Hostname: page-component-745bb68f8f-cphqk Total loading time: 0 Render date: 2025-02-04T22:46:52.485Z Has data issue: false hasContentIssue false
  • English
  • Français

Sceptical theism and the evil-god challenge

Published online by Cambridge University Press:  19 March 2018

PERRY HENDRICKS
PERRY HENDRICKS*
Affiliation:
Trinity Western University, Langley, British Columbia V2Y 1Y1, Canada
*
e-mail: ampchendricks@gmail.com
Rights & Permissions [Opens in a new window]

Abstract

This article is a response to Stephen Law's article ‘The evil-god challenge’. In his article, Law argues that if belief in evil-god is unreasonable, then belief in good-god is unreasonable; that the antecedent is true; and hence so is the consequent. In this article, I show that Law's affirmation of the antecedent is predicated on the problem of good (i.e. the problem of whether an all-evil, all-powerful, and all-knowing God would allow there to be as much good in the world as there is), and argue that the problem of good fails. Thus, the antecedent is unmotivated, which renders the consequent unmotivated. Law's challenge for good-god theists is to show that good-god theism is not rendered unreasonable by the problem of evil in the same way that evil-god theism is rendered unreasonable by the problem of good. Insofar as the problem of good does not render belief in evil-god unreasonable, Law's challenge has been answered: since it is not unreasonable to believe in evil-god (at least for the reasons that Law gives) it is not unreasonable to believe in good-god. Finally, I show that – my criticism aside – the evil-god challenge turns out to be more complicated and controversial than it initially appears, for it relies on the (previously unacknowledged) contentious assumption that sceptical theism is false.

Type
Original Article
Information
Religious Studies , Volume 54 , Issue 4 , December 2018 , pp. 549 - 561
Copyright
Copyright © Cambridge University Press 2018 

The evil-god challenge

Stephen Law's evil-god challenge (Reference Law2010) is directed at good-god theistsFootnote 1 who think that belief in evil-god is unreasonable: if a good-god theist thinks that belief in evil-god is unreasonable because of the problem of good (explained below), then they must also think that belief in good-god is unreasonable because of the problem of evil. Law's challenge is for good-god theists of the above stripe to show that this conditional is not true – or, if it is true, that they are not affected by it since neither the antecedent nor consequent are true. In this article, I will first explicate Law's challenge, making its premises clear. After showing that premise (2) (explained below) is motivated by the problem of good, I argue that a version of sceptical theism undermines the problem of good, which renders premise (2) unmotivated. Thus, we have no reason to affirm the conclusion, and hence the evil-god challenge has been answered; sceptical theism renders the evil-god challenge a failure. I conclude by considering several objections to sceptical theism, and argue that they do not work.

Though Law does not do so, it will be of use to put his challenge in syllogistic form, for we will then have a more concrete target to aim at. We may put it as follows:

  1. (1) If belief in evil-godFootnote 2 is unreasonable, then belief in good-god is unreasonable.

  2. (2) Belief in evil-god is unreasonable.

  3. (3) Therefore, belief in good-god is unreasonable (modus ponens, (1) and (2) ).Footnote 3

In support of premise (2), Law first notes that the good in the world constitutes ‘overwhelming evidence against [evil-god's] existence’ (Law (Reference Law2010), 357); that is, belief in evil-god is rendered unreasonable because of the amount of good there is. Law calls this – the problem of reconciling the existence of evil-god with the amount of good in the world – the problem of good, and says that it is parallel to the problem of evil that good-god theists face. Next, he argues that (almost) any response that the good-god theist can give to the problem of evil, the evil-god theist can give to the problem of good.Footnote 4 This leads him to affirm

The symmetry thesis: there is a rough symmetry between the reasonableness of belief in evil-god and belief in good-god. (ibid., 359)

Law thinks that the symmetry thesis is true because (a) evil and good-god theism face a parallel evidential problem about values (i.e. the problem of evil and the problem of good) and (b) there is no (cogent) argument that supports good-god theism that does not also support evil-god theism. In what follows, I will assume that the symmetry thesis – and therefore premise (1) – is true.Footnote 5

Law's lapse

Law claims that when ‘presented with the evil-god hypothesis, most of us immediately dismiss it as absurd, typically because we consider the problem of good decisive’.Footnote 6 So, most people (supposedly) think that the problem of good renders belief in evil-god unreasonable, and hence affirm premise (2) of the evil-god challenge, and – since we have granted the symmetry thesis and therefore premise (1) – are thereby committed to its conclusion. In what follows, I will show that Law does not give us good reason to affirm premise (2).

The reason that Law affirms premise (2) is because of the problem of good. That is, he thinks that the amount of good in our world renders belief in the existence of evil-god unreasonable. While his argument is not explicit, it appears to be something like the following:Footnote 7

  1. (4) For some actual goods G we know of, we cannot think of any immorally justifying reasonsFootnote 8 for permitting them.

  2. (5) Therefore, probably, there are not any immorally justifying reasons for permitting them.

  3. (6) If evil-god exists, then he would not permit G if there were no immorally justifying reasons for permitting them.

  4. (7) Therefore, probably, evil-god does not exist.

The inference from premise (4) to (5), a ‘noseeum inference’,Footnote 9 is key here.Footnote 10 Is there any reason to think that it is a good inference? It does not appear so, for it is susceptible to criticisms stemming from sceptical theism.Footnote 11 The term ‘sceptical theism’ denotes a variety of positions that have been developed in response to evidential arguments from evil. Some types of sceptical theism deny that evil is evidence at all for atheism (or they deny that it appears that there is gratuitous evil),Footnote 12 others claim only that we lack good reason to think that our knowledge of values and the entailment relations between them are representative of the actual values and entailment relations there are.Footnote 13 Different types of sceptical theism target different aspects of different arguments from evil.Footnote 14 What I will argue here is that the principles that drive a certain type of sceptical theism can be used to undermine (at least one version of) the problem of good.

Following Hud Hudson (Reference Hudson and Kvanvig2014b)Footnote 15 we may think of the search for an immorally justifying reason (i.e. a reverse-compensatory state of affairs, see note 8) for G as like searching for a rabbit in a garden. If our garden is small, uniform in nature, wholly accessible to us, and we have good vision, then if we do not find a rabbit after searching for it, we are justified in inferring that there probably is not one. This is because if there were a rabbit, then we would (at least probably) know about it and recognize it as such. The sceptical theist, however, thinks that the garden is (perhaps infinitely) large, that parts of the garden are not accessible to us, that we have no good reason to think it is uniform in nature, and that our vision is subpar. The garden is large because the number of states of affairs is infinite – or, at least, unimaginably high. Therefore, even if all the states of affairs that we examine do not immorally justify G, it does not follow that there is no such reason, or even that there probably is no such reason. Our sample size is too small to come to any significant conclusion about this.Footnote 16

Further, parts of the garden are inaccessible to us: there are no doubt states of affairs that are so complex that humans cannot comprehend them. And since we are not able to access such states of affairs, we are in the dark about how large this group is and about its contents, and this means that we have no good reason to think that the immorally justifying reasons that we know of are representativeFootnote 17 of those that there areFootnote 18 – we have no good reason to think that the garden is uniform in nature. But if we have no good reason to think that the immorally justifying reasons that we know of are representative, then the inference from (4) to (5) is unwarranted. In other words, we know that a section of the garden of immorally justifying reasons is inaccessible to us, and we are unsure about how large the section is and whether the contents of that section resemble the contents of the section of the garden that we are able to search. And this prevents us from inferring that there (probably) is not an immorally justifying reason for G in it. Still further, our vision is subpar, for ethics is notoriously murky water, and an immorally justifying reason for G falls into the ethical category. So, it could be that we already know of an immorally justifying reason for G, but do not recognize it as such – perhaps free will really does justify the amount of good in the world, but we do not recognize its proper value; we cannot be confident in our ability to recognize abstract ethical truths or weigh values. In summary, because of the nature of immorally justificatory reasons, it is not true that if there is an immorally justifying reason for G, we probably would know about it and recognize it as such, and hence we are not warranted in making an inductive inference from the immorally justifying reasons that we know of to the immorally justifying reasons that there are. Therefore, the inference from premise (4) to (5) is unwarranted, and the problem of good fails.Footnote 19

Now, since Law supports premise (2) with the problem of good, it follows that – since the problem of good fails – premise (2) is groundless: Law has lapsed in affirming it. Thus, the evil-god challenge has been answered: there is no reason to affirm premise (2) – or, at least, Law's reason for affirming it is dubious – and hence the conclusion is avoided.

Some objections and replies

An objection that Law might make to my argument is that sceptical theism is false, and hence my critique of the problem of good is undermined since it relies on it. Indeed, Law (Reference Law2015) has made clear that he thinks that sceptical theism leads to absurd consequences: it (supposedly) entails that a person cannot reasonably hold beliefs about the past, about the external world, and about religion, and this is true even if they have knowledge of such beliefs. (Call these beliefs ‘common-sense beliefs.’) Before summarizing his argument, it will help to explicate briefly the externalist epistemological position known as proper functionalism.Footnote 20 A proper functionalist claims that a belief is warranted and amounts to knowledge if the following conditions hold: S believes p, p is true, S's belief was formed by properly functioning cognitive faculties successfully aimed at producing true beliefs, and S is situated in an appropriate cognitive (mini/maxi) environment.Footnote 21 Now that we have a grasp of proper functionalism, we may return to Law's argument. He argues (very roughly) as follows: if a person sees an orange on a table then they may reasonably believe that there is an orange there. However, if they come to believe that God would deceive them about there being an orange on the table if there is an (im)morally justifying reason for doing so and that they are in the dark about whether there is such a reason, then – so the argument goes – it is unreasonable for that person to hold a belief about the orange. Since, Law claims, this is precisely what sceptical theists affirm, it follows that it is unreasonable for the person, if they are sceptical theist, to believe that there is an orange on the table. The same goes for common-sense beliefs: since we are in the dark about whether God has (im)morally justifying reasons for deceiving us about our common-sense beliefs, we cannot reasonably hold them. Further, Law claims that even if the sceptical theist's common-sense beliefs amount to knowledge via proper functionalism, that such beliefs are nonetheless unreasonable.Footnote 22 Thus, sceptical theism is false: it entails that beliefs that are clearly reasonable are unreasonable.Footnote 23

There are many responses available to Law's objection, but I will only (and briefly) rehearse three here. First, it is important to understand what exactly Law's charge of unreasonableness amounts to. He identifies (un)reasonableness with (ir)rationality, and holds that a person's belief that p is irrational if they have good reason to suppose that the method by which they formed p is untrustworthy. The crucial question is this: does Law's argument give the sceptical theist good reason to think that their common-sense beliefs are formed in an untrustworthy manner? Suppose that the sceptical theist is a proper functionalist who thinks that their common-sense beliefs are produced by properly functioning cognitive faculties (etc.). Suppose further that Law's argument does not compel them to abandon their common-sense beliefs; that is, they are not shaken of their common-sense beliefs by Law's argument. If that is the case, does the sceptical theist have good reason to think that their common-sense beliefs are untrustworthy? The answer to this question is contingent on what ‘good reason’ amounts to. If it amounts to compulsion, then the sceptical theist's beliefs are not unreasonable, for they have not been compelled to give up their belief. So, Law must mean something different. Indeed, it is more likely that he is thinking of ‘good reason’ as an alethic defeater; that is, he thinks that a person has a good reason to think that a method of belief is untrustworthy if they have an alethic defeater for their beliefs about said method. Alvin Plantinga characterizes alethic defeaters as follows:

D is a purely [alethic] defeater of B for S at t iff (1) S's noetic structure N at t includes B and S comes to believe D at t, and (2) any person S* (a) whose cognitive faculties are functioning properly in the relevant respects, (b) who is such that the bit of the design plan governing the sustaining of B in her noetic structure is successfully aimed at truth (i.e., at the maximization of true belief and minimization of false belief) and nothing more, (c) whose noetic structure is N and includes B, and (d) who comes to believe D but nothing else independent of or stronger than D, would withhold B (or believe it less strongly). (Plantinga (Reference Plantinga and Bielby2002), 209)

Here D represents the belief that God would deceive us if he had good reasons for doing so and that we are in the dark about whether there is such a reason, and B represents one's common-sense beliefs. So, the question is whether any person whose truth-directed and properly functioning cognitive faculties responsible for sustaining B (etc.) would withhold (or, at least, hold less strongly) B after coming to hold D. So far as I can see, Law has not given us reason to affirm this, nor is it clear what such a reason would look like. But this means that the sceptical theist's common-sense beliefs have not been shown to be untrustworthy and therefore have not been shown to be unreasonable.Footnote 24 (Indeed, if the sceptical theist's cognitive faculties have been successfully aimed at truth, then it is trivially true that their belief forming methods are trustworthy, for they would – by definition – produce mostly true beliefs. Thus, whether the sceptical theist has an alethic defeater appears to come down to factors that they do not have access to.) Hence Law's objection is unsuccessful – he has more work to do to show that the sceptical theist's common-sense beliefs are untrustworthy or unreasonable.

Indeed, we may go further and offer a concrete counterexample to a crucial premise of Law's argument against sceptical theism (i.e. that if the sceptical theist recognizes that God would deceive them about their common-sense beliefs if he has a good reason to do so and that they are in the dark about whether he has such a reason, that it is no longer reasonable for them to hold their common-sense beliefs. Call this premise ‘P.’). Near the end of William Alston (Reference Alston and Morris1994), he reveals that he has intentionally deceived the reader with the title of his essay (his exact reasons are unimportant). This opens (or, at any rate, should open) the reader's eyes to the fact that all authors might have reasons to use their titles to deceive their readers about the content of their work. Indeed, the reader now knows that if an author has good reason to use the title of their work to deceive their readers, that they will do it, and they know that they are in the dark about whether an author has such a reason – they recognize that they do not know a priori whether (or how probable it is that) the author has a good reason to deceive them with their title. (Call these truths ‘P*’ and call a person who recognizes P* a ‘sceptical reader’.) However, the sceptical reader – despite these facts – is surely reasonable in believing that e.g. Hudson's book The Fall and Hypertime (Reference Hudson2014a) is (at least generally) about hypertime and the biblical story of the fall of humanity – they are not unreasonable in holding such a belief.Footnote 25 Now, if the sceptical reader's belief about the content of Hudson's book (and books in general) is not rendered unreasonable by P*, then – since P and P* are structurally identical – neither are the sceptical theist's common-sense beliefs rendered unreasonable by P. Thus, either the sceptical reader's beliefs about the content of books are unreasonable, or the sceptical theist's common-sense beliefs are reasonable. If the former, then everyone, including Law, faces a sceptical challenge; if the latter, then Law's objection to sceptical theism fails. He may pick his poison.

Another, simpler objection to Law one might make is to deny that there can be unreasonable knowledge; that is, one can affirm the (to my mind, plausible) thesis that knowledge entails reasonableness. This would force Law to drop his weaker claim that even if sceptical theists have knowledge of their common-sense beliefs, they are nevertheless unreasonable. From here, the sceptical theist can argue as follows: if our common-sense beliefs are produced by properly functioning cognitive faculties successfully aimed at truth (etc.), then our common-sense beliefs amount to knowledge, and this is knowledge regardless of whether we are in the dark about God's reasons to deceive us. Since knowledge, on this view, entails reasonableness, it follows that – if proper functionalist conditions obtain – one's common-sense beliefs are also reasonable. Since reasonableness on this view is contingent on whether proper functionalist conditions obtain, Law will have to give us reason to think that such conditions have not obtained. But he has not done so. Further, this response will be available to (almost) any externalist: so long as one's common-sense beliefs are produced in the right way, they will amount to knowledge. Thus, Law's objection will only – if at all – pose a problem to non-externalists.Footnote 26

Finally, it is worth pointing out that this discussion shows that Law's evil-god challenge is not as simple as he thought: it requires him to affirm, along with the symmetry thesis, the following:

The sceptical theism thesis: sceptical theism is false.

Such a thesis is highly controversial, to say the least. Thus, the evil-god challenge relies on multiple theses, at least one of which is highly controversial, and this makes the challenge far weaker than it initially appeared.

Appendix: the symmetry thesis and EG2

In this article, I have, for the sake of argument, assumed the symmetry thesis is correct. In what follows, I will drop that assumption, and, indeed, argue against it. This, I will show, has implications for EG2 (explained below and in note 3). Though my comments here are admittedly brief, I hope to have sketched out some plausible routes for denying the symmetry thesis and answering EG2.

As previously noted (see note 3), Law states the evil-god challenge in two different ways: one as I have stated above (i.e. premises (1)–(3) ), and another in which he says: ‘the challenge is to explain why the hypothesis that there exists an omnipotent, omniscient and all-good god should be considered significantly more reasonable than the hypothesis that there exists an omnipotent, omniscient and all-evil god’ (Law (Reference Law2010), 353). So, the second version of the evil-god challenge – EG2 – challenges the good-god theist to show why it is more reasonable for them to accept good-god theism over evil-god theism. It should clear, then, that if the symmetry thesis is rejected, EG2 will have be answered as well – if the good-god theist shows that they have more evidence for, or are more justified in, accepting good over evil-god theism, then EG2 is answered.Footnote 27 In what follows, I will illustrate several ways that good-god theists can reject the symmetry thesis and answer EG2.

If a good-god theist affirms – is compelled by – the modal ontological argument (as advocated by Plantinga (Reference Plantinga1979) ) then they are committed to the existence of a perfectly good god. Hence the good-god theist who affirms the argument has significant grounds or evidence for their belief that they do not think the evil-god theist has, and hence they ought to reject the symmetry thesis. Law responds to this move by saying that he does not know of many philosophers who endorse the argument and that one can construct a parody of the argument to support a maximally evil-god (Law (Reference Law2010), 370). This response is wholly inadequate. It appears that Law has forgotten the dialectic situation: he is claiming that good-god theists cannot think that belief in good-god is more reasonable than belief in evil-god, but any good-god theist who affirms the modal ontological argument can simply cite it as non-symmetric and cogent evidence for good-god. The popularity of the argument is irrelevant: all that matters is that the good-god theist accepts it, for that would give them reason to reject the symmetry thesis. Further, they can cite the argument to explain why it is more reasonable for them to accept good-god theism over evil-god theism, thereby giving them an answer to EG2.Footnote 28

But endorsement of the modal ontological argument is not the only way for a good-god theist to reject the symmetry thesis. Perhaps, for example, they follow Alston in thinking that ‘a belief is . . . justified . . . provided it stems from a socially established doxastic practice that is not discredited by the total output of such practices’ (Alston (Reference Alston1991), 182).Footnote 29 A doxastic practice is socially established, according to Alston, if it has been practised for a non-negligible amount of time by a community of persons, and a doxastic practice is not discredited by its total output if it is not massively incoherent. Further, a doxastic practice garners more support – more justification – if it successfully carries out its aim; that is, a doxastic practice gains justification if its goal is met.Footnote 30 Alston argues at length that beliefs formed within the Christian tradition about God have this justification.Footnote 31 I will not defend that thesis here, but will merely assume that beliefs formed within the Christian doxastic practice have such justification. (I refer the reader to Alston (Reference Alston1991) for a full-blooded defence of this thesis.) This is significant since the Christian tradition affirms the existence of an all-good god. Furthermore, there is – to my knowledge – no socially established (etc.) doxastic practice that evil-god theists partake in. Thus, we have a non-symmetric form of justification that supports good-god theism over evil-god theism, and hence the symmetry thesis is false – the good-god theist who partakes in the Christian doxastic practice and follows Alston in respect to justification should view good-god theism as significantly more reasonable than evil-god theism.Footnote 32 Further, this gives the good-god theist a straightforward answer to EG2: Alstonian justification provides them with reason for affirming good-god theism over evil-god theism.

Another way for the good-god theist to reject the symmetry thesis is to endorse phenomenal conservatism. If the good-god theist is a phenomenal conservatist, then they think (very roughly) that if it seems to them that p, that this gives them justification (or evidence or reason) for believing p (see e.g. Michael Huemer (Reference Huemer2007).)Footnote 33 If that is the case, then, if it seems to them that good-god exists and it does not seem to them that evil-god exists (or it seems to them that he does not), then they can cite this fact as rendering belief in good-god significantly more reasonable for them than belief in evil-god. Hence, they may reject the symmetry thesis. Further, they may cite their seeming as reason or evidence for them to affirm good-god theism over evil-god theism, thereby answering EG2.Footnote 34 Or perhaps the good-god theist is a reformed epistemologist, and thinks that their belief in good-god is properly basic,Footnote 35 and they lack belief (basic or otherwise) in evil-god. While this may not give them reason to reject the symmetry thesis, it would at least allow them to answer EG2: the reason why it is reasonable for them to accept belief in good-god over evil-god is because, they think, their belief in good-god is properly basic. Finally, perhaps the good-god theist finds arguments or evidence for the resurrection of Jesus to be compelling (for two very different versions of resurrection arguments, see Richard Swinburne (Reference Swinburne2003) and Michael Licona (Reference Licona2010) ). Since evidence for the resurrection of Jesus is evidence for Christianity, and Christianity entails the existence of an all-good god, the good-god theist of the above stripe should reject the symmetry thesis; the evidence for the resurrection, for them, is significant and non-symmetric evidence for good-god theism. Further, this gives them an answer to EG2: it is more reasonable for them to affirm good-god theism over evil-god theism because the evidence supporting the resurrection of Jesus favours the former and not the latter.

In this appendix, I have briefly sketched out several ways in which the good-god theist can reject the symmetry thesis and answer EG2. I do not claim that these responses are exhaustive, nor do I hold that all are equally plausible. However, it seems to me that the above sketches suffice to defang the symmetry thesis and EG2. I remind the reader again that they are sketches, and I do not pretend to have fleshed out all the relevant details.

Concluding remarks

We have seen that Law supports premise (2) with the problem of good, and that the problem of good fails. Hence, we have no reason to affirm premise (2): belief in evil-god is not unreasonable – at least for the reasons that Law gives. (However, it is important to note that one can accept that belief in evil-god is not unreasonable (or, at least, that it is not rendered unreasonable by the problem of good) and that (only) good-god exists. In other words, one can affirm a proposition p while also affirming that an argument against ~p (or argument for q which entails ~p) fails.) Thus, a good-god theist need not worry about the evil-god challenge. Now, this does not show that premise (2) is false, for there could be other ways to support it.Footnote 36 However, it does show that Law's challenge needs to be revised. Finally, we saw that Law's challenge turns out to rely on the falsity of sceptical theism, and therefore turns out to be more complicated and controversial than it initially appeared.Footnote 37

Footnotes

1. A good-god theist is a theist who affirms that God is omnipotent, omniscient, and omnibenevolent.

2. Evil-god is omnipotent, omniscient, and omnimalevolent.

3. Law construes the evil-god challenge in two different ways in his article. One way is as I have done above. However, in his abstract, Law states it differently, saying that ‘the challenge is to explain why the hypothesis that there exists an omnipotent, omniscient and all-good god should be considered significantly more reasonable than the hypothesis that there exists an omnipotent, omniscient and all-evil god’ (Law (Reference Law2010), 353). Call this version ‘EG2’. In other words, EG2 challenges the theist to justify their belief that God is good and not evil. In this article, I will not address this version of the challenge, except for in a brief appendix before the conclusion.

An anonymous referee objects to my passing over EG2, saying ‘if good-god theism is not [shown to be] more reasonable then it seems Law has won’. There are several issues here. First, as I pointed out above, Law states his challenge in two different ways. I do not regard it as a strike against me to only address one of the ways that he states it. Moreover, if the only thesis that a person establishes is that good-god theism is more reasonable than evil-god theism, then they still have a lot of work to do, for this conclusion is consistent with good-god theism nevertheless being unreasonable. To see this, consider the following. Suppose that a person showed that the thesis that the moon is made of cheese is more reasonable than the thesis that no human has ever set foot on the moon. Would the defender of Cheese Moon have made a significant accomplishment? No, for belief in Cheese Moon is nonetheless unreasonable. It seems to me, therefore, that EG2 is of secondary importance. (Though, again, I do address it in the appendix.)

Furthermore, if neither good nor evil-god theism are shown to be unreasonable by Law's argument (as I argue), then both versions of Law's challenge lose much of their bite; if neither position is unreasonable, or if both are reasonable, then it does not seem that the good-god theist must explain why their version of theism is significantly more reasonable than evil-god theism. That is, the good-god theist might be content knowing that their theistic belief is not unreasonable, and feel no pressure to explain why good-god theism is more reasonable than evil-god theism. The referee anticipates something like this objection, saying ‘this cannot be dismissed by admitting that both are equally reasonable, since it is a de facto truth of theology that almost all theists do believe in one and not the other, which requires justification from these theists’. Note, however, that I have not claimed that both are equally reasonable. In my article, I argue only that neither is rendered unreasonable by Law's argument. Hence, I am not committed to both being equally reasonable. Perhaps the referee would then enquire why the good-god theist accepts good-god theism over evil-god theism. To that question, I direct the reader to the appendix for a brief sketch of some possible ways in which the good-god theist could respond to this. (The appendix is not directed explicitly at answering this question. But it should be clear how the responses included in it could be given here as well.)

4. It is worth noting that at least one theodicy does not appear to be able to be reversed and used to respond on behalf of evil-god to the problem of good – or, if it can be, it is not obvious how it can be. The theodicy I have in mind is the Felix Culpa theodicy, most notably defended by Alvin Plantinga (Reference Plantinga and van Inwagen2004) (see too McNabb & Baldwin (Reference Tyler, Baldwin, Arbour and Gilhooly2017b) for a defence of it). Law briefly discusses and flippantly dismisses something akin to this theodicy, but he does not appear to be aware of Plantinga's recent powerful defence of it. (Law thinks that something like the Felix Culpa stands or falls on whether there was a literal Adam and Eve (Law (Reference Law2010), 358), but this a dubious assumption to make.)

5. But see the appendix for reasons that one may have to reject it.

6. Law (Reference Law2010), 357. How he knows how most people react when presented with the problem of good, we are not told. For my part, I have never once encountered a person who thinks that the problem of good is a powerful argument or strong evidence against evil-god. It is, therefore, unclear to me why Law says that most people think this way – has there been a study on this subject that he is privy to, or is he just assuming that this is the case since it would be convenient for his thesis? There are several places in Law's article where he seems to be trying to pass off his personal psychological inclination as the consensus view, and this is one of them.

7. This formulation is roughly parallel to Michael Bergmann's formulation of the evidential argument from evil. See Bergmann (Reference Bergmann, Clark and Rea2012, 11).

8. There are different kinds of immorally justificatory reasons. However, I will only focus on reverse-compensatory states of affairs here. A state of affairs reverse-compensates for G if it is sufficiently worse for the subject or subjects who benefited from G, or, perhaps, if it makes the world as a whole sufficiently worse. (Entailment relations between immorally justifying reasons are also important, but I will pass over them here. It is easy enough, however, to see how that would factor into my brief explanation (below) of the sceptical theist's scepticism.)

9. A noseeum inference (roughly) infers from our lack of knowledge of X, to the conclusion that probably not X. See Wykstra (Reference Wykstra and Howard-Snyder1996) for more on noseeum inferences.

10. Perhaps Law would argue that he does not rely on the noseeum inference from (4) to (5) to establish that there is no immorally justifying reason for G. Rather, he can just see that there is none. There are (at least) two problems with this response: (a) the fact that he can just see that there is no immorally justifying reason for G does not mean that anyone else can just see it. Thus, anyone who does not share his intuition is not going to be convinced by this. Consequently, this means that Law is not offering us an argument at all: he is merely reporting his personal intuition about the matter. Again, this will not convince anyone who does not already share his view – he is merely preaching to the choir. And (b) if it is acceptable for Law to say that he can just see that there is no immorally justifying reason for G, then there is no reason why the good-god theist cannot respond by saying that they can just see that there is no immorally justifying reason for G and that they can just see that there is a morally justifying reason for the evil in our world. Hence the symmetry thesis is false, and the evil-god challenge is avoided. In other words, if we allow just see intuitions into play, then anything goes, and this enables the good-god theist to circumvent Law's challenge. Thus, Law – if his challenge is to be, well, challenging – must rely on something other than a just see intuition. However, there is good reason to think that Law is using something like a noseeum inference, for, when discussing the problem of good, he examines and dismisses several possible ‘reverse theodicies’, and appears to think that this entails that there is (at least probably) no successful reverse theodicy; he seemingly concludes from our lack of knowledge of a successful reverse theodicy that there probably is no such reverse theodicy. (If he is neither relying on a noseeum inference, nor a just see intuition, then he should make clear what method he is using.)

11. In light of this, we can see that the evil-god challenge will only be as successful as the problem of evil: if a noseeum inference about values is unwarranted, then it will be unwarranted whether it pertains to morally or immorally justifying reasons.

12. See e.g. Wykstra (Reference Wykstra1984). Though, he has since softened his view, see Wykstra & Perrine (Reference Wykstra and Perrine2014).

13. See e.g. Bergmann (Reference Bergmann2001), (Reference Bergmann, Flint and Rea2009), and (Reference Bergmann, Clark and Rea2012) and Hud Hudson (Reference Hudson and Kvanvig2014) and (Reference Hudson2017). For another, newly developed version of sceptical theism, see Cullison (Reference Cullison2014). And see Dougherty & McBrayer (Reference Dougherty and McBrayer2014) for a good discussion of many issues pertaining to sceptical theism.

14. For different statements of the problem of evil, see e.g. Rowe (Reference Rowe1979), Draper (Reference Draper1989), and Michael Tooley's contribution to Tooley & Plantinga (Reference Tooley and Plantinga2008), 70–151.

15. To be clear, Hudson is addressing the problem of evil, not the problem of good, in his article.

16. Consider a different analogy: I own a 500-acre strawberry field and suspect that some of my strawberries are rotten. To alleviate my worry, I check 5 acres of my field, but find no rotten strawberries. From this, I am clearly not warranted in inferring that there probably are no rotten strawberries in my field. Obviously, if my strawberry field has an infinite number of acres, then my situation is even worse.

17. For a good discussion of the relevance of representativeness, see Bergmann (Reference Bergmann, Flint and Rea2009).

18. This is essentially Bergmann's second sceptical thesis. See Bergmann (Reference Bergmann2001).

19. For further arguments for the truth of the sceptical theist's position as advocated in these paragraphs, see e.g. Bergmann (Reference Bergmann2001) and (Reference Bergmann, Flint and Rea2009), and Hudson (Reference Hudson and Kvanvig2014b) and (Reference Hudson2017).

20. For statements and defences of proper functionalism, see e.g. Bergmann (Reference Bergmann2006), Boyce & Plantinga (Reference Boyce, Plantinga and Cullison2012), and Plantinga (Reference Plantinga1993a), (Reference Plantinga1993b), and (Reference Plantinga2000).

21. An appropriate mini environment is, roughly, a state of affairs in which the exercise of S's cognitive faculties in respect to p is not misleading. A maxi environment is just a more global version of the mini environment. For more on mini/maxi environments, see Boyce & Plantinga (Reference Boyce, Plantinga and Cullison2012, 132–133).

22. He takes this view, that unreasonable knowledge is possible, from Aarnio (Reference Aarnio2010).

23. In his article, Law does not claim that sceptical theism is false; he is only concerned with showing that it entails radical scepticism – at least for theists (he thinks that an atheist who embraces the sceptical component of sceptical theism can avoid radical scepticism). I have portrayed his criticism as an argument against sceptical theism merely for the sake of convenience.

24. Of course, this does not entail that the sceptical theist's common-sense beliefs are trustworthy. The point is that Law is the one claiming that they are not (or that the sceptical theist should think that they are not) trustworthy, so it is up to him to show this. But he has not shown this. Not by a long shot.

25. If this example strikes the reader as too controversial, then frame it instead in terms of a book entitled Dogs that is about dogs. Surely, we can know from the title (and perhaps the table of contents) that Dogs is a book about dogs.

26. There is much more to be said on this topic, but this is a response to Law's evil-god challenge, not to his views on sceptical theism, so I will stop here. It has also been drawn to my attention that my reply to Law here is similar to the reply that McNabb and Baldwin (Reference McNabb and Baldwin2017a, esp. 300–302) offer.

27. More exactly, the symmetry thesis will have to be rejected in a certain way: it must be shown that good-god theism has significantly more evidential support – is significantly more reasonable – than evil-god theism.

28. See also Nagasawa (Reference Nagasawa2008) for reasons to doubt that the ontological argument can be parodied in support of an evil god.

29. It should be noted that this is Alston's view of internalist justification. In Alston (Reference Alston1991), he devotes far more space to defending an externalist view of justification.

30. The goal of sense perception as a doxastic practice, for example, is to help one navigate the world.

31. The bulk of his book is spent arguing that the Christian mystical doxastic practice enjoys stronger, externalist justification.

32. If one accepts Alstonian justification, then they may reject the symmetry thesis even if they do not themselves partake in the Christian mystical practice. This is because one need not partake in the Christian mystical practice to recognize that (a) the Christian mystical practice has Alstonian justification and (b) there is no parallel practice that evil-god theists partake in.

33. For various issues pertaining to phenomenal conservatism, see Tucker (Reference Tucker2013).

34. Indeed, as Tucker (Reference Tucker, Clark and VanArragon2011) points out, phenomenal conservatism makes evidence extremely easy to acquire. Hence it makes it extremely easy for good-god theists to reject the symmetry thesis and answer EG2.

35. Very roughly meaning that the belief has epistemic warrant and is basic (i.e. not inferred from argument or reason). See Plantinga (Reference Plantinga1981) and (Reference Plantinga2007) for more on proper basicality.

36. Perhaps, for example, Law could mirror the argument from horrendous evils and argue that particularly glamorous goods make the existence of evil-god improbable. However, it is my view that this argument would only be as successful as the argument from horrendous evils; that is, it would not be successful at all.

37. Thanks to Ben Bavar, Cameron Bertuzzi, Brett Lunn, and Wade Tisthammer for comments on earlier drafts of this article. And thanks especially to G.L.G. – Colin Patrick Mitchell – for particularly insightful comments.

References

Aarnio, Maria (2010) ‘Unreasonable knowledge’, Philosophical Perspectives, 24, 1–21.Google Scholar
Alston, William (1991) Perceiving God: The Epistemology of Religious Experience (Ithaca: Cornell University Press).Google Scholar
Alston, William (1994) ‘A philosopher's way back’, in Morris, T. (ed.) God and the Philosophers (New York: Oxford University Press), 19–30.Google Scholar
Bergmann, Michael (2001) ‘Skeptical theism and Rowe's new evidential argument from evil’, Nous, 35, 278–296.Google Scholar
Bergmann, Michael (2006) Justification without Awareness: A Defense of Epistemic Externalism (Oxford: Oxford University Press).Google Scholar
Bergmann, Michael (2009) ‘Skeptical theism and the problem of evil’, in Flint, T. & Rea, M. (eds) The Oxford Handbook of Philosophical Theology (Oxford: Oxford University Press), 375–399.Google Scholar
Bergmann, Michael (2012) ‘Common sense skeptical theism’, in Clark, K. & Rea, M. (eds) Reason, Metaphysics, and Mind: New Essays on the Philosophy of Alvin Plantinga (New York: Oxford University Press), 9–37.Google Scholar
Boyce, Kenny & Plantinga, Alvin (2012) ‘Proper functionalism’, in Cullison, A. (ed.) The Continuum Companion to Epistemology (London: Bloomsbury Academic), 124–141.Google Scholar
Cullison, Andrew (2014) ‘Two new versions of skeptical theism’, in Dougherty & McBrayer (2014), 250–263.Google Scholar
Dougherty, Trent & McBrayer, Justin (eds) (2014) Skeptical Theism: New Essays (Oxford: Oxford University Press).Google Scholar
Draper, Paul (1989) ‘Pain and Pleasure: an evidential problem for theists’, Nous 23, 331–350, reprinted in D. Howard-Snyder (ed.) The Evidential Argument From Evil (Bloomington: Indiana University Press, 1996), 12–29.Google Scholar
Hudson, Hud (2014a) The Fall and Hypertime (Oxford: Oxford University Press).Google Scholar
Hudson, Hud (2014b) ‘The father of lies?’, in Kvanvig, J. (ed.) Oxford Studies in Philosophy of Religion: Volume 5 (Oxford: Oxford University Press), 147–166.Google Scholar
Hudson, Hud (2017) A Grotesque in the Garden (n.p.: Xerxes Press).Google Scholar
Huemer, Michael (2007). ‘Compassionate phenomenal conservatism’, Philosophy and Phenomenological Research, 74, 30–55.Google Scholar
Law, Stephen (2010) ‘The evil-god challenge’, Religious Studies, 43, 353–373.Google Scholar
Law, Stephen (2015) ‘The Pandora's box objection to skeptical theism’, International Journal for Philosophy of Religion, 78, 285–299.Google Scholar
Licona, Michael (2010) The Resurrection of Jesus: A New Historiographical Approach (Downers Grove: InterVarsity Press Academic).Google Scholar
McNabb, Tyler & Baldwin, Erik (2017a) ‘Divine methodology: a lawful deflection of Kantian and Kantian-esque defeaters’, Open Theology, 3, 293–304.Google Scholar
Tyler, McNabb, & Baldwin, Erik (2017b) ‘From theology to theodicy: a defense of Felix Culpa’, in Arbour, Benjamin & Gilhooly, John R. (eds) Evil and a Selection of Some of its Theological Problems (Newcastle upon Tyne: Cambridge Scholars Publishing), 162–172.Google Scholar
Nagasawa, Yujin (2008) ‘The ontological argument and the devil’, The Philosophical Quarterly, 60, 72–91.Google Scholar
Plantinga, Alvin (1981) ‘Is belief in God properly basic?’, Nous, 15, 41–52.Google Scholar
Plantinga, Alvin (1979) The Nature of Necessity (New York: Clarendon Press).Google Scholar
Plantinga, Alvin (1993a) Warrant and Proper Function (New York: Oxford University Press).Google Scholar
Plantinga, Alvin (1993b) Warrant: The Current Debate (New York: Oxford University Press).Google Scholar
Plantinga, Alvin (2000) Warranted Christian Belief (New York: Oxford University Press).Google Scholar
Plantinga, Alvin (2002) ‘Reply to Bielby's cohorts’, in Bielby, J. (ed.) Naturalism Defeated: Essays on Plantinga's Evolutionary Argument against Naturalism (Ithaca NY: Cornell University Press), 204–276.Google Scholar
Plantinga, Alvin (2004) ‘Supralapsarianism; or “O Felix Culpa”’, in van Inwagen, P. (ed.) Christian Faith and The Problem of Evil (Grand Rapids: Eerdmans Publishing Company), 1–26.Google Scholar
Plantinga, Alvin (2007) ‘On proper basicality’, Philosophy and Phenomenological Research, 75, 612–621.Google Scholar
Rowe, William (1979) ‘The problem of evil and some varieties of atheism’, American Philosophical Quarterly, 16, 335–341.Google Scholar
Swinburne, Richard (2003) The Resurrection of God Incarnate (Oxford: Oxford University Press).Google Scholar
Tooley, Michael & Plantinga, Alvin (2008) Knowledge of God (Malden: Wiley-Blackwell).Google Scholar
Tucker, Chris (ed.) (2013) Seemings and Justification: New Essays on Dogmatism and Phenomenal Conservatism (Oxford: Oxford University Press).Google Scholar
Tucker, Chris (2011) ‘Phenomenal conservatism and evidentialism in religious epistemology’, in Clark, K. & VanArragon, R. (eds) Evidence and Religious Belief (New York: Oxford University Press), 52–76.Google Scholar
Wykstra, Stephen (1984) ‘The human obstacle to evidential arguments from suffering: on avoiding the evils of “appearance”’, International Journal for Philosophy of Religion, 16, 73–94.Google Scholar
Wykstra, Stephen (1996) ‘Rowe's noseeum arguments from evil’, in Howard-Snyder, D. (ed.) The Evidential Argument from Evil (Bloomington: Indiana University Press), 126–150.Google Scholar
Wykstra, Stephen, & Perrine, Timothy (2014) ‘Skeptical theism, abductive atheology, and theory versioning’, in Dougherty & McBrayer (2014), 142–163.Google Scholar