Tuesday, December 6, 2022

The Contradictory God - Five Incompatible Properties Arguments against Orthodox Theism

Note; This was originally a reddit post from nearly a year ago. I've made some edits to it and added some stuff. 

Introduction

This is the second in a collection of posts I plan to make in arguing against orthodox theism. The first post was a defense of a logical problem of evil, originally conceived by Richard R. La Croix. In this post, I will be presenting and defending a set of arguments for the incoherence of orthodoxly conceived theism, known as incompatible properties arguments. What the heck is an incompatible properties argument, you ask? An incompatible properties argument is an argument that seeks to demonstrate the falsity of orthodox theism via proof by contradiction in how God is conceived. An argument of this sort will take a form which looks something like this;

1. If God exists then He exemplifies properties X and Y
2. It's not possible to jointly exemplify properties X and Y
3. Therefore God does not exist

It's important to point out here that not all theists will have the same commitments. As such, a real limitation of incompatible properties arguments is that they will target only certain conceptions of God. A theist might well avoid these kinds of arguments by affirming that God does not possess one or both of the properties in question. This would involve a denial of what would be premise 1. However, the theist must tread carefully here. For they may only be able to deny that God has a certain property on pain of giving up certain widely held religious beliefs about God, as well as a motivated conception of God. I take it that only God as conceived by classical or neo-classical theisms have philosophical justifications. Additionally, if the theist goes too far in jettisoning God's attributes, then God may no longer be a being that is apt to be worthy of worship. In that case, it's not clear that such a being would count as God at all.

However, most responses to incompatible properties arguments will involve a denial of what would be premise 2. It is objections of this kind that I will address in my presentation of the following arguments.

To close the introduction off, I will first note that the idea of this post is inspired by Theodore Drange's 1998 article
[1].  I will not be including most of the arguments Drange presents in this discussion, and will include some not mentioned by Drange. Secondly, I will not be defending single attribute arguments against theism, such as the stone paradox and Patrick Grim's proofs for the incoherence of omniscience. I will only focus on arguments that purport to show that 2 or more of God's attributes are incompatible.


Argument 1 - Timelessness and Personhood

The God of classical theism is taken to be a timeless person, and indeed most religious conceptions see God as a timeless person. The following is an incompatible properties argument for the incoherence of a timeless person;

1. If God exists, then there is a timeless person
2. All persons have a succession of mental states in time
3. A timeless being cannot have a succession of mental states in time
4. Thus a timeless person does not exist (2, 3)
5. Thus God does not exist (1, 4)

As mentioned, premise 1 is analytically entailed given the classical theist conception of God. Importantly, some non-classical theists will not accept 1, in which case this argument will be ineffective for them. Premise 3 is just an analytical entailment from what it is to be timeless. So, premise 2 is the only premise that is up for dispute. But there is a case to be made that it seems overwhelmingly plausible. Upon reflecting on what it is to be like a person, it seems that it presupposes temporality, it is to have a sequence of phenomenological experiences.

For instance; at one moment, I am typing this paragraph and having that experience, and at another moment I am not and that experience becomes a memory. When I have thoughts, at a prior moment I think about some proposition or state of affairs, such as what I'll be eating tonight, and then at a later moment I think about some other proposition or state of affairs such as what my favorite video game is. Thinking inferentially, such as forming the line of inference "if P then Q, P therefore Q" seems to be a temporal sequence of mental events, we have the prior mental event of my thinking about the premises, and then we have a later mental event, where I derive the conclusion. I might change my beliefs as a result of inference-making, in which case there was a moment where I fail to believe P and a later moment where I believe P (P being a generic stand-in for some proposition or other). When I actualize my desires, that seems to be a temporal process, there is a moment when my desire is not realized, and then a later moment when I take some action to make it the case that my desire is realized. These are all features of my personhood which I take it are inextricably linked to temporality. While this may not be particularly convincing to theists, and while it might be argued that the premise may not be a necessary truth - I think my confidence in it is quite strong, as my understanding here stems from an introspection of my own internal phenomenology, and there isn't much I can be more sure of then that! Given how I understand my phenomenology on a very fundamental level, I literally cannot conceive of what it would be like to be a person with no sequence of phenomenal states in time. It is constitutive of my very understanding, and experiential knowledge of what it is, and what it's like to be a person. But perhaps I'm just being incredulous, perhaps you can conceive it, perhaps you have a very different internal phenomenology than I do. If so, please do let me know in the comments!

The common objection from the classical theist here will be to point out that God's being a person is merely an analogical predicate. That is, God is not a person in the same way we are persons He does not have a succession of mental states. When we refer to God as a person we are referring to him by the doctrine of analogy. The issue is that for something to be an analogy, there has to be common features. If I tell you that last night I had a burger, but then I tell you that it's not a burger in any sense you understand the term, but it was still a burger in an analogical sense. I have a feeling you would give me an odd look and rightly believe I was speaking gobbledygook. This is because for something to be analogical to a burger, while there may be some sense in which it is unlike a burger, there is also some sense in which it is like a burger. So, in the same way, in order for God's personhood to be an intelligible attribution, there has to be some feature it has in common with our human understanding of personhood. Yet, as mentioned, it seems that, constitutive of our human understanding of personhood, is a notion of temporal order and continuance of mental states through time, so it's not clear what that would be. If the theist wants to say that God is analogically a person, it's on them to intelligibly explain what common features it shares with the personhood humans possess, otherwise, they aren't really doing an analogy at all are they?

Argument 2 - Omnipresence and Personhood

Another incompatible properties argument that utilizes God's attribute of personhood, is one that pits it against God's attribute of omnipresence[2]. It can be formulated in the following way;

1. If God exists, then an omnipresent person exists
2. Necessarily, persons have intentionality
3. To have intentionality, is to be able to perceive external objects independent of the self
4. An omnipresent being, cannot perceive external objects independent of the self
5. Thus an omnipresent person does not exist (2, 3, 4)
6. Thus God does not exist (1, 5)

Premise 1 is an analytical entailment, 2 is also an analytic truth. So, what needs to be discussed is 3 and 4. The basic idea here is an appeal to content externalism, or related views. A feature of intentional states is that they are content-full mental states, they are representational, they are about things. We don't merely desire, we desire something, we don't merely have beliefs, we believe propositions. So there must be something external to our consciousness which furnishes mental content, but for an omnipresent being there is no external content. The theist may deny externalism and affirm that God's mental states can just be about other mental states in God's mind, but I think this seems insufficient for intentionality.

To see why, let's lay some groundwork by considering Kant's famous refutation of material idealism. Kant argues that we cannot doubt the external world without doubting the consciousness of our own existence in inner sense. The outer sense is how we represent things (e.g. empirical objects) as distinct from us in space, and the inner sense is where our mental states and representations reside such as thoughts and sensations. The idea here is that the way we distinguish ourselves from other objects is by comparing our experience to external objects outside of our consciousness. If there is no external content, then we could not in principle grasp that mental states are representations of something different from itself. So we must be able to make a conceptual distinction between the self and external objects or else we cannot form judgements about objects.

To put it another way, the thought is, higher order consciousness and intentionality includes the ability to form judgements about objects, and the ability to discern the difference between an object and our representation of it. For example I must recognize the difference between my mental representation of my keyboard, how I perceive it, and the keyboard itself, the properties it has etc. But this presupposes the ability to grasp a conceptual distinction between the self and the non-self, I could not judge the keyboard as an object with certain properties if I do not have an understanding of the keyboard independent of my mental representation of it. The issue being, an omnipresent being cannot make such a conceptual distinction, because for an omnipresent being there is no non-self. Now, it has been criticized that Kant's Transcendental argument goes too far here, and that all Kant establishes is that we must be able to conceive of our experiences as being related to external objects and events, not that they actually are. However, this more modest conclusion still threatens theism. God can not think of objects as external from Himself if they are not, as that would involve a false belief and God by definition cannot have false beliefs.

Once again the theist might object that God's having intentionality is merely analogical, but this runs into the same issue discussed in the previous section. If a being cannot form judgements about objects, it's very unclear that it would be right to say such a being is an intentional one at all. That's just how I understand intentionality. I leave it to the theist to give an intelligible account of God's supposed analogical intentionality.

The theist may object that even if God is everywhere He can still distinguish between parts of Himself, after-all, surely I can distinguish between my hands as parts of myself and recognize the distinction between my mental representation of my hands and my hands themselves. Firstly, it's unclear that God can be composed of parts, the traditional view is that God is purely simple. But more to the point, it seems that in order to make judgements about my hands, indeed in order to judge them as *my* hands, and not someone else's hands, I must be able to distinguish between them and other objects in the world which are not me. Further, I must be able to judge my hands as objects with properties, which are external to my consciousness, thus there is still an object-self distinction involved.

One last objection I will consider is that the theist could give a modified version of omnipresence, God is omnipresent not in virtue of being located everywhere and the whole of reality being in Him or part of Him, but perhaps He is co-present with everything in reality. Or perhaps He is omnipresent insofar as He can be anywhere at once.  We are still able to run a similar argument without using omnipresence. For, another essential property of God is that He is the ex nihilo creator of the world, thus He existed prior to creation with no external world to constrain His mental contents, or for His thoughts to be directed at. We can formulate the argument like so; 

1. If God exists, then there was a person prior to the creation of an external world.
2. Necessarily, persons have intentionality.
3. To have intentionality, is to be able to perceive external objects independent of the self.
4. Prior to the creation of the external world, it is not possible to perceive external objects independent of the self.
5. Thus there cannot be a person prior to the creation of an external world (2, 3, 4)
6. Thus God does not exist. (1, 5) 


Argument 3 - Omniscience, Omnipotence and Personhood

One more incompatible properties argument that focus's on God's personhood is one that uses two of God's "omni" attributes, His omnipotence, and omniscience. It can be formalized as follows;

1. If God exists, then an omnipotent, omniscient person exists
2. Necessarily, persons are intentional agents
3. Intentional agency entails uncertainty and/or the possibility of failure
4. An omnipotent, omniscient being cannot be uncertain and/or fail
5. Thus an omnipotent, omniscient being cannot be an intentional agent (2, 3, 4)
6. Thus God does not exist (1, 5)

1, 2 and 4 are true by definition. 3 is the only controversial premise. There are a couple reasons to accept 3. One reason is, it seems intentional states are things one can be in or out of accord with (where out of accord with is what is meant by "failure"). This seems especially true in the case of desires, if there is a blizzard and I desire to not be cold, then I acknowledge that there is the possibility that I can be out of accord with this desire by experiencing cold sensations. We can say that if I go outside without the proper winter attire then I am failing to live up to my commitment as a rational agent. Indeed plausibly part of how we understand the concept of desires, is that they have possible conditions of satisfaction and dissatisfaction. If there is no possibility of being out of accord with one's desires, that is if some behavior is strictly governed by the fact that it could not have been out of accord with it's nature or something such like, then it seems to reduce to a mere causal tendency, such as a calculator following inputs, or a rock falling down a hill. At the very least, it's not without some plausibility that, part of what distinguishes intentional acts from mechanical events is that intentional acts involve responsiveness to normative reasons that one can be in or out of accord with. That is to say, intentional states are unlike dispositional states in that intentional states have a normative, or more precisely, rule-following component built in. (I also suspect that many if not most theists especially will be uncomfortable with reducing intentionality to mere causal tendencies, since theists tend to like libertarian free will). However, in the case of God, it doesn't make sense to speak of conditions of satisfaction, there is no possibility of being out of accord with His desires. I
n virtue of God's omnipotence His will is unfailingly efficacious, so God "desires" X, logically entails that X obtains or is the case. The theist can conceivably bite that bullet however, and accept that intentional states can reduce to mere causal tendencies or entailment relations, as bizarre as that initially seems. Perhaps more convincingly, then is it seems that in order to actualize our desires, we must pick out a state of affairs that is not currently in accord with our desires, and then take some action to make it the case that the state of affairs accords with our desire. For example, suppose I desire to lose weight, what that means is that I recognize that the current state of affairs is that I am more overweight then what I want to be, and so I must take some action, say exercise and diet, in order for the state of affairs to satisfy or to be in accord with my desire to weigh less. At the very least, I must be able to perceive that the state of affairs diverges from my goal/desire.


But God is omnipotent, so the only constraint on His power is logical possibility. God is omniscient, so I take it this means that God knows what He can achieve, and He cannot have non-rational desires, such as the desire to actualize logical impossibilities. Additionally, if God's desires are not in sync with wh
at He can do, and if He can have conflicting desires, then it seems He has a less than perfect will which is surely not how God is conceived. Thus it seems, for any time t, and any possible state of affairs which is the object of God's desire S, God is omnipotent omniscient and desires S at t. So all the necessary and sufficient conditions for S's obtaining are met at t since God desire
s S, and God necessarily succeeds in actualizing His desires. But for any time when the necessary and sufficient conditions for S obtains, then S necessarily obtains, thus S necessarily obtains at t. So it seems there can be no state of affairs that fails to be in accord with God's desires, and in virtue of God's omniscience/infallibility He cannot falsely perceive states of affairs as out of accord
 with His desires. Thus it seems like, since there is no possibility of God being out of accord with His desires, God cannot act on His desires, since any object of desire He would have, necessarily, could not be unactualized at any point in which He has the desire and He cannot wrongly perceive states of affairs as being out of accord with His desires in virtue of His infallibility. If we take it, as is plausible, that acting towards ones desires is a necessary constitutive feature of agency, then this is a problem. 


We can also argue for 3 on the grounds that intending requires that one is uncertain about the future[3]. Intending involves a process of deliberation, weighing normative reasons, reflecting on which of ones desires is a good reason to act. However, when one is deliberating their mind is not yet made up, they are in a state of uncertainty about what they are going to do, otherwise there would be no point in deliberating. But an omniscient being cannot be uncertain. Further, deliberating between options seems to entail at-least having a belief that both options remain open to you. In other words, we view the state-of-affairs of our Φ-ing or not Φ-ing at t as epistemically contingent. However, if an omniscient being knows that she will Φ at t, then she infallibly knows that not Φ-ing at t is not an open option for her. The theist might object by positing open theism and suggest that there is no truth value for future contingents, however there is a worry that open theism is incompatible with anselmian conceptions of God and undermines widely held views about God's absolute sovereignty, and,  it by my lights seems obvious that propositions about future contingents are true or false (Although that's a whole other topic which we can put aside). The theist might object that God is timeless so God cannot believe that He Φ's at t. But this does not diffuse the argument, if 'God Φ's' is true, then it is true now that God Φ's. But if it is true that God Φ's now then it could be true that God Φ's at t. (These are all compatible with His being timeless).


Argument 4 - Omniscience and Timelessness

Next we have an incompatible properties argument which utilizes the divine attributes of omniscience and timelessness[5]Here is the formulation;

1. If God exists, then an omniscient and timeless being exists
2. A being that is omniscient knows what time it is at present
3. A being that always knows what time it is at present cannot be timeless
4. Thus a timeless, omniscient being does not exist (2, 3)
5. Thus God does not exist (1, 4)

I think this is one of the weaker arguments, and in fact I think the argument can be safely diffused by rejecting presentism about time, as there would be no "what time it is at present". But for the theist committed to presentism, it seems both 1 and 2 are true. The thought behind 3 is, in order to know what time it is, or to know all time-indexed propositions, one must be situated in time, and must know different things at different times. But a timeless being, by definition, cannot be situated in time. Let's consider some objections the presentist theist might levy.

The theist might try to object that, God always knows what time it is, but there is only a changing succession of time in God's object of knowledge not God Himself. This is a mistake though, if God knows that it is such and such time, and the state of the universe changes such that it is no longer such and such time, then God must change His belief to be in accord with what the time is otherwise He will have a false belief.

The theist might object that God can have knowledge of all temporal states of the universe at once and not successively. However, given presentism, it seems that having knowledge of all temporal states is not sufficient for omniscience. God may know the time of the heat-death of the universe, when I will die, what I will do tomorrow, what you will do tomorrow, but on this view God cannot know what it is I or you are doing presently, because what I or you are doing presently constantly changes. That is, He cannot know, as I do, that in the present I am typing this paragraph, or as you know, that you are reading it now, for all He knows the present is thousands of years in the future far past both our deaths. Thus it seems an omniscient being cannot merely know the entirety of the universes temporal states, He must also know what time it is presently, otherwise we would be admitting that I, a non-omniscient knower, can know something an omniscient being cannot, but this seems absurd.

The theist might take omniscience to be knowing everything that is logically possible for God to know, and since it is logically impossible for God to know what time it is in virtue of His timelessness, God cannot know what time it is and still be omniscient. The problem is this runs into reductios such as Michael Martin's McNose objection[4], suppose I define a McNose as a being that logically necessarily can only know how to scratch it's nose, and I further stipulate that the McNose has perfect knowledge of how to scratch it's nose. Since a McNose knows everything that is logically possible for it to know, it follows that the McNose is omniscient despite only knowing how to scratch it's nose, but surely that is absurd. The theist might then modify there conception of God's omniscience to knowing everything which does not otherwise decrease God's perfection. Since timelessness is a perfection, God cannot know what time it is. However, this seems like an ad hoc move designed to avoid the objection, there doesn't seem to be any independent motivation for such a conception of omniscience. Further, the condition "which does not otherwise decrease God's perfection" is not one that is necessary for maintaining the coherence of omniscience. Thus, absurdly, it seems this entails that we can conceive of a more knowledgeable being, than an omniscient being.

Argument 5 - Omnipotence and Essential Moral Perfection

The final incompatible properties argument that I will present will be one that argues for the incompatibility of omnipotence and moral perfection. Specifically, necessary moral perfection, which is the attribute of being morally perfect in all possible worlds. This attribute is essential to the MGB (Maximally Great Being) conception of God which entails God's having maximal excellence (which includes moral perfection) in every possible world. Following Wes Morriston[6] the argument can be formulated in the following way, let E be some all-things-considered evil state of affairs such that a morally perfect being would not actualize it;

1. If God exists then God is necessarily morally perfect and omnipotent.
2. If God is necessarily morally perfect, then there is no possible world in which He actualizes E.
3. If God is omnipotent, He has the power to actualize E.
4. If God has the power to actualize E, then there is a possible world in which God actualizes E.
5. Therefore God cannot be necessarily morally perfect and omnipotent. (2, 3, 4)
6. Therefore God does not exist. (1, 5)

The theist may object to 2 on the grounds that while God may not be able to strongly actualize E, He can still weakly actualize E. Strong actualization means God directly causes some state of affairs to obtain with His actions. Weak actualization is a bit more complicated if you are not familiar with Plantinga, but the basic idea is that God indirectly causes some state of affairs to obtain by actualizing a state of affairs where a contracausally free creature is in a situation where she freely chooses to cause some other event. In any case, as Morriston points out, we can revise the premise to "If God is necessarily morally perfect, then there is no possible world in which He strongly actualizes E." and the worry remains. If I tell you to lift a box for me because I am not strong enough to lift the box, and this causes you to lift the box, I weakly actualize the state of affairs of you lifting the box, but it's still the case that you have superior box-lifting power than me. So it's clear weak actualization in this case, is not sufficient for omnipotence.

A theist might deny 3 and construct a conception of omnipotence that is such that a being is omnipotent if they are able to actualize any logically possible state of affairs that is consistent with it's nature. The problem is that this runs afoul of the famous McEar objection (Similar to the McNose objection above), suppose we have a being that is so defined such that it's nature is that it can only scratch it's ear, it cannot tie it's shoes, it cannot lift a feather, it can do only one thing namely scratch it's ear. Yet it can do anything that is logically compatible with it's nature, thus it follows, absurdly, that the McEar is omnipotent. There have been replies to the McEar objection most notably, "But it might be replied that an agent such as McEar is impossible. It can be cogently argued that, necessarily, if McEar has the power to scratch his ear, then he also has the power to move a part of his body to scratch his ear, for instance, his arm (Wierenga 1989, pp. 28–29)". But this doesn't seem particularly convincing, one can reconstruct the problem by adding that the McEar can necessarily only scratch it's ear and whatever necessary and sufficient conditions must obtain for scratching it's ear. Further it doesn't seem to engage with the heart of the issue. The problem that the McEar objection underlies is that just because something is impossible for you, doesn't mean that it is impossible tout court. It seems like an omnipotent being must be able to do anything that is logically possible for someone or other to do. A case can be made that omnipotence at time t requires being able to actualize any logically actualizable state of affairs at t.

Finally, the theist can dispute 4 on the basis that, God cannot actualize E but not by any lack of power on His part. If He (per impossible) wanted to actualize E He would succeed. But in virtue of Him being morally perfect across all possible worlds, there is no possible world where He would ever bring Himself to actualize E. This view still lends itself to reductio's. Suppose we imagine a being, let's call it a McCheese, a McCheese is extraordinarily powerful, so powerful such that if it decided to actualize some state of affairs, it will always succeed (Morriston calls this conditional power). However, the McCheese is in a state of mental paralysis, such that it can only choose one thing, to eat cheese. It seems clearly that the McCheese is not omnipotent, yet absurdly, since it has maximal conditional power, there is nothing to suggest it is not on this analysis. Thus it seems, the theists objections all seem to make us view beings as omnipotent that we wouldn't otherwise take to be omnipotent.


Conclusion

These are 5 of among, I think, the strongest incompatible properties arguments against orthodox theism. There are many other incompatible properties arguments that I did not cover here, for instance there are worries of omniscience being incompatible with moral perfection, after-all it seems an omniscient being would know what it's like to sin, or to have perverted thoughts, or to be envious. There are worries that God cannot be timeless and the efficient cause of the universe, since it seems causation is a temporal notion. These too, are concerns I have some sympathies with. Overall, I think a priori deductive arguments such as these aren't going to tend to be the most convincing for theists, however, they succeed at least in raising my credence in atheism. They may serve as a symmetry breaker for the possibility premise in modal ontological arguments, since if we have a reason to think God's attributes are contradictory then we have a reason to think the possibility premise is false. They also might be effective in terms of blocking stage 2 cosmological argument inferences, as we'd always have a strong a priori reason to favor a necessary being that isn't incoherent, or against Bayesian fine-tuning arguments, since if one's prior probability for theism is 0 then Bayesian fine-tuning arguments are dead on arrival.I


Inspirations & Further Reading

[1] Theodore Drange - Incompatible-Properties Arguments: A Survey
[2] Matthew Mccormick - Why God Cannot Think: Kant, Omnipresence, & Consciousness
[3] Tomis Kapitan - Agency and Omniscience
[4] Michael Martin - Atheism: A Philosophical Justification
[5] Norman Kretzmann - Omniscience and Immutability
[6] Wes Morriston - Omnipotence and necessary moral perfection: are they compatible?

66 Comments

Resurrecting the Logical Problem of Evil

 Note; This was originally a reddit post from around a year ago. Don't read it, I think it's bad and this one is an improvement. I realize now that I've made serious errors in my original post, so I've done some semi-major editing. Further feedback is appreciated 

I will start with some preliminary remarks. Firstly, it is important to note that this argument is not completely my own, this post is heavily inspired by Richard R. La Croix's paper Unjustified Evil and God's choice, and by J.L Schellenbergs paper A New Logical Problem of Evil. Which are highly recommended reads. I've made some modifications to the argument, and made it more accessible. Second, I am not a professional philosopher, I am a college drop out who is interested in philosophy of religion. Do not present this post to theists with the implication that it has decisively refuted theism. Chances are, there are responses to this type of argument in the literature, and then responses to those responses, etc. Philosophy is hard and, believe it or not, it is not a video game! Rarely has any position been "won" through a single argument.

Introduction

Nowadays logical formulations of the problem of evil are fairly unpopular. A common talking point among theists is that the logical problem of evil is dead. And many, if not most, contemporary atheologians tend to favor evidential arguments from evil in favor of logical arguments (If you aren't aware of the distinction between logical and evidential formulations of the argument from evil, it may be worth it to check here and here). William Rowe, Michael Tooley, and Paul Draper, just to name a few. But why is that? Some insight from Alvin Plantinga who is famously credited for having refuted J. L. Mackie's logical problem of evil will be of use here. In God and other Minds he writes;

The authors referred to above take the following five propositions to be essential to traditional theism: (a) that God exists, (b) that God is omnipotent, (e) that God is omniscient, (d) that God is wholly good, and (e) that evil exists. Here they are certainly right; each of these propositions is indeed an essential feature of orthodox theism. And it is just these five propositions whose conjunction is said by our atheologians to be self-contradictory. The first point to note is that of course these five propositions do not by themselves formally entail a contradiction; to get a formally contradictory set the atheologian must add some proposition or other. But of course he cannot add just any proposition he pleases. What conditions must be met by the proposition he adds (which I shall call (f)) if his accusation is to be made good? First, the conjunction of (f) with (a)-(e ) must formally entail a contradiction. But what further condition must it meet? If (f) were necessarily true then (a)-(e ) would formally entail the denial of (f); and perhaps we could say of any proposition which formally entails the denial of a necessarily true proposition that it is self-contradictory, at any rate in a broad sense of that term. On the other hand, if (f) were an essential part of theism, then, although it would not follow that there is a contradiction in (a)-(e), there would be one in some larger set of beliefs accepted by any theist. So to make good his claim the atheologian must provide some proposition which is either necessarily true, or essential to theism, or a logical consequence of such propositions. And none of the atheologians I quoted above seems to have realized the difficulty of that task.


In other words, logical problem of evils seek to show that orthodoxly conceived theism is self-contradictory. So, the difficulty here is finding a set of propositions, that are either necessarily true or that a theist is committed to, which together jointly entail a contradiction. This is by no means an easy task, which can explain why many atheologians aren't optimistic about it's chances. To see more clearly why this is no easy task, let's look at the following standard formulation of a logical problem of evil taken from Michael Tooley's fantastic SEP article on the Problem of Evil;

1. If God exists, then God is omnipotent, omniscient, and morally perfect.

2. If God is omnipotent, then God has the power to eliminate all evil.

3. If God is omniscient, then God knows when evil exists.

4. If God is morally perfect, then God has the desire to eliminate all evil.

5. Evil exists.

6. If evil exists and God exists, then either God doesn’t have the power to eliminate all evil, or doesn’t know when evil exists, or doesn’t have the desire to eliminate all evil.

7. Therefore, God doesn’t exist

This argument is certainly valid, if the premises are true then the conclusion necessarily follows. However, are all the premises necessarily true, or commitments of theism? It seems not, 4 is not a commitment of theism nor is it a necessary truth. In fact, it's unclear why, given no argument, a theist would be compelled to accept 4 at all. Some evils might entail greater goods we aren't aware of and so God would permit such evils for the sake of greater goods which entail them, such an answer may not be likely or even plausible but all the theist needs is that it is logically possible and they've at-least avoided the logical problem of evil.


The question remains; Can Plantinga's challenge to the atheologian be met? Is there a set of propositions that are necessarily true, or commitments of theism which jointly entail a contradiction? I suggest that perhaps there is.

The Argument

From here I shall present my preferred logical argument from evil, and give some brief comments and justification for the premises.

(1) Evil Exists

This premise is relatively uncontroversial among both theists and atheists. At-least a commitment of Abrahamic theism is that sin exists and sin is evil. However, it may be noted that a theist could be tempted to deny this premise by suggesting that evil is merely a privation of good. But for one, this view is implausible. Surely pain is not merely a lack of some good, it is positively bad. It is bad, not in virtue of a privation, but because of it's positive phenomenal content. It is bad because humans, or at-least rational self-interested agents recognize pain as states that are intrinsically undesirable, what it's like to be brutally tortured makes me, and presumably other self-interested humans desire to avoid being in those states. The view also has problems with accounting for moral obligations, for a couple reasons. First it doesn't logically follow that if something is not good, or lacks good-making properties that I'm morally obligated to not do it. Second, it seems to be subject to a problem of moral demandiness, it's not good to choose not to be a doctor, but surely I'm not morally obligated to choose to be a doctor.

Finally, and perhaps most importantly, even if evil is a privation of good, we can still run the same argument, as a logical argument from the privation of good. So it's clear a privation of good theodicy won't do.

(2) God exists

Obviously, a theist is committed to this premise.

(3) If God exists then God is omnipotent.

This is an analytic truth. In other words, it is true in virtue of the definition of God. The following premises, as well, are analytically true;

(4) If God exists then God is omniscient.

(5) If God exists then God is morally perfect.

Slightly more controversially,

(6) If God exists then God is maximally free.

God is free in the sense of not being subject to causal laws, He is omnipotent and He is the creator of causal laws, that much is clear. However, a theist might deny this by suggesting that God is necessarily morally perfect. So that while God has external and internal freedom (nothing external causes Him to act, and He is not subject to causal influences) He does not have maximal logical freedom, as there is no possible world wherein God does evil. I think there are serious issues with this position, it seems to seriously undermine omnipotence. For a good discussion, check out Wes Morriston's paper, Omnipotence and necessary moral perfection. are they compatible? For the purposes of this discussion, however, we could revise the premise to a more modest;

(6*) If God exists then God is maximally free with respect to any non-moral choices.

Now, it's unlikely but a theist might yet still deny this by claiming that all of God's choices, including His choice to create the universe are necessary. But this view is extremely implausible. For one, this completely undermines omnipotence even more severely than the denial of (6), since it entails that God cannot do otherwise than what He does at all. For two it entails modal collapse, which is to say it entails that there are no contingent truths and all truths are necessary truths. This is so because God Himself, is a necessary being, and God makes the same choices across all possible worlds. Modal collapse has alot of unpalatable consequences, the free will defense would no longer be available to the theist since there could not be libertarian free will, our modal intuitions such as "It is possible that I would have never been born" and "It is possible that my left foot would be placed 1 millimeter to the right" would all be necessarily false. Finally, one of the main motivations for theism is that it explains why there is something rather than nothing, why contingent things exist, and in particular, why the actual world exists with life, consciousness etc. But, if modal collapse is accepted, this motivation disappears it seems to me. Since the answer to those questions is "it's necessary", so why not just say that, without invoking an extra metaphysical postulate such as God?

Next;

(7) An omnipotent, omniscient person is morally perfect only if He eliminates every evil which is such that eliminating it does not entail preventing an equal or greater good.

The purpose of this premise would be to side-step any greater good theodicy/defense. At first blush, this seems to be true simply via conceptual analysis i.e from what it is to be morally perfect and omnipotent. An omnipotent, omniscient being can prevent any evil state of affairs, there is no restriction on their abilities aside from what is logically possible. A morally perfect being would prevent evil, so long as preventing it does not entail eliminating an equal or greater good. However, as Plantinga correctly points out in God and Other minds, there may be a conjunctive state of affairs E&G, where E is a lesser evil, and G is a greater good, such that G outweighs E. This state of affairs would be a greater good, and eliminating E entails eliminating E&G. Thus, by the premise, God could permit E. To quote Plantinga;

Consider any evil state of affairs E such that there is a good state of affairs G which outweighs it. G-and-E (the state of affairs that obtains just in case both G and E obtain) is a good state of affairs, and it is also a logically sufficient condition of E. Hence, any evil E which is out-weighed by at least one good G, is a logically necessary condition of a good state of affairs (G-and-E) which outweighs it. But this means that an omnipotent and omniscient being could permit as much evil as he pleased, without, under (f 3 ), forfeiting his claim to omnibenevolence, so long as for every evil state of affairs he permits, there is a greater good.


This is helpful, this will serve us. Clearly, just because the conjunctive state of affairs E&G entails E, and E&G is a greater good, surely it remains the case that a morally perfect being would not permit E, if it is the case that the being can actualize G without actualizing the conjunctive state of affairs E&G. For if we suppose it does, then any evil is justifiable, including murder, rape, and genocide, so long as we can trivially, and arbitrarily create a conjunctive state of affairs which is a greater good, and includes the evil as one of it's conjuncts. This is absurd. What this shows, following Plantinga's suggestion, is that the premise must be revised to;

(7*) An omnipotent, omniscient person is morally perfect only if He eliminates every evil which is such that for every equal or greater good which entails it, there is a greater good which does not entail it.

This premise might yet still be disputed on the grounds of the free will defense as Plantinga states;
This assumption is by no means self-evident, however, and apologists for traditional theism have often denied it; they have suggested that perhaps there are certain good states of affairs that an omnipotent God cannot bring about without permitting evil, despite the fact that these goods are not a logically sufficient condition of any evil at all. This suggestion on their parts is sometimes called the free will defense

So, God would not be morally blameworthy for failing to eliminate evil just in case eliminating the evil entails preventing the possibility of an equal or greater good, that being, creatures with morally significant freedom. I think the free will defense is a failure even against the logical problem of evil see here, for more details. However, we can actually side-step the free will defense all together by revising the premise to;

(7'') An omnipotent, omniscient person is morally perfect only if He eliminates every evil which is such that for every equal or greater good that entails it, He does not thereby eliminate a greater good which does not entail it, or the possibility of a greater good which does not entail it.

Then (1)-(5) and (7'') jointly entail.

(8) ) Every evil state of affairs is such that for any equal or greater good that entails it, it's not the case that there is a greater good or possibility of a greater good which does not entail it.

Next we've got;

(9) If God did not create there would be nothing but God

This should be fairly straightforward. A commitment of orthodoxly conceived theism is to take God to be the (ex nihilo) creator of the universe. So there was a state of the world, logically prior to God's act of creation, wherein only God existed, and if God had not created that would be this world. Perhaps some abstracta existed prior to God's act of creation as well, but that is neither here nor there for our purposes.

(10) God is the greatest possible good.

This is a commitment of Anselmian theism. God's goodness is the greatest possible goodness, higher than any possible earthly or humanly goods.

From (9) and (10) it follows

(11) If God had not created there would be nothing but the greatest possible good.

Next we've got;

(12) God is not morally obligated to create a world with evil.

Unlike much of the other premises, this one is less immediately obvious. However, this premise seems plausibly to be a commitment of orthodoxly conceived theism for a few reasons.

For one, it may be argued a maximally great being would be the ground of moral obligations otherwise they would not be maximally great. So that moral obligations just are constituted by God's commands, or by God's telos for human beings or some other such story. God is the ultimate ground of moral obligations, His will or desire or nature is logically prior to them not vise versa. If God were not the ground of all moral duties, and there are independent duties which bind God, then surely God is not the greatest being conceivable, since we can imagine a greater being which does satisfy such a description.

For two, suppose we grant that God can be bound by obligations, still, it doesn't seem like prior to creation God can be bound by obligations. Obligations arise not in a vacuum, but to others. For example, I have an obligation to the government, and to the people of the country I live in, to pay taxes, I have an obligation to my employer, I have an obligation to respect other peoples privacy, autonomy, and property. But all of my obligations are obligations to others. Yet, prior to creation there is nobody which exists to be the object of God's obligation. God would not be wronging anyone by not creating them, since they would not yet exist!

For three, even if God has obligations prior to creation, it seems highly implausible that, in particular, God would be obligated to create a world with evil. Why would God have to create a world with evil, rather than a world with nothing but positive axiological value?

Finally, I raise a challenge to the theist who would reject this premise with a dilemma. Is God's moral obligation to create this world contingent or necessary? If it is necessary, then this seems to undermine omnipotence, since God necessarily cannot not create. It seems utterly bizarre for a being that is omnipotent to have no choice but to create, an omnipotent being should have complete control over their actions and the possibilities they can actualize. If it is contingent, then that is puzzling, prior to creation nothing but God existed, so what possible reasons could God have in possible worlds where He exists prior to creation, such that in some possible worlds He is obligated to create, and in others He is not? It seems there could be nothing which could account for the difference in God's obligations. Further, it seems like moral obligations are the kinds of things which must be necessary, plausibly, if it's true I ought not torture babies all else equal, then there is no possible world where it's not the case that I ought not torture babies all else equal.

(13) If God is not morally obligated to create a world with evil, and God is free with respect to any non-moral choices, then there is a possible world where God did not create a world with evil.

This is true, in virtue of what it is to be free. As noted in my brief discussion of (6), freedom here includes logical freedom (the ability to do otherwise in some possible worlds).

It then follows from (6`), (12), and (13).

(14) There is a possible world where God did not create a world with evil

Next;

(15) If there is a possible world where God did not create a world with evil, the greatest possible good does not entail the existence of any evil.

This is trivially true, given (11), and the definition of entailment. E.g P entails Q iff there is no possible world where P obtains and Q does not obtain.

From 14 and 15 we get;

(16) The greatest possible good does not entail the existence of any evil.

Finally, we have;

(17) If evil exists, and the greatest possible good does not entail the existence of any evil, then there is some evil which is such that for any equal or greater good that entails it, there is a greater good or possibility of a greater good which does not entail it.

This is a long and scary-sounding premise, but it can be shortened to;

(17') (1) & (16) entails ~(8)

The premise is true because what it means to be the greatest possible good, is that there is no good greater than it. So, since the greatest possible good entails no evil, it follows that any possible greater good any evil state of affairs entails, will just be a lesser good compared to the greatest possible good. So for every good that entails, or is possibly instantiated by any evil state of affairs, there is a greater good, namely, God, which does not entail it.

Then, (1) (16) and (17) jointly entails;

(18) There is some evil which is such that for any equal or greater good that entails it, there is a greater good or possibility of a greater good which does not entail it.

Using first order predicate logic (19) can be translated to ∃xPx where P is the property of being an evil state of affairs which is such that for any good that entails it, there is a greater good or possibility of a greater good which does not entail it. ∃xPx is logically equivalent to, ~∀x~Px. Which is the negation of (8), which would be ∀x~Px. So we have successfully generated a contradiction.

Conclusion


It seems Plantinga's challenge can plausibly be met, we have here a formally contradictory set containing nothing but necessary truths, and claims which are essential to orthodoxly conceived theism. The theist could only consistently deny the conclusion on pain of rejecting one of the premises. In other words, (1), (2), (3), (4), (5), (6'), (7'), (9), (10), (12), (13), (15), and (17). The strength of this argument, I think, is it's immunity to greater good type objections, and the fact that it survives the free will defense (due to premise 7``). If such responses are not available to the theist here, then we must ask, what resources does the theist have to challenge this argument? There are probably some not-totally-crazy, rejoinders available to the theist that I have missed, I do not wish to make a strong claim here. But at-least I hope to have shown that the logical problem of evil is not dead.

I will end my concluding section by pre-empting one last objection that I've seen theists use in response to this type of argument. A theist may argue that while a world where God exists sans creation might entail the exemplification of all the great-making qualities qua being, this does not entail that all the great-making qualities qua world are exemplified. A great-making quality qua world might entail something like a diversity of beings. So God would be doing better by actualizing such a world rather than the world where God exists alone. It's unclear what premise this objection is addressing. But, first, it just seems to me that what the objector here is calling "great-making qualities qua world" is just going to be lesser goods. Since, after-all, the greatest possible good obtains sans creation. But it's hard to see how the addition of lesser/finite goods, can improve a world where infinite goodness, and the greatest possible goodness is already exemplified. Second, when we consider the world where God exists alone, it's difficult to see how the theist can claim that such a state of affairs can be improved, afterall, the world is coextensive with God, it refers to nothing over and above God Himself, and surely God cannot be improved. Third, I would think that on the theists view such created worldly goods are not good in themselves, they are only good in virtue of resembling God or God's telos that He created for them. If that's right then what sense does it make to say the addition of created worldly goods adds to the greatness of a state of affairs wherein God, whom maximally resembles God and God's telos, already exists? So, if the theist wants to affirm that such goods add value that wouldn't be there sans creation, then it looks to me that they could do so only on pain of denying that God is the greatest possible good, and denying that God is the ground of all good which seems to be contrary to orthodoxly conceived conceptions of God. Finally, even if God can improve the world by creating stuff other than Himself, by premise 7, it seems there wouldn't be any point at which He would be justified in doing that, since for any state of affairs He could create which contains evil, even if the evil entails a greater good, there will always be a greater good which does not entail the evil, which is God himself.