The Axiology of Theism

Everything so far has been about whether God exists. This rung asks a question that can be separated from that one entirely, and which almost nobody in the field asked directly until recently: supposing we do not know the answer, would it be better if God did exist?

The obvious response is that the answer is too obvious to be worth a rung. A perfectly good being who loves each person, defeats death, and secures justice would plainly be an improvement on a universe indifferent to whether anyone suffers. But the obvious response has been disputed by people on both sides, and the arguments turn out to be more interesting than the question sounds.

The question, stated precisely

Three distinctions have to be made before the positions can be stated, because a good deal of apparent disagreement dissolves once they are in place.

The first is scope. The question can be asked about a particular person, about some restricted group, or about the world as a whole. Kahane's terms for the ends of this range are the personal and the impersonal, and the answers can differ: it might be better for me that God exists while being worse in some impersonal respect, or the reverse.

The second is whether the comparison is with the actual world or with the nearest world in which the other answer holds. Asking whether it would be better if God existed given everything else stays fixed is not the same as asking whether a world designed by God would be better than a godless world, since a world with God would presumably differ in many other ways.

The third is that the question is comparative and admits of partial answers. Something can be better in one respect and worse in another, and the interesting claims in this literature are usually about respects rather than about totals.

The two positions

The independence of these from the existential question is what makes the field interesting. There are theists who accept some anti-theistic conclusions, holding that the goods lost are real losses which are nonetheless outweighed. There are atheists who are strong pro-theists, holding that the truth is regrettable. And a person's position on the axiological question can be a datum about them rather than about the world, which is a point taken up below.

The case for pro-theism

Each item on the list can be pressed independently, and some are stronger than others.

The claim about waste is among the strongest. On a godless picture, a person who dies in obscurity after a life of unrelieved difficulty has simply had that life, and nothing further happens. On the theistic picture the same life is held in the awareness of a being who knows it completely, and rung 16's demand for defeat within a life becomes at least intelligible. That is a real difference, and it is the one to which the axiological literature keeps returning.

The claim about justice is nearly as strong. Many wrongs are never addressed, and the people who committed them frequently do well. A universe in which that is the final state of affairs is different from one in which it is not, and it is difficult to argue that the difference is not a difference in value.

The claim about intention is more disputable. Being deliberately made is not straightforwardly better than coming about unintended, and some people report finding the second more attractive, which is where the case for the other side begins.

There is also an impersonal version of the pro-theistic case, and it is worth separating because it survives objections that damage the personal version. On this version, the relevant good is not what anyone gets out of the arrangement but simply that a being of unsurpassable value exists. If a perfectly good, perfectly knowing being is itself the greatest conceivable good, then a world containing one contains more value than a world without one, whatever else is true and whoever benefits. The considerations gathered at rung 18 about maximising value bear directly on this, since a world with God is on this reading closer to the ideal by definition.

The impersonal version has a corresponding weakness. It settles nothing about whether any creature is better off, and it is compatible with the human situation being worse. Someone who accepts it can still be an anti-theist on the personal scope, and the two claims do not compete.

The case for anti-theism

The privacy consideration is the one most often stated and the easiest to feel. A person who is never unobserved, not in any room and not in any thought, lacks something that people ordinarily value a great deal, and the value in question does not look reducible to the practical benefits of not being watched. Part of what makes a private thought private is that it is one's own in a way nothing else is.

The autonomy consideration goes further and is harder to state without exaggeration. The claim is not that divine sustenance is a form of coercion; it is that a certain kind of standing, being the ultimate author of one's own life within an indifferent universe, is not available if the universe is not indifferent. Whether that standing is a good is exactly what is at issue, and this is where intuitions divide most sharply.

Thomas Nagel supplied the remarks that prompted the discussion. In the final chapter of The Last Word (Oxford University Press, 1997), he reported that he does not want there to be a God and does not want the universe to be that sort of place, described this as a fear of religion rather than a conclusion from evidence, and named the underlying difficulty the cosmic authority problem. He also thought the attitude widespread among his colleagues and partly responsible for the appeal of certain reductive programmes in philosophy, which is a striking thing for a naturalist to say in print.

Two further anti-theistic considerations are worth noting, since they draw on earlier rungs.

If cosmic justice is guaranteed, then the human work of building institutions that deliver justice loses a certain significance. It remains worth doing, but it is no longer the only thing standing between wrongdoing and impunity, and the moral seriousness of that situation is itself something some people value.

And if every loss is restored, then sacrifice changes character. Giving up something permanently for another person is a different act from giving it up in the confidence that it will be returned, and the first is arguably the more admirable.

The most natural theistic response to the privacy argument is that it misdescribes the observer. Being watched by an indifferent or hostile party is an intrusion; being fully known by someone who loves you completely is what people spend their lives trying to achieve with each other and mostly fail at. On this reading the theistic situation is not universal surveillance but universal understanding, and calling it a loss mistakes the best available good for the worst.

The reply has force, and it disposes of the cruder versions of the privacy complaint. It does not quite dispose of the argument, for two reasons. The intimacy people value with each other is chosen and mutual, and it is valuable partly because it is granted rather than unavoidable; a state of being fully known which no one can decline is a different thing, whatever its other merits. And the same asymmetry that made rung 15's parent analogy strain applies here: an intimacy between parties of unequal knowledge and unequal power is not the relation the analogy trades on.

Whether the preference is even coherent

An objection more fundamental than any of these has been pressed against both camps, and it draws on the modal material from earlier in the tier.

If God exists, God exists necessarily. That was the upshot of the discussion of necessary being: on classical theism God is not the sort of thing that happens to be there, and a being who might have failed to exist would not be the being under discussion. But then exactly one of the two scenarios the axiologist wants to compare is impossible. Either there is no possible world without God, or there is none with God, and in either case the anti-theist or the pro-theist is expressing a preference over something that could not have been.

Whether that is a real problem depends on what preferences require. Preferring one impossible state to another is not obviously incoherent, since people manage to have attitudes toward mathematical impossibilities and toward counterlogical suppositions all the time. But comparative evaluation normally works by holding some things fixed and varying others, and if the variable in question cannot vary, it is not clear what the comparison is doing.

Two responses are available. The first accepts the difficulty and reinterprets the whole question as epistemic: not what would be better, but which of two live hypotheses one should hope is true, which requires no comparison of worlds at all. The second denies the necessity claim, either by adopting a conception of God as a contingent being or by treating the axiological question as concerning conceptions of God rather than possible worlds. Neither response is free, and the first substantially reduces what the field can claim to have shown.

What "better" is being asked

The word doing the work in every position above is ambiguous in a way that explains some of the persistent disagreement.

It can mean better in the sense of containing a greater quantity of value. On that reading the pro-theist has an easy case, since a being of unlimited value plus everything else outweighs everything else on its own, and the anti-theist is reduced to arguing that the addition subtracts.

It can mean better in the sense of being a better kind of situation to be in. That reading makes the anti-theistic considerations much more powerful, since they are all claims about the character of a position rather than about totals, and character does not aggregate.

And it can mean better for a particular person, which reintroduces the scope distinction and makes it clear why the two camps so often talk past each other. Someone arguing that reality is better with God in it and someone arguing that the human position is worse are not contradicting each other, and both may be right.

The lesson is that the axiological question is not one question. It is at least three, and progress in the literature has mostly consisted of separating them.

What it does to the earlier arguments

The reply just given deserves expanding, because the axiological question turns out to be presupposed in several places already visited.

The pragmatic argument of the beginner tier assumed that if God exists, believing is enormously better for the believer, and that if God does not exist, little is lost. That is a strong pro-theistic premise on the personal scope, and it was never defended. An anti-theist who accepts every other step of the wager can decline it at that point.

The argument from evil requires that we can identify what a good being would want. The axiological literature shows that this is harder than it appears, since informed and careful people disagree about whether particular features of a theistic world are goods at all. That is not a defence of theism, but it is a complication for both sides.

The debunking arguments of the intermediate tier explained religious belief by the operation of biases that do not track truth. The axiological findings make the symmetric point available: if there are people strongly motivated to find that no God exists, the debunking move runs in both directions, and neither side has clean hands. Nagel's admission is the best evidence in the literature that this is not merely a theoretical possibility.

What follows for what one should feel

The clearest practical result concerns attitudes rather than beliefs, and it applies to the agnostic most directly.

An agnostic holds that the evidence does not settle the matter. The axiological question asks what they should hope, and unlike the existential question this one has an answer available now. If pro-theism is right on the personal scope, then hope is the appropriate attitude for someone in that position, and the absence of hope is a kind of failure to appreciate what is at stake. If anti-theism is right on that scope, then the appropriate attitude is closer to wariness, and the person who hopes is hoping for their own diminishment.

That is not nothing. Attitudes toward an unsettled question shape how a person conducts an inquiry, how long they persist, and what they count as good news. A researcher who wants a result and a researcher who dreads it will not read the same paper the same way.

The same holds for the committed on both sides. A theist who accepts that something is lost has a different religious life available than one who denies it, and the tradition's own literature on submission and self-surrender suggests that the loss was noticed long before the philosophers named it. An atheist who accepts that something is missing has a different intellectual position than one who insists that nothing is, and it is a more honest one if the arguments above are sound.

Complications worth keeping

Three refinements have emerged, and each blocks a tempting oversimplification.

The first is that pro-theism and anti-theism are not exhaustive, and the neutral position is not silly. Someone may hold that God's existence would change the world enormously without making it better or worse on balance, or that the two sets of considerations are incommensurable in the sense discussed at rung 18.

The second is that the personal and impersonal answers can diverge, and that the divergence is not a mark of confusion. A person may consistently hold that a world with God is better overall while wishing that they personally were not in the position a creature occupies, in something like the way a person may think a well-run institution is a good thing and still not want to work inside one.

The third is that anti-theistic considerations, if sound, tell against the strong form of pro-theism only. A theist can accept that privacy and independence are genuine goods, accept that a theistic world lacks them, and hold that they are outweighed many times over. That position is coherent, and it is more honest than denying that anything at all is lost.

Where it stands

Three observations are fair.

The field has established that the question is not trivial. That result alone justified the enterprise, since the initial reaction of most philosophers was that the answer required no argument.

The strongest anti-theistic considerations are the ones about standing rather than the ones about outcome. Arguments that a theistic world contains less of some good tend to fail, because a God can supply goods as well as remove them. Arguments that certain positions are unavailable to any creature whatever, no matter how well things go, are harder to answer, because they concern the structure of the relation rather than its consequences.

And the practical payoff is mostly diagnostic. The value of the discussion has been to make explicit the evaluative commitments running underneath the existential arguments, and to give both sides a way of naming a bias that had previously been noticed only as an accusation.

Next

The last several rungs have treated theism as a single hypothesis. It is not: the traditions that affirm a God disagree with each other about almost everything else, and their claims conflict. The next rung asks whether one tradition can be uniquely correct, whether all of them might be responses to a single reality described differently, and what the persistence of disagreement among informed and sincere people does to either answer.