Stephen Law’s Evil God Hypothesis

Posted on 08/02/11 13 Comments

There were a number of points that Stephen Law made in his book Believing Bullshit which I didn’t have time to address in my already bloated review. One of those was his so-called “Evil God hypothesis” (henceforth the EGH), a clever, if abortive, attempt to undermine theodicies (see pp. 24-27).

First off, what is a theodicy? Theodicies attempt to reconcile the existence of a loving and omnipotent God with the existence of evil in the world by pointing out reasons that God could have for allowing evil. Law’s EGH points out that the same type of reasoning used by the theodicist to reconcile a maximally good God to the evil we find in the world can be used to reconcile a maximally evil God to the good in the world. Given that such arguments don’t persuade us that an evil god exists, neither should they persuade us that a good God exists.

Law has a delightful little section where he quickly summarizes three “evil theodicies” which could be used to defend the existence of a maximally evil God. He begins with the so-called “reverse free will theodicy” which he summarizes like this:

“Evil god could have made us puppet beings that always did the wrong thing, so that we always acted to maximize pain and suffering. but then the world would have lacked one of the most profound and important forms of evil–moral evil: evil freely done of our own volition for which we can be held morally responsible.” (26)

Law drives the point home with a “character-destroying theodicy” and a “reverse laws of nature theodicy”.

And yet, despite all these impressive theodicies for a maximally evil deity, we don’t seriously consider that such an entity might exist. This brings us to the punchline of the EGH:

“The fact is, the amount of good that exists clearly is sufficient to place beyond reasonable doubt the conclusion that there is no evil god, notwithstanding such ingenious and convoluted attempts to try to explain it away. But if it remains fairly obvious that there is no evil god, given the available evidence, why isn’t it equally obvious that there is no good God either?” (27)

Frankly I was surprised when I read this argument because it evinced a rather glaring misunderstanding of what it is theodices attempt to do. The point of a theodicy is not to provide evidence for the claim that God exists to persuade those who do not already accept that claim. Rather, the point is to deal with putative defeaters to the claim by demonstrating that those defeaters do not work.

Here’s an illustration of what I mean. Let’s say that Jones devises an experiment in order to prove that he can generate phlegm in a test tube under controlled conditions. Jones runs the experiment and sure enough afterward a substance appears in the test tube. However, Jones’ detractors argue that the substance is not phlegm because it is brown and  yellow, not green. By pointing this out the detractors have presented a defeater to Jones’ claim that he produced phlegm. After all, phlegm is supposed to be green, isn’t it?

Well, maybe not. Jones then provides a defeater to their defeater. “Phlegm isn’t necessarily green,” he says. “It can be yellowish and/or brownish.” Let’s carefully note what Jones’ rebuttal accomplishes. It doesn’t establish the truth of the claim “Jones’ experiment did produce phlegm.” But it does undermine our conviction in the truth of the claim “Jones’ experiment did not produce phlegm.” In other words, it keeps Jones’ claim that he did produce phlegm in the running and should keep open minds listening.

That’s precisely what a theodicy aims to do. It doesn’t seek to establish the truth of the claim “God exists” but it does seek to undermine our conviction in the truth of the claim “God does not exist”, at least with respect to the existence of evil. (A person may have other reasons to reject the existence of God of course. But then no argument can deal with everything.)

Now let’s think about the EGH again. If there were in fact people who believed in the existence of a maximally evil god then they could develop precisely the kinds of arguments that Law outlines in his book. Such arguments, if successful, would not establish the truth of the claim “an evil god exists” but they would undermine at least one defeater to that claim (i.e. the defeater that there is too much good in the world for such a god to exist.)

Finally, let me note that atheists argue this way as well. At issue is this proposition:

Moral Realist Thesis (MRT): “There are moral facts which obtain irrespective of whether any human individual or community believes them.”

Many people are moral realists who believe that MRT is true. Many of those people also argue that MRT is inconsistent with atheism and thus they argue from MRT to the existence of God. (Even the great atheist philosopher J.L. Mackie famously conceded the force of such an argument.) But here’s the interesting thing. Many atheists have argued that MRT is consistent with atheism and thus that MRT does not constitute a defeater for atheism. If Law’s EGH critique of defeater-defeater arguments like theodicies had any merit then the theist could use it against the atheist’s attempt to reconcile atheism to MRT. So in retrospect it is actually a good thing for the atheist that Law’s argument is without merit.

Share
  • Christian Missionary

    There is a god of evil. He is the god of this world. His name is Satan.

  • toryninja

    Thanks, that’s helpful! I feel kind of goofy because I know what theodicies are suppose to do and yet I never noticed what Law was really doing there.

    Too bad you don’t review all the books I read :p

  • The Atheist Missionary

    Randal, the academic version of Law’s thesis is presented in his paper The evil-god challenge in the journal Religious Studies, Volume 46, Issue 3, September 2010.

    I understand your point about the intent of a theodicy. However, you have failed to respond to Law’s suggestion that his reverse theodicy shows that, if there is a god, it is just as likely that he is malevolent.

    • randal

      “However, you have failed to respond to Law’s suggestion that his reverse theodicy shows that, if there is a god, it is just as likely that he is malevolent.”

      It doesn’t show that at all! It shows that people who believe there is a maximally malevolent God can defeat defeaters to their view based on the type and amount of good in the world. That’s all it does. Perhaps you’d like to spell out step-wise how you think his argument establishes what you claim.

      • FutureMike

        This is where I think you haven’t gone far enough, which I was addressing in my previous comment–

        You’re right, the argument is not intended to show that ” if there is a god, it is just as likely that he is malevolent.” But it also isn’t intended to show “that people who believe there is a maximally malevolent God can defeat defeaters to their view based on the type and amount of good in the world.”

        Yes, it raises some “anti-defeaters”, but only to show that they DON’T work–that they don’t actually defeat the defeaters, and we almost universally recognize this since (almost) no one really believes in an evil god.

        The question is what’s different about the seemingly parallel good-God theodicies such that reasonable people should believe they actually do defeat their defeaters.

  • Eric

    Law’s claim, as i understand it, is that every reason we can provide to defend the existence of a ‘good’ god can be used (either exactly, as with, say, cosmological arguments, or in parallel forms) to defend the existence of an evil god. But we all think that the notion of an evil god is nonsense; hence, we should, if we wish to be consistent, conclude that the notion of a good god is nonsense as well.

    I don’t think Law’s argument works (Professor Ed Feser has outlined some great reasons here for the classical theists among you), but that is what I take him to be saying.

    • randal

      “Law’s claim, as i understand it, is that every reason we can provide to defend the existence of a ‘good’ god can be used … to defend the existence of an evil god.”

      Exactly. But theodicists are not attempting to provide reasons to accept the existence of a good God. They’re attempting to demonstrate that the world as it is is merely consistent with the existence of a good God (so that the world as it is does not provide a defeater for that God). The fact that a person can also develop arguments that show the world as it is is consistent with the existence of an evil god is not of relevance for the Christian theodicist.

      Thanks for the Feser link.

      • Eric

        “But theodicists are not attempting to provide reasons to accept the existence of a good God. They’re attempting to demonstrate that the world as it is is merely consistent with the existence of a good God (so that the world as it is does not provide a defeater for that God).”

        Oh, I agree. But I took Law to be referring to his evidential problem of evil argument there. That is, it seems to me that he’s saying that just as the (evidential) problem of good defeats the evil god hypothesis, the (evidential) problem of evil defeats the good god hypothesis.

        If I understand you correctly, you’re suggesting that Law thinks that evil god reverse theodicies would defeat arguments from the problem of good, whereas I think he’s saying that they don’t. Further, he seems to be saying that reverse theodicies are rather transparent rationalizations, and once we see this, he thinks, we’ll see that our traditional theodicies (which are, he claims, just parallel versions of evil god reverse theodicies) are equally weak. So, this whole exercise is an attempt on Law’s part to clear away some of what he takes to be the clutter that prevents us from seeing the (decisive?) strength of the evidential problem of evil.

  • The Atheist Missionary

    I’m currently reading Harold Kushner’s When Bad Things Happen to Good People. He touches on this issue early on when he writes:

    “The misfortunes of good people are not only a problem to the people who suffer and their families. They are a problem to everyone who wants to believe in a just and fair and livable world. They inevitably raise questions about goodness, the kindness, even the existence of God.”

    I guess the question back at Randal is what he relies on (aside from Scripture) to arrive at the conclusion that his deity is benevolent.

  • Brad Haggard

    Law has brought this up a couple of times on the “Unbelievable” podcast. He seems very confident of it, actually, even though it missed the mark both times.

  • FutureMike

    Randal-

    I doubt you’re checking comments to this post anymore, but wanted to leave this anyway.

    I appreciate your clarification of the purpose of theodicy, but I’m not sure how it rebuts Law’s point. You write:

    “Such arguments, if successful, would not establish the truth of the claim “an evil god exists” but they would undermine at least one defeater to that claim.”

    But what if such arguments are *unsuccesful*? Isn’t that what Law is getting at? If the arguments are unsuccesful, the defeaters stand, and the grounds for belief in the god in question are undermined.

    Law observes that the anti-defeater arguments for an evil god are pretty much universally considered unsuccesful, thus we conclude we don’t have sufficient evidence to believe in an evil god.

    But if the anti-defeaters for a good God are essentially the same as those for an evil god (but in “reverse”), why don’t we make the same conclusion about the good God?

    I’m not sure you answered that. If you did, I’d appreciate if you could elaborate. Thanks.

  • Pingback: Response to Randal Rauser’s Response to my response to his review… | Believing Bullshit

  • Pingback: Response to Randal Rauser’s response to my response to his shoddy review…