ManM

Thomas Metcalf,
Would an omnipotent being have the ability to contradict its own nature?

bd-from-kg

Thomas Metcalf:

OK, I’ve finally had time to read your entire paper. Although I have some other problems with it, I want to concentrate on the “freely” problem.

The reason that God cannot cause someone other than Himself to “freely do X” (if indeed He cannot) is that an intrinsic part of the concept of “acting freely” in any sense is that the act in question cannot be caused by anyone or anything external to the agent. Thus God cannot cause someone (other than Himself) to act freely for the simple reason that it’s not possible for an act to be both caused by an external agent and not caused by an external agent.

Now of course there are many senses in which X might be said to be a cause of Y. The most general sense is that Y would not have occurred had it not been for X, other things being equal. (I’m trying to be general enough here to allow X to be an event, a state of affairs, or whatever.) But this allows for any number of things to be causes of the same event. In this general sense God might be said to cause someone to freely do Y simply by creating conditions such that he freely chooses to do Y. (Both God and the agent would then have caused Y.) But this is what you call in your paper “weak actualization”, which you explain is not the kind of causation you have in mind. An omnipotent being, you say, should be able to directly cause the state of affairs in question (namely, in this case, the state of affairs that someone has freely done Y). Whatever this means exactly, you clearly don’t consider X’s doing Y to be free if God “directly” brings about (or “strongly actualizes”) the state of affairs “X does Y” (unless X is God, of course), and “weak actualization” isn’t good enough for omnipotence. [More about “strong/weak actualization” later,]

You seem to be arguing that regardless, “someone freely learning” is a bring-out-able state of affairs, so that an omnipotent being should be able to bring it about. But this is a rather special kind of bring-out-able state of affairs because by definition only the “someone” in question can bring it about. Thus “X brings about the state of affairs ‘for some Z, Z freely does Y’” is logically equivalent to “X brings about the state of affairs ‘X freely does Y”. which in turn is logically equivalent to “X freely does Y”. Hence “God cannot bring about the state of affairs ‘someone freely does Y’” is logically equivalent to “God cannot freely do Y”. But the whole point of S-omnipotence was to avoid this kind of problem – i.e., to avoid saying that because God cannot Himself do certain things because they imply a “defect” incompatible with His perfect nature, He is therefore not omnipotent. So it seems to me that all you’ve done (at most) is to find a subtle flaw in the definition of S-omnipotence.

There’s another problem, but explaining it requires a short discussion of the nature of free will. As I pointed out in an earlier post, the very notion of “acting” seems to involve causation: X can only be properly said to have done Y if he caused Y. If the cause is external to X, or if there is no cause, then Y is not an act of X but merely something that happened to X. And to say that X caused Y means (at least) that given X (i.e., X’s precise state at the time) Y necessarily occurred – i.e., that “Given X, necessarily Y” is true. But the notion of libertarian free will (LFW) requires that X “really could”, given the precise state he was in at the time, have done something other than Y – i.e., that “Given X, necessarily Y” is false. Other versions allow that X could not have done anything but Y after he willed Y, but that given his precise state at the time he could have willed something other than Y. Neither this nor any similar change in the “moment of truth” where the act becomes inevitable changes anything. The point is always that at this “moment of truth” something happens that determines what X will do, and X cannot properly be said to have “done” Y unless he caused the “something” to happen, whereas LFW requires that something else could have happened instead. These requirements are logically incompatible, so the notion of LFW is logically incoherent. Thus the notion of someone “freely” doing something is only intelligible in terms of “compatibilist” freedom.

Whether or not you agree with this, you claim that your argument “works” under either libertarian or compatibilist conceptions of freedom, so it’s relevant to examine whether it does indeed work under the compatibilist conception.

Now we’re ready to discuss the problem. It seems clear that God must exist outside time, or at any rate this space-time continuum’s time, since He created this STC. So (aside from the original creation event) God’s “actualizing a state of affairs” in this STC would consist of changing all or part of the STC. It doesn’t even make sense to ask whether He does so “all at once” or (for example) makes a change at one point and allows the effects to “ripple forward” in time. An intervention is a change in the STC, and there’s really nothing more that can be said. Now let’s suppose that what He does is to change the STC from a given point in time forward in such a way that aside from this one point all of the usual natural laws hold and (as part of the “new” version of the STC) X freely does Y in the compatibilist sense. This clearly makes sense. Among other things, compatibilist freedom allows for the possibility that the world is deterministic (or deterministic at all but one point or a discrete set of points) and whatever conditions one specifies for an action to be “free” in this sense, they cannot include conditions that do not allow God to be fully “in control” of the course of events. So God can clearly arrange that the conditions hold for X’s doing Y to be “free”.

But does this constitute “weak actualization” or “strong actualization”? I really don’t see how there can be any difference between the two here. (The supposed difference seems to hinge on the existence of LFW.) God has simply actualized a change in the STC such that, in the new version, X freely does Y. In what other sense is it possible to imagine God “actualizing” a state of affairs in this STC? What would constitute actualizing the requisite state of affairs in a “stronger” sense? It seems to me that this kind of actualization of a state of affairs is as “strong” as it gets.

So it would seem that, under a compatibilist conception of freedom (which I consider the only coherent one), God can bring about the state of affairs “someone freely does Y” for any Y that it is logically possible for someone to freely do. And He can do so in the strongest possible sense. So this kind of example does not represent a constraint on God’s omnipotence.

Thomas Metcalf

Originally posted by Normal :

Okay. And God still isn't learning, so God isn't omnipotent. I'm not sure how you think this preserves compatibility; all you've done is reiterate that God doesn't learn. I agree.

Okay. You say:

"If god necessarily exists, then it is logically incoherent for him to end his existence."

And I assume you mean this to show that God, as an omnipotent being, need not be able to end his own existence, because that wouldn't be logically coherent. I agree that it wouldn't be logically possible for God to end his own existence, but for someone to end her own existence is perfectly logically coherent.

Imagine McEar, the being who can only scratch his ear. It's logically impossible for him to do anything else. Suppose you told me he's not omnipotent, because he can't scratch his nose. I could reply:

"If McEar is necessarily only able to scratch his ear, it is logically incoherent for him to scratch his nose."

If your analogous statement above saves God's omnipotence, then my statement just above saves McEar's omnipotence. But no one wants that result. McEar is definitely not omnipotent.

Now, you say that existing and scratching one's ear are different species of concepts. Sure. But ending one's existence and scratching one's nose (or ear, or whatever) are both actions, and an omnipotent being ought to be able to do both of them.

A living person, at least as a necessary condition. I have no doubt that I could cause myself to cease to exist if I tried.

Thomas Metcalf

Originally posted by ManM :

I'm not sure what that means, although I have a suspicion about where you're going with this. I'll say that it can't. No being can contradict its own nature, if that means something like McEar scratching his nose, when he's only able to scratch his ear and that's part of his nature.

ManM

Perhaps "contradict" was a poor choice for a word. My nature currently includes the ability to point at an object with my left pinky finger. However, I also have the ability to remove my left pinky (ouch). Such an action would essentially change the set of my abilities. In other words, I can change an attribute of myself that would modify the set of my abilities. Would an omnipotent being be able to do likewise?

Wyz_sub10

My take is that "its own nature", and what would constitute a contradiction, would indicate whether it was omnipotent to begin with.

If "it's own nature" was blue in colour (so it could not be "not blue"), then I would argue this being was not truly omnipotent.

If "it's own nature" was "all powerful", so it could not stop being "all powerful", then it could not contradict it's own nature.

But in the second case, the definition, IMO, demonstrates the impossiblity of an omnipotent being. Most of Thomas' example seem to illustrate the impossible state of being omnipotent and omniscient.

Thomas Metcalf

Originally posted by bd-from-kg :

Right.

I'm not inclined to rate this such a small flaw. Personally, I think T-omnipotence is just as good as S-omnipotence; I think an omnipotent being really should be able to perform any logically possible task. S-omnipotence was proposed to correct it, but I've shown that any task that precludes T-omnipotence can be redescribed as a state of affairs that precludes S-omnipotence.

The obvious thing to request now is a new definition of omnipotence. And I'm quite sure that no sensible one will take care of the problem with all those "Some person freely does task T" states of affairs. I'm not sure you're correct with your derivation of a logical equivalency, but even if you are, it seems to help my side just as much, by forming a very strong tie between certain tasks in T-omnipotence and isomorphic states of affairs in S-omnipotence.

Now, you have two more critiques, one relative to LFW and one relative to CFW. I will say that my argument seems to work if LFW were true, so to say LFW is incoherent is still to allow that if it were true, my argument would be sound. The fact that most apologists accept LFW means my argument has genuinely important consequences.

On to CFW. I'm going to make my own brand of CFW depend upon a lack of distinct agents' causation. Let's partially analyze (i.e., only give a necessary condition for) "freely" with CFW as follows:

(s)(t) (s was brought about freely by a at t --> it is not the case that: (there is some agent b such that if b did not choose for s to happen at t, s would not have happened at t), and (b != a) )

s ranges over states of affairs, and t over points in time.

If God puts me in a situation in which I learn some fact F (and I wouldn't have otherwise been in that situation), then I didn't CFW choose to learn fact F, because if God hadn't chosen to put me in that situation in which I learned F, I wouldn't have learned F right then.

But if I put myself in a situation in which I learn F, then I did (or at least may have) CFW chosen to learn F, because if I hadn't chosen to put myself in that situation in which I learned F, I wouldn't have learned F right then, but I am identical to myself so the second conjunct of the negation in the consequent isn't satisfied.

How's that?

I have another pair of questions: Does it make sense to say that sometimes, people freely learn some facts? And if God decided to make me learn some fact, and I therefore learned it, I wouldn't have learned it freely, right?

Thomas Metcalf

Originally posted by ManM :

I'd say yes. If a being is omnipotent, it can modify what abilities it has, because the state of affairs "some person freely modifies its abilities" is logically possible and bring-about-able.

Normal

God doesn't forget either, but his creation forgets. God doesn't commit evil, but his creation does. God doesn't eat, but his creation eats. All are signs of deficinacy.

It is no more an action then "committing evil" is an action a perfect omnipotent god would not be able to perform, in that it would contradict it's own nature.

You define yourself as "a living person", but that is like defining yourself as "a person with long hair". If you shave your head, under these terms, you "commit suicide", but all you are doing is ending the existance of yourself with long hair. You can cause yourself to become not-alive, that is without question, but to cause yourself to "cease to exist" is another thing.

rainbow walking

rw: You are, in effect, accusing this being of not being able to learn because he already knows everything.


thomas: Yes, just like a man in a bakery cannot eat donuts without leaving the bakery after all the donuts have been eaten, and a woman permanently locked in a room with no books cannot perform "to read a book." I dealt with all this in the paper.

rw: Yes, you made availability an important constraint on omnipotence. It just doesn't follow that un-availability equates to in-ability, especially when the task requires one avail oneself of something whose availability has already been exhausted. While it is trivially true that such a god could not perform the task of learning, whereas you can, it doesn't render omnipotence/omniscience incoherent. It simply renders learning logically unnecessary. It certainly doesn't make you more powerful or knowledgable than a god, so it doesn't appear to be a devastating reduction to his abilities.



rw: Another unsupported, but implied assumption here is that such a being acquired his omniscience via learning in the first place.


thomas: Nowhere do I assert or imply this.

rw: No? Your T-omnipotence implies "learning" as a valid test, but if such a being created all knowledge, learning becomes a moot point. Such a being, if he so desired, could simply create additional knowledge and then proceed to learn it in the conventional way that you and I acquire our knowledge.



rw: And finally, the "task" of learning is not generally accomplished via "power" but intellect anyway.


thomas: The definitions of omnipotent that I offered would all require the ability to learn. Choose a different definition of "omnipotent" if you don't like them.

rw: Then your definition needs more substantiation. How does the ability to perform the task "to learn", an ability that requires anything but intellect and available knowledge?



rw: For example, is oxygen logically necessary to sustain human life in any logically possible world?




thomas: Only if human life is defined to require oxygen.

rw: The point being, have you established any of the tasks in your argument as being logically necessary to render omnipotence/omniscience less incoherent?



rw: Thus a theist is likely to argue that eating and counter-feiting are not "logically necessary" to sustain omnipotence in any logically possible world...especially when you posit the additional attribute of immateriality.




thomas: Huh? What does this objection mean? Why don't I say that for any task T I can't perform, it's not logically necessary to sustain omnipotence in any logically possible world, and therefore I'm omnipotent?

rw: Because that's not the point of my criticism here. I can say there is no logically necessary task T an omnipotent being can't perform. This puts the burden back on you to establish the logical necessity and not just logical possibility.



rw: For such a being to be moral he must have an availability of choices that include evil and malevolent alternatives, where he chooses, instead, good and beneficial alternatives. Further, to be morally perfect, he must consistently choose good and beneficial options.


thomas: By the definition of capability I offered, God is incapable of doing evil. Choose a different definition of "capability" if you want. God is incapable of having an evil will, and I am, and moral perfection doesn't seem to me to require the ability to have an evil will, only the ability, if you're correct, to do evil actions.

rw: Agreed, but an evil action also requires a choice prior to the act...yes?