically formed definitive preference states The formation of a definitive preference itself (which Ekstrom regards as also being an "act") need not be construed as a free act (since such a requirement would lead to an infinite regress).
I don't know if a lot of LFWer's would like this account, however, since it allows that LFW free acts can be (must be, at some point in the causal chain) determined by acts which themselves are not free. Ekstrom's account is an interesting twist on the compatibilist theme that an agent always acts in accordance with her strongest desires, but with indeterminism coming in before the formation of those desires.
As far as Ekstrom's account relates to control, I think a meaningful sort of control we can be said to have over our volitions is that of "value directed control," in which our set of values, conjoined to a set of well formed beliefs concerning which actions best serve those values, acts as a higher order selection mechanism directing our volitions. It seems that on Ekstrom's account, we may exercise maximal value directed control over which non-definitive preference/reason states we form (my value system may allow, as an option, for me to do A for R1, but disallow me to do A for R2 – e.g. maybe stealing the money to buy a fancy new car isn't an acceptable option, but, under the right conditions of desperation, stealing the money to supply a family member with a life-saving operation is), but then we still have disjunctive non-definitive preference/reasons states left over from this process. Consequently, after this maximal control process screens out certain possibilities, it still leaves open the possibility of doing A or not A for various disjunctive reasons. If the disjunctive reasons are such that either doing A or doing not A are both equally acceptable on my value system, then my value system does not definitively favor either outcome and so I am exercising no value directed control at that moment. If my value system does definitively favor one outcome over the other, and the outcome is still indeterminate, then my value directed control is diminished, since there is a chance that I could fail to do what I most value. Ekstrom concedes that there is diminished control on account of indeterminism, but she says that someone who desires not to have this indeterminism on that account shows that he values being in control more than he values being free.
Perhaps Ekstrom has me nailed there, but it seems to me that the ability to fail to do what one most values is a determent to freedom rather than an enhancement to it. I think that cases of addiction prove this point. The alcoholic may value being sober over being drunk, but he is not free to do what he most values. He is enslaved to his addiction (note: I also have in mind Paul's language in Romans chapter 7 here). I'm not denying that we do have the ability to act against what value the most (I think we sometimes do so), nor am I denying that we are morally responsible when we do so (often we are), but our doing so proves a detriment to our freedom nonetheless (which I think suggests that the relationship between freedom and moral responsibility is more complex than others, as well as myself previously, have presupposed).
As a compatibilist, though, I wonder what value such indeterminism conveys with respect to our freedom and responsibility. What would be the practical difference, for example, if (as Daniel Dennett has asked), such genuine indeterminism were replaced with something analogous to a deterministic "pseudo-random number generator." In either case, whether the outputs are genuinely indeterminate or determined, the degree of control exercised by the agent remains the same.
From the standpoint of the Calvinist/Arminian debate, what value would such soft versions of LFW add with respect to our choice to respond to God's grace? My initial answer would have to be "none," since if our response to God's grace were indeterminate, our control over our response would be just as diminished as it would if the final resolution of that choice were sovereignly determined by God.
Perhaps the soft LFWer might say that such indeterminism provides the agent with a certain causal "originality" regarding her actions and that such originality is valuable for reasons other than those associated with control or responsibility. I'll leave it as an open question as to whether that's the case (I would like to know why, exactly, it would be though) but even if it were so, would such a value be enough to justify God's leaving our eternal destinies up to chance when God could have just as easily (on these accounts) determined the outcome in a way that would no more diminish our control than if such chance were present?
In Christ,
Kenny