Introduction

There seems to be an undeniable tension with both atheists and theists, regarding the problem of our experience and the belief in a classical God-- that is, a God that is wholly good and wants the best for his creation. Atheists wonder how, in the face of great pain and adversity and evil, religious believers can believe in a wholly good God. By the same token, theists wonder how, in the face of great pain and adversity and evil, that the God they believe in can justify it. Atheists think a wholly good and wholly powerful God would want to, and be able to stop, any evil that occurred in the world, and since evil occurs constantly in the world, such a God must not exist. Theists, while many pointing to a higher divine plan, ponder what exactly that divine plan might be, and why evil must be an element of such a plan.

Therein lies the problem of evil: The existence of evil in our world experience seems to indicate, on some level, that God is not preventing this evil. Why wouldn't God prevent this evil, if he were wholly good, and wholly powerful? Australian philosopher J.L Mackie posits the problem of evil as a simple case of logical inconsistency. There are three statements which the believer in God is assumed to accept, and the so-called problem of evil illustrates the problem of inconsistency in the statements all being true, at the same time. The three facts the religious believer assumes are: 1.) God is omnibenevolent (wholly good), 2.) God is omnipotent (wholly powerful), and 3.) Evil exists. The first two facts are definitional, at least for the religious believer; what it means for a being to be God, at least in the religious sense, is to be, among other things, both omnibenevolent and omnipotent. The third fact is an a posteriori truth; we can evidently extract that evil exists from our experience in the world. The problem of evil is such: An omnibenevolent God would want to stop evil if it existed. If an omnibenevolent God could prevent evil but wasn't willing to, then God wouldn't be omnibenevolent. If an omnibenevolent God wanted to prevent evil but was unable to, then God would not be omnipotent. If God was omniscient, then God would know every evil. So, we are faced with the following: Evil exists. Therefore, either God is not omnibenevolent, because God is not willing to prevent evil, God is not omnipotent, because he is unable to prevent evil, or God is not omniscient, because he doesn't know the vil that is occurring. In either case, one of the definitional truths for God must be false, in which case a "perfect" (in the sense of having absolutely maximal properties for Goodness and Power) doesn't exist. Yet, the believer in God wants to accept all three propositions as true.

There are many possible replies to the problem of evil, two of which I will evaluate here. The first response is the response from higher-order and redeeming goods, which postulates that, on at least some level, it is impossible for good to exist without evil. The second response is the response from free will, which postulates that God allows evil for the sake of human free will, and that evil occurs because God does not want to compromise this free will by preventing evil. I will examine Mackie's response to each of these objections, and, in the second case, I will examine Alvin Plantinga's rebuttal to Mackie's free will response, and whether that rebuttal adequately solves Mackie's concerns about the response from free will.

The Higher-Order and Redeeming Goods Defense

The response from higher-order and redeeming goods is two-fold. Firstly, it is impossible for good to exist without evil. That is to say, it seems plausible to say that there are a certain class of goods (second-order goods) that cannot exist without an evil or deficient counterpart. These second-order goods would seem to exist to promote another class of good (first-order good) and minimize evil (first-order evil). It would seem that an omnibenevolent God would want to maximize the second order goods, insofar as to promote the first order goods. It would thusly seem that God's goodness is in another class, we shall say, third-order good.

For example, steadfastness is considered as an admirable good. But it seems impossible for steadfastness to exist without trial, which we would consider an evil. Because steadfastness is a response to some type of evil, namely, trial, we cannot say that it would be able to exist if trial didn't exist, because there would be no way to measure it. Consider, as another example, forgiveness, a good, and offense, an evil. It would be impossible to have forgiveness without offense, because the good of forgiveness logically pre-supposes that offense exists, for if offense didn't exist, then forgiveness couldn't exists as a response to offense. Since an omnipotent God, many argue, can only bring about intrinsically possible states of affairs, and it is intrinsically impossible to have certain types of goods without certain types of evils, then God could not have created a world with, for example, steadfastness and forgiveness but no trial or offense. It seems, then, that God had two options: create a world with these goods but also with the corresponding evils, or create a world that contained neither these goods nor the corresponding evils. As such, it would also seem that God deemed the first case more desirable than the second case. In this sense, it would also seem that there are a certain class of goods that are inseparable from evils that also outweigh the evils. That is to say, it would be a better case for a world to contain both a good and its corresponding evil than for it not to contain either. We can thusly say that such a good would outweigh, or redeem, its evil, if and only if it is better to have both the good and the evil than to have neither the good nor the evil.

Mackie raises three objections to this response. Firstly, It would seem that, in some sense, the "redeeming" goods we speak of, such as steadfastness, or forgiveness, or courage, or redemption, are considered good only insofar as they act as coping strategies for evil. It would not seem that these "goods" have the property of "goodness" in and of themselves, which seem to be the kinds of goods an omnibenevolent God would want in the world. Allowing evil to exist for the sake of higher-order goods is much akin to making the argument that a world with a world with polio and an polio vaccine would be better than a world that contained neither polio nor polio vaccine. That is to say, it is perfectly legitimate to say that a polio vaccine is good, but it is good only insofar as at combats polio. This objection raises serious doubts about the response, but at the same time does not create a logical contradiction. Secondly, it would seem that a omnibenevolent God would not only want to promote good, but also minimize evil. Since, in this case, God's goodness would merely be the will to promote second-order goods, God would not seem to be omnibenevolent, for an omnibenevolent being would also be concerned with preventing evil. Much like the first response, it raised serious doubts for the response, but does not logically refute it. Mackie's third response, which argues that there are perhaps second-order evils that contrast with second-order goods, third-order evils that contrast with third-order goods, etc. ad infintum, does not logically refute the response either. However, a rather simpler response is this: If God is omnipotent, then he can create a world which contains only the best kinds of goods, ie. the goods that are good in and of themselves, much like a religious conception of heaven might sound like. The higher-order goods argument assumes that the best possible world would contain the maximal number of goods, whether or not they require a corresponding evil. However, this assumption from addition does not seem plausible. It may perhaps be countered that the best possible world has the maximal number of the best kinds of goods, rather than having the maximal number of any type of good which also may pre-suppose evil, which sounds like the conception of heaven. Thus, we have to make a distinction, and, if a proponent of the higher-order goods defense wants to defend their thesis, they must adequately demonstrate the first of the following questions: Is a possible world a better possible world insofar as it contains more goods even with corresponding evils, or is a possible world a better possible world insofar as it is more perfect? This is a subtle, yet crucial, distinction, much akin to arguing whether a quarterback who throws 4 touchdowns and 0 interceptions had a better performance than the quarterback who throws for 6 touchdowns and 2 interceptions. Is player B better, because he threw more touchdowns, or is player A better, because his performance was more "perfect"? We also must ask the proponent of this defense: Does God want to create a world with X number of goods, or does God want to create a world with an X degree of perfection? If the proponent argues that the former determines the latter, he must demonstrate precisely why this is the case.

The Free Will Defense

However, the more common response to the problem of evil is the argument that evil is due to human free will. That is to say, God is not responsible for the evil that occurs in the world, but rather, humans are because they, on some occasions where there is a significant moral choice and they have the ability to exercise significant free will, will choose evil. In this case, the argument seems to be that God knows that evil occurs, and God does not want evil to occur, and God has the ability to prevent evil, but evil still exists because God wants humans to have significant free will. In such a case, significant free will would seem to be a good such that it would be better for the world to contain it and the evil that occurs as a result from it rather than to have a world that contains no evil but at the same time no significant free will.

However, this response is not adequate, for in some sense it assumes that free will is logically inseparable from evil, which it clearly isn't. For example, I can, at the first moral decision I face in my life, choose to do good. Likewise, on the second choice I face, I can choose to do good again, and so on ad infinitum. It is much like tossing a fair coin. It is certainly highly unlikely that a coin could come up heads on every single toss that ever occurs, but nevertheless it is not logically impossible for it to do so. Likewise, it may be highly unlikely, but not logically impossible, that for every human, and for every significantly free choice, good will occur. Since God is omnipotent, God could have created a world in which humans have free will, and evil doesn't exist. If so, why didn't God create such a world?

The preceding argument can be structured as follows:

1.) There is a possible world that contains both significant free will and no evil

2.) For any possible world, an omnipotent God could have created it

C.) Therefore, God could have created a world that contained both significant free will and no evil

This, essentially, is the crux of Mackie's response to the free will defense, and if one accepts the argument as sound, then one is forced to conclude that the free will defense is not an adequate solution to the problem of evil. Because free will and evil are logically separable, and the former need not necessitate the latter, it is legitimate to say that it is possible to have significantly free will and no evil. The only manner in which to refute this argument is to deny one of the premises, which is precisely what Plantinga does, in his thesis for trans-world depravity.

Plantinga does not argue that a perfect God and evil both exist, as others would argue, but merely that the concepts of a perfect God and the existence of evil are logically consistent, which is the crux of Mackie's entire argument. Plantinga thus does not solve the problem of evil per se, but merely demonstrates that one who formulates the problem as Mackie did will run into problems. Plantinga raises issues with the second premise, namely that for any possible world, an omnipotent God could have created it. Plantinga calls this fallacious assumption "Leibniz' Lapse", and demonstrates that, in some cases, there are worlds that are certainly possible, at least in a theoretical sense, but unable to be actualized by an omnipotent God.

Here is a rough sketch of Plantinga's argument for trans-world depravity: Suppose I face a significant moral choice, and I can choose to do X (good) or or choose to do Y (evil). Given certain conditions, I will always choose X. If I choose X under certain conditions, God can actualize a world with the same conditions that doesn't include the action of me making that choice (which Plantinga refers to as a "maximal world segment"), but can't compel me to choose Y without compromising my free will. This seems to be true, because for any world which God could actualize that included a maximal world segment but didn't include my moral choice, there always exists a possible world in which I will do wrong, and thus I will do wrong on at least one occasion (or at least the possibility exists, one that God cannot prevent without compromising my free will). Thus, it certainly seems possible that a world with free will and no evil could exist, but only we, insofar as we are significantly free creatures, and not God, can actualize such a world If God were to actualize such a world, the God would be compromising our free will. Therefore, even though God is omnipotent, God cannot actualize such a world, should God maintain that free will is a good such that it is imperative to have it.

However, Plantinga's argument raises significant issues regarding God. It might be argued that the argument would only seem to apply to creatures, but it seems that the argument applies particularly to creatures insofar as they have the ability to exist in different possible worlds. As Plantinga's position may be interpreted, God also has the ability to exist in different possible worlds (see p. 229, Beginning Metaphysics). Thus, it would seem fair to be able to apply to argument to God, because God can exist in different possible worlds and can make different possible moral choices in these worlds---God can choose to make the heavens and the earth, or God could create something entirely different. He could choose to endow humans with free will, or he could choose not to.

God is omnipotent, and thus God can conceivably and freely choose evil. Some theologians have countered that God cannot have the ability to freely choose evil, due to God's supreme goodness, and because it represents an imperfection. But clearly it does not, for it seems evident that it is a more maximal perfection to be able to choose evil and not choose it rather than to not do it merely because one doesn't have the ability to. Much in the same way, it would seem that it is much better to freely choose not to steal without any disincentives than to not steal because the store's owner is threatening you at gunpoint.

But, if God indeed cannot choose evil, then God wouldn't be free, and thus God wouldn't be omnipotent, because an omnipotent God would be able to conceivably and freely choose anything that is logically consistent. If God can conceivably and freely choose evil, but never does, then it seems possible that God can actualize creatures that can conceivably and freely choose evil but never do. This seems possible given that a supremely good being, even with significant free will, will never choose evil and an omnipotent God, who is such a supremely good being, would be able to create a supremely good being, who as such never chooses evil. Then, there would be no possible world in which a supremely good being chose evil, for if there were such a world, then it would not be the case that the being was supremely good. If God cannot actualize such a creature, it would seem that God is not omnipotent, and thus, not perfect.

Suppose, however, that even for a supremely good being, that there still existed a possible world in which that being chose evil on at least one occasion. Then it would be fair to say that if there exists a possible world in which God chooses evil on at least one occasion, then it seems, by Plantinga's trans-world depravity argument, that God could not actualize a world in which God can conceivably and freely choose evil but never does. God, it seems, can conceivably and freely choose evil but never does; but if God does choose evil on at least one occasion, then God is responsible for at least one evil, in which case God is not perfect. Therefore, God cannot actualize the actual world, in which case God wouldn't exist at all. If God, however, could actualize a world with a significantly free God but no evil, then it also seems that God would be able to actualize a world with significantly free creatures, but no evil, which would contradict the essence of Plantinga's argument. As I have mentioned above, significantly free creatures and evil are logically separable, and thus we can have the former without the latter.

Conclusion

In conclusion, Mackie's argument against a perfect God remains relevant and unanswered. The response from the perspective of higher-order goods fails because of the fallacious assumption that more good means better, even if it comes with corresponding evil, and even if the good redeems its evil. The response from free will fails because of its insistence on maintaining that free will and evil are logically inseparable, which they clearly are. Plantinga's support of the free will defense does not fail, necessarily speaking, but his account for evil, if applied to an omnipotent God that has free will, raises concerns. If his thesis of trans-world depravity is to fully succeed in rejecting the second premise of Mackie's argument against the free will defense, Plantinga must give a better account of precisely why we can't apply this thesis to God, if God does indeed have free will. If we do apply his thesis to God, then it seems that it becomes self-refuting, resulting in the conclusion that God doesn't exist at all.