It has been suggested that to hold to a greater good defense (GGD) in response to the problem of evil, one must hold to a consequentialist theory of ethics. If the GGD proponent is not committed to consequentialism and is committed to an alternative ethical theory, a divine command theory, for example, then the GGD proponent is guilty of an inconsistency within his or her system of beliefs. This is a shame. The GGD is eminently plausible. So, is this right? The following is an attempt to provide a way out for the GGD proponent. Roughly I shall argue first that the GGD proponent does not have to appeal to any causal thesis about the production of greater goods. Feinberg’s objection relies, in part, on taking “production” to mean cause and by removing references to causes in her defense, the GGD proponent can blunt the objection. I will suggest that causal-consequentialism, the thesis that actions are justified by their effects either lets in too many or too few consequences. Therefore, consequentialism should avoid appeal to any causal thesis.
The Problem of Evil, the Greater Good Defense, and Consequentialism
The challenge of the problem of evil to the theist lies in supplying the objector with reason to think that the world we live is sufficiently good to outweigh evil. For example, William Rowe invites the theist to show that there are goods that outweigh evil that cannot be obtained without evil. He suggests that an incidence of an evil without a greater good is grounds for skepticism about the existence of God. His argument depends on a moral premise that states that God would remove evil unless evil could not be removed without the loss of some outweighing good or allowing an equally bad or worse evil.[1] In order to achieve this aim, the theist is charged to supply some morally sufficient reason for God to permit evil.
A common strategy is to suggest that God allows evil in order to produce some goods that would not be possible without evil. The greater good defense argues that there are some goods that logically entail the presence of evil in order for them to obtain. For example, compassion, mercy, forgiveness, grace, and courage all appear to require the presence of evil. One cannot forgive anyone if no one has done anything wrong. Some GGDs suggest greater goods such as a more complete knowledge of and relationship with God or simply the glory of God.[2] Roughly, the argument proceeds as follows: Some goods are only possible if there is evil. A world with greater goods plus evil is a better world than one with no evil and no greater goods. Therefore, God is morally justified in allowing evil.
The GGD sometimes involves a second argument to show how particular evils produce greater goods. The reason for a second argument is that though some goods logically entail evil, there is the possibility that evil won’t produce those goods. Perhaps there is a battle in which no one displays any courage and everyone dies brutally. In fact, it could be argued that there are possible worlds in which no courage is produced by any of the battles. So, God has to ensure that goods are produced from evil in the world he chooses to create. God can ensure that the good obtains through causing them to obtain. Therefore, God is not obligated to remove evil if he ensures that the goods obtain from evil.
According to John Feinberg, a GGD proposes that “evil is causally and logically tied to some subsequent good which justifies the evil’s existence.”[3] Feinberg argues that if the theist is committed to a non-consequentialist theory of ethics, then she should not use the GGD:
Those who use this defense or some version of it invariably justify evil’s existence in terms of something it produces for which it is the logical precursor. This, at least in part, is how one knows that the greater good defense incorporates consequentialist ethics. If a greater good defense justified evil’s existence in terms of something subsequent to the evil, but the greater good defender held a nonconsequentialist ethic, that would generate a contradiction in the theologian’s system.[4]
A consequentialist ethic is a form of teleological ethic that suggests, “the moral status of actions depends on their relationship to the accomplishment of some end, or goal, rather than some feature of the actions themselves.”[5] For example, a utilitarian ethic is a form of consequentialism and determines the moral status of an action by gauging which actions produce happiness: “actions are right in proportion as they tend to promote happiness, wrong as they tend to produce the reverse of happiness.”[6] In sum, the consequentialist holds that “the moral status of an act is exhausted by the moral status of its consequences.”[7]
In contrast, a deontological ethic proposes that the moral status of an action is exhausted by the intrinsic moral status of the action itself. The essential feature of this view is that the outcome of an action is not determinative of its moral status. Consequently, an action may be considered good even if by its being carried out a subsequent evil occurs. Feinberg concludes that the GGD entails a form of consequentialism since the principle of an outweighing good is an appeal to a subsequent good that requires some evil in order for it to obtain.
The problem, then, is for those who want to hold to a GGD and a non-consequentialist theory of ethics. There may be some GGD proponents who are also consequentialists or those who think God’s moral justifications are consequentialist but that human beings should not be consequentialists or who hold to a mixed theory of ethics.[8] Whether holding to any of these views is inconsistent is not in the purview of this paper. However, those who reject consequentialism entirely are open to Feinberg’s objection. There is, therefore, an apparent contradiction within the GGD proponent’s system if she holds to a non-consequentialist ethical theory. This is seldom explicit in the defense itself. However, Richard Swinburne adopts a form of GGD and holds to a non-consequentialist ethic. In reference to his defense, Swinburne argues:
God… could not achieve some of his good purposes except by means of a delay before they are achieved, and these and other good purposes except by means of allowing evil to occur… some of the good states cannot be achieved without delay and suffering, and that the evil of this world is indeed necessary for the achievement of these good purposes.[9]
Key to Swinburne’s defense is an appeal to a moral justification akin to Feinberg’s characterization: Evil is necessary for subsequent goods. However, Swinburne insists he does not hold to a consequentialist ethic:
Now clearly the goodness of its consequences makes for the goodness of an action, and the badness of its consequences makes for its badness. But I side with the anti-consequentialist in holding that other considerations are also important in determining the goodness or badness of actions; for example, lying and some forms of deception are bad, independently of any good consequences they might have.[10]
Is Swinburne committed to a contradiction? It is plausible to suggest that Swinburne provides a way out within his defense. Swinburne argues that God has a different set of rights than human beings. For Swinburne, rights are generated by the duty to care. Since God is responsible for the care of all that he has created, he is within his rights to allow some harm to come to creation for the purpose of some good. In the same way parents have rights to allow harm to occur to their children for some greater good. Perhaps, then, Swinburne can justify God allowing evil not only on the basis of subsequent goods but also in the authority, or right, of God to allow evil to happen to those under his care. If so, then he could argue that there is a prior ethic of duty that is necessary for the posterior evaluative claim. However, David McNaughton refers to Swinburne’s inconsistency as “Swinburne’s Lapse” and has argued that Swinburne’s argument requires a fundamental revision in order to avoid the consequentialism objection.[11]
Swinburne’s solution, if indeed this is what he considers the solution, appears to fail to avoid the consequentialism objection. While considering God’s entitlement to allow harm to befall his creation introduces a right God has it does not diminish the underlying moral justification for allowing evil. Consider the greater good principle: an evil, E, is morally permitted iff it is logically necessary for some good and causally sufficient to bring it about. One can rephrase the principle in order to account for Swinburne’s principle of rights. God has the right to allow evil, E, to befall those under his care iff E is logically necessary for some good, G, and causally sufficient to bring about G for those under God’s care. The additional claim that God has the right to allow evil does not remove the fundamental consequentialist assumption of Swinburne’s argument.
Some philosophers have argued that the problem of evil itself assumes a consequentialist ethic. Consequently, in order to supply any defense one must assume a consequentialist ethic. Eric Reitan argues that since William Rowe’s problem of evil presupposes a consequentialist ethic, the theist is within his rights to deny the moral premise and work instead to revise the premise to assume a deontological ethic and then to answer that problem.[12] The result of Reitan’s proposal amounts to the abandonment of both the problem of evil as presented and the greater good defense itself.
Though this may be a fruitful avenue, the GGD proponent is not confined to it. In this paper I propose two possible responses that the non-consequentialist GGD proponent can use to stave off Feinberg’s objection while retaining the essential thrust of his argument. The main conclusion is that the GGD need not entail a consequentialist ethic. First, I will suggest that Feinberg’s causal assumption is not true of some forms of the GGD. Second, I will argue that the evaluation of consequent states of affairs does not entail a consequentialist ethic. The latter claim will also involve a suggestion for the GGD proponent to use evaluative criteria to press the atheist to consider her own ethical assumptions.
Although not always stated by the proponent, the GGD sometimes relies not only on logical necessity but causal sufficiency. The greater good in question is usually said to be the effect of some evil. The consequentialism objection suggests that the GGD implies that some good is caused by evil. If the GGD proponent can show that logical necessity justifies God without entailing causal sufficiency, she can claim that she is not guilty of appealing to a consequentialist ethic. In other words, the GGD proponent can argue that the GGD does not entail consequentialism.
The appeal to cause is motivated by the concern that logical necessity is not enough to bring about the desired good. The truth of the claim, “if courage, then evil” is circumstance independent. One needs no instance of evil or courage for it to be true. A GGD, however, has to deal with the actual presence of evil in the world. Logical necessities don’t produce anything. This is where one needs a causal claim. A causal claim is something like, “a battle in which Jim’s friend is injured causes Jim to courageously rescue his friend from the battlefield.” If God is justified in allowing evil because i) an evil is logically entailed by some good and ii) God can ensure that the desired good obtains, then he has to cause the good outcome from the evil act. In a GGD, the proponent makes the case that evils are not only logically necessary, but producers of subsequent goods. I.e. Some Evil (a battle) causes a person’s or some peoples’ courage. God is justified in allowing evil iff evil is both the logically necessary condition for the desired good and those evils that are in the world produce or result in those desired goods. The battle, its horror and death, produces a state in Jim that results in his courageous actions.
The most troubling part of any consequentialist ethic that relies on causation is that causation either includes too few or too many consequences. Carolina Sartorio suggests that whichever direction one takes it is difficult to determine what should be included in the roster of consequences. Consider the following: “An assassin murders a child’s parents. When he does, the child becomes an orphan.”[13] On a restrictive view,
the assassin’s murdering the parents doesn’t cause the child’s orphanhood, for the fact that the child’s parents were murdered entails the fact that the child is an orphan, and entailment is not causation. Thus causal consequentialism would imply that the child’s orphanhood is not among the consequences of the act of murdering the parents.[14]
Sartorio asks us to consider the consequence of considering entailment as causation. The result would be a chain of consequences too numerous to quantify. Every subsequent state of affairs would, in part, be not only contingent on the murder but caused by it. If so, then whatever happens to the orphan after the event could be said to be caused by the murder of the orphan’s parents.
Consequently, the GGD proponent can argue that though some goods are entailed by evil the defense need not appeal to any causal thesis. Instead the defense rests on the logical entailment of evil for some goods and the appeal to the intuition that the world with evil and those goods is better than a world without evil and without those goods. The revision would enable a slightly reduced version of the greater good principle. Call it, the greater good principle 2: Evil is morally permitted iff it is logically necessary for some greater good.[15]
Though removing reference to cause may blunt the objection it does not successfully remove the charge of consequentialism. For the charge may be rephrased without an appeal to the notion of causation. Instead, the objection could seek to show that the evaluative premise assumes a consequentialist ethic. The GGD proponent remains committed to demonstrating that the goods entailed by evil in the world outweigh or, more precisely, outvalue the total evil that God permits. Since only consequentialists determine the moral status of an action in terms of the value of goods, the GGD proponent remains illicitly committed to consequentialism. The GGD proponent, the objector will suggest, leaves no room for a non-consequentialist ethic since the GGD proponent is committed to evaluating the moral status of actions in terms of the value of posterior states of affairs.
In order to defeat this claim, the GGD proponent has to show that there is no such entailment, that valuation of states of affairs does not entail a view of the moral status of actions. Rather, evaluations of states of affairs do not require consideration of any logical or causal connections between greater goods and evil in order to produce a judgment. In fact, there may be a way for the GGD proponent to reverse the order and suggest that since deontological claims would have to precede evaluative claims, an action’s moral status would have to be determined prior to any evaluative claim. The latter strategy leads to my promised suggestion as to how to use evaluative claims in order to suggest that the atheist has an obligation to prefer a world with greater goods over a world without them.
It is commonly suggested that the distinction between evaluative claims and moral claims is what bears them. Actions bear moral statuses whereas states of affairs bear evaluative statuses.[16] However, there is good reason to reject this view. One evaluates a state of affairs by what is contained within the state of affairs. The state of affairs itself is not under scrutiny but the facts about the state of affairs in virtue of which the state of affairs is deemed good or bad.[17] Within the battle Jim has the courage necessary to rescue his injured friend from the battle field. The relevant feature is not the state of affairs but the constituting component of Jim’s courage. However, for the sake of simplicity, I will continue to refer to evaluative claims in terms of states of affairs. I don’t think it makes much difference to my argument.
The most vital thing to note is that a moral claim or justification is not the same as an evaluative claim. A moral justification attempts to provide some criteria by which one can judge an action to be morally right or wrong. A claim about value is is a judgement about the goodness of a possible or actual state of affairs (or, more fundamentally, its features). The moral status of an action may contribute to the overall value of a given state of affairs, but it is not identical to it. Consequentialists often appeal to prospective states of affairs that are dependent upon an action. An action is morally justified if it brings about the most goods. Thus, consequentialists are committed to an evaluative premise in order to provide a moral judgment.
Deontological ethical theories are said to go in reverse. They generally begin with a judgement about the moral status of an action before making any value judgments. The goodness of an action is evaluated in terms of the action having a sufficient reason or moral justification. On a divine command theory of ethics, an action is morally justified if it accords with a command of God. The best action, then, is to obey a command. Crucially, the action is good because it is morally right. It is not morally right because it is good.[18] Philip Petit rejects this characterization. He argues that all ethical theories commence with what is good and then move to discern what is right. On a deontological ethical theory, Petit argues that the properties in question are properties that can be analyzed in terms of goodness, the goodness of obedience to a divine command or the goodness of acting in accordance with natural laws.[19] An action, even on a deontological theory can be morally right because it is good.
The evaluative claim of the GGD, the claim that a world in which there is evil and greater goods is better than a world without evil and those goods, is a claim that is index linked in so far as it applies evaluative claims to a person. This is an important claim. Any answer to the problem of evil appeals to an intuition about what is of a high enough value to justify God not removing evil. The question is: valuable for whom? If the evaluative claim is linked to the theist, then the theist has the responsibility of providing criteria for any evaluative claim. However, a GGD relies not only on what the theist may value but on what the atheist values. Indeed, a common response to any defense that suggests that a good-making property of the world justifies God allowing evil is to suggest that the good-making property is not so good after all. Someone might respond to the free will defense in this way by saying that he would rather lack the capacity of libertarian free will and experience no evil.
Consider the following: G is valuable for A. If A values G and G entails E should A value G? The answer is: it depends if A thinks the cost of G is worth paying. However, this assumes a relationship between G and E that is instrumental not logical. On a consequentialist account if E is the means to G, then one should evaluate E in terms of the consequent G. However, the GGD does not have to appeal to this premise. Rather, the appeal is to the value of G in terms of intrinsic goods. Given this account of evaluation, it is open to the GGD proponent to claim that the evaluative premise in the defense refers to intrinsically good features of states of affairs the bringing about of which is logically dependent on evil. Crucially, the GGD proponent can detach the evaluative claim from the defense and argue that any state of affairs that includes intrinsically good features such as compassion, mercy, and courage but also has evil in it is more valuable than a state of affairs devoid of those features.
It is worth noting that precisely the same evaluative process is produced by alternative defenses. Most notably, the free will defense invites the objector to evaluate the value of a world in which human beings have libertarian free will and evil compared to a world with no libertarian free will and no evil. If the atheist concludes that a world with libertarian free will and evil is a better world that one without those features, then the free will defense succeeds in its evaluative claim. This line of reasoning is the same for free will defenses and for greater good defenses. Importantly, the same line of reasoning is used in Feinberg’s own defense.[20] On Feinberg’s view, a world with non-glorified human beings and evil is better than a world without those features. The atheist is asked to consider the value of human beings in a non-glorified state in the same way he is asked to evaluate a world with libertarian free will or greater goods (strictly speaking, libertarian free will, non-glorified human beings, and second-order goods all count as greater goods). The point is that the evaluative process takes place independently of any logical relationship the intrinsic good has with evil and does not assume any ethical theory.
If this is right, then there is a way to establish the evaluative premise without appeal to consequences or the logical entailment between evil and greater goods. The atheist is asked to consider two possible worlds, two maximal states of affairs. He is then asked to consider which one is of the most value. This can be done without appeal to a consequentialist ethical theory.
The cost of the consequentialist objection now carries the same price that would have to be paid by nearly all defenses. The free will defense and the GGD alike account for God allowing evil in terms of what God can logically achieve. If it turns out that God cannot remove evil without removing something of sufficient value in the world, then he is not obliged to remove evil.
Fitting Attitudes
While this is sufficient, I think, to defeat the consequentialist objection, there is another line of argument the GGD proponent may pursue in order to strengthen the force of her argument. What I have in mind is a claim about the priority of deontic claims over evaluative claims whereby the GGD proponent can argue that we are morally obliged to value one state of affairs, specifically a state of affairs that includes the aforementioned greater goods, over an alternative state of affairs, one without those greater goods.
So far we have assumed that any evaluative claim is up to the one making the evaluation. However, the GGD proponent can claim that there is a prior obligation to desire certain states of affairs over alternative states of affairs. On a fitting attitudes view of evaluative claims, the good or value in question is not determined by a resulting state of affairs but a prior obligation to desire certain states of affairs over others.[21] The outcomes in question are intrinsic goods such as compassion, courage or mercy. The obligation in question is rationality. It is rational to prefer a state of affairs that includes features that are intrinsically good. The fitting attitudes view has a useful feature. One can use it to suggest that the evaluative premise, if it is person linked, assumes a prior obligation – the obligation to desire certain states of affairs over others. Desires, on this view are considered in terms of duties.
John Stuart Mill argued that there are higher order goods and lower order goods. He proposed that there is an obligation to value higher order goods over lower order goods and that those who had experience with higher order goods should train those who had not had similar experiences to appreciate finer pleasures. Though Mill’s use of this argument was to further a utilitarian ethic, the basic strategy can be adapted for our purposes. One can suggest that there are some states of affairs that ought to be valued above others. One could argue that greater goods are of such intrinsic worth that it is fitting for us to desire their presence. It is rational and right to value greater goods. Greater goods cannot be obtained without evil. Therefore, it is rational and right to value a world in which there are greater goods and evil.
Some work may need to be done on such an argument. For example, it is not clear whether this helps or hinders the GGD proponent’s defense against the consequentialist objection. Fitting attitude theories are largely advocated by consequentialists. That aside, it is not clear how the GGD proponent would ground such an argument. What is it, exactly, that would obligate the atheist objector to desire those goods as opposed to other goods, goods that entail no evil? I don’t think there is much to go on here. As has often been argued, atheists have very little to ground any obligation apart from a consideration of the consequences of their decision to desire certain goods and this would amount to asking the objector to chase his own tail.
However, for the Christian, especially one committed to non-consequentialism, the obligation can be grounded in God’s will. The obligation to prefer a state of affairs in which there are greater goods—forgiveness, mercy, compassion, atonement for sin, hope of redemption and all the rest—is grounded in the wants and desires of God himself. To have a fitting attitude to a state of affairs—this world—is an obligation to align one’s attitude to the attitude of God. Since this is actually the world which accords with the will of God it is the world which the believer is obligated to desire.
The most obvious objection to such a line of thought is that to desire a world in which there is evil is to desire evil and that appears to run contrary to a basic moral intuition – it is wrong to desire evil. However, what is entailed by my suggestion is not that evil ought to be desired but that a state of affairs in which there are greater goods ought to be desired and that those greater goods entail the presence of evil and mutatis mutandis for God and his relationship with evil.
Alvin Plantinga has suggested that the “towering” good of the incarnation and atonement outweigh the value of any amount of evil and that they entail the existence of evil. Further, Plantinga argues that all the very good possible worlds contain some version of incarnation and atonement. If so, then all the very good possible worlds contain evil.[22]
Plantinga notes the obvious objection that sufferers are treated as a means to an end. What if the suffering we undergo does not lead to some greater good for us but for someone else or for the total value of a consequent state of affairs? Plantinga argues that if God is indeed using our suffering as a means to a greater good we should desire that greater good even if it is for the benefit of others:
suppose further yet that God knows that I would not accept the suffering in question, but only because of disordered affections; if I had the right affections (and also knew enough), then I would accept the suffering: in this case too, as far as I can see, his being perfectly loving would not preclude his allowing me to suffer.[23]
Plantinga argues that if we freely accept our role as a sufferer, then God is justified in allowing us to suffer. To freely accept the role as sufferer one must either actually know that the suffering is entailed by the greater goods and desire those greater goods and accept suffering, or had one known that the suffering is entailed by greater goods one would desire those goods and accept the suffering. However, in cases where the sufferer does not desire the greater goods it is not that God has done wrong but that the sufferer has wrong affections. The sufferer, in other words, is wrong to value a world without those greater goods to which his or her suffering is linked. According to Plantinga, our desires, and thereby our evaluation of this world, ought to accord with God’s desires. When they don’t it is not God who has gone wrong but us.
Conclusion
I have argued that the greater good defense against the problem of evil is not guilty of necessarily assuming the truth of a consequentialist ethical theory and, therefore, does not hold to a contradiction within her set of beliefs. I have suggested that the GGD proponent avoid appeal to any causal thesis on the grounds that it is difficult to either limit or include relevant consequences into the causal chain. Further, I have argued that the evaluative claim found within the GGD does not entail consequentialism and if it does, then nearly all defenses appeal to consequentialism. Finally, I have suggested that the evaluative claim within the GGD can include an argument for the obligation of Christians to prefer a world with greater goods than a world without them. If that final argument goes through, then not only is the GGD proponent clearly not guilty of illicit consequentialism, but she also has a further line of argument to make about the moral obligations of human beings to favor God’s will for the world.
[2] For example, John Piper argues that evil occurs in order to manifest the glory of Christ. John Piper, Spectacular Sins (Wheaton: Crossway, 2008).
[3] John Feinberg, The Many Faces of Evil (Wheaton: Crossway, 2004), 135.
[4] Ibid., 137–138.
[5] Louis Pojman and Peter Trammel, Moral Philosophy: A Reader (Indianapolis: Hackett, 2009), 156.
[6] John Stuart Mill, Utilitarianism in Pojman and Trammel, 158.
[7] Carolina Sartorio, “Causation and Ethics,” in The Oxford Handbook of Causation, ed. Helen Beebee, Christopher Hitchcock, and Peter Menzies (Oxford: Oxford University Press, 2012), 575–591.
[8] “An end-justifies-the-means ethic is fallacious and therefore wicked for finite men (who can neither control nor know all the results of their choices), but it is perfectly fitting for the infinite God (who both controls and knows all the results of His choices)–and, after all, God being supreme need not justify His choices to anyone.” E. Calvin Beisner and Chad Meister, “Point-Counterpoint: How Should Christians approach the problem of evil?” Christian Research Journal 30, no 5 (2007), 5.
[9] Richard Swinburne, Providence and the Problem of Evil (New York: Oxford University Press, 1998), x.
[10] Ibid., 254.
[11] David McNaughton, “The Problem of Evil: A Deontological Perspective,” in Reason and the Christian Religion: Essays in Honor of Richard Swinburne, ed. Alan Padgett (Oxford: Clarendon Press, 1994), 329–351 and David McNaughton, “Is God (almost) a Consequentialist?: Swinburne’s moral theory.” Religious Studies 38, no. 3 (2002): 265-281.
[12] Eric Reitan, “Does the Argument from Evil Assume a Consequentialist Morality?” Faith and Philosophy 17, no. 3 (2000): 306–319.
[13] Sartorio, 3.
[14] Ibid.
[15] Interestingly, the argument Feinberg critiques, Keith Yandell’s version of the GGD, does not appeal to causation. “Every evil is logically necessary to some good which either counterbalances or overbalances it, and some evil is overbalanced by the good to which it is logically necessary” Keith Yandell, “The Greater Good Defense” Sophia 13, no. 3 (Oct, 1974), 4.
[16] For example, Jeremy Evans, The Problem of Evil: The Challenge to Essential Christian Beliefs (Nashville: Broadman & Holman Publishers, 2013), 133.
[17] Mark Schroeder, “Value Theory,” The Stanford Encyclopedia of Philosophy (Summer 2012 Edition), Edward N. Zalta (ed.), URL = http://plato.stanford.edu/archives/sum2012/entries/value-theory.
[18] Ibid.
[19] Philip Pettit, “Consequentialism,” in A Companion to Ethics, ed. Peter Singer (Oxford: Blackwell Publishers, 1991), 230.
[20] This was pointed out to me by Greg Welty. Greg Welty, email message to the author, January 22, 2016.
[21] Henry Sidgwick and Marcus George Singer. Essays on Ethics and Method (Oxford: Clarendon Press, 2000).
[22] Alvin Plantinga, “Supralapsarianism, or ‘O Felix Culpa’”, in Christian Faith and the Problem of Evil, ed. Peter van Inwagen (Grand Rapids: Eerdmans, 2004), 1–25.
[23] Ibid., 24.
