I recently mentioned Howard-Snyder's important new paper, "The Logical Problem of Evil: Mackie and Plantinga". Given the importance of the paper, I thought I'd give a rough sketch of the core argument. Plantinga's Free Will Defense (FWD) depends on his claim that there is a possible world at which every creaturely essence suffers from transworld depravity (<>TWD). However, <>TWD depends on a controversial picture of the the distribution of the counterfactuals of freedom to creaturely essences. In particular, it depends on the thesis Howard-Snyder calls Interworld Plenitude, which is (very roughly) the view that while there are an infinite number of creaturely essences and an infinite number of differing bundles of counterfactuals of freedom for each creaturely essence to have, and while each possible bundle is had by one or more essences, the plenitude of essence/bundle pairs is diffused across a large stretch of the space of possible worlds. To be more specific: while each bundle is had by some essences at some possible world or other, not every bundle is had by some essence or other at every possible world. Given Interworld Plenitude, then, (i) there are possible worlds at which every creaturely essence suffers from transworld sanctity, (ii) there are possible worlds at which every creaturely essence suffers from transworld depravity, and (iii) in between these two extremes, each other combination and quantity of essence/bundle pair exists at some possible world or other. If Interworld Plenitude holds, then, TWD is possible, as there will be a possible world at which every creaturely essence suffers from transworld depravity.
However, there is another basic account of the distribution of counterfactuals of freedom, which Howard-Snyder calls Intraworld Plenitude. Very roughly, this account states that a token of each such bundle is had by at least one (and indeed perhaps infinitely many) creaturely essence(s) at every possible world. But if so, then even if infinitely many creaturely essences suffer from transworld depravity at any given possible world, it's also true that at least one (and indeed perhaps infinitely many) creaturely essence(s) enjoy(s) transworld sanctity at every possible world ([]TWS).[1] And if that's right, then at no possible world are God's hands tied: at every possible world, there are creaturely essences God can instantiate that never freely go wrong. And if so, then <>TWD is false, in which case Plantinga's FWD fails to establish the compatibility of God and evil.
Now here's the rub. As Howard-Snyder puts it:
which picture (if either) accurately represents the distribution of counterfactuals of freedom to essences? Each picture is internally consistent; and each is consistent with everything we know or reasonably believe. So which is it? I submit that none of us is in a position to answer that question. We are in no position to tell which picture (if either) is accurate. But in that case, we are in no position to tell whether S [ i.e., []TWS] or D [i.e., <>TWD] is true. And if we are in no position to tell whether S or D is true, then it is no more reasonable for us to believe D than S and, therefore, it is reasonable for us to refrain from believing D, in which case Plantinga’s FWD fails.
Now those who follow the literature on the logical problem of evil know that Howard-Snyder has made this point before (with John Hawthorne). This paper furthers the discussion in at least two important ways: (i) by fleshing out and motivating both Interworld Plenitude and Intraworld Plenitude (and by distinguishing weak and strong versions of both. See the paper for the details), and (ii) by responding (in my view, decisively) to the most important replies to the argument, viz., those from Plantinga and Rowe. Howard-Snyder points out that the core of their replies is the same, and that both replies beg the question by assuming that Interworld Plenitude is true and Intraworld Plenitude is false. As such, Howard-Snyder's (and Hawthorne's) defeater for Plantinga's FWD remains undefeated.
In short, FWD succeeds in showing the compossibility of God and evil only if <>TWD is true. But <>TWD is true only if Interworld Plenitude is true and Intraworld Plenitude is false. But we have no reason to prefer one of these two pictures of the distribution of the counterfactuals of freedom to creaturely essences over the other, in which case we have no reason to think <>TWD is true. And if that's right, Plantinga's FWD fails to show the compossibility of God and evil.
My intuitions sometimes incline me toward a stronger construal of Howard-Snyder's (and Hawthorne's) core claim, and Rasmussen has argued persuasively for the latter. I'm grateful that Howard-Snyder (and Hawthorne) have done the hard work to flesh out the case for the weaker claim, and that Rasmussen has done the hard work to flesh out the case for the stronger construal.
------------------
[1] It's crucial to note that transworld sanctity, like transworld depravity, isn't construed as an essential property of a given essence; rather, both are contingent, world-indexed properties of creaturely essences.
My intuitions sometimes incline me toward a stronger construal of Howard-Snyder's (and Hawthorne's) core claim, and Rasmussen has argued persuasively for the latter. I'm grateful that Howard-Snyder (and Hawthorne) have done the hard work to flesh out the case for the weaker claim, and that Rasmussen has done the hard work to flesh out the case for the stronger construal.
------------------
[1] It's crucial to note that transworld sanctity, like transworld depravity, isn't construed as an essential property of a given essence; rather, both are contingent, world-indexed properties of creaturely essences.
6 comments:
Hi exapologist,
I haven't read any of the papers you've linked to (just your post on the subject), but I have a couple of thoughts.
First, in "Supralapsarianism or O Felix Culpa", Plantinga gives an argument for why he thinks TWD is true. If I recall the argument correctly, it goes like this: for a possible world to be good enough for God to actualize, it has to meet two conditions: (1) God has to exist in it; and (2) it has to have an incarnation and an atonement. Now, if God necessarily exists, then every possible world meets (1). On the other hand, the only possible worlds that satisfy (2) are those worlds that have a fall, for the reason that an incarnation and atonement is necessary is if there is a fall. But any world with a fall is also a world where all the rational, finite creatures (save for beings like angels, but we'll leave them aside, and save for whatever rational, finite creature God incarnates in) suffer original sin, which means they all are guilty for at least one wrong thing. So, TWD is true of the creatures in any world good enough for God to realize. Obviously, one can disagree with (to say the least) (2), but I'm just pointing out that Plantinga has argued for TWD to some extent.
Second, and I think more important for your post, you write:
"In short, FWD succeeds in showing the compossibility of God and evil only if <>TWD is true. But <>TWD is true only if Interworld Plenitude is true and Intraworld Plenitude is false. But we have no reason to prefer one of these two pictures of the distribution of the counterfactuals of freedom to creaturely essences over the other, in which case we have no reason to think <>TWD is true. And if that's right, Plantinga's FWD fails to show the compossibility of God and evil."
It seems to me right that if we don't know whether Interworld or Intraworld Plentitude are true, then we have no reason to think that FWD is true. And from that it follows to show that the FWD fails to show the compossibility of God and evil. But on the other hand, it also seems to me that for the logical PoE to work, we need to know whether Interworld or Intraworld Plentitude are true. But as you said above, we don't know that. So it seems to me that, even if we don't know whether the FWD works, we also don't know whether the logical PoE works, in which case the atheist can't rest on it.
(BTW, this reminds me of a response PvI gave to Plantinga's modal ontological argument. In The Nature of Necessity, Plantinga argues that, since we don't know whether a greatest conceivable being (GCB) is possible or impossible, it's reasonable to believe that it is possible, and so it's reasonable to think that the modal ontological argument works. But in review of that work, PvI points out, rightly, it seems to me, that if you don't know whether a GCB is possible, then you can't reasonably believe an argument that depends on the possibility of the GCB for its conclusion to work. That seems to me structurally similar to the criticism I'm making of the argument in your post.)
Hi Bobcat,
Plantinga's "O Felix Culpa" is an interesting piece, but the line of reasoning you mention seems to have some dialectical deficiencies with respect to the point at issue (to which you nicely gesture).
About the upshot of Howard-Snyder's argument: No you're right: it doesn't justify that God and evil are not compossible; it just undercuts Plantinga's case that they are compossible. Perhaps it's worth noting, though, that the papers by Rasmussen, Pruss, Schellenberg, et al. I mentioned in my subsequent post argue for the stronger claim.
Hi exapologist,
I see the links to the papers by Rasmussen and Schellenberg, but where is the paper/post by Pruss?
Rob
Hi Bobcat,
A draft of Pruss's paper can be found here.
Best!
EA
Hi, Bobcat
Regarding what an atheist might do (or generally a non-theist who denies the existence of the specific kind of entity the LPoE is against), as long as she accepts at least libertarian free will, that there are essences, and counterfactuals of freedom of the kind Plantinga proposes (plus perhaps some other hypotheses), she shouldn't rest on the LPoE as long as she also accepts Howard-Snyder's arguments.
However, an atheist might be, say, a compatibilist, and accept Howard-Snyder's argument under the assumption (made just for the sake of the argument) that libertarianism is true (plus perhaps other assumptions she doesn't actually accept).
In cases like that, it seems to me that accepting H-S's argument under those conditions should not significantly affect the atheist's assessment of the LPoE, as long as she assigns a sufficiently low probability to the assumptions she's granting for the sake of the argument.
Thanks EA, and thanks Angra.
Post a Comment