As I mentioned in the last section, the Jewish and Christian traditions hold that the essence of ethics consists in love for particular people, not in the maximizing of an impersonal quantity of happiness or pleasure. This kind of person-directed love is expressed by the Greek word "agape" (ah-gopp-ay) in the New Testament. The argument from evil can be reformulated so that it presupposes an agapeistic instead of a utilitarian ethic.
Some theists (the really tough-minded) would reject premise 1. I will call these the hard-boiled theists. It is hard to see how any form of the argument from evil could touch hard-boiled theism. For many of us, hard-boiled theism comes quite close to dystheism, since, by limiting the scope of God's love, it drives a wedge between ethics and God's character. I will, therefore, consider how a tough-minded but only soft-boiled theist might respond to the agapeistic argument, without challenging premise 1.
Premise 6 seems a weak link. It assumes both : (1) God has no conflicting wants, and (2) God operates under no deontic constraints. A deontic constraint is an ethical prohibition that forbids doing something, even if its consequences would be very desirable. For example, there are deontic constraints that forbid breaking a promise or telling a lie, even when doing so would produce the greatest happiness for those affected.
We could take care of the first assumption (the assumption that God has no conflicting desires) by strengthening premise 1 to read: God's paramount desire is to love everyone. This new premise is considerably less plausible than the original. Even a soft-boiled theist might hesitate to assume that God's desire to love human beings takes precedence over all other values. Nonetheless, even if we accept the strengthened premise 1, we must still deal with the even more difficult matter of deontic constraints.
In order to close this loophole, we would have to add two additional premises:
13. God is bound by no deontic constraints except those He has voluntarily undertaken.
14. If God wants to love x, God would not voluntarily undertake any deontic constraints that would prevent Him from making x eligible for His love.
Premise 13 seems quite reasonable, since God is under no one else's authority. Premise 14 seems plausible, but it is in fact probably false, since it overlooks a very significant possibility. It might be that the very existence of some human beings presupposes the fact that God has already undertaken certain deontic constraints. When God undertakes deontic constraints, when, for example, He enters voluntarily into binding commitments, like promises or vows, this shapes the subsequent course of history in many profound ways. Deontic constraints undertaken by God long ago are inextricably connected with the actual history of humankind. If, as seems plausible, my identity is bound up with my place in human history as it had unfolded at the time of my conception, then my identity presupposes those prior deontic constraints. Had God not undertaken the vows and commitments He has, then human history would have gone very differently, and it would have been logically impossible for me to exist. Someone very much like me might have existed in those very different circumstances, but that person would not be numerically identical to myself. (For a comprehensive treatment of this issue, see Robert M. Adams, "Existence, Evil and Self-Interest", in The Virtue of Faith.)
In order to take this possibility into account, we need to add another proviso to premise 14 and add yet another premise to the argument:
14R. If God wants to love x, God would not voluntarily undertake any deontic constraints that would prevent Him from making x eligible for His love, unless God's so undertaking is a logically necessary condition of x's existence.
15. For no human being x, and for no undertaking by God of a deontic constraint is it the case that that undertaking both prevents God from making x eligible for God's love and is a logically necessary condition of x's existence.
Now, premise 14R seems correct (at least, under the assumption that God's desire to love is His paramount desire). However, premise 15 is plainly false, for the reasons I suggested above. Significant features of the past, up to and including the event of my conception, are logically necessary conditions of my existence. After my conception, things could have gone quite differently without excluding my existence since, once I have come into existence, I can continue to exist under a wide varietyof circumstances. However, any hypothetical change to the conditions of my origin bears directly on the question of whether the resulting person is me or only someone very much like me.
This thesis that the origin of a thing is essential to its numerical identity is defended quite ably by Saul Kripke in his classic, Naming and Necessity. Even if Kripke's theory is incorrect, the agapeistic argument from evil is in trouble unless Kripke's theory (and any relevantly similar theory) is obviously false.
There is another argument that reinforces the point made in the last section. Even if we reject premise 1 of the agapeistic argument, denying that God's paramount desire is to love human beings, we still face a problem raised by Ivan Karamazov, a character in Fyodor Dostoevsky's The Brothers Karamazov. Ivan marshalls several stories of horrific suffering inflicted on innocent children. Ivan argues that no amount of good at the cosmic level could possibly outweigh (to one who loves these children) the horrors besetting them.
As in the agapeistic argument, it is clear that premise 4 needs to be modified by adding two provisos:
4*. ... unless doing so would result in the loss of some greater good, or x's doing so would violate some prior deontic constraint.
If God has entered into binding deontic constraints, commitments that prohibit God's interference in human affairs except in special circumstances, then these constraints can explain why God, despite His love for the children involved, does nothing to prevent the horrors described by Ivan. This answer again raises the further question: why would God enter into such commitments, knowing that they would interfere with His preventing the affliction of His loved ones by horrific evils?
As I argued in the last section, the answer to this latter question may lie in the presuppositions of the existence of the specific human beings involved. If a deontic constraint is a necessary condition of the existence of a particular child, then God's undertaking this constraint cannot be inconsistent with His love for that child. Love presupposes the existence of the beloved, and, therefore, nothing required by that existence can be incompatible with a love relationship.
Are there cases (cases involving horrific evils, such as those put forward by Ivan Karamazov) in which love for a person requires one to prevent that person's existence? Could there be cases in which my love for X demands that I act so as to prevent X's coming into existence? I would argue that the answer to these questions must be, No. As I suggested above, love for X must always be compatible with any condition necessary for X's existence. This truth is expressed by the following principle, the PEPL:
Here is a simple argument for PEPL:
In summary, both the standard arguments and the agapeistic arguments from evil fail. In demonstrating this failure, it was not necessary to bring in the role of free will or other limitations on God's omnipotence. Nonetheless, I believe that there are relevant limitations to God's omnipotence, that provide us with independent grounds for rejecting the arguments from evil.