S: All evil in the world ultimately traces back to human sin.
Some people say things like this, but it is fairly clearly false; Christian theology has never been so anthropocentric as to think that all the evil in the cosmos, or even all the evil on earth was due to human free will. S is, in fact, what Dembski says mainstream theology considers it to be: preposterous. Part of Dembski's mistake is confusing S with the following:
S1: Evil is the result of a will that has turned against God.
On Christian principles, S1 is more plausible than S; in part because Christian theology has never assumed that human wills are the only wills capable of turning against God. Whatever one may think of talk of angels and demons, its place in Christian theology has always and ever prevented major Christian theologians from assuming that important things -- whether evil or anything else -- are all about human beings.
But I think even S1 is not really tenable. For one thing, if evil is the result of a will that has turned against God, then it would follow that a will turning against God is not evil unless it is the result of a will that has already turned. And this is just not tenable. So we need to qualify S1 further by saying that evil involves either (a) a will turned against God; or (b) the intended effect of a will turned against God; or (c) both. However, even here we seem to run into a question. Does the modified S1 include natural evil?
It turns out to be difficult to fit natural evil into the modified S1. For one thing, it's rather difficult to pin natural evil down. I'm reminded in this regard of a paper by Hugh McCloskey arguing from natural evil that God does not exist. McCloskey's argument involves a very crude notion of natural evil. Thus deserts are evil because they can kill or hurt you; rattlesnakes are evil because they can kill or hurt you; etc. There is a fairly obvious problem with the criterion that anything is a natural evil if it can kill or hurt you: (1) the first is that by this criterion everything is a natural evil if it has causal powers. For unless it is countered by other causal powers (antidotes, immune systems, etc.) there are few things in the world that cannot kill or hurt you if it can do anything. So we would have to say that fruit is evil because you can choke on it; and water is evil because you can drown in it; and fire is evil because it can burn you; and books are evil because they can give you painful papercuts; and science is evil because it can lead to things that are deadly. This is not a rational viewpoint.
The problem with a criterion of this sort is that it is too quick to identify things as natural evils. Such a perspective not only cannot be reconciled with Christian theology, which regards every substantial thing in creation as good and the union of them together into a world or cosmos as very good; it cannot even be reconciled with ordinary common sense. However, if we take a more traditional view, we can regard as evil any privation of good. In this sense, what is evil in being killed by a rattlesnake is not the rattlesnake. The rattlesnake, in fact, is good in its kind (as any passionate herpetologist can tell you). What is evil is the privation of good that results incidentally from the rattlesnake (namely, being killed, in which we are deprived of life). And so it is with everything else.
Well and good; now we have a view of natural evil that is at least reasonable. Now the question is this: is every privation of good an act of will, a directly intended effect of will, or both? And the answer seems to be: obviously not. The rattlesnake's killing me is not an act of will -- my being deprived of life is not an act of will; it is simply this: that I am deprived of life. Likewise, unless the rattlesnake has been sent by man, God, or another intelligence to punish or torment me, neither is it a directly intended effect of will. Now, it is true that mainstream theology has tended to regard most evils experienced by human beings that are not faults (moral evils) to be penalties (punishments); but it's very clear that this is not a general principle. For everyone is very clear that the distinction between penalties and faults only divides one class of evils -- voluntary evils -- and not all evils are voluntary. For instance, if a cat leaps on a beetle and kills it, there is (perhaps) a natural evil involved: a beetle has been deprived of a good, namely, life. (Strictly speaking, the beetle can only non-figuratively be deprived of life if its life is due to it in some way. This would be controversial, but we can grant it. Of we deny this generally for this sort of case, we would be committed to saying that there are no natural evils that are not penalties. Some people hold this, but it is not generally accepted.) But there is no penalty or fault involved.
But since the modified S1 was as follows:
S1': Every evil is either a will turned from God or the directly intended result of such a will
we would be committed to saying that every natural evil is the directly intended result of an evil will. For us to know that this is consistent with the case of the cat killing a beetle, we would have to know that there is an evil will directly intending that the cat should deprive the beetle of life. And so on for every other case. This certainly doesn't follow directly from what makes such cases fall under the label 'evil': all that is required for that is lack of due good. And it isn't clear how we would know it to be true short of being shown this evil willer, or being told of him by a trustworthy source. But not even to Satan has there ever been attributed a will intending every natural evil directly. But maybe S1' is a little too strong, and we should remove the qualification 'directly intended'. But if we remove it entirely we get a statement that is inconsistent with the Christian doctrine of providence -- which tells us, among other things, that God is able to make some of the results of evil wills good. And even if we were simply to replace 'directly intended' with 'indirectly intended', i.e., intended under a very general description, this wouldn't seem to solve the problem. So S1', although more plausible than S1 or S, is not consistent with Christian principles, either.
So I don't think Dembski is right that the reason we are not inclined to accept S is that we are not inclined to believe in a historical Fall; rather, the reason we are not inclined to accept S is that there are good reasons not to do so, and always have been. Even the legend in Genesis does not suggest S -- it has a tempter, whose evil is not explained at all, and could not possibly be the result of human sin, because it precedes it. Further, it isolates the story of the Fall to a single garden of delights -- there is no suggestion in the text that everywhere was a garden of delights (as Dembski, to his great credit, explicitly recognizes). Further, there is no suggestion in the story that some things that could be called evil were not there -- certain kinds of pains and deaths, for instance. To get something like S we would have to read all of this into the story and take it as straightforward fact rather than, as it has usually been considered in Christian history, as a story told in the manner of a folktale to convey certain genuine truths, a case of God condescending to speak to us in our own way so that we might understand His basic points. (Dembski recognizes that we can do this with Genesis 1; but he doesn't appear to extend the point, as it should be extended, to Genesis 2.)
Dembski further makes the claim that the doctrine of redemption presupposes S; but this seems to me to be fairly obviously false, for reasons similar to those I have already discussed.
Dembski is more successful with two points: that (some) natural evils could exist as a preemptive response to human sin; and that Genesis 1 should be seen as primarily a way of describing divine purposes rather than the natural events that result from them. The latter is found, in a different variant, among the Greek Fathers. The former is technically true, but seems to me to get us nowhere, except into a terrain of mere speculation; I don't see any way we could know which natural evils exist in this way (and in what way they are brought into the divine plan for this purpose) unless we were explicitly told by God. There is no serious argument, as far as I know, that this has been revealed. So it remains a speculative possibility. Dembski in recognizing this is right in recognizing that there are certain sorts of evils that can be willed without fault -- just punishment of the guilty, for instance, always involves willing that someone experience penalty. It is a point usually forgotten. But recognizing that natural evils can be willed faultlessly is itself enough to deal with all but a handful of the theological issues involved in natural evil; there's no need for such an elaborate apparatus beyond that, particularly given that there is no reason to attribute all natural evils to human will.
I think Alejandro goes a bit far at certain points in his criticism (e.g., I don't think Dembski's point is that the Garden of Eden was any sort of God-induced illusion), but I agree with him on his primary points: (1) S, the foundation of the problem Dembski is trying to solve, really isn't very plausible at all. (2) Dembski's solution seems a Red Queen's race. In the Red Queen's country we have to run as fast as we can to stay in one place. I do think certain parts of the solution are clever; but it seems an awfully elaborate apparatus to keep things exactly one way, particularly given that the problem for which it is a solution doesn't seem to be genuine. (3) There is no serious theological need for claims that human beings are the crown of creation. Unless it's used in a qualified sense, I don't see any way to make it consistent with Christian theology, which has always tended to regard human beings as (to use C. S. Lewis's phrase) creatures of the margin -- we are not the center of the universe or crown of creation in any evaluative sense. Of course, by reason we have dominion over creation; but this has usually been considered a dominion of use -- it's just the straightforward fact that through reason we have some ability to use things as we please. So it does not matter one way or another whether there is any extraterrestrial intelligence or not; nothing about the human place in the universe really changes.
Theologically there are two obvious reasons to make use of the story of Adam and Eve: (1) the unity of the human race; and (2) original sin, for which I direct you to the answers to Rebecca Stark's recent quiz on original sin. However, for theological purposes, it does not (as far as I can see) matter whether you take the whole story of Adam and Eve in a realistic or an instrumentalistic fashion. As long as the story makes the relevant theological points, it really doesn't matter.