Augustine on Free Choice

| | Comments (10)

Augustine gives an argument (City of God Book V chapter IX, among other places) that I've always had little patience with. Here is R.W. Dyson's translation of the City of God occurrence of it:

Moreover, even if a certain order of causes does exist in the mind of God, it does not follow that nothing is left to the free choice of our will. For our wills are themselves included in the order of the causes which is certain to God and contained within his foreknowledge. For the wills of men are causes of the deeds of men, and so He Who has foreseen the causes of all things cannot have been ignorant of our wills among those causes, since he foresaw them to be the causes of our deeds.

The reason I find such reasoning frustrating is because it comes across as if Augustine is trying to respond to the foreknowledge problem by saying that God foreknows our free choices, and if God knows our free choices, and God can't know something false, then they must be true. So foreknowdge of free choices actually establishes them as free rather than undermining it. The problem with such an argument is that it's question-begging. The opponent of foreknowledge will insist that God can't foreknow a free choice. So the very assumption of the argument is what the argument is trying to prove.

As I re-read the sections of City of God that I taught this semester in Dyson's translation (now that I've finally managed to get a copy), it occurred to me that Augustine might actually be doing something different in this text, something much less problematic. It looks to me as if what he's saying is that, even if there is this order of causes leading up to our wills, that's compatible with our choices being free, and then he gives a reason. The reason is that our wills are the causes of our actions. God's foreseeing of what we choose is God's foreseeing of our causing our actions. It's not God foreseeing our freedom that makes freedom compatible with foreknowledge, as the bad argument above has it. It's that God's foreseeing our freedom is God's foreseeing our own causing of our actions. Such causing is what explains our freedom.

Thus Augustine is making the Stoic point that our choices do happen even if there are causes of them that God can see ahead of time, and it's that they happen as choices that makes them free. Augustine does later distinguish his view from the Stoic position, but at this point he seems to be giving basically the same argument they give for compatibilism about being caused to do something and being free in doing it.

[Completely as an aside, what is going on with Dyson's capitalization in that passage? He capitalizes not just the personal pronoun but even the relative pronoun when it refers to God, but then he leaves even the personal pronoun in lower case in the very net clause. It's almost as bad as some Bible translations when trying to deal with psalms that don't clearly refer to just a messanic figure who thus to a Christian refers to Christ.]

10 Comments

The remarks that follow occur within a context where it is granted and agreed that God's foreknowledge does not cause what occurs. That said, it still remains anything but apparent that human freedom to choose is compatible or compossible with this foreknowledge. But, it is not so much the foreknowledge which is the source of the problem; rather, it is the truth of the matter which is the problem.

As discussed here, a truth - certainly expressions of truth - reference what can be described in terms of a determinate matter, state, condition, or context – which is to say that what is being referenced is described as being set or settled or definite. Qualifiers such as likely, possibly, and probably – in and of themselves – essentially deny that what is being referenced is a determinate or settled matter. For a condition to be possibly true, it has to be a matter which is to some extent indeterminate (this applies just as much in a metaphysical context as it does in an epistemological context).

If what God knows are (expressible as) truths, and if God foreknows, then this is to say that, with regards to the future, God knows it as determinate. Now, this can be either an utter determinateness wherein there is no indeterminateness, no unsettledness whatsoever, or this can be a determinate condition of which indeterminateness is constituent.

For instance, the conjoining of contraries (i.e., a person does or does not do some act at some future) can exhaustively account for all of the possibilities pertaining to some situation; this exhaustive accounting presents a determinate condition, but it is a determinate condition in which there is indeterminateness.

This is in sharp distinction to the (presumably) utter determinateness of the past, for instance, wherein there is now no indeterminateness regarding a person's doing some act. Likewise, if it is true - if it is a determinate matter - that a person does some act at some future, then that would be an instance of utter determinateness, determinateness without any indeterminateness whatsoever.

In a discussion, it is often at such a point as this that it will be noted that, although determinate, the person's act is contingent. But, where precisely does the contingency enter the picture? Is it not indeterminateness which effects contingency?

If it is indeterminateness which effects - which is necessary for - contingency, then for the contingency to be located at the person's act there would have to be indeterminateness with regards to what is to be done by the person. But, if there is indeterminateness, then there is incoherence in any claim that the matter is both indeterminate and lacking indeterminateness, and this is precisely what is claimed when there is a truth about what a person will contingently do.

In a related matter, as discussed here, if indeterminateness is necessary for freedom-to and choice, and if there is no indeterminateness with regards to the future, then how are choice and freedom-to available to human persons?

Michael, I agree with you that the truth of future contingents is really the issue. I think it's a mistake to think that the truth and falsity of future contingent choices has anything to do with whether those choices are free. You think they have everything to do with each other.

As philosophers use the term, something is contingent when it doesn't have to be that way. It could have been otherwise. Something is possible when it's one of the ways things could be. Something is necessary when it's the only way something could be.

You want to make matters of contingency and possibility instead about whether there is determinate truth about them. That, by definition, rules out contingent truths. Everything is necessary, by your definition. That's not what I mean by those terms. Barack Obama is necessarily president, by your definition. But it strikes me that things could have gone very differently, and Hillary Clinton or John McCain could have ended up winning the 2008 election. But since Obama won, it became necessary by your terms. That just seems to me to be a misuse of the philosophical jargon. It's certainly true that he won. There's no longer anything anyone could do to influence that outcome. But it's not necessary just because it happens to be in the past. The events that led up to it could have gone differently, and he might not have won. So it's a contingent truth about the past.

Why, then, is it any different about the future? Suppose he loses in 2012. We would then look back and say that it's true that he lost. Does that mean we weren't free at the time in who we voted for, because it would eventually become past? Surely not. It's a contingent matter, and it never ceases to be contingent just because we eventually end up afterward. So when I look forward to it as a possible future (not knowing if it's the actual one), and I claim that it will happen, I'm speaking truly (if it will happen) or falsely (if it won't). That doesn't make the actual outcome necessary, because the factors that influence it are still contingent matters. They just haven't happened yet. But to someone able to detect them, there's a ground for understanding what makes them true. That allows foreknowledge of contingent matters.

So you say that you're granting that foreknowledge isn't predetermination. But I think your view still confuses the two. I don't see how you can claim that it's not contingent simply because it's true unless you think being true means being predetermined in some problematic sense. If contingency is defined the way philosophers typically define it, then all that matters is that there are various possibilities, not that none of them is true. Contingency is defined in such a way that it's compatible with being true, as is evidenced by the contingency of the past. Your claim that the past is necessary is, I think, where your confusion of these two things starts.

As to your last question, I think the answer is that the sense in which the future is indeterminate to me is that the facts that ground its truth are not present in the world at the time of my decision. It's not my current state or the state of the world that determines my act. It's whatever grounds libertarian freedom that explains my act (and different theories would answer that differently). What determines my act, then, is the same thing any open-futurist would say determines my act. But what grounds the truth of future contingents is their happening in the future. The truth-conditions of future contingents depend on the reality of the future, not on events in the present guaranteeing an outcome. I think your argument only makes sense if you move to determinism, the very thing you're saying you're not relying on in your criticism of foreknowledge. You need determinism to undermine libertarian freedom. Mere truth about future contingents is not enough.

I think it's a mistake to think that the truth and falsity of future contingent choices has anything to do with whether those choices are free. You think they have everything to do with each other.

If it is claimed that a choice is made, then the conditions which might be necessary for there to be a choice at all are most certainly pertinent. Is it going to be claimed that any contingency at all is sufficient for humans to have choices which are free? That, of course, would be preposterous.

As philosophers use the term, something is contingent when it doesn't have to be that way.

That's right, but the matter to also consider is how is it that the contingent condition need not be that way. This means that the contingency must be located, so to speak, and the contingency must be considered in terms of what is necessary for there to ever be contingency. For instance, the strictest of physicalist determinists could readily agree to the notion that the universe is contingent, but that contingency could be located at the initial condition when what would be the constituents of this universe first cohered in such a manner that this physical reality was thenceforth determined and utterly determinate. In such a determinist manner of thinking, this reality is contingent and, yet, there is no contingency with regards to any single human action.

You want to make matters of contingency and possibility instead about whether there is determinate truth about them. That, by definition, rules out contingent truths.

No. That is not correct. So, let me clarify. The determinist example above discussed contains contingent truth. According to that determinism, it is a determinate, a settled, a definite matter that a person will do some act; it is, however, a contingent act inasmuch as the entire context of reality is contingent, could have been otherwise. It is just that the contingency in that example is far distant from the human action itself; it might be said that the human action is only indirectly or derivatively contingent rather than contingent itself.

Now, let us consider a different situation. Let us assume that there is no perspective from which some human action could be seen as being already settled, definite, determinate. Even that indeterminate condition could be described as determinate either by positing each and every possible way in which the unsettledness might get settled or by simply noting that the action either will occur or it will not. The point is that there is a determinateness when all of the possibilities are exhaustively taken into account; rather than utterly determinate, it just happens to be a determinate condition in which there is indeterminateness. The determinate condition is amenable to expression as truth, but that truth is, of course, a contingent truth not because it includes indeterminateness but, instead, because either the context in which it occurs is contingent (as per the determinist discussion above) or because there is an even more immediate and more relevant contingency.

Here is where it would be best to revisit some of my remarks concerning contingency. I said that it is indeterminateness that effects contingency. It could just as well be said that contingency indicates indeterminateness -- meaning a condition which is or was to some extent unsettled.

Everything is necessary, by your definition.

No. Not everything is modally necessary by what I have argued. Let me make that more obvious with the discussion that follows.

But since Obama won, it became necessary by your terms. That just seems to me to be a misuse of the philosophical jargon. It's certainly true that he won. There's no longer anything anyone could do to influence that outcome. But it's not necessary just because it happens to be in the past. The events that led up to it could have gone differently, and he might not have won. So it's a contingent truth about the past.

The determinist discussed earlier would say that Obama's victory was contingent, but that contingency was located at the initial conditions of this physical reality rather than anywhere near the election itself. So the story goes. That determinist is just denying all other indeterminateness with regards to reality other than what there might have been prior to initiation of this physical reality.

Whether that determinist is right or not, most people speak as if there is indeterminateness and contingency located more immediate to events such as elections or even many ordinary human actions. If there is indeterminateness or contingency relevantly immediate to some future act by some person, then would it be in any way correct to assert that it is already a determinate matter what that person will do? Of course not. Yet, when it is said that it is true that some person will do some act, what is being expressed and claimed is that it is a determinate matter that the person will do the act. The truth being claimed is not being deemed modally necessary; it is contingent, but it is not merely possible once it is a determinate matter. Similarly, if it is a determinate matter that some person will do some act, then it is also a determinate matter that the person will not do something else, and that means that once it is a determinate matter that the person will do some act, it is thereafter not possible that the person does otherwise -- even though it is a contingent matter that the person does as he does.

So when I look forward to it as a possible future (not knowing if it's the actual one), and I claim that it will happen, I'm speaking truly (if it will happen) or falsely (if it won't).

Consider the following statements:

1. I think such and such person will do some particular thing.
2. It is possibly true that such and such person will do some particular thing.
3. It is likely the case that such and such person will do some particular thing.
4. It is true that such and such person will do some particular thing.
5. It is true that such and such person will do some particular thing and not something else.
6. It is necessarily true that such and such person will do some particular thing.
7. It is a modally necessary truth that such and such person will do some particular thing.
8. It is true that such and such person will or will not do some particular thing.

Examples 1, 2, and 3 are essentially identical. In effect, those three examples are versions of example 8 as well. All four of these statements express the condition as inclusive of indeterminateness or unsettledness. Example 4, which is essentially identical to example 5, effectively denies any indeterminateness to the future at issue; examples 4 and 5 express that the future at issue is wholly determinate by the time that those statements are made even though they are made before the event at issue occurs. Example 6 differs from 4 and 5 inasmuch as 4 and 5 are not even addressing the matter whether the future event at issue follows necessarily from the state of affairs at the time the statement is made. Example 7 is presented simply to emphasize the fact that examples 4 and 5 are not alleging modally necessary truths.

With all that in mind, let us consider you example regarding "speaking truly". If you were to say in 2010 that it is true that Obama will lose the 2012 presidential election, then you would be claiming that in 2010 it is an utterly determinate matter that Obama will lose the 2012 presidential election. This means that you are speaking truly only if it is a determinate matter that the loss will occur. If it is not a determinate matter, then you speak truly only when you say that it is possible Obama will lose, or it is likely Obama will lose. If you say it is likely Obama will lose, or you think Obama will lose, then the 2012 election will determine whether you were correct in 2012; it will not determine whether you spoke truly.

If contingency is defined the way philosophers typically define it, then all that matters is that there are various possibilities, not that none of them is true.

What is it that ultimately distinguishes possibility from modal necessity, and what is it that multiplies possibilities? Indeterminateness. At the very least, indeterminateness is necessary even if not sufficint. Where there is no indeterminateness, there is only determinateness. Where there is only determinateness, alternative possibilities are no longer -- even if the determinateness is only determinate contingently such that there had been other ways in which the determinateness could have been set, if there had been indeterminateness.

Your claim that the past is necessary is, I think, where your confusion of these two things starts.

But, I never claimed the past is necessary. I only claim that the past is determinate, and condition can be determinate without having to be modally necessary.

It's whatever grounds libertarian freedom that explains my act (and different theories would answer that differently). What determines my act, then, is the same thing any open-futurist would say determines my act.

If libertarians were to claim that indeterminateness is necessary for freedom(-to), then that is the issue to address before moving on to notions about what does the determining. If the future contains no relevant indeterminateness despite being contingent (as per the determinist discussion above), then it remains to be established that you do the determining if there has been no indeterminateness subject to being determined during your lifetime. This is where we tie back into the earlier expressions that claim utter determinateness for the future. Sure, we all feel that we are swimming in a reality replete with indeterminateness, but the issue regards those claims, those expressions which put forth the future as being actually devoid of indeterminateness despite our experiences of reality.

To put it another way, it is absolutely irrelevant to the issue at hand regarding determinateness just what it is and when it is that does the determining. That is because the notion of timeless truth is the notion of eternal determinateness, especially when expressed in terms of it being true that some person will do some particular act.

You need determinism to undermine libertarian freedom. Mere truth about future contingents is not enough.

I hope I have made it clearer that determinism is not needed to undermine the notion that indeterminateness is necessary for the freedom-to. What I have emphasized is the expressive incoherence of asserting an utterly determinate future while also claiming that there is a freedom-to which depends on there being indeterminateness with regards to what is to be done.

You don't have to get into theories about what libertarian freedom consists of to show that a particular argument against the compatibility of foreknowledge and freedom fails. The standard historical fatalist argument is:

1. If S knows that P, then P is true.
2. If P is true, then P is necessary.
3. Therefore: if S knows that P, then P is necessary.

There's another way of putting the argument that's clearly invalid (but it's found in all the historical open-future authors). It assumes that the necessity of statement 1 means the necessity of the consequent of statement 1, which conflates the de re/de dicto distinction. So I've put it in a way that's at least logically valid. But the second premise strikes me as pretty clearly false. It assumes the necessity of the past and present but openness of the future, and therefore it's question-begging (in the milder sense, at least; it's not explicitly circular). Someone who doesn't accept that view would deny 2. So this argument should convince no one who doesn't already accept the conclusion. But the way philosophers generally use their terms doesn't fit with 2 anyway. There's a difference between truth and necessity. To point that out and therefore resist the fatalistic argument does not require having any view on what freedom actually consists of. The point is whether truth about future choices requires the necessity of the future, and it simply doesn't. You don't need an account of freedom to resist that argument.

Now you're right that contingency alone doesn't establish the existence of libertarian freedom. I sure agree with that, as a compatibilist. I don't think there is any libertarian freedom, but I'm happy to speak of contingency. But my point is that libertarian freedom is compatible with foreknowledge, and my resistance to the primary argument for contrary views is that the argument is simply fallacious.

To have a positive account of libertarian freedom, you would need to say more. But Augustine, for example, does, and his view is certainly not compatible with efficient-cause determinism. He locates freedom in the final causes of our actions, actions that aren't caused by efficient causes at all but are fully explainable in terms of our reasons for action and therefore fully predictable by God without interfering with our freedom. You could critique his view, but that would be another whole argument. All I was responding to in your comment was the claim that truth about future choices is automatically necessary and therefore means no freedom. I was explaining why I don't think that follows.

Now as the determinateness issue, I see that you want to keep that distinct from the necessity/contingency issue.

If we replace "determinate" with "necessary" in my argument above, we get:

1. If S knows that P, then P is true.
2. If P is true, then P is determinate.
3. Therefore: if S knows that P, then P is determinate.

I'm not sure anyone would resist that, though. So there just isn't a version of this argument in your terms that anyone would find problematic. The question is not whether the future is determinate if it's known. We'd need another premise to get the conclusion that S's knowledge means P isn't a truth about someone's free choice. So the argument is entirely elsewhere from the classical fatalist argument.

So the issue under dispute is whether something's determinateness is compatible with its being the sort of thing that allows for freedom. You might have to get into what kind of freedom is under discussion here, but I actually think much of the disagreement lies in what makes it determinate.

So take an action where I choose in a libertarianly-free way to do something, and I do it. According to libertarianism, I could have done otherwise. There was nothing at the time of my choice that guaranteed that I would do it. It wasn't physically caused by exhaustive efficient-cause explanations in terms of the laws of nature and the past (or present). What explains my action is entirely due to whatever freedom is explained by in your favored account. If it's agent causation, then it's the agent cause. If it's Augustinian final causes, then it's your reasons that explain why you choose it. If it's Epicurus' swerving atom or quantum indeterminacy, then that's your explanation. Whatever the explanation is, that's what gets you to do it, and the account of freedom in question takes it to be a free act.

What I'm saying is that someone ahead of time who managed to get information about your choice would then know it, but as long as that doesn't influence what leads to the action or explains why it happens then there's no threat to freedom. The issue of knowledge doesn't in principle require any interference with what makes the act free.

That's why truth-makers in the future make such a big difference. Cicero's argument against foreknowledge isn't the fatalist argument that there can't be future contingent truths. It's that God couldn't have a mechanism of knowing those truths without predicting it based on deterministic processes. Atemporality gets around that (as does middle knowledge, as does Augustinian final-cause determinism sans efficient-cause determinism. If there's a mechanism for how someone gets the information, then it doesn't mean there's no freedom, because the account of freedom explains how freedom occurs without the explanation for the determinateness interfering with that.

So that's why I think determinateness doesn't remove freedom. It does only in cases where what makes it determinate interferes with the process of what would normally explain the freedom. (And it must interfere to prevent whatever the freedom-inducing processes are from doing what they do to provide freedom. Simply affecting them but retaining those processes doesn't destroy freedom. It just affects the circumstances in which you exercise that freedom.)

Jeremy, thank you for your replies.

Now as the determinateness issue, I see that you want to keep that distinct from the necessity/contingency issue.

Nothing I argue with regards to freedom or truth depends on or asserts modal necessity. I have noted that contingency depends on indeterminateness; contingency indicates indeterminateness; indeterminateness is necessary for contingency, but a contingent status does not have to locate the indeterminateness necessary for the contingency at the matter being described as contingent. In this way, contingency clearly relates to the determinateness issue, but, as I have noted, contingency does not itself establish any sort of human freedom-to whatsoever; this is to say that the insufficiency of contingency pertains to more than just libertarian freedom.

So there just isn't a version of this argument in your terms that anyone would find problematic. The question is not whether the future is determinate if it's known.

I also think someone would be hard pressed to come up with an objection to my points about contingency as it relates to indeterminateness.

So the issue under dispute is whether something's determinateness is compatible with its being the sort of thing that allows for freedom. You might have to get into what kind of freedom is under discussion here, but I actually think much of the disagreement lies in what makes it determinate.

In the blog to which I linked previously, the freedom is described in terms of being free-to as distinguished from being free-from. There it is noted that the experience of seeming to be free-to depends on there seeming to be indeterminateness with regards to the future. It is also noted that the past seems to be utterly determinate.

At no point is it claimed that there is mind-independent indeterminateness with regards to the future, and this is to say that this is not an argument for what is usually referred to as libertarian freedom.

This means that the point of contention is not the matter of what makes the future determinate. And here is why. Something can be made determinate only if it is first indeterminate. On the other hand, something can be determinate without ever having been indeterminate (at least conceivably). If the past not only seems but actually is determinate, and if it is claimed that some future is also determinate, then there is no indeterminateness to be had with regards to that future. And, that is to say that there is nothing between the past and that future which "makes it determinate." All that is relevant is that the futire is determinate.

So take an action where I choose in a libertarianly-free way to do something, and I do it.

To the extent that what I describe is something which some self-described libertarian would embrace, then for you to "choose in a libertarianly-free way" there has to be indeterminateness with regards to the action at issue. If you chose in the past, that means there was indeterminateness with regards to that past act although there is now no such indeterminateness with regards to your past action. You did, after all, bring forth determinateness from the indeterminateness which was at hand (we are, at least, assuming so based on our experiences and for the sake of argument).

According to libertarianism, I could have done otherwise.

The phrase "could have done otherwise" indicates the presumption that there had been indeterminateness.

If it's agent causation ... If it's Augustinian final causes ... If it's Epicurus' swerving atom or quantum indeterminacy, then that's your explanation. Whatever the explanation is, that's what gets you to do it, and the account of freedom in question takes it to be a free act.

Any "account of freedom" which depends on the notion of "could have done otherwise" is a sort of freedom for which indeterminateness is necessary. But, here is the problem. As explained above, any claim which effectively puts forth a determinate future is a claim which denies the relevant indeterminateness which is necessary for the freedom-to at issue.

On the other hand, if there is some notion of there being freedom-to despite there never being any relevant indeterminateness, then not only is the experience of seeming indeterminateness brushed aside but, also, there is the problem of considering whether and how it would be that humans are any more free than are clocks, computers, or any other sort of merely mechanistic devices.

What I'm saying is that someone ahead of time who managed to get information about your choice would then know it

This remark of yours would actually lead to a far more interesting -- and a far more practically relevant -- separate topic. In order to give just a peak into that issue, I will grant that even if there is the very sort of indeterminateness necessary for the freedom-to under discussion, it may well be that the actions on the part of the vast majority of people would be extraordinarily highly probable (virtual certainties) if most people are not especially aware of the scope of the possibilities - the indeterminateness - available to them. Some people can be freer than others -- but not in an Animal Farm sort of way. Please excuse the digression; now back to the topic at hand.

but as long as that doesn't influence what leads to the action or explains why it happens then there's no threat to freedom. The issue of knowledge doesn't in principle require any interference with what makes the act free.

In my first comment, I noted that God's foreknowledge does not cause what occurs. A variation on that same point is that someone else's knowledge, in itself, in no way interferes with or affects another person's actions.

That's why truth-makers in the future make such a big difference.

This issue was dispensed with above in the discussion regarding "what makes it determinate".

God couldn't have a mechanism of knowing those truths without predicting it based on deterministic processes. Atemporality gets around that (as does middle knowledge, as does Augustinian final-cause determinism sans efficient-cause determinism. If there's a mechanism for how someone gets the information, then it doesn't mean there's no freedom, because the account of freedom explains how freedom occurs without the explanation for the determinateness interfering with that.

In all of these approaches, determinateness with regards to the future is claimed and indeterminateness is effectively and essentially precluded. We can grant that people are free-from interference with regards to the knowledge obtained about these determinate futures, and we can grant that people do the actions it is known that they will do. However, being free-from interference is not sufficient for there to be a freedom-to, and freedom-from mereinterference does not even appear to be necessary for a freedom-to.

Such compatibilist notions become especially expressively incoherent when they assert that the future is utterly determinate and then also assert that people are free-to despite this utter determinateness since they can do - or could have done - otherwise. Indeterminateness with regards to (or contingency located with) the person's action event is necessary for this "otherwise" assertion, and the claim for utter determinateness is expressively incompatible with the claim that there is the indeterminateness necessary for doing otherwise.

the freedom is described in terms of being free-to as distinguished from being free-from. There it is noted that the experience of seeming to be free-to depends on there seeming to be indeterminateness with regards to the future. It is also noted that the past seems to be utterly determinate.

Right, but even the compatibilist has an explanation of all this. Freedom requires epistemic possibility, epistemic indeterminateness about the future, and there's no such epistemic possibility about the past, at least with respect to anything I can do now (leaving aside retroactive prayer and time travel, the first of which is actual and the second of which is probably non-actual but seems to me to be metaphysically possible). I would argue that the past is no more or less necessary than the future in any sense except epistemic but that the fact that the arrow of causation is almost always past-to-future leads us to believe falsely that (1) it is always past-to-future and (2) the past has a different modal status than the future. I don't think the past can be changed, but I don't think the future can be changed either. I do think we can influence the future and the past.

This stuff about being made determinate is confusing me. It sounds like you're talking about becoming determinate, which in fact couldn't occur without once being indeterminate, but I was talking about what makes it determinate in the sense of grounding or explaining the determinacy, not what makes it come into being (as if it weren't there before). I'm talking about truth-makers of determinate facts being in the future. This stuff about making it come to be determinate is completely in a different direction. I don't believe in anything like that.

I'm not sure most libertarians are with you on the indeterminateness issue. They insist that it must be up to them for them to be free, and they insist that it not be caused in a deterministic way, but most libertarians don't insist that there be no determinate truth about what they will do. It's exactly such a view that Carneades, Cicero defended, and since then there's been a long strain of indeterminists about causal determinism but determinate-ists about future contingent truths. You think such a view makes no sense, but a large number of libertarians hold their libertarianism in a way that they think it's compatible with that.

The phrase "could have done otherwise" indicates the presumption that there had been indeterminateness.

In the epistemic sense, even compatibilists would accept that. The variety of libertarianism I have in mind would also accept it in one metaphysical sense, if indeterminateness just means modal contingency. But they would not accept it if indeterminateness means no fact of the matter about what will happen.

There is relevant indeterminateness according to each of the three views. The disagreement is over which kind of indeterminateness is relevant. You have one view. Many libertarians have another. Compatibilists have yet a third. But that strikes me as where the disagreement lies.

Jeremy, I hope you had a merry Christmas. I certainly did.

Freedom requires epistemic possibility, epistemic indeterminateness about the future, and there's no such epistemic possibility about the past, at least with respect to anything I can do now ...

The indeterminateness which I have been discussing with regards to freedom-to is firstly a non-epistemic indeterminateness. I would agree that, given such a non-epistemic indeterminateness, awareness about this indeterminateness would most definitely be relevant to the freedom-to at issue.

I would argue that the past is no more or less necessary than the future in any sense except epistemic ... the past has [no] different modal status than the future.

The epistemic possibility to which you refer pertains to the future seeming as if it is at all indeterminate. That aside, I agree "that the past is no more or less necessary than the future". That is why my contention in no way depends upon any notion about the past being in any way "necessary". I discuss this matter of a past necessity here.

This stuff about being made determinate is confusing me. It sounds like you're talking about becoming determinate, which in fact couldn't occur without once being indeterminate, but I was talking about what makes it determinate in the sense of grounding or explaining the determinacy, not what makes it come into being (as if it weren't there before).

What you recognize as the fact of "couldn't occur without once being indeterminate" was addressed as part of the process of disambiguating the phrase "what makes it determinate". You, of course, posit a determinateness (or a determinacy) which is or which obtains without there having been the indeterminateness necessary in order for there to be a transformation into determinateness. I understand that, and it is one of the positions to which the determinateness notion I have been discussing applies. The "sense of grounding or explaining" amounts to -- is identical to -- a description of what occurs, and this would be the case even if there were any indeterminateness that got transformed into determinateness at the occurrence.

The issue I bring out pertains to what differences there are or might be between the descriptions:

(1) a person performs a particular act, and
(2) a person is free-to perform a particular act (or a person freely chooses to perform a particular act).

In order to preserve the sense of freedom to which you referred earlier, statement (1) above should be understood as something along the lines of : being free-from any interference that is effectively controlling or which precludes a freedom-to, a person performs a particular act.

The problem is that such an extensive freedom-from does not itself provide for a freedom-to.

There are, after all, eternalistic views (or perspectives) according to which all events and all times always are, wherein events only seem to occur when viewed from an incomplete (or effectively blinkered) temporal perspective. Human acts are free-from an effectively-controlling interference; it is the humans who do the acts they do without being controlled or made to do the acts they perform, and while humans can be free-from such interference, they are not free-to (they do not freely choose; they are not free-to choose or free-to act) inasmuch as there is never any non-epistemic indeterminateness with regards to their actions.

As I have discussed, the distinction between the freedom-to and the freedom-from is made more stark by the fact that a relevant non-epistemic indeterminateness would itself provide not only for the freedom-to do some act but also for the freedom-from an effectively-controlling interference.

Do compatibilists hold that there is freedom-to (rather than just freedom-from) if there is no non-epistemic indeterminateness? If so, then in what sense does a compatibilist think that there is freedom-to (rather than just freedom-from) if there is no non-epistemic indeterminateness?

Before it is said that a person performs a particular act is identical or equivalent to a person is free-to perform a particular act or a person freely chooses to perform a particular act, surely it is best to take into consideration those conditions posited as necessary for there to be the freedom-to as distinguished from the freedom-from.

The indeterminateness which I have been discussing with regards to freedom-to is firstly a non-epistemic indeterminateness. I would agree that, given such a non-epistemic indeterminateness, awareness about this indeterminateness would most definitely be relevant to the freedom-to at issue.

That sounds like a recipe for the denial of free will, then. Whatever else is true in this debate, the difference between a world in which I have libertarian freedom and a world in which I have compatibilist freedom are indistinguishable from the point of view of the person making choices. I'm more convinced of that than I am of any of the metaphysical claims any of the positions might make. So if freedom requires being aware of some kind of non-epistemic indeterminism, then no one is free. Yet we do seem to be aware of our freedom. Such freedom must, therefore, not require being aware of any non-epistemic indeterminacy.

The "sense of grounding or explaining" amounts to -- is identical to -- a description of what occurs, and this would be the case even if there were any indeterminateness that got transformed into determinateness at the occurrence.

Are you just talking about truth corresponding to reality here? That's what I meant by grounding. There must be some facts that serve as truth-makers for any true statements. On the view that there is ever any indeterminacy, those truth-makers must not exist until the statements are true (which is a hard to fit with what our best physics tells us, since there's no absolute present frame of reference in the 4D manifold, leading me to think we've empirically discovered that the A-theory of time is false).

What I keep seeing is assertions that freedom-to requires indeterminacy. I haven't seen anything I recognize as an argument, though. I understand that this is your view, but repeating the view doesn't convince me of its truth. Since I think the view is false, I'm going to need more than just making the distinctions you've been making. If freedom isn't defined in terms of non-epistemic indeterminacy, then a number of accounts of freedom appear.

My basic thought experiment type comes from Walter Stace. He asks us to consider several cases where we assume determinism is true, and then we discover there are still distinctions to make between someone fasting because of being in the desert with no food vs. fasting for political purposes, lying on the floor because you're tied up and lying on the floor because you enjoy the feel of your carpet, etc. In all such pairs of cases, we would say that in one case there's a freedom not possessed in the other case, and that difference seems important even if determinism is true. In fact, what seems to explain the freedom is something that doesn't have to do with predetermination to begin with.

For incompatibilist libertarian views, what I would say is required for freedom is either agent causation (which means the agent is the cause; nothing is said about whether there is determinacy about what the agent will cause) or that there's an ability to do otherwise (meaning that nothing in the present guarantees the outcome, not that no future choice grounds the truth of what will be chosen).

Michael, thanks for alerting me to what looks like a terrific blog.

I agree with everything Jeremy says!

Suppose I pick two people. One of them says, “Tomorrow, it will rain.” The other says, “Tomorrow, it will not rain.”

At the time these two speak, it is indeterminate whether it will rain or not. That it will rain, or fail to rain, will become determinate in the fullness of time; i.e., with the arrival of tomorrow.

Nevertheless, one of them has already spoken truly.

One must keep in mind the distinction between propositions, which are abstract entities, and events and acts, which are concrete. Propositions may abstractly be true in advance of the events/acts that they describe, but those events and acts happen (become determinate) at a time; i.e., when they happen.

Propositions take their truth from the way the world is. The way the world is, obviously does not take its truth from descriptive propositions about the world, for that would make propositions prescriptive and not descriptive, and that’s incoherent.

With all this in mind, the whole argument to fatalism of any form simply evaporates.

With all respect, Michael, you (and Timothy in our ongoing discussion of Bob the Pancake Saying Guy) are repeatedly committing the modal fallacy, notwithstanding your claim that you are not invoking modal necessity. You are invoking it, whether you believe yourself to be doing so or not.

Your argument logically reduces to the following form:

If it’s true today that tomorrow it will rain, then it must rain tomorrow.

Whereas, of course, the proper formulation is:

If it’s true today that tomorrow it will rain, then it will (not must!) rain tomorrow.

It really is not more complicated than that.

Given quantum indeterminacy and a regularity theory of the laws of nature, causal determinism is false. And logical determinism/fatalism and epistemic (foreknowledge) determinism/fatalism are also false, for invariably one detects in them the modal fallacy. That God knows in advance what you will do does not mean you have to do that thing; it just means that He knows what you will do because you (freely) it, and God’s perfect foreknowledge and your act must (necessarily) match. This does not mean that you must do what you do, because that ascribes necessity to the consequent of the antecedent of the argument and this is plainly invalid.

You (and Timothy in the Bob thread) keep arguing along the lines that while something is “modally possible” it’s not “actually possible.” Timothy’s latest post at the library has the following statement: “It's clearly modally possible, but just as clearly it's not a thing it's possible for him to do.”

The only way this makes sense is to realize that the totality of possible worlds means all logically possible worlds. Physically possible worlds are a subset of logically possible worlds, and the actual world is a subset of physically possible worlds.

At some logically possible worlds pigs fly, just not at our (actual) world. However, at our actual world, any contingent act could have been otherwise, by definition. To say that an act is contingent but could not “really” have been otherwise is logically incoherent.

Jermy has correctly noted (as I have been harping on at The Galilean Library) that one cannot change the past, but one cannot change the present or future, either. One actualizes the past, present and future, and does not change anything at all. When I actualize history by putting on my hat, I made it true at all times – past, present and future – that I put on my hat at some time t. It was physically indeterminate that I would put on my hat before I put it on, but if I put it on then someone who said a thousand years ago, “ x will put on his hat a thousand years hence" spoke truly. And none of this implies fatalism, for the reason that propositions take their truth from the way things were, are and will be, and not the other way around.

Leave a comment

Contact

    The Parablemen are: , , and .

Archives

Archives

Books I'm Reading

Fiction I've Finished Recently

Non-Fiction I've Finished Recently

Books I've Been Referring To

I've Been Listening To

Games I've Been Playing

Other Stuff

    jolly_good_blogger

    thinking blogger
    thinking blogger

    Dr. Seuss Pro

    Search or read the Bible


    Example: John 1 or love one another (ESV)





  • Link Policy
Powered by Movable Type 5.04