The Argument from Pain and Pleasure

From its name, the ‘problem of evil’ might give the mistaken impression that there is one singular problem of reconciling a particular observation with theism. In reality, there have not only been several variations upon the argument within philosophical discourse, but also a multitude of different but related experiences and observations buttressing the general discourse which has inspired debates and dialogues for many centuries. Paul Draper’s argument from pain and pleasure traces some of its roots back to David Hume, and uses an hypothesis of indifference to make sense of empirical evidence that has long seemed to cry out for an explanation on a theistic perspective. In this article, I will consider Draper’s case for the superiority of the Hypothesis of Indifference, as well as Peter van Inwagen’s response. I argue that van Inwagen is unsuccessful in his defense against the argument from pain and pleasure.

I. Pain and Pleasure

In his paper “Pain and Pleasure: An Evidential Problem for Theists”, Paul Draper argues that certain facts about pain and pleasure (O) provide us with prima facie good reason not to accept theism. The Hypothesis of Indifference (HI) is the claim that neither the nature nor the condition of sentient beings on earth is the result of benevolent or malevolent actions performed by non-human persons. According to Draper, HI constitutes a better explanation of the biological role of pain and pleasure than theism. Three observations about this are given for consideration, which are as follows:

O[1]: moral agents experiencing pain or pleasure that we know to be biologically useful;
O[2]: sentient beings that are not moral agents experiencing pain or pleasure that we know to be biologically useful; and
O[3]: sentient beings experiencing pain or pleasure that we do not know to be biologically useful (or that we know not to be biologically useful).1

Pain or pleasure that serves some biological usefulness is goal-directed in that it contributes to an organism’s biological goal, and it does not do so in a biologically accidental way. The common example is how the pain we experience from touching a hot stove suits our biological goal of survival by alerting us to harm.

Regarding O[1], we may notice that one important difference between the pain and pleasure system and other biological systems is that pain and pleasure have intrinsic value. If an omnipotent, omniscient, and morally perfect God exists, however, this association is surprising. HI predicts that pain and pleasure contribute systematically to biological goals like survival and reproduction, but given theism, we would not be surprised to find just the opposite to be true. We might imagine that God uses pain to warn us of threats to our survival, or that he uses pleasure to encourage us to reproduce, but a being such as God does not need pain and pleasure to be biologically advantageous in order to make organisms that are goal-directed in ways like these. In addition, the morally sufficient reasons God would have for permitting pain and pleasure could well be inconsistent with contributing systematically to the biological goals of human organisms. Thus, we have less reason to expect O[1] on theism than we do on HI.

Some theists have argued that the reason why God allows some suffering like that encapsulated by O[1] is because it plays a role in the lives of the moral agents who experience it. The second observation, O[2], points out that there are sentient beings who are not moral agents and still seem to experience pain and pleasure in a manner very similar to how moral agents experience them. HI predicts only that pain and pleasure serve biological functions, not moral ones, but theism faces a challenge here. A morally perfect God would seek to alleviate pain as much as possible, and so we would expect there to be an absence of gratuitous pain in the lives of non-moral agents.

O[3] reports the facts about sentient beings experiencing pain and pleasure we know to be biologically gratuitious, but also some that is neither known to be gratuitous or to be useful. Draper argues that much of what falls under O[3] is pathological pain and pleasure, resulting from the failure of an organic system to function properly, or it is biologically appropriate in that it happens in situations where it is biologically useful to the organism. Again, this is what’s to be expected if HI is true. On theism, though, we would expect an all-powerful, all-knowing, and all-good God to give some greater sense of balance to the moral and biological functions of pain and pleasure, perhaps where there would not be such strong appearances of ‘malfunctioning design’ or moral indifference.

Finally, Draper considers three theodicies in his paper to show that they do not significantly raise the probability of theism against HI. Plantinga claims that one of God’s final ends is a favorable balance of freely performed actions over morally wrong actions. While this does predict pain, pain also influences people to perform morally wrong actions (it is a necessary condition of many morally wrong actions), and so we actually have more reason to be surprised by O[1]-O[3] on Plantinga’s theodicy than we do on theism alone. Swinburne claims that one of God’s final ends is human freedom to make important moral decisions, but Draper finds this wanting in terms of amoral pain that doesn’t give us new knowledge, and even most of it that does. If God were really like a parent, he would withhold weighty responsibility from us until we proved ourselves ready for it. Last of all, Wykstra’s skeptical theism is vulnerable to the suggestion that God might just as well have reasons to prevent an evil as to allow it. HI makes better sense of the inability to provide a plausible theistic explanation of the facts about pain and pleasure.

II. Evil, Air, and Silence

Peter van Inwagen takes a different approach to Draper’s argument in “The Problems of Evil, Air, and Silence”, endorsing skeptical theism on other grounds than Wykstra. If HI were a better explanation than theism, he says, then we would know whether we could expect patterns of pain and suffering like we find in the actual world in a world that is created by God. This is something we cannot know, though, according to van Inwagen. We can invent stories that are true, for all we know, and that entail both theism and O, and if these stories are also false, for all we know, then our only rational option is going to be that we suspend our judgment altogether. There is a story with the following three propositions:

(1) Every possible world that consists of higher-level sentient creatures either contains patterns of suffering morally equivalent to those recorded by O, or else is massively irregular.
(2) Some important good depends on the existence of higher-level sentient creatures; this good is of sufficient magnitude that it outweighs the patterns of suffering recorded by O.
(3) Being massively irregular is a defect in a world, a defect at least as great as the defect of containing patterns of pain and suffering like those contained in the actual world.2

Patterns of suffering are morally equivalent for van Inwagen so long as there are no morally decisive reasons for preferring one to the other. A massively irregular world is a world where the laws of nature fail in some massive way. In light of the story above, we have reason to doubt our considerations about HI.

For a theistic story to support the skeptical thesis, three conditions must be met: it must be (i) aprobable, (ii) good, and (iii) undefeated. A theistic story is aprobable if one is unable to assign any specific range of values to the antecedent probability of the story given theism. If some story has a very high antecedent probability and it is a good story, then the skeptical theist claim – that humans are not in a position to compare the probability of O on HI with the probability of O on theism – is undermined. A “good” story is one that accounts just as well for O as HI, or that is at least not much worse in accounting for it than HI is. The idea behind this is roughly that a defensive strategy needs to be built on something better than a story that’s just ‘true for all I know.’ Lastly, a story can be called “undefeated” if it is not undermined by a parallel story that is antecedently as probable. In other words, if we have just as much reason to accept some other story as the theistic defense story, then the theistic story is not undefeated.

In his paper on “The Skeptical Theist”, Draper is quick to note that these three conditions alone are not sufficient. What is also needed are multiple, independent stories that meet these conditions, otherwise the theist is at risk of being in a worse position having found evidence for a theistic story than they would have been prior to finding that evidence. If these don’t seem like promising odds, Draper goes further to point out that good theistic stories tend not to be aprobable or undefeated. Many of them are particularly susceptible to counter-defenses which we have no less reason to accept than we do the theistic defenses.

Take for example the story that an omnipotent and omniscient being has a morally sufficient reason for permitting O that is beyond our ken. This story seems both good and aprobable. However, it isn’t undefeated because there is the alternative story that permitting O entails an unknown evil worse than any we are familiar with. Or that preventing O would entail an unknown good that’s better than anything we are familiar with. We can even bring free will into the picture and suppose that the best way for God to obtain a favorable balance of freely performed right actions over wrong actions is to utilize pain and pleasure in more or less the way O describes. An obvious counter-defense to this is the story that says the best way for God to achieve that balance is to utilize pain and pleasure in a way that differs from O. Without trying to examine every theistic story, we can already see the problem Draper mentions. Finding even one story that fits the three conditions is trouble enough, let alone finding multiple, independent ones.

III. Judgment and Skepticism

The moral skepticism van Inwagen advocates in propositions (2) and (3) stems from his reservations about our ability to make value-judgments concerning cosmic states of affairs. “My position,” he states, “is that we cannot be sure, and that for all we know our inclinations to make value-judgments are not veridical when they are applied to cosmic matters unrelated to the concerns of everyday life.”3 It is not entirely clear what argument is offered for this view in the paper. Professor van Inwagen does toss out a few suggestions about our moral intuitions being either a gift from God, a product of evolution, the result of social inculcation, or some combination of all three, but these do not seem to endorse fallibilism in the strong sense that would be needed for the sort of extreme moral skepticism being defended. There is also the controversial idea that only a source of our value-judgments (whatever it may be) that bears some appropriate relation to the concerns of everyday life can be considered trustworthy.

We are left with a number of questions here, like how we should define a cosmic state of affairs, what it is that makes the source of a judgment appropriately connected with everyday concerns, and whether cosmic states of affairs like those touched on in the argument from pain and pleasure really are as unrelated to everyday concerns as van Inwagen suggests. These are not easy questions to address, admittedly, and it does not seem that van Inwagen’s theistic defense really provides answers. This exemplifies Draper’s claim that theistic stories tend not to be aprobable. Whether or not we accept the modal skepticism of (1), it strikes me that (2) and (3) proceed from an argument for moral skepticism that is ultimately unconvincing.

Even so, there is reason to find fault with (1), too. As a counter-defense to the conjunction of (1)-(3), Draper offers A6:

Many possible worlds containing higher-level sentient creatures are not massively irregular and are such that there are morally decisive reasons for preferring the patterns of suffering in them to those reported by O.4

It does appear, as Draper notes, that we have antecedently no more reason to accept the theistic defense story than to accept this counter-defense. An example of such a world as he describes would be one that differs from the actual world in only one respect: absent from it is a single case of suffering. This might seem like a very minor difference, and it is when considering that a world with one less instance of suffering can hardly be thought to be massively irregular in comparison to our own. Yet this minor difference does not mean there is no good reason for preferring this alternative world to ours. The fact that such a world only differs in that it lacks one case of suffering is reason enough to favor it above the reports of suffering in O.

The discomfort some theists seem to have with making judgments about how God would create, intervene in the world, or interact with human beings is understandable from a certain theistic perspective. Particularly where we start to ponder things far beyond our ordinary range of experiences, it can be wise to hold fast to modesty and humility. On the other hand, we don’t want to pull the proverbial rug out from under our own feet, either. The less we think ourselves capable of even knowing what to expect from a theistic worldview, the less reason we may have to believe in it. Especially where evidential arguments are concerned, this looks like a risky gambit. Facts about pain and pleasure support the notion of a universe that is indifferent to us, and skeptical theism seems only to compound that silence with yet further mystery.



1. Paul Draper, “Pain and Pleasure: An Evidential Problem for Theists,” Nous 23 (1989): 331-50.
2. Peter van Inwagen, “The Problem of Evil, Air, and Silence,” Philosophical Perspectives 5 (1991): 135-65.
3. Ibid.
4. Paul Draper, “The Skeptical Theist,” in The Evidential Argument from Evil, ed. Daniel Howard-Snyder (Bloomington: Indiana University Press, 1996), p. 175-192.

%d bloggers like this: