Perfect and Best
May 16, 2010 — 10:40

Author: Michael Almeida  Category: Uncategorized  Comments: 16

Lots of people believe that a perfect being must actualize the best possible world if:
(i) there is a best possible world and
(ii) the best possible world is actualizable.
To paraphrase David Johnson, give me a little time and I’ll give you an argument with no discernible flaw that this common belief is false. Assume for simplicity that, necessarily, the best *possible* world is also the best *actualizable* world. I make no assumptions about the nature of free action. The argument works as well under the assumption of compatiblism (determinism) as it does under the assumption of libertarianism. More precisely, the widely believed claim is in P.
P. Necessarily, if there is a best possible world W, an (essentially) morally perfect, omnipotent, omniscient, necessarily existing being actualizes W.
P is not merely false, it’s *impossible*. It is impossible that a perfect being is such that he necessarily actualizes the best possible world. It is impossible that *any being* is such that he necessarily actualizes the best possible world. Here’s the proof that P is (necessarily) false.
1. Necessarily, a perfect being actualizes the best possible world. Assume for Reductio
2. W is the best possible world. Assumption
3. A perfect being exists. Assumption
4. W includes a great deal of natural and moral value. From def. of ‘best world’
5. W is the only possible world. From 1,2,3
6. Everything possible is actual in W. From 5
7. W is a fatalistic world. From 6
8. No moral agent is libertarian or compatibilist free in fatalistic worlds. Fact
9. No moral agent is free in W. From 7, 8
10. There is no moral value in W. From 9
11. W is not the best possible world. From 10, 4. *Contradiction* 11,2
12. ∴ It is not necessary that a perfect being actualizes the best possible world. From 11,2
That concludes the reductio ad absurdum. It is easy to see that P is not merely false, but necessarily false. So, if there is a best possible world, a perfect being *could not* be such that necessarily he actualizes that world. So, the commonly held belief is false. It is P’ that is true, not P.
P’. *If there is a best possible world and that world is actualizable, it is possible that a perfect being does not actualize it.*

Comments:
  • Mathis

    It’s late where I’m writing, so excuse me if this sounds not very well thought-out (and by that I mean “incoherent”).
    – P begins with “Necessarily[…]”, so it follows under S5 that if P is even possibly false, it’s necessarily false.
    – P seems to say that “If a perfect world is possible then it’s necessary that God exists”
    – This argument against P only works if God exists (see assumption #3), so an atheist might say, “I will run a similar argument, but I will use (1),(2) and P and will get that (3) is false, yay for me!” so it doesn’t look a theist is any better off with this argument, but how would that argument work?
    (A perfect world is possible & P & this isn’t a perfect world) -> (God does not exist) but as I said, (P & a perfect world is possible) entails that God exists

    May 16, 2010 — 16:39
  • Mike Almeida

    Hi Mathis,
    Right, there is that reply on behalf of the atheist. But my argument really concerns the properties a best possible world could have. A best possible world could not be the only possible world. Were it the only possible world, it would not have the properties that are characteristic of best possible worlds: viz. lots of natural and moral good. This is true whether or not God exists. So the argument really concerns the nature of best possible worlds. It has nothing to do with the nature of God. This is why I say in the post,
    It is impossible that a perfect being is such that he necessarily actualizes the best possible world. It is impossible that any being is such that he necessarily actualizes the best possible world.
    The problem is the same whether God exists or near-Gods exist, no nothing like a God exists.

    May 16, 2010 — 17:16
  • May 16, 2010 — 18:40
  • Hello Mike,
    Why could there not be equally *best* possible worlds?
    God Bless,
    NPT

    May 16, 2010 — 21:07
  • John Pittard

    Perhaps I’m not up on my philosophy of free will, but I’m not sure I grasp why you are entitled to assert (8) while maintaining that you make no assumptions on the nature of free action. I understand that a fatalistic world (which is the same as a necessitarian world, right?) is not the same as a determinist world, but wouldn’t at least some arguments for the compatibility of freedom and determinism also show, if they are sound, the compatibility of freedom and necessitarianism?
    Second, for (5) to follow I think you need to replace (3) with a stronger assumption, namely: *necessarily* a perfect being exists. Of course the common conception of a perfect being is that such a being necessarily exists and is necessarily perfect. So maybe you took the necessity to be implied. But this conception of a perfect being may very well be mistaken. If perfection requires being morally praiseworthy for doing some specific action A, and if a morally praiseworthy act must be free in the libertarian sense, then it may be impossible for a being to be necessarily perfect. For God’s necessary perfection would imply that God necessarily does A and is morally praiseworthy for doing A, but libertarianism implies that any being who necessarily does A is not morally praiseworthy for doing A. So if moral perfection requires any specific morally praiseworthy actions, then necessary perfection would seem to be incoherent. While we could still insist in this case that a perfect being would exist necessarily, the perfection itself would have to be contingent. But if God’s perfection could be contingent, you can’t infer (5) and your argument doesn’t go through. So I think you need to replace (3) with (3*): necessarily, a perfect being exists. And I’m suggesting that one could plausibly contest (3*) even while affirming theism. In fact, the theist might have use for your argument as a reductio against (3*). For (1) is arguably more plausible than (3*) even if we accept theism! At least I am inclined to think so.

    May 16, 2010 — 22:13
  • Mike Almeida

    but wouldn’t at least some arguments for the compatibility of freedom and determinism also show, if they are sound, the compatibility of freedom and necessitarianism?
    No, when it comes to compatibilisim, necessitarian worlds are worse than fatalistic worlds which in turn are worse than determnistic worlds. Compatibilists and incompatibilists disagree about whether deterministic worlds are fatalistic. If they are, then I can’t see how we get any free action at all. Since the world I’ve described is in fact necessitarian (and not merely fatalistic) there won’t be any free action in it.
    Second, for (5) to follow I think you need to replace (3) with a stronger assumption, namely: *necessarily* a perfect being exists.
    Perfect beings necessarily exist, so I don’t need the additional premise.
    but libertarianism implies that any being who necessarily does A is not morally praiseworthy for doing A.
    Libetarianism doesn’t imply that. It says nothing about the conditions under which an agent is praiseworthy or not. It’s a theory about free will, not a theory about appraisability.

    May 17, 2010 — 7:52
  • Mike Almeida

    Why could there not be equally *best* possible worlds?
    Hi Nathanael,
    Right, there might be. But I’m working with the assumption that there is a best world in the sense of “better than any other world”, not in the sense of “at least as good as any other world”. I concede as well that the best possible world is also the best actualizable world. So I’ve conceded about as much as I can to those who contend that God must actualize the best world.

    May 17, 2010 — 7:56
  • Alan Rhoda

    Mike,
    I’m worried about (4) and (8).
    Regarding (4), I don’t see why we should think it’s part of the definition of ‘best world’ that it includes any specific kind of objective value, whether moral or natural. All that the notion of ‘best world’ seems to imply is more overall objective value, of whatever sort, than any other world. Moreover, since the notion of a ‘best world’ precludes incommensurable objective values, moral value cannot be sui generis in such a world unless objective value just is moral value.
    (8) strikes me as clearly false. Even the strongest forms of fatalism are consistent with versions of compatibilism that reject the principle of alternative possibilities.

    May 18, 2010 — 11:09
  • Alan:
    I worried about (4), too, but Mike doesn’t say it’s part of the definition. It’s just a fact that the best world, if there is one, includes moral value. Why? Well, our world includes moral value, and it is plausible that any world better than ours will also include it.
    As for (8), it’s not obvious that compatibilism is compatible with modal fatalism. For instance, suppose that our compatibilist version of the principle of alternate possibilities says that I freely did A only if had I wanted not to do A, I would have been able not to do A. But on modal fatalism, this is a per impossibile conditional, and those are problematic. So one might qualify the compatibilism by requiring the logical possibility of wanting otherwise.
    One might also supplement the argument with this principle: If all of x’s actions are intentionally determined by an agent y, then x is not free. But in the scenario in question, the perfect being intentionally actualizes the best world, and all the actions of every other agent are constitutive parts of the best world, presumably, and hence are intentionally determined by God.

    May 18, 2010 — 11:32
  • Alan Rhoda

    Alex,
    You’re right re (8) that some versions of compatibilism are inconsistent with the modal-cum-theological fatalism under discussion. But some extant versions of compatibilism remain clearly consistent with that sort of fatalism. I’m thinking specifically of Frankfurt here. Not only does he not reject the need for any possibility of doing (or wanting) otherwise, but he also rejects the idea that intentional determination (or even manipulation) by other beings precludes free will and moral responsibility.

    May 18, 2010 — 12:41
  • Sure, there are such views. But aren’t they false?

    May 18, 2010 — 14:22
  • Mike Almeida

    Not only does he not reject the need for any possibility of doing (or wanting) otherwise, but he also rejects the idea that intentional determination (or even manipulation) by other beings precludes free will and moral responsibility.
    Lots of views seem to be mixed together here. Frankfurt’s examples show (conceding for the moment, what I don’t believe, that they actually work) that moral responsibility (not freedom) does not require alternative possibilities. His examples do not show (or aim to show) that genuine freedom does not require alternative possibilities. That aside, I’m not sure what any of this his to do with moral value. Nothing Frankfurt shows has anything to do with the moral value of actions. Nothing Frankfurt says shows that the moral value of actions doesn’t require alternative possibilities. But of course I’m concerned in the argument with morally valuable actions.
    On a different score, the debate between compatibilists and libertarians is the debate about whether deterministic worlds are fatalistic. Compatibilists want to say that determinism does not entail that, for instance, I could not have raised my hand, if in fact I did not raise my hand. That is, determinism does not entail, as fatalism does, that whatever I do I necessarily do. So, if fatalism is true, compatibilism is false: if I did not raise my hand, I necessarily didn’t. If that doesn’t count as being unfree, I don’t know what does. What’s worse is that, if there is one world, it is not merely fatalistic, it is necessitarian! Certainly, there’s no credible view that makes freedom compatible with necessitarianism.
    For what it’s worth, do Frankfurt examples show that moral responsibility is compatible with fatalism? Not by a long shot. It would have to be true in the Frankfurt case, not merely that I decide to do what is right, but that I necessarily decide to do the right thing. That’s clearly not compatible with moral responsibility.
    I do assume that the best possible world would include lots of moral value. This is the kind of thing that would be denied not because it is remotely plausible to deny it, but because any philosophical view can be questioned for any old reason. But there’s another side to that coin. Any implausible objection can be prudently ignored. 🙂

    May 18, 2010 — 15:38
  • Alan Rhoda

    Alex,
    I agree with you that such views are false, but that they have been defended by serious thinkers means that (8) stands in need of argument. That’s all I’m claiming.

    May 18, 2010 — 15:42
  • Alan Rhoda

    His examples do not show (or aim to show) that genuine freedom does not require alternative possibilities.
    I never claimed that Frankfurt’s examples were intended to show this. Frankfurt’s positive views about the nature of free will are based on his analysis of persons, not his arguments against PAP.
    The debate between compatibilists and libertarians is the debate about whether deterministic worlds are fatalistic.
    That’s a highly idiosyncratic characterization of the compatibilist / incompatibilist debate. Historically, ‘fatalism’ is simply a denial of future contingency. It is the claim that whatever happens in the future is now-unpreventable or causally necessary.
    Compatibilists want to say that determinism does not entail that, for instance, I could not have raised my hand, if in fact I did not raise my hand.
    Which compatibilists? Classical compatibilists (those who accept PAP) might way to say this, but there is nothing that requires compatibilists as such to say this. New compatibilists (those who reject PAP) in general wouldn’t.

    May 18, 2010 — 16:04
  • Mike Almeida

    I never claimed that Frankfurt’s examples were intended to show this. Frankfurt’s positive views about the nature of free will are based on his analysis of persons, not his arguments against PAP.
    And this view informs the “new” compatibilists? There’s certainly not much new about the all the hierarchy stuff. In any case, I’ve never found that approach plausible.
    Historically, ‘fatalism’ is simply a denial of future contingency. It is the claim that whatever happens in the future is now-unpreventable or causally necessary.
    You might want to take a look at M. Bernstein on Fatalism, maybe the only book-length work on the topic worth studying. By ‘fatalism’ I mean ‘metaphysical fatalism’.
    Of course, as I’ve said, if there’s just a single world, then logical fatalism is true, and there is no free action compatible with that. Well, maybe the “really, really new” compatibilists will take that on. 🙂

    May 18, 2010 — 17:03
  • Doesn’t Frankfurt simply commit the error of thinking that the higher the order of a desire, the more the desire is reflective of who one is? That principle is surely mistaken. Some higher order desires are in no way reflective of who one is. They have no special authority for determining things of moral significance to the person. We can easily imagine someone of a good and settled first-order character, who upon reading a moral philosopher comes to have all sorts of second-order desires, without significantly changing in “who she is”, and if she is truly convinced by this moral philosopher, these second-order desires can be quite settled.
    In fact, this is enough to generate a counterexample to a hierarchalist view of freedom. Sally is a loyal friend, of strong character, who leads an apparently flourishing life, though little if any of her life is covered by second-order desires. (I’ve heard this said about Richard Gale, and Richard agreed with this assessment, and seemed to like the idea. The fact that he liked the idea, though, seems indicative of a higher order desire.) She makes great sacrifices for her friends, however, and she is a philosopher–a logician, to be precise. One day, she branches out and starts reading Aristotle and Aquinas. Never having previously reflected on it, she comes to believe that it’s valuable having various first-order desires, like the desire to make sacrifices for her friends. Completely as a consequence of the reading, she forms some fairly abstract, but quite settled (the view was intellectually quite compelling), desires to have such first-order desires. But this does not change her life in any significant way, because she already had the requisite first-order desires, and she is still moved by them in the same way. She just has some added beliefs, which have triggered some added desires.
    On Frankfurt’s view, she became free after reading Aristotle and Aquinas. (In this case, I guess, the truth set her free. But philosophical falsehood could also have set her free.) But I think this is mistaken. The kind of person I am describing does not change significantly as a person after acquiring the second-order desires, and her sacrifices for her friends were not free earlier if and only if they are not free now. We can imagine her later coming to be convinced that there is no viable concept of proper function, and almost immediately losing all her second-order desires to have the right first-order desires–without this in any significant way affecting her character.

    May 19, 2010 — 10:23