The fawn in the fire: evil makes God improbable

The evidential problem of evil

William L. Rowe 1931 · 2015Seemingly pointless suffering makes it likely that there is no God.

See profile

What it claims

It is reasonable to believe that there is no being who is at once all-powerful, all-knowing and wholly good.

Why

Some intense suffering brings no visible good, and a perfectly good God would prevent it.

The same idea through history

David Hume1779In the Dialogues, Philo takes up the dilemma attributed to Epicurus: if God can and will prevent evil, whence comes evil?

John Leslie Mackie1955"Logical" version: an all-powerful and wholly good God would simply be inconsistent with the existence of evil.

William L. Rowe1979"Evidential" version: evil does not prove that God does not exist, but it makes it reasonable to believe so. This is the most discussed form today.

In short

"The fawn in the fire: evil makes God improbable" is an argument by William L. Rowe ("The Problem of Evil and Some Varieties of Atheism", 1979, 1979) on the question "Does God exist?" (answer: atheism). Its claim: It is reasonable to believe that there is no being who is at once all-powerful, all-knowing and wholly good. According to AskPhilosophy's logic check (Z3 solver), the argument is valid only if one grants a premise the author leaves unstated (P2). AskPhilosophy presents 3 objections, 3 of them with a documented reply, and the original text behind each premise, checked word for word.

"The Problem of Evil and Some Varieties of Atheism", 1979

Valid with an implicit premise

The American philosopher William Rowe sets aside the old version of the problem of evil, according to which God and evil would be contradictory: in his view, no one has succeeded in showing that. He offers a more modest argument that is harder to dismiss. Lightning strikes a dead tree, the forest burns, a fawn is trapped in the fire and lies dying for several days. We can see no good this suffering served. Rowe does not claim to prove that God does not exist: he says that suffering of this kind, every day and by the millions, makes it reasonable to believe so.

click a premise to reject it

If you accept every premise, logic forces you to accept the conclusion. So the debate is about the premises.

Checked by a logic solver: the premises cannot all be true while the conclusion is false.
Without the implicit premise: Without P2, the computer finds a world where we see no reason for the fawn's suffering, but where such a reason exists all the same, out of our sight. In that world, God can exist.

Valid with P2. Rowe defends this step in his paper (he finds it "altogether reasonable" to believe that pointless sufferings exist), but does not make it a separate premise: it is the step "we see no reason, so there probably is none". That is exactly the step Wykstra's major objection targets. P3, by contrast, is accepted by almost everyone, believers included: Rowe notes it expresses a shared morality. So the argument is very different from Mackie's: it does not say that God and evil are contradictory, it says that evil as we observe it is strong evidence against God. Its conclusions are reasonable beliefs, not certainties.

Analogy. You look in your garage and see no dog: you are right to conclude there is none, because a dog you would see. You see no flea: there you can conclude nothing, because a flea you would not see. Rowe thinks God's reasons, if they existed, would be more like the dog. His opponents think they are like the flea. This garage example is often used in this debate.

Objections

Each objection is a conversation: click to see who replies and where the debate stands.

Stephen Wykstra (skeptical theism) 1984 · attacks premise 2

We are entitled to say "it appears there is no reason" only if, were there one, we would be likely to see it (Wykstra calls this rule CORNEA). But if God exists, his wisdom exceeds ours at least as much as a parent's exceeds that of a one-month-old baby. A baby does not understand why its parents let it go through certain pains. Likewise, our not seeing God's reasons is exactly what we should expect if he exists: so it is no evidence against him.

Rowe replies

see the argument

William Rowe (reply to Wykstra, 1984, as he summarizes it in 2006) 1984

That God's mind grasps goods beyond our ken does not mean that the goods for which he permits our sufferings are beyond our ken. When these goods occur, they involve those who suffered: we should be able to know them. To suppose they will come only in another life, one must add particular religious beliefs to bare theism: Wykstra has, without noticing, changed the question.

William Rowe (on the parent comparison) 2006

The parent comparison turns against the believer. A sick child does not understand why it is left at the hospital, but loving parents comfort it and assure it of their love the whole time. No loving parents use their child's hospital stay to go on holiday. Yet many humans have gone through horrendous suffering with no sign of God's love.

The debate goes on (our summary) The debate has gone on for decades under the name "skeptical theism" (Wykstra in 1996, Michael Bergmann in 2001, among others). Its critics point to a price: if we can never judge that a suffering is pointless, how can we know we must prevent the ones we see? Skeptical theists answer that our moral duties do not depend on what God knows.

Where the debate stands (our summary)This thread is developed further in the argument on evil in Hume (Bergmann, Almeida and Oppy, Bergmann and Rea). Core disagreement: can a good we do not see justify specific sufferings? Rowe finds it reasonable to conclude it cannot; skeptical theists answer that we are not in a position to judge.

The "G. E. Moore shift" (which Rowe himself states for the believer) 1979 · attacks premise 2

The believer can turn the argument around, as the philosopher G. E. Moore turned around those of the skeptics. Rowe says: "there are pointless sufferings, so God does not exist". The believer answers: "God exists, I have good reasons to believe it, so there is no pointless suffering, even if I cannot see the reason". Both arguments are equally valid: everything depends on which is stronger, Rowe's premise or the reasons to believe in God.

Rowe replies

William Rowe ("friendly atheism") 1979

Rowe accepts that this is the believer's best answer, and draws a lesson in modesty from it. An atheist may think that some believers have good reasons to believe in God, while thinking they are wrong: he calls this "friendly atheism". But for him, those reasons do not weigh as much as the believer thinks, against the mass of observed suffering.

The debate goes on (our summary) So everything is decided elsewhere: in the strength of the other arguments about God (the first cause, design, the ontological argument...). For those who find them weak, Rowe's argument keeps all its force; for those who find them strong, Moore's reversal is reasonable.

Where the debate stands (our summary)Everything turns on the relative weight of the other arguments about God: for those who find them weak, Rowe's argument keeps its force; for those who find them strong, the reversal is reasonable. Rowe speaks of "friendly atheism": he allows that the believer may be reasonable.

John Hick (the soul-making theodicy) 1966 · attacks premise 1

There is a good that suffering serves. God did not want a paradise where people would only be happy, but a world where persons become good, brave and compassionate through their own efforts. That requires a real world, governed by stable laws, with real dangers and real pain: Hick borrows from the poet Keats the idea of a "vale of soul-making".

Rowe replies

William Rowe 1979

Some suffering makes us grow morally, but it is fairly clear that suffering often goes far beyond what character formation requires. The believer cannot show, suffering by suffering, what good each one serves.

The debate goes on (our summary) It is also often noted that the fawn, alone in the forest, forms no one's character: Hick's idea applies mostly to humans. Rowe, for his part, would grant in 2006 that the theodicies of Hick and Swinburne go some way towards reconciling God and evil, while not being enough in his view.

Where the debate stands (our summary)Core disagreement: does suffering, in its amount and kind, serve the good of forming persons? Hick holds that it does; Rowe judges that it often goes beyond what is needed. Open point: the fawn, far from any witness, is often cited as the hardest case for this theodicy.

What today's voices say

Defend it

Alex O'ConnorArgues that meaningless or unnecessary suffering seems incompatible with a God who loves us and has the power to prevent it.

In the same family · The problem of evil

Evil is not a thing: privation and free willAugustine of Hippo