Monthly Archives: February 2012

Pascal’s Wager and Newcomb’s Box

When I teach philosophy of religion to undergraduates, they consistently respond positively to Pascal’s Wager. I suspect it appeals to the practical, self-interested side of most young people, and that’s clearly what it’s designed to do. Pascal’s Wager is an argument in favor of belief in God on the basis of practical reason. Assuming that the theistic arguments are not rationally compelling, Pascal argues that the agnostic ought to believe in God for pragmatic, rather than evidential, reasons. In other words, one minimizes loss and maximizes gain by believing in God. After all, if we believe in God and are right we gain eternal reward. If we believe in God and are wrong, we ‘break even’ as it were. If we don’t believe in God and are right, we also ‘break even.’ But if we don’t believe in God and are wrong, we lose eternal reward and reap eternal damnation.

There are numerous arguments against the Wager, but most of them involve an appeal to considerations external to the Wager itself. I’m thinking specifically of the “other gods” challenge, in which the objector recalculates the odds on the basis of religious options other than atheism or Christian theism. While such an approach is interesting, I’m inclined to think that something like atheistic naturalism or something like Christian theism just are the live options, philosophically speaking. Now, I don’t have time to justify that assertion, but if the current state of the field is any indication, I think it’s a safe assumption. So let’s grant Pascal the point that these are the only live options available to us. How ought we to think about the Wager from the standpoint of rational decision-making? It occurs to me that we might usefully compare the Wager to another classic rational decision-making problem.

In my critical thinking course last semester, we had some fun with a thought experiment called “Newcomb’s Box.” This an interesting test case for rational decision-making models because philosophers, and students, disagree about what the most rational course of action is. It occurs to me that there are some similarities between Newcomb’s Box and Pascal’s Wager. But first, let’s outline Newcomb’s famous problem and then compare and contrast the two. Newcomb’s Box has various formulations. The one below is adapted from  Introduction to Philosophy, John Perry et al (Oxford University Press, 2010) “Newcomb’s Problem” pp 818 – 19.

Let’s say you’re participating in an experiement designed by a psychologist with a reputation for being brilliant and well-funded. On a table are two boxes. One of them, labeled A, is transparent; in it you can see an enormous pile of $100 bills. The other, labeled B, is opaque. She tells you that there is $10,000 in transparent box A and that in box B there is either $1,000,000 or nothing. She tells you that she is going to give you a choice between:

Taking just what is in box B.

Taking what is in both boxes.

She then informs you that she has made a sophisticated profile of your personality and character traits. On the basis of your profile, she made a prediction about what choice you would make, and she decided what to put in Box B on the basis of this prediction:

If she predicted you would take both, she put nothing in Box B.

If she predicted you would take only Box B, she put $1,000,000 in it.

At this point you ask her how accurate her predictions have been. She says that 1,000 participants have been given the choice, and she has only been wrong once. In all other cases, participants who chose both boxes got only $10,000, whereas those who chose only Box B got $1,000,000. You now have to choose.

In the case in which she was wrong, the participant either received $1,010,000 or nothing depending on whether s/he chose both boxes or Box B only. So if you choose both boxes and she’s right, you get $10,000. If you choose both boxes and she is wrong, you get $1,010,000. If you choose Box B only and she is right, you get $1,000,000. If you choose Box B only and she is wrong, you get nothing. These are the logical possibilities. Granted, not all outcomes are equally probable. But the probability of your getting $10,000 if you take both boxes is 1. So it’s the ‘safe’ bet. Most of my students recognized this and opted for both boxes.

Regarding the Wager, my students’ intuitions were that believing in God is the safe bet. However, thinking about both Newcomb’s Box and Pascal’s Wager, it seems to me that believing in God is analogous to ‘Box B only.’ You’re wagering on a big pay-off by foregoing the ‘sure bet.’ Now even assuming that the odds of that pay-off are very high, they aren’t 1. In fact, the odds of ‘Box B only’ paying off are significantly higher than the odds Pascal places on the existence of God (he estimates these to be no better than .5). Yet most of my students thought that ‘Box B only’ was a bad risk. However, when presented with the Wager, most thought that theism was a better bet than atheism.

Perhaps this has to do with the way the Wager is usually presented. It’s typically presented as having no cost up front. You simply believe or don’t believe, and believing is in your best interests. However, there is a cost involved in belief. Religious belief of the kind Pascal has in mind isn’t simply intellectual assent. It also involves certain commitments, the cultivation of certain virtues, and foregoing certain temporal pleasures (sleeping in on Sunday mornings and casual sex seem to rate pretty high among university students). Perhaps it would be nice if we could have our cake and eat it too (something analogous to taking the $10,000 and hoping there’s $1,000,000 in the other box), but that’s not the way the Wager works. You either opt for the religious life or you don’t. Agnosticism or nominal religiosity are not live options; they amount to practical atheism.

Since there is a cost involved in opting for the religious life (various spiritual disciplines and foregoing various temporal pleasures) and the odds in favor of theism (and the hypothetical pay-off) generously construed are no better than .5, should we reconsider the rationality of opting for theism on the basis of the Wager? Should we reason more on analogy with Newcomb’s Box? I’m inclined to think we should and I’m puzzled that people’s intuitions differ so widely in the two cases. I’d be interested in doing a more scientific study to see if the statistics bear out my informal experience. But then I’d be involved in some kind of social scientific enterprise that would seriously threaten my status as an armchair philosopher.


Leave a comment

Filed under Uncategorized

Universalism and Omnipotence

I’ve been thinking about the coherence of divine attributes lately, particularly omnipotence. At the very least, it’s a counter-intuitive notion. This issue comes up, I think, with respect to universalism. Rob Bell recently incited some controversy with his book Love Wins, in which he suggests that all might be saved. Now, I haven’t read the book, so I’m not going to say much about it. However, I’ve heard Bell talk about it and he seems to base his view on an intuition about omnipotence: how could an omnipotent being (God) fail to get what he wants (the salvation of all)? The core intuition here seems to be that an omnipotent being can’t fail. At first blush, this seems right. But as a fine-grained analysis of omnipotence shows, this intuition might not be so obvious. Since this issue interests me, I’m going to use Bell’s intuition as a springboard to talk about omnipotence and tie it back to the subject of universalism.

Let’s think about the question: can an omnipotent being fail? At this point, we might recall the famous Stone Paradox, familiar to most undergraduate philosophy students. Can an omnipotent being create a stone so heavy that he cannot lift it? If one says ‘no’ then there’s something an omnipotent being can’t do, namely create the stone. If one says ‘yes’ then there’s something an omnipotent being can’t do, namely lift the stone. However, this isn’t necessarily a problem for omnipotence per se. After all, one of the things an omnipotent being could do (presumably) is to cease to be omnipotent. So an omnipotent being could create the stone, cease to be omnipotent, then fail to lift the stone. This is a bit of a cheat though, because the failure itself would not be the failure of an omnipotent being. But the point is, that the Stone Paradox is not a problem for what we might call non-essential omnipotence, i.e. a being whose powers include being able to lay aside his omnipotence. (I believe that Q from Star Trek was omnipotent in this sense.) However, the Stone Paradox is potentially a challenge for an essentially omnipotent being, i.e. a being who cannot cease to be omnipotent. The God of classical theism is said to be omnipotent in this sense. So can God create a stone that he can’t lift?

Some, like William Alston, have argued that ‘a stone heavier than infinite power can lift’ is a contradiction in terms. Since omnipotence is usually defined in the literature as the ability to do anything that is logically possible, if ‘a stone heavier than infinite power can lift’ is a logical impossibility, we can hardly expect God to accomplish it. However, we might think of other actions that clearly are logically possible that God cannot perform. Again, these are not necessarily problems for omnipotence per se, but for a being with God’s attributes. For example, due to God’s moral perfection, he can’t perform evil acts. Also, due to God’s incorporeality, he can’t swim or dance a waltz (without becoming incarnate, if that notion is coherent).  These acts are clearly logically possible and a morally flawed or corporeal omnipotent being could perform them, but God could not. So if God is omnipotent, he is not what we might call ‘act-omnipotent.’ There are several acts that God could fail to perform.

Perhaps this is why philosophers of religion have moved away from defining omnipotence as act-omnipotence. Rather, omnipotence is defined as the power to bring about any state of affairs the description of which is logically possible. A being who is omnipotent in this sense need not be act-omnipotent. (Kenneth Pearce and Alexander Pruss have recently argued that omnipotence doesn’t entail act-omnipotence. It’s very technical and I’m not sure I understand it, but let’s grant for the sake of argument.) Nevertheless, it’s still omnipotence worthy of the name. In the case of God, we have a being who is necessarily omnipotent (omnipotent in every possible world) and can bring about any logically possible state of affairs. This certainly seems like a lot of power. Can an omnipotent being, defined in this way, fail?

Plausibly, yes. This issue comes up with respect to the problem of evil. Alvin Plantinga’s famous “Free Will Defense” claims that even an omnipotent being could not give creatures free will and guarantee that those creatures would never do evil. However, as J.L. Mackie pointed out, a world in which free creatures always freely choose the good is a logically possible state of affairs, and as such, it’s one God could have actualized. Here the argument gets rather technical. Plantinga replied that although that description is a logically possible state of affairs, it’s not a feasible state of affairs for God to bring about. In other words, once God gives free creatures a say in the world, it’s no longer exclusively up to God which logically possible state of affairs will obtain. Furthermore, Plantinga argues, it may be the case, for all we know to the contrary, that for any world containing free creatures that God creates, those free creatures will go wrong. Plantinga calls this feature ‘transworld depravity.’ In other words, maybe God can’t, despite omnipotence, create a world containing free creatures without moral evil. Another way we might say this is that here we have a case of an omnipotent being failing to get what it (presumably) wants, i.e. a world in which free creatures always freely choose the good.

Now to bring the issue back to universalism. A world in which all free creatures are freely saved is a logically possible state of affairs. Moreover, if this is a state of affairs that God wants, then this state of affairs should obtain because God is omnipotent and could not fail to accomplish what he wants. This seems to be what Bell is arguing. However, parallel to the case of transworld depravity, is what some have called ‘transworld damnation.’ For all we know to the contrary, a world in which all are freely saved, although a logically possible state of affairs, is not a feasible state of affairs for God to bring about. So here’s another example of how an omnipotent being might fail.

I don’t really have a horse in this race. As an outsider (albeit one who used to be an insider) I don’t take a position on universalism vs. particularism and that hasn’t been the point of this little essay. Moreover, I’m not sure where I stand on the coherence of the concept of an omnipotent being, although I suspect the burden of proof is on those who say it’s incoherent. I haven’t sought to take up that issue either. Rather, I’ve sought to show how our intuitions about these issues can lead us astray. What seems like a commonsensical intuition — that an omnipotent being cannot fail — seems to be false regardless of which of our definitions of ‘omnipotence’ one chooses. Bell may have to give up his intuition on this issue. What implications that might have for his theology, I leave to others.

Leave a comment

Filed under Uncategorized

Is the Argument from Hiddenness a Stronger Challenge to Theism than the Argument from Evil?

I’ve been teaching some philosophy of religion to my intro to philosophy students over the past couple of weeks. We’ve had some great discussions, especially about the natural theological arguments and the problem of evil. Also, because the textbook has a focus on Canadian content, it presents an atheistic argument from Canadian philosopher J.L. Schellenberg, the argument from divine hiddenness.

Recently, I’ve come to believe that J.L. Schellenberg’s argument against God’s existence from divine hiddenness is a better atheistic argument than the argument from evil (logical or evidential). I’ll defend my reasons for believing this shortly. First, let’s look at Schellenberg’s argument. It goes like this:

1) If no perfectly loving God exists, then God does not exist
2) If a perfectly loving God exists, then there is a God who is always open to a personal relationship with each human person.
3) If there is a God who is always open to a relationship with each human person, then no human person is ever non-resistantly unaware that God exists.
4) If a perfectly loving God exists, then no human person is ever non-resistantly unaware that God exists (from 2 and 3 by hypothetical syllogism).
5) Some human persons are non-resistantly unaware that God exists.
6) No perfectly loving God exists (from 4 and 5 by modus tollens).
7) God does not exist (from 1 and 6 by modus ponens).

The only candidates for premises we might deny are 2 and 5. But these look very defensible. And this is why I think that the problem of hiddenness is a more serious challenge to theism than the problem of evil: the most common strategies employed against the problem of evil are not as plausible with respect to hiddenness.

For example, one might use the ‘greater good’ strategy in denying premise 2. One might say that God isn’t always (at every time) open to a personal relationship with each human person because God can only achieve certain goods if some persons lack explicit awareness of His existence (at least for a time). Again, this ‘greater good’ strategy is a common approach to the problem of evil. But what would a candidate for such a greater good be in the case of divine hiddenness? As with the problem of evil, sometimes freedom is suggested as the greater good that divine hiddenness serves to accomplish. Perhaps if God’s existence were too obvious, that would compromise our moral freedom; we’d do things or refrain from doing things because Big Brother is watching. However, this move doesn’t seem as plausible with respect to hiddenness as it does with respect to evil. God doesn’t have to reveal Himself in earth-shattering Hollywood fashion; He could simply give every non-resistant person a profound religious experience such that they were assured of God’s presence. Certainly some people claim to have such experiences, and it’s not obvious that such experiences would obliterate moral freedom. So why not give everybody such an experience?

Some have suggested other goods that might obtain from divine hiddenness, such as deep awareness of our spiritual deficiencies or cooperation with a spiritual community in seeking knowledge of God. But it seems as though those goods could be acheived in the context of awareness of God’s existence. Sometimes it might be necessary for a person in a relationship to withdraw for a time, and the same might be true of God (what Christians sometimes call a ‘dark night of the soul’). But that is compatible with an awareness of God’s existence, even if His immediate presence is not felt. So it seems that any good God could achieve through hiddenness, could be achieved in some other way. Again, I find the argument from hiddenness to be stronger than the argument from evil on this point.

Some theists would deny premise 5. Although a minority position among philosophers, this is probably the majority response among ordinary believers. Popular theology tells us that we all resist God. Maybe the people who are unaware of God’s existence are resisting a relationship with God. Maybe God is open to relationship, but they are not. I find this line of reasoning implausible. After all, most believers would acknowledge that the evidence for God’s existence falls far short of what it would take to convince a neutral audience. Some would cite the cognitive effects of sin here, but if God can open the eyes of some, why not all? (Calvinists give a very implausible answer to this question in my judgment.) Moreover, it seems that there are many people, Schellenberg included, who have done their homework on this issue and very much wish a loving God  existed. In other words, some unbelievers are emotionally open to such a relationship, but lack the evidence to believe. Are all of these people just culpably dishonest? That strikes me as implausible.

A parallel can be drawn here with respect to the problem of gratuitous evil (evil which seems to produce no greater good), except the epistemic burdens of  believers and non-believers are reversed. In the case of allegedly gratuitous evil, believers can opt for skepticism: we can’t know for sure that evils that appear gratuitous really are gratuitous, but the non-believer has to say, at the very least, that it’s implausible to deny that they are gratuitous. It’s difficult in this case (at least for me) to assign the burden of proof. In the hiddenness case, however, it’s much easier. The believer has to claim that he knows, or at least that it’s implausible to deny, that persons who appear to be non-resistantly unaware are actually resistant. Whereas all the non-believer has to say is that we can’t possibly know that, and judging from appearances, it’s unlikely to be true — an added advantage that the theist doesn’t have in the case of gratuitous evil since appearances in that case favor the atheist. In other words, this looks much less like a stalemate than the gratuitous evil case, which is another reason why I think the hiddenness argument is stronger than the problem of evil.

One more reason why I suspect that the problem of hiddenness is a more serious challenge to theism than the problem of evil: hiddnness is a more fundamental problem than evil. What I mean by this is that the problem of evil gains much of its purchase from the fact of hiddeness. While they are separate problems and, as we have seen, strategies for dealing with one don’t necessarily work for the other, evil would not be an epistemic problem for belief in God’s existence if God’s existence were more obvious. If that were the case, we might still have the question of why God allows certain evils, but evil would not be a barrier to belief in God per se. The more basic problem, then, is the hiddenness problem. While both evil and hiddenness are formidable challenges for theistic belief, I’m beginning to think hiddenness is the stronger of the two.


Filed under Uncategorized

Introducing dude ex machina

Welcome to dude ex machina, a blog for the musings of a philosophically inclined guy who is under the delusion that other people might be interested in what he thinks. I teach philosophy part time, so I already have a captive audience, but I’m always thinking about things after class, and things I should’ve said in class. That’s where this blog comes in. I’ll share what’s on my mind about philosophy, report what’s going on in philosophy, offer some reflections on teaching, confess my frustrations with academia (the blog is anonymous for a reason), and flag anything else that interests me in the news or in popular culture. Full disclosure: I’m an avid sci-fi geek and like to draw on it to make philosophical ideas a bit easier to understand. My main areas of interest at the moment are philosophy of religion and philosophy of mind, so if you don’t like those subjects, you’ve been warned.

Leave a comment

Filed under Uncategorized