A Logical Argument from Evil (2013)

Ryan Stringer

1. Introduction

According to traditional theism, there is a God who is eternal, omniscient, omnipotent, completely free, perfectly rational, and maximally good. Logical arguments from evil attempt to demonstrate that the God of traditional theism does not exist based on the fact that evil or suffering exists. In order for such arguments to succeed, there must be a logical inconsistency between the existence of God and the existence of evil or suffering. Because it is widely held that attempts to find such an inconsistency have failed, defenses of atheism on the basis of evil or suffering have largely shifted from logical arguments to evidential ones. Evidential arguments from evil maintain that the existence of the world’s evil or suffering makes the existence of God very unlikely. While evidential arguments from evil are very important, I think that a case for a logical argument from evil can still be made. In this paper I make such a case.

2. The Argument

The argument I offer runs as follows:

(P1) If God exists, then only God, necessary things, and things necessarily coexisting with God exist prior to an initial creation event.
(P2) There are no necessary things, or things necessarily coexisting with God, that are evil.
(P3) If God exists, then the world consists of the highest level of goodness prior to an initial creation event. (from P1 & P2)
(P4) If the world consists of the highest level of goodness prior to an initial creation event, then nothing that comes about from an initial creation event can make the world better.
(P5) If nothing that comes about from an initial creation event can make the world better, then there is no initially created evil or initially created thing that could produce evil that can make the world better.
(P6) If there is no initially created evil or initially created thing that could produce evil that can make the world better, then God will not initially create any evil or anything that could produce evil.
(P7) If God exists, then he will not initially create any evil or anything that could produce evil. (from P3-P6)
(P8) If (a) God exists, (b) the world consists of the highest level of goodness prior to an initial creation event, and (c) God will not initially create any evil or anything that could produce evil, then God will never create any evil or anything that could produce evil.
(P9) If God exists, then he will never create any evil or anything that could produce evil. (from P3, P7, & P8)
(P10) If God exists, then there is no evil in the world. (from P1, P2, & P9)
(P11) There is evil in the world.
(C) Therefore, God does not exist. (from P10 & P11)

Now that I have presented the argument, I want to discuss its premises. P1 is a necessary truth about what exists prior to an initial creation event given that God exists. First, it is trivially true that God’s existence implies that God exists prior to an initial creation event. For if God exists, then it follows that he is the one responsible for any (potential or actual) initial creation event; and to be responsible for any such event, God must exist prior to it. Furthermore, anything that necessarily exists in itself—perhaps abstract objects such as sets and propositions—has always existed, and thus exists with God before an initial creation event. Finally, anything that God’s existence entails, or that necessarily coexists with God, must exist alongside God, and thus would have existed before an initial creation event. So if God exists, then God, necessary things, and things necessarily coexisting with God must exist prior to an initial creation event. But since these things are the only things that necessarily exist in a theistic world, anything else that could exist would exist contingently, and thus would have to follow an initial creation event. So if God exists, then only God, necessary things, and things necessarily coexisting with God exist prior to an initial creation event.

P2 is a necessary truth asserting that there are no necessary things, or things necessarily coexisting with God, that are evil (and those are the only things that could be evil prior to an initial creation event). For example, abstract objects such as sets, numbers, or propositions, which might plausibly be thought of as necessary things, cannot be evil. Also, the thought that something evil must necessarily coexist with a maximally good God is both odd and repugnant.

P3 then follows from P1 and P2. To see how, suppose first that God exists. P1 says that from this supposition it follows that only God, necessary things, and things necessarily coexisting with God exist prior to an initial creation event. Now God is of course not evil, and P2 rules out any evil among the other things that exist with God prior to an initial creation event; so there is no evil in the precreation world. Overall, then, God’s existence entails that the precreation world consists of only God, necessary things, things necessarily coexisting with God, and no evil. But what are we to make of this world? It appears to be a world consisting of the highest level of (worldly) goodness. First of all, consider some of the things that are true about this precreation world:

There is no evil.
It is perfectly just.
It is perfectly peaceful.
There is love (God is love) and unspeakable beauty (God is unspeakably beautiful).
There is no irrationality.
Every creature possesses absolute metaphysical freedom.
Every creature is a perfectly good agent.
Every creature is maximally knowledgeable and powerful.
Every creature is infinitely joyous, completely self-respecting, and without need.
Every creature believes in God and respects him to the highest possible extent.

Such things obviously make this theistic, precreation world an extremely good one, and thus a strong candidate for a maximally good one. The fact that there is no evil in this precreation world means that there is nothing evil that needs to be subtracted from it to make it better, and this removes the first potential barrier to this world being the highest in goodness. Furthermore, the only necessary things that could exist in such a world are abstract things such as numbers, sets, propositions, the laws of logic, and so on; and I know of nothing that necessarily coexists with God other than these necessary things. But such things are either neutral with respect to the world’s goodness, or actually contribute to it; their presence certainly does not make the world less good than it would have been without them. This then removes yet another potential barrier to the theistic, precreation world being maximally good. Finally, God is supposed to be a completely self-sufficient being—a being that is both ontologically and psychologically independent of all else. As such, there is nothing missing from the theistic, precreation world that God needs, nothing that can increase his infinite joy or boost his psychological well-being and thus improve this world. In light of these things, it seems that this precreation world consists of the highest level of goodness.

P4 is a necessary truth. If the world consists of the highest level of goodness prior to an initial creation event, then the world cannot get any better, for there is nothing better than the highest level of goodness. But if the world cannot get any better, then nothing that comes about from an initial creation event can make the world better. So if the world consists of the highest level of goodness prior to an initial creation event, then nothing that comes about from such an event can make the world better.

P5 is a rather straightforward necessary truth: any initially created evil, or initially created thing that could produce evil, would have to come about from an initial creation event. So if nothing that comes about from an initial creation event can make the world better, then no initially created evil, or initially created thing that could produce evil, can make the world better.

P6 is another necessary truth. Since God is maximally good, he will not initially create any evil, or anything that could produce evil, if there is no benefit or improvement to be gained from such things. But if there is no initially created evil or potential producer of evil that can make the world better, then there is no benefit or improvement to be gained from such things. Therefore, if there is no initially created evil or potential producer of evil that can make the world better, then God will not initially create any evil or anything that could produce evil.

P7 follows straightforwardly from P3 through P6. Then we have P8, which is another necessary truth. For suppose that God exists, the world consists of the highest level of goodness prior to an initial creation event, and God will not initially create any evil or anything that could produce evil. Since God will not initially create any evil or anything that could produce evil, either of which would reduce the world’s goodness, it follows from our supposition that the world will still consist of the highest level of goodness after an initial creation event. But if the world still consists of the highest level of goodness after an initial creation event, then nothing that could come about from another creation event could make the world better, which implies that no created evil or potential producer of evil could make the world better this time around. So again, God will not create any evil or anything that could produce evil, which in turn means that the world will still consist of the highest level of goodness after a second creation event. But if the world still consists of the highest level of goodness after a second creation event, then nothing that could come about from a third creation event could make the world better, which implies that no created evil or potential producer of evil could make the world better this time around. And so on indefinitely. For any potential creation event, there is no created evil or potential producer of evil that could make the world better; so God will never create evil or anything that could produce evil.

P9 follows straightforwardly from P3, P7, and P8. For suppose, once again, that God exists. This is the first conjunct of P8’s antecedent. We can then derive, via modus ponens, the second conjunct from our supposition and P3; and we can derive the third conjunct from our supposition and P7. But since our supposition entails P8’s antecedent, it also entails P8’s consequent—that God will never create any evil, or anything that could produce evil.

P10 then follows from P1, P2, and P9. Suppose, yet again, that God exists. As was shown earlier in the explanation of P3, it follows from this supposition, P1, and P2 that there is no evil in the world prior to an initial creation event. Furthermore, P9 tells us that our supposition entails that God will never create any evil, or anything that could produce evil. But if there is no evil in the theistic, precreation world, and God will never create any evil—or anything that could produce evil—then there will be no evil in the world. So if God exists, then there is no evil in the world.

P11 is the obvious fact that there is evil in the world. The nonexistence of God then follows from P10 and P11 via modus tollens.

3. Responding to Possible Objections

P1 might be contested, but potential critics are unlikely to find it to be an appealing target. For rejecting this premise amounts to rejecting God’s ultimate responsibility for the existence of all things that he could in principle be responsible for. But theologically, if not conceptually, God is the ultimate creator of, and thus ultimately responsible for, everything that is contingent in a theistic world.[1] So a critic of P1 would have to maintain that there might be something that does not necessarily exist in itself and does not necessarily coexist with God, but is nonetheless a feature of a theistic world’s precreation state, even though this something could in principle have been, yet was not, created by God. Because this potential line of criticism conflicts with the concept of God as ultimately responsible for all that contingently exists in a theistic universe (i.e., all existence for which he could in principle be responsible), it is not a promising threat to P1.

Another premise that could be contested is P2. A critic could argue that evil is either a necessarily existing thing in itself, or something that necessarily coexists with God. To my knowledge, the only way to argue that evil necessarily coexists with God is to utilize some form of the “logically necessary contrast condition” argument, which can be formulated existentially or epistemically. The initial claim of every formulation is that good logically requires a contrast condition, which is evil. The existential formulation states that, since good logically requires the contrast condition of evil, the mere existence of good requires the existence of evil; and because God is a maximally good thing, evil must exist if God exists. The epistemic formulation, on the other hand, goes in one of the following three ways. Since good logically requires the contrast condition of evil, either: (1) the experience of good requires the experience of evil; (2) the recognition (or knowledge) of good requires the recognition of evil; or (3) the appreciation of good requires the recognition of evil.[2] And because God necessarily experiences, recognizes, and appreciates goodness, there must be something evil for God to experience or recognize that necessarily coexists with him.

However, none of these options work. First of all, evil is not a necessarily existing thing: we can easily conceive of a world in which there is no evil, there is no incoherence in there being a world with no evil, and there is nothing about the nature of evil that suggests that it must exist. At this point, the onus is on the critic to show why this view is mistaken; and until we have such a demonstration, we have no reason to think otherwise.

Even if we grant the first part of the initial claim of “logically necessary contrast condition” arguments—that good logically requires a contrast condition—evil is not required to satisfy it. For the logically necessary contrast condition of “good” can simply be “not good,” and this can be satisfied by the complete lack or absence of good. So the initial claim of all versions of such arguments is false. With regard to the existential formulation of the argument, we might say that sets, propositions, numbers, and the like are necessarily existing, abstract things that are not good in the sense that they completely lack goodness (while also completely lacking evil). Because they are not good, they provide the logically necessary contrast condition for good. And if this is insufficient for some reason, the nothingness or emptiness preceding God’s initial ex nihilo creation event can provide a sufficient contrast condition for good. Thus, evil need not necessarily coexist with God. With regard to the epistemic formulations, we can say that experiencing a neutral state that lacks good—perhaps one that is induced from the perception or contemplation of abstract objects, empty space, or nothingness—can provide the contrast for the experience of good; and that recognizing either these objects or the neutral state as lacking good can provide the contrast for recognizing and appreciating good. Thus, we need not postulate any experience or recognition of evil as necessarily coexisting with God.

For the sake of argument, however, let’s assume that I am wrong and that evil must be the logically necessary contrast condition of good. Even so, these arguments still face serious problems. Let’s begin with the existential formulation. The fundamental problem here is that the argument seems to be invalid: the alleged fact that evil is the logically necessary contrast condition of good, which if true is a conceptual truth, does not seem to entail that there must be evil if there is good (and thus that evil must exist if God exists). In other words, the fact (if it is a fact) that the concept “good” requires its contrasting concept “evil” does not seem to entail that the instantiation of the one requires that of the other. In order to see this, consider the following two demonstrations of the invalidity of similar entailments. First, even though the logically necessary contrast condition of the property “being self-identical” is “not being self-identical,” it certainly does not follow that if there are things that are self-identical, then there must be something that is not self-identical. In fact, it is logically necessary that everything is identical with itself, so it is logically impossible for the property “not being self-identical” to be instantiated. Second, even though the logically necessary contrast condition of “being visible” is “being invisible,” this does not entail that there must be invisible things if there are visible things (and vice versa). After all, we can conceive of a reality with only visible things, or one with only invisible things; and neither reality seems to be incoherent.[3] These examples strongly suggest that the necessity of a contrast condition for a given property that is instantiated does not entail the instantiation of the contrasting property. So, once again, we need not postulate evil as necessarily coexisting with God.

The first two epistemic formulations are also logically invalid. Consider the first formulation. The properties “being oneself” and “being alive” have the logically necessary contrasts “not being oneself” and “not being alive,” respectively. While we obviously experience being ourselves and being alive every day, we certainly have never experienced not being ourselves or not being alive, which means that we do not need to experience the latter in order to experience the former.[4] Thus the fact that a property P has a logically necessary contrast condition does not entail that the experience of P requires the experience of not-P. Now consider the second formulation. The property “being self-identical” has the logically necessary contrast “not being self-identical.” Yet we certainly do not need to recognize something that is not identical with itself in order to recognize the self-identicality of things. So the fact that a property P has a logically necessary contrast condition does not entail that the recognition of P requires the recognition of not-P.

Furthermore, the second and third epistemic formulations do not establish what is needed to refute P2. For even if the recognition and appreciation of good requires the recognition of evil, it does not follow that this evil must be actual in order for it to be recognized as evil. Since God has epistemic access to all possible worlds, he can “peer into” any one of them containing evil and recognize its evil as evil before creating anything. He thus can recognize evil without it actually existing, which means that neither of these epistemic formulations requires that evil necessarily coexist with God.

Finally, there is the problem of what particular evil thing would either necessarily exist, or necessarily coexist with God, prior to an initial creation event. Neither option for undermining P2 will be very compelling without a plausible candidate, and I can only think of the following possibility. As one who recognizes and appreciate goodness, God must recognize evil. This evil need not be actual—God could simply think about possible evil in order to recognize it as such (or “peer into” a possible world containing evil in order to recognize it). But when he does this, he will have a negative experience in virtue of his maximal goodness. For as a maximally good being, he will be affected by evil things in an appropriate way—namely, he will experience negative feelings (like disgust) toward them. But this negative experience is itself an evil; so there is indeed evil necessarily coexisting with God.

However, there are two major problems with this attempt to undermine P2. First, it is not clear that God must have a negative experience from recognizing evil. Since this evil would only be possible evil, it need not have the same effect on God as actual evil would need to have. And even if God would have to have a negative experience from recognizing evil, it seems false that such an experience itself must be an evil. For although evil is certainly a negative thing, it need not be the case that negative things are evil. In fact, the negative experience of recognizing a merely possible evil—as mild as it would have to be—seems to be a good thing, for it is an appropriate reaction to evil. Such a reaction is precisely what we would expect from a maximally good being, and thus it is a symptom of God’s maximal goodness. As such, it seems to be a good thing, not an evil one. So the problem of what evil thing undermines P2 still remains.

Besides P2, another possible point of contention is P3. One way to contest it is to argue that even if the theistic, precreation world is extremely good, this world does not consist of the highest level of goodness because there is no highest level of worldly goodness. For even if there is an extremely high level of worldly goodness G, there is an indefinite amount of higher levels G + 1, G + 2, and so on.

Though natural and apparently plausible, the objection is nonetheless highly questionable. It assumes that worldly goodness “behaves” like numerical quantity does—that there is always a higher level of worldly goodness than any given level, and thus no highest level of worldly goodness (just like there is always a higher numerical quantity than any given one, and thus no highest numerical quantity). But this assumption is dubious in virtue of the following disanalogy between numerical quantity and worldly goodness. In the case of numerical quantity, there is always a higher numerical quantity than any given one, and thus no highest one, because the numerical quantity of any number can always be increased by adding any positive number to the original number. Yet the corresponding claim for worldly goodness—that the goodness of a world can always be increased by adding any good thing to it—appears to be false. For example, adding a good car to the theistic, precreation world would not increase its goodness since it would not improve the world in any way (God would certainly have no use for it, and would gain no other recognizable benefit from its presence). And the same would be true of good food, good drink, and many other good things.[5]

Even if my examples here are flawed, the corresponding claim for worldly goodness is still questionable in virtue of G. E. Moore’s principle of organic unities. This principle states that the value of a whole may not be the same as the sum of the values of its parts.[6] Moore’s famous illustration of this principle is that the value of someone being conscious of a beautiful object (a complex whole) is not reducible to the sum of the values of its parts—namely, the value of consciousness and the value of the beautiful object. Consider my being in a room by myself and being conscious of a beautiful painting hanging on the wall. If the value of this whole were given by simply adding the value of my consciousness to the value of the painting, then this whole would be just as valuable as the whole in which I am not looking at the painting at all. But these two wholes are not equally valuable—the scenario in which I am conscious of the painting is better than the scenario in which I am not looking at it. So the value of the former whole is not the same as the sum of the values of its parts.

Regardless of whether Moore’s famous example works here, his principle is independently plausible with respect to complex wholes like worlds because of (a) emergent properties and (b) interaction effects among the parts. When we place individual objects into a world to form a whole, the result is a distinct object (or more accurately, a state of affairs) with its own set of properties, some of which will be emergent properties.[7] But these properties, which do not result from adding the properties of the individual objects together, can affect the value of the whole. More important are the interaction effects between individual objects, which are similar to the interaction effects that drugs can have when they are simultaneously present in the body. Two different drugs can each have a certain effect when present by themselves, yet when combined the effect of both together is not equal to the sum of the effects that each of them have when present by themselves. Something similar probably explains why the value of someone being conscious of a beautiful object is not simply given by adding the values of the parts: the beautiful object and consciousness “interact” to produce a higher value than that of their sum. It is therefore quite plausible, in virtue of emergent properties and interaction effects, that the value of a world may not be the same as the sum of the values of its parts. But if so we cannot assume that adding a good thing to a world will automatically increase its goodness. Doing so might make a world better, but it might not. It therefore may be false that the goodness of a world can always be increased by adding any good thing to it.

Besides the disanalogy between numerical quantity and worldly goodness, the objection’s assumption can be called into further question by the fact that there are other things besides worlds that have a highest level of goodness. Perhaps the best example here is that of moral agents: since God is a morally perfect agent, there is no way for a moral agent to be morally better than God. So there is a highest level of goodness with respect to moral agency. This alone opens up the possibility that there is a highest level of goodness with respect to worlds. Of course, the more examples that we have, the stronger this possibility becomes. It would also be strengthened if we could find examples of complex wholes that have a highest level of goodness. A decent example here is that of bedrooms: for each person, some bedrooms will be better than others, and there is at least one bedroom that admits of no improvement. Another decent and closely related example is that of homes, for we are all familiar with the concept of a “dream home” (and that of a “dream wedding,” for that matter).[8]

As it stands, then, this first objection to P3 is not compelling. Instead of arguing that “there is no highest level of worldly goodness” by dubiously abstracting this goodness away from the possible worlds that must realize this goodness and treating it as if it “behaves” like numerical quantity does, it is much more promising for a potential critic to try to show that the theistic, precreation world specifically does not consist of the highest level of goodness. But since this world contains only God, abstract objects, and no evil, there is nothing present in this world that could be taken away to make it better. Thus, the only remaining option is to argue that this world lacks some good thing(s) whose presence would make the world better. One such argument goes something like this: the theistic, precreation world lacks the very important goods of friendship, loving relationships, forgiveness, courage, and compassion, and the presence of these goods would make this world better. But then the theistic, precreation world could be better, and thus it does not consist of the highest level of goodness. Consequently, P3 is false.

Although this objection to P3 is much stronger than the first one, it is hardly compelling because certain conditions must be in place in order for the presence of the goods in question to make a world better than it would be without them. Consider first the goods of forgiveness, courage, and compassion. Since God is the ultimately forgiving, courageous, and compassionate person, the goods missing from the theistic, precreation world must be acts of forgiveness, courage, and compassion. But the presence of such acts make a world better than it would be without them only if there are already bad or negative things meriting these acts. For example, if we assume that forgiveness has the value that religious people tend to give it, then when people have been wronged by others, the world will be better if the perpetrators are forgiven for their transgressions. In this case, where the condition of people being wronged is present, we are comparing worlds on the basis of forgiveness alone—comparing a world where people have been wronged and their perpetrators subsequently forgiven versus one where people have been wronged but their perpetrators have not been forgiven. As such, forgiveness will make the world better.

However, when the condition of being wronged is no longer present (and thus forgiveness is absent), it is no longer true that the presence of forgiveness will make the world better. For then we are no longer comparing worlds on the basis of forgiveness alone. Instead, we are comparing them on the basis of forgiveness and people being wronged—we are comparing a world without wronged people and without forgiveness versus one with both. But since forgiveness comes with the costs of wronged people and wrongdoings, its presence will not make the world better, but will in fact make it worse. For when one person has forgiven another, their relationship has been repaired or restored to its original condition. In other words, forgiveness itself aims to bring things back to the way that they were before the transgression, that is, to a state without wronged people. The very nature of forgiveness, then, suggests that the absence of wronged people is the real value here. Thus a world without wronged people and without forgiveness is better than a world with both. And since the theistic, precreation world lacks both forgiveness and wronged people, it would be made worse, not better, by the presence of forgiveness. Something similar can be said for compassion and courage; though the details will be different, the results will be the same. So the absence of these goods from the theistic, precreation world does not show that it could be better and thus that P3 is false.

The same is also true when it comes to the absence of friendship and loving relationships in the theistic, precreation world. Again, certain conditions must be in place in order for their presence to make a world better than it would be without them. In particular, there must be beings that need or will benefit from friendship and loving relationships—that is, beings whose welfare will be promoted by having friends and loving relationships. For when there are beings (such as humans) that require friendship and loving relationships to have a good life, then it is most definitely true that the presence of these things makes the world better than it would be without them. But this is only because these goods promote the welfare of the beings that are already there; they would not make the world better if they did not promote anyone’s welfare.[9] Recall from earlier that the theistic, precreation world would not be improved by the addition of cars, good food, or good drink because God, the only being in the theistic, precreation world, has no need for such things, and would receive no benefit or improvement in his lot from such things. In a similar fashion, then, the addition of friendship and loving relationships will not improve this world because God is a psychologically independent, completely self-sufficient being that does not need friends and loving relationships like we do. They would provide no benefit or improve his lot in any way.[10] This is not to say that God would not enjoy or want such things if there were other personal beings with which he could form personal relationships. It is only to say that, as a psychologically independent and completely self-sufficient being, God would be just as well off by himself, which is perfectly compatible with God’s enjoyment of or desire for friends or loving relationships if he were not by himself. Since the addition of friendship and loving relationships will not improve the theistic, precreation world[11], the lack of these goods does not show that this world could be better and thus that P3 is false.

However, one more argument could be leveled against P3: Even though the theistic, precreation world is beautiful in virtue of God’s unspeakable beauty, it would be even more beautiful if it had more beautiful things in it. Wouldn’t the addition of the numerous beauties here on Earth and in outer space make the theistic, precreation world more beautiful, and thus better? Surely it would; so once again this world could be better and thus does not consist of the highest level of goodness. Therefore, P3 is still false.

While clever, this argument against P3 is unsuccessful. Although the theistic, precreation world could consist of more beautiful things, it does not necessarily follow that the world itself would be more beautiful if it had more beautiful things in it. For one thing, the aesthetic value of something is not necessarily increased by adding more beautiful things to it (the principle of organic unities tells us this much). For example, a beautiful painting is not necessarily made more beautiful by packing more beautiful things in it—this may actually decrease its aesthetic value. This seems particularly true of portraits, where the beauty lies in the person being portrayed against a modest background. Packing more beautiful things in it would not necessarily make it more beautiful, but might instead make it worse, or make it into a different painting altogether with comparable aesthetic value. Something similar could very well be true about the theistic, precreation world: its beauty, like that of a portrait, could lie in consisting of an unspeakably beautiful creature against a modest background. Perhaps an even better example here is that of rooms in houses or buildings, which are, like worlds, complex wholes. A beautiful room is not necessarily made more beautiful by packing more beautiful things in it, and it could even be made less beautiful by doing this. In light of these examples, then, it is dubious to assume that the theistic, precreation world would be more beautiful if it had more beautiful things in it. So it remains to be demonstrated that the theistic, precreation world could be better, and thus that P3 is false.

Besides the goods considered above that are absent from the theistic, precreation world, yet another absent good is art. And wouldn’t this world be better if it had a collection of the most exquisite artworks imaginable? If so, then the theistic, precreation world does not consist of the highest level of goodness, and thus P3 is false.[12] But, once again, it is questionable that the theistic, precreation world would be better if it had such an exquisite art collection. For how, exactly, would this collection improve that world? I see only two options here. The first is that the collection would purportedly make that world more beautiful; but this would be subject to a similar critique as that leveled against the previous aesthetic argument against P3. The other option is that the collection would benefit God in some way. However, this benefit would have to come from either (a) God producing the collection himself or (b) God positively experiencing the collection. Yet God is not an artist that needs to express himself artistically; so he would not benefit, like some humans do, from producing the collection himself. He also could, in virtue of his epistemic access to every possible world, experience any exquisite collection of art by “peering into” the possible worlds that contain them; so he would gain no benefit from the collection being actualized. Consequently, it still has not been shown that the theistic, precreation world could be better and thus that P3 is false.

The last argument against P3 that I will consider is the following. The only living creature in the theistic, precreation world is God—there are no angels or other purely good spirits. But wouldn’t this world be better if it contained, in addition to God, a troupe of morally perfect angels who all loved and respected each other? Surely it would, so the argument goes, thus the theistic, precreation world does not consist of the highest level of goodness, and therefore P3 is false.[13] Now this argument is definitely intuitively plausible. However, upon reflection it is not clear exactly how the theistic, precreation world would be improved by the addition of these angels. As previously noted, God would not really benefit from this addition. Though he might enjoy it, he will be just as well off by himself. Moreover, the two worlds in question are evil-free, hate-free, irrationality-free, need-free, perfectly just, and perfectly peaceful. They both contain great beauty, and every creature loves and respects every creature. Every creature in each world is morally perfect and believes, respects, and loves God to the highest possible extent. The addition of angels can, at best, merely preserve the absolute metaphysical freedom, infinite joy, and complete self-respect of every creature. And while this addition of angels increases the amount of beings that have knowledge and power, it does not preserve the fact that every creature is maximally knowledgeable and powerful. So where, then, is the worldly improvement? Without an explanation of how these angels would actually improve the theistic, precreation world, it remains questionable that they would improve it. And this, in turn, means that P3 has not been shown to be false.[14]

The final premise that might be contested is P11, even though it seems quite obvious that the world contains evil. One could argue that there only appears to be evil in the world, but there really is none. In other words, we are misperceiving the world when we perceive it to contain evil. However, in addition to this being as implausible as the claim that there only appears to be good in the world, but there really isn’t any, the argument seems to be incoherent (or self-refuting). For if evil is really an illusion, then its appearance is a deceptive one, which means that we are being massively deceived. But such massive deception, whether intentional or not, is itself an evil; so there is still evil in the world in virtue of this deception. Thus the claim that “there is no evil in the world, only the faulty appearance of it” implies that there is evil in the world, which is incoherent. So this attempt to undermine P11 does not work.

But a potential critic of P11 might still argue that since evil is really just the absence of goodness, there technically is no evil, only the absence of goodness. This argument does not work, however, for at least two reasons. First, conceptualizing evil as merely the absence of goodness does not make a substantial difference here. For even if the so-called “evil” features of our world are really just completely lacking in goodness, they still cannot coexist with God because (a) he would not create them or anything that would produce them unless either would make the world better, and yet (b) neither can make the world better. The alleged fact that these features really are not evil, but only lack goodness, does not change the fact that God, as the kind of being that he is, will never create them or anything that could produce them. Redefining “evil” as the “absence of goodness” is, at best, a slight change in the way that we conceptualize the phenomena that is problematic for traditional theism—they each have no positive quality of evil, as we previously thought, but only lack the positive quality of goodness. However, this change does not eliminate the phenomena altogether.

Moreover, even if it would make a substantial difference to conceptualize evil as merely the absence of goodness, it is absurd to do so. For starters, such a conceptualization does not allow for degrees of evil, or things that are more (or less) evil than others, because true absences do not admit of degrees. For example, it does not make sense to say that strangling a child to death, which is a greater evil than punching him or her in the face, is a greater absence of goodness since a true absence of something cannot be greater than another such absence. To use mathematics here as an analogy, all absences are essentially zeroes, yet all zeroes are equal and thus cannot be said to be greater or lesser than another zero. Therefore, if evil is nothing more than zero goodness, then all evil things are equally evil, and thus cannot be said to be greater or lesser than other things in their degrees of evil.

Furthermore, if evil is nothing but the absence of goodness, then (1) evil things must completely lack goodness and (2) things that completely lack goodness must be evil. Yet both of these consequences are false. For counterexamples to the first consequence, consider complex things like persons or states of affairs, which can be evil even though they have good aspects. Adolf Hitler is the quintessential evil person, yet he surely had some good characteristics. Similarly, a state of affairs that contains, among other things, an extremely unequal distribution of resources and extreme poverty can be evil even if it contains the good of happy, affluent people as a part. For counterexamples to the second consequence, note that abstract objects and empty space appear to completely lack goodness[15], yet surely do not count as evil. For in what way are these things evil? How do they belong to the same category of terrible things as stalking, rape, sex trafficking, abduction of people and pets, child and spousal abuse, brutal murders and frightful deaths, war, genocide, torture, slavery, disease and mental illness, poverty and severe unevenness in resource distribution, dehydration and starvation, severe beatings, burnings, and broken bones, racism, heterosexism, sexism and misogyny, oppression and despotism, and cruelty to nonhuman animals? While abstract objects and empty space completely lack goodness, it is absurd to group them with such evil things.[16]

Of course, it is true that evil things lack goodness. For instance, evil people and evil states of affairs surely lack at least some of the good things had by good people and good states of affairs. But this is not the whole story: evil things—whether they be experiences, events, desires, character traits, people, actions, states of affairs, or what have you—are true existential presences that lack goodness. They characteristically involve either the presence of negative things like pain and suffering, cruelty, violence, disrespect for persons, violations of rights, and so on, or the taking away of positive things like freedom, happiness, and so on. In fact, these two characteristics not only allow us to make sense of degrees of evil, but also explain why (a) the two consequences of viewing evil as nothing but the absence of goodness are both false and (b) it is still true that evil things lack goodness. As for the two consequences, evil things need not completely lack goodness, for good parts can accompany the presence of negative things that, together, form an evil whole; and things that contain no good need not be evil since they need not take away anything positive or present anything negative. Evil things lack goodness not because evil is just the absence of goodness, but because the presence of negative things and the taking away of positive things both logically preclude the presence of certain positive things.

Since it is false that evil is just the absence of goodness, the second attempt to undermine P11 fails. If there is one thing that we can call a fact, it is P11—that there is evil in the world.

4. Conclusion

In this paper I have attempted to vindicate a logical argument from evil. After formulating the argument and discussing its premises, I examined several objections that could be leveled against it and argued that none of them succeed. I therefore conclude that the existence of evil entails the nonexistence of God.

Notes

[1] By “contingent in a theistic world,” I mean those contingently existing things that God could be ultimately responsible for because they do not necessarily exist in themselves and do not necessarily coexist with God, and are therefore capable of existing or not existing in a theistic world.

[2] I differentiate these formulations of the epistemic argument because it seems to me that experiencing, recognizing, and appreciating good are all different things.

[3] The possibility of such realities will of course depend on the beings in relation to which we are defining “visibility.” For example, there are some things that are visible, and others that are invisible, to human perception; so a world with only visible objects, and another with only invisible ones, both seem to be possible. On the other hand, if there was nothing that could be invisible to a certain kind of being’s perception, then only a world with visible objects would be possible. In fact, God appears to be the kind of being that can in principle perceive anything; so an “invisible object” as defined in relation to God is an impossible object. Now this has a very interesting result. For according to the existential argument’s form, the necessity of a contrast condition for a given property P entails that if P is instantiated, then not-P is also instantiated. This means that since being visible to God (P) logically requires the contrast condition of being invisible to God (not-P), then, according to the existential “logically necessary contrast condition” argument, if there are things visible to God, then there are things invisible to God. However, from this and the previously established fact that things invisible to God are impossible objects, it follows that there can be nothing visible to God. So if the existential argument were valid, the only possible worlds would be those with no objects in them at all—including God!

[4] If my examples here are not satisfactory, consider the fact that the experience of love does not require the experience of hate. While most (if not all) people hate someone, have hated someone, are (or have been) hated, or have experienced hate second-hand, many people have surely loved and been loved by others (like their parents) before they experienced hate in any form.

[5] One could object to my examples here on the grounds that they only concern instrumentally valuable things, not intrinsically valuable ones. If this is right, then I have only shown that it is false that the goodness of any world can always be increased by adding any instrumentally good thing to that world. It could still be the case that the goodness of any world can always be increased by adding any intrinsically good thing to it; and this would entail that there is always a higher level of worldly goodness than any given level of it, and thus that there is no highest level of worldly goodness. However, it is not the case that cars, food, and drink are only instrumentally valuable goods. Though some of these things are only instrumentally valuable, many cars, foods, and drinks are both instrumentally and intrinsically valuable, while some foods and drinks are only intrinsically valuable. For people that love certain cars do not just value them because they take them where they need or want to go—they value them as objects that are worth having for their own sakes as well. And some foods and drinks are valued for their ability to eliminate hunger and quench thirst, respectively, as well as for their distinctive taste; while others are valued only for their distinctive taste (e.g., desserts and soda). So my examples here seem to be fine.

[6] G. E. Moore, Principia Ethica (New York, NY: Dover Publications, 2004), p. 28.

[7] Very roughly, an emergent property is, as the concept suggests, a property of a whole that emerges when its parts, each of which lacks that holistic property, are put together to form the whole. The typical example of such a property is the wetness of water. The individual H20 molecules of water are not wet, but when they a grouped together into water, the complex whole is wet.

[8] One could object to my examples of bedrooms, dream houses, and dream weddings on the grounds that the measure of their goodness is, at least in large part, whether they fulfill certain functions, which is not true of worlds. But even so, these examples are only meant to be examples of complex wholes that admit of highest levels of goodness and thereby strengthen the possibility that worlds, which are also complex wholes, admit of a highest level of goodness. The cited difference between my examples and worlds certainly weakens the support that these examples provide, but it does not obliterate it. Instead, the examples still provide some support for the possibility that worlds, insofar as they, too, are complex wholes, admit of a highest level of goodness.

[9] I suppose that these goods could improve the world without promoting anything’s welfare, but the burden of proof is on the potential defender of this claim to show exactly how this would work.

[10] I do not intend to suggest here that friendship and loving relationships would not improve the theistic, precreation world because they are only instrumentally valuable goods that God has no use for. To the contrary, I regard these goods as some of the most valuable intrinsic goods that there are. But even though these goods are intrinsically valuable in that they are not valued or sought for the sake of anything other than themselves, it is still the case that they have intrinsic value because of the benefits that they provide, or due to their contribution to our well-being. This does not make them instrumentally valuable, as if they are mere means to benefits or improvements in our well-being. On the contrary, they constitute particularly weighty benefits or improvements to our well-being. However, they are still only beneficial for beings like us, and so they will improve a world only when there are beings like us that can benefit from them. Therefore, since the theistic, precreation world does not contain any such beings, friendship and loving relationships would not improve that world.

[11] This does not mean that the addition of such goods would make the theistic, precreation world worse. All that follows from this addition is that the world would be different.

[12] I would like to thank an anonymous reviewer for pointing out this objection.

[13] I would like to thank the same anonymous reviewer for pointing out this objection in addition to the previous one.

[14] Nothing that I have said here reduces angels to less than intrinsically valuable things. For even though they presumably are intrinsically valuable, an explanation of how they would actually improve the theistic, precreation world is called for—it cannot simply be assumed that they do because of the principle of organic unities. Moreover, the fact that angels would not benefit God only shows that their addition would not improve the theistic, precreation world by benefiting God—it does not imply that angels are only instrumentally valuable things. It is perfectly coherent to maintain that such angels are intrinsically valuable things that would not improve the theistic, precreation world insofar as they would not improve the lot of any being already in that world (an obvious way to improve a world).

[15] This appearance could be contested on the grounds that, as things lacking evil, abstract objects and empty space do not completely lack goodness, but instead are good precisely because they completely lack evil. Fortunately, the truth of this objection would not cause any major problems for my logical argument from evil. It would undermine my first line of reasoning against the logically necessary contrast condition arguments above, which relied on the neutrality of abstract objects and empty space, but would not touch my alternative lines of reasoning against these arguments. Moreover, if it were indeed true that completely lacking evil implied goodness, then the falsity of conceptualizing evil as the mere absence of goodness would actually be vindicated by this objection. For if completely lacking evil implied goodness, then the absence of goodness would imply the absence of completely lacking evil—that is, it would imply the presence of evil. Evil would thus be a true existential presence, not a mere absence of goodness.

[16] If abstract objects and empty space do not work here, we could instead appeal to useless things like broken appliances that merely take up space. Surely such things are neither good nor evil, and thus refute the consequence that things completely lacking in good must be evil. (They also refute the idea that completely lacking evil implies goodness.)

Cookie	Duration	Description
cookielawinfo-checkbox-analytics	11 months	This cookie is set by GDPR Cookie Consent plugin. The cookie is used to store the user consent for the cookies in the category "Analytics".
cookielawinfo-checkbox-functional	11 months	The cookie is set by GDPR cookie consent to record the user consent for the cookies in the category "Functional".
cookielawinfo-checkbox-necessary	11 months	This cookie is set by GDPR Cookie Consent plugin. The cookies is used to store the user consent for the cookies in the category "Necessary".
cookielawinfo-checkbox-others	11 months	This cookie is set by GDPR Cookie Consent plugin. The cookie is used to store the user consent for the cookies in the category "Other.
cookielawinfo-checkbox-performance	11 months	This cookie is set by GDPR Cookie Consent plugin. The cookie is used to store the user consent for the cookies in the category "Performance".
viewed_cookie_policy	11 months	The cookie is set by the GDPR Cookie Consent plugin and is used to store whether or not user has consented to the use of cookies. It does not store any personal data.