by Robert Howell, SMU
"We call predestination God's eternal decree, by which he determined within himself what he willed to become of each man. For all are not created in equal condition; rather, eternal life is foreordained for some, eternal damnation for others" (Calvin, Institutes 3. 21. 5).
Calvinists and their ilk believed that there are a group of people, the Elect, who are predestined by God to partake of the fruits of heaven, while the rest are headed for less salubrious fates. One can not tell who the elect are, except by gleaning a hint from the fact that the elect lead perfectly Christian lives. Nevertheless, Calvinists typically continued to adhere to an extremely strict and inconvenient religious regimen. Why would they do so? It is not as if they were earning their salvation, since whether or not they would be saved or damned was already determined and could not be affected by any of their actions or thoughts. Consequently, the obedience of the Calvinists is often rationalized by appealing to something like "epistemic salvation": in fear of either ostracism or self-loathing (or anguish, fear, etc.) Calvinists act pious so as to preserve the appearance that they are among the elect. This is not a particularly charitable interpretation of Calvinist motives, however, as it seems to mislocate their reasons for being pious. If epistemic salvation were all that motivated them, then it would be entirely possible, through self-deception or deception of others, to satisfy that goal while being impious. While there were surely no hypnotists available to aid in self deception, or Rings of Gyges to aid in deceiving others, it seems dubious that these contingencies were the only reasons Calvinists remained truly pious. It behooves us to find an explanation for their actually being pious, and not merely seeming so, that doesn't ascribe to them a blatant irrationality.
An important puzzle in decision theory can help, I think, in sorting out intuitions behind these matters. In Newcomb's problem, we are asked to imagine the following scenario.1 An immensely intelligent fellow, baptized by reputation as The Predictor, is able with astounding accuracy (approaching perfection, say) to predict the actions of other people. The Predictor sets you the following game. There are in front of you two boxes, one opaque the other translucent. The game allows you two options: you can either take both boxes (and keep the contents of both) or you can take only the opaque box keeping only its contents. In the translucent box, there is $50,000. In the opaque box, the Predictor (who has presumably amassed quite a fortune by his forecasting excellence) will either have placed $1 million or nothing at all.2 (Although it is inessential, we can say that he places the money within the box an hour before you make your choice.) His decision about what to place in the opaque box is determined by the following: if he predicts that you will take only the opaque box, he will place $1million within it; if he predicts you will take both, he will place nothing inside the opaque box. Now: should you take only the opaque box, or both boxes?
People have radically conflicting intuitions on this matter: there are many one-boxers, and many two-boxers. Both sides adduce persuasive reasons in their favor. In general, though, most people-at least those unschooled in the problem-side with the one boxers.3 Why? The Predictor is extremely good at predicting whether people take one or two boxes, and you know this. We can imagine that you have seen him play this game with hundreds of people, and everytime people take one box they receive a million dollars, and everytime they take two they receive only the $50,000. When it is your turn to choose, why think that you can buck the odds? In general it seems you should act in such a way that maximizes expected utility, and the probabilities are extremely high that if you take one box you will walk off with a cool million, while if you take two you'll have to gripe about your only receiving 50 grand.
As I said, it seems to me that most people are one-boxers.4 The interesting thing is that the rationale for being a one-boxer is exactly the same as the rationale for being extremely well-behaved if you are a Calvinist! God is the predictor, heaven might or might not be in the opaque box, and sinful pleasures on earth are in the translucent box. Heaven is in the opaque box only for the elect, but God chooses the Elect based upon his infallible prediction as to whether or not they partake of earthly sins.5 Thus by the same utility maximization strategy as before, it seems quite rational to be very well behaved indeed! Calvinists were just one-boxers ahead of their time!
I think it is a sufficient defense of the Calvinists that there is this plausible a defense of their position.6 If they were making a mistake, it is an easy one to make, and it certainly leaves them with quite a bit of rational justification for their pious obedience. They are not out of the woods, however, for the old complaint against the Calvinists still resonates. Consider yourself confronted with the two boxes once again. The money is either already in the opaque box or it is not. Nothing you will do can change that! How can you go wrong by taking both boxes? The Predictor either put the money in the opaque box or he didn't. Suppose he did. Then, if you take both boxes you wind up with $1,050,000, 50 grand more than you would have otherwise. Suppose he didn't. Well, then you go home with $50,000, which is again 50 grand more than if you had only taken the second box. It would seem that whatever the Predictor actually did, you would be better taking two boxes! How could this be anything but the rational thing to do? One can imagine that the Predictor, who is also a truth-teller it turns out, tells your buddy whether or not the million is in the box. What would your buddy, who has all the information about the potential payoffs, recommend that you do? If he had your financial interests at heart, he would always, no matter what the Predictor did, recommend you take two boxes. It would seem completely irrational to go against the advice of your well-informed buddy.
These considerations, of course, support the original doubts about the motivations driving pious Calvinists. God already knows if someone is a member of the elect: he has already decided the matter. He's not going to take away that status-it is irrevocable. Suppose Satan, a two-boxer to the bitter end, knows of Jacob whether or not he is a member of the Elect. It seems that no matter whether Jacob was a member of the elect or not, Satan would recommend that he partake of whatever carnal sin tickled his fancy. Despite his wily ways, it seems that Satan in this case would have Jacob's best interests at heart, and that any angel who wasn't completely under the thumb of the Old Man would recommend the same. It furthermore seems that pious Jacob, once in heaven, would be right to kick himself and say "I was one of the Elect all along! I should've gone for Goody Whitfield when I had the chance!" So, anyway, suggests the Protestant two-boxer.
Whether or not the Calvinists were right, or whether one should be a one-boxer or not is a troubled question, and if it is ever solved conclusively, it will not be here. In any case, it seems without a doubt that the Calvinists are provided with a more rational justification than is usually ascribed to them. After all, in response to the two-boxer, they can always say, "Thanks for the lesson in decision theory, but I've noticed that whenever someone acts on your argument they wind up in hell. Excuse me, but I think in light of that, it is quite rational to remain pious." At this point, one suspects that the Calvinist and his critic must simply part paths, perhaps in more ways than one.7
Footnotes and References
1. For an introduction and explanation of Newcomb's problem, one could not do better than Robert Nozick's "Newcomb's Problem and Two Principles of Choice" and "Reflections on Newcombs Problem" published as chapters 2 and 3 of his Socratic Puzzles.
2. These numbers can of course be altered so as to make the amount in the opaque box seem a marginally and subjectively sigificant gain over the amount in the translucent box.
3. I have found an overwhelming majority of my students are one-boxers as are most of the people to whom I describe the problem. When Martin Gardner published a version of the problem in Scientific American, he received an outpouring of letters in response which were forwarded to Robert Nozick. According to Nozick, the ratio of one-boxers to two-boxers in the 148 letters was about 2.5 to 1.
4. Jeremy Fantl recently asked me in a discussion on this matter, "I know you are a two-boxer, but if it happened would you take two boxes?" While this question might tell more about his conception of the relationship between rationality and action than it does about Newcomb's problem, it does seem likely that many self-reporting two boxers would take only one box if the situation actually arose. Whether there is really prediction here is inessential. David Lewis emphasizes this in his "Prisoner's Dilemma is a Newcomb's Problem".
5. Whether there is really prediction here is inessential. David Lewis emphasizes this in his "Prisoner's Dilemma is a Newcomb's Problem".
6. This is assuming one had all those other beliefs of theirs, which might be rather less defensible.
7. It should be noted that predestination is a Newcombe's problem on the condition that God's foreknowledge is not a result of something like backwards causation. If something like this is the case, the matter gets more complicated, but the one-boxer's strategy is probably much more justified. Thanks to Thane Weedon for helping me see the urgency of this point. [/html]