Suppose some evil appeared to us to be gratuitous. How far would that fact count against belief in God?

Suppose some evil appeared to us to be gratuitous. How far would that fact count against belief in God? The problem of evil is one of the oldest questions of religion and philosophy- the two oldest books in the Bible, Job and Genesis with its Fall narrative, contain important contributions to the reconciliation of an omnipotent and good God with the existence of suffering. On the other side, Hume quotes Epicurus, from the 3 rd Century BCE, whose questions are yet unanswered. Is he [God] willing to prevent evil, but not able? Then is he impotent. Is he able, but not willing? Then is he malevolent. Is he both able and willing? Whence then is evil? 1, advancing a logical problem of evil. More recently, the evidential problem of gratuitous evil, that the abundant existence of apparently preventable and horrific ( in other words gratuitous) evil counts as evidence against the existence of an omnipotent, omnibenevolent God, has been touted as a sufficient defeater for the God of classical theism. Andrea Weisberger calls evil, particularly gratuitous evil, the most fundamental threat 2 to classical theism, and J.L. Mackie argues this makes theism positively irrational 3. How far do these examples of apparently gratuitous evil count as evidence against belief in God? In this essay, I shall argue that these examples of apparently gratuitous evil hold no weight as evidence against belief in God, for to advance such an argument, one must assume moral knowledge. Moral knowledge itself, however, is not underpinned by naturalism, but rather by theism, providing strong confirmation for the existence of a deity. This leaves the atheist with a dilemma; either we have moral knowledge, in which case theism is confirmed over atheism and gratuitous evil counts as evidence for belief in God; or we do not have moral knowledge, in which case the gratuitous evil must be abandoned as evidence entirely. Either way, the evil, however apparently gratuitous, cannot serve as evidence against belief in God. 1 Hume, D., 2008, Dialogues Concerning Natural Religion, Oxford: Oxford University Press. 2 Weisberger, A., 2006, The Argument from Evil in: M. Martin, ed. 2006, The Cambridge Companion to Atheism, Cambridge: Cambridge University Press. 3 Mackie, J.L., 1955, Evil and Omnipotence, Mind, Vol. 64, No. 254, p. 200-212 1

1. Initial considerations concerning the evidential problem of evil Gratuitous evil has long been considered a key problem for the existence of God- since an omnibenevolent and omnipotent God seems both willing and able to prevent gratuitous (horrific and unnecessary) evil, natural or moral. When we perceive many instances of apparently gratuitous evil, this therefore seems to count against belief in God. By counting against, I refer to evil providing evidence against God, rather than showing that God is logically impossible; or advancing the evidential problem of evil, not the logical problem. This takes the problem of evil at its strongest- the logical problem of evil seems to fail, for we cannot know that God does not have morally sufficient reasons for permitting [this] evil 4. Instead, the evidential problem argues that atheism is a more reasonable hypothesis than theism when considering gratuitous evil (GE), for we would not expect this evil to exist with theism, but we would expect the gratuitous evil to exist with atheism. In probabilistic terms, P(GE/Theism) is far lower than P(GE/Atheism): so a Bayesian analysis would show that gratuitous evil provides evidence for atheism over theism. Evil that appears gratuitous is widespread- W.L. Rowe considered a fawn [who] is trapped, horribly burned, and lies in terrible agony for several days before death relieves its suffering. 5 Weisberger gives more examples in a bold rhetorical flourish- Where was the all powerful, all knowing, wholly good being whose very essence is radically opposed to evil, while millions of children were starved to death by Stalin, had their limbs chopped off with machetes in Rwanda, were turned into amputees by the diamond trade in Sierra Leone, and worked to death, even now, by the child slave trade that, by conservative estimates, enslaves 250 million children worldwide? 6 Paul Draper further illustrates the problem with biologically useful pain 7, a widespread phenomenon. These examples- both moral and natural evil- certainly appear to be gratuitous, and so the extent to which they count as evidence against belief in God is an important question for the theist. 4 Craig, W.L., The Problem of Evil, Available at: http://www.reasonablefaith.org/the- problem- of- evil [Accessed 30 July 2015] 5 Rowe, W.L., 1979, The Problem of Evil and Some Varieties of Atheism, American Philosophical Quarterly, Vol. 16, No. 4, p. 335-341 6 Weisberger, The Argument from Evil, 7 Draper, P., 1989, Pain and Pleasure: An Evidential Problem for Theists, Noûs, Vol. 23, No. 3, pp. 331-350 2

In premise and conclusion form, the argument is as follows: P1. If God exists, then gratuitous evil does not exist. P2. We perceive many apparent instances of gratuitous evil. P3. We are justified in inferring from these that gratuitous evil does exist. Therefore: P4. God probably does not exist. Premise 1 is universally agreed, for reasons expounded above. Theists have instead mainly attacked P3 and the implication to P4. Either theists have attacked the inference from apparently gratuitous to actually gratuitous, through theodicies and defences (e.g. Plantinga s free- will defence or Hick s soul- making) or on the inference to 4, using theistic arguments to raise the prior probability of theism (e.g. cosmological or design). I shall instead attack 2- arguing that if naturalism is true, we cannot know that there are even apparent instances of evil. One must assume theism to affirm this premise, therein providing an undercutting defeater for the argument. Forthwith, I shall only consider the options of theism and atheism, and particularly evolutionary naturalism, and thus shall equivocate between atheism, naturalism and evolutionary naturalism. While these positions are not the only options (e.g. deism), they are the most prominent options; and for the atheist, evolutionary naturalism has been argued by many (Alvin Plantinga, Richard Dawkins, Daniel Dennett, Paul and Patricia Churchland, Alex Rosenberg) to be the only game in town 8. The force of the upcoming objections may not apply to intermediary positions; but discussing other positions remains beyond this essay s scope. 2. The undercutting defeater of moral knowledge The most prominent undercutting defeater for the problem of evil has been objective moral values; to say something is evil or gratuitously evil, we must have an objective standard of evil and good, possible only on theism (as the argument goes). C. S. Lewis is most famous for this line of argument- My argument against God was that the universe seemed so cruel and unjust. But how had I got this idea of just and unjust? A man does not call a line crooked unless he has some idea of a straight line 9 - and this is an argument beloved of 8 Linville, M., 2012, The Moral Argument in: Craig, WL, and Moreland, JP, 2012, The Blackwell Companion to Natural Theology, Oxford: Blackwell Publishing. 9 Lewis, C.S., 2012, Mere Christianity, London: Collins Publishing 3

popular Christian apologists 10. I believe this argument can be strengthened: namely, the evidential problem of evil requires not only moral values but also moral knowledge. To use apparently gratuitous evils as evidence against the existence of God, one must know that some events are evil e.g. the Rwandan Genocide or the Holocaust, and that goodness would entail preventing such evils, if it were an attribute of God. I suggest three arguments that moral knowledge cannot exist if naturalism is true. Firstly, naturalism does not explain how there could be objective moral facts. Evolutionary naturalism argues that we are products of matter and energy, and have evolved entirely by chance, as is all else within this universe. In such a mechanistic universe, it seems that If there were objective [moral] values [or facts], then they would be entities or qualities or relations of a very strange sort, utterly different from anything else in the universe 11. Not only is the ontology of moral values and facts strange, but the very content of these facts is bizarre naturalistically- that humans ought to do something, despite it not being necessary that we act accordingly, unlike a scientific or logical fact about humans. If these entities are so different, however, it is unlikely that these entities exist, and so the queerness 12 of objective moral facts, such as the Holocaust was wrong, is difficult to square with the remainder of the naturalistic worldview. Furthermore, it is commonly assumed that moral facts and obligations apply only to humans (as moral agents) and not to other animals (who still may be considered moral patients). But this presents two problems; firstly, as William Lane Craig has argued, it seems an utterly incredible coincidence 13 that these moral facts obligations simply happened to apply to humans, who evolved after 14 billion years, over other animals or rocks or plants. Secondly, 10 See Keller, T., 2009, The Reason For God, London: Hodder and Stoughton Publishing p. 26; Turek, F, 2015, Stealing from God, Colorado Springs: NavPress, p. 118; The Veritas Forum, 2013, The Problem of Suffering and the Goodness of God - Ravi Zacharias at Johns Hopkins, [video online], Available at: < https://www.youtube.com/watch?v=t7- gp1gc8gm>, [Accessed 30 July 2015]; Geisler, N., 2012, From Apologist to Atheist: A Critical Review, Available at: http://normangeisler.com/from- apologist- to- atheist- a- critical- review/, [Accessed 30 July 2015] These provide an example, and are by no means exhaustive. 11 Mackie, J.L., 1977, Ethics: Inventing Right And Wrong, London: Pelican Books 12 Mavrodes, G., 1986, Religion and the Queerness of Morality in: R. Audi and W. J. Wainwright, ed. 1986, Rationality, Religious Belief and Moral Commitment, Ithaca: Cornell University Press 13 Craig, W.L., 2006, Theistic Critiques of Atheism in: M. Martin, ed. 2006, The Cambridge Companion to Atheism, Cambridge: Cambridge University Press 4

the moral responsibility of humans over animals is questionable; whether human or animal, all are entirely determined by the laws of nature, merely dancing to [the] music 14 of our DNA. We would not consider a table, robot or a cow morally responsible, if they are entirely determined - so why should we consider humans morally responsible? If there is no moral responsibility, there can be no moral obligations and no moral facts. The absence of moral facts renders moral knowledge impossible, and so Alvin Plantinga rightly argues that Naturalism can perhaps accommodate foolishness and irrationality, acting contrary to what are or what you take to be your own interests; it can t accommodate appalling wickedness 15. Secondly, naturalism fails to explain rationality. Plantinga argues that if naturalism (and thus materialism) is true, then belief is a neuronal event [and] mental properties are determined by physical properties 16, therefore a belief s content properties are determined by physical properties. These physical, or neurophysiological (NP), properties, will be adaptive and selected if they cause adaptive behaviour conducive to survival. If, however, the NP properties are survival- focussed, there is no reason that the content of our beliefs should be correct. Firstly, if materialism is true, then it is by virtue of the [NP] properties of a belief B, not by virtue of its content, that the belief causes the behaviour it does cause 17, meaning that the content properties are causally irrelevant to behaviour- i.e. If the belief had had the same NP properties but different content, it would have had the same effect on behaviour 18. Moreover, adaptive behaviour will not adapt in accordance with truth, but survival; and a priori, there is no necessary conjunction between the two. If our beliefs need not be true, then evolutionary naturalism undermines our cognitive faculties and their propensity for rationality. This provides one of many arguments against rationality within naturalism: indeed, Victor Reppert lists six arguments against the possibility of rationality given naturalism, from intentionality truth mental causation the psychological relevance of logical laws the unity of consciousness [and] the 14 Dawkins, R., 1996, River Out of Eden, London: Phoenix Books 15 Plantinga, A., 1993, A Christian Life Partly Lived, in: Kelly James Clark, ed. 1993, Philosophers Who Believe, Illinois: InterVarsity Press 16 Plantinga, A., 2011, Where the Conflict Really Lies, New York: Oxford University Press 17 Ibid. 18 Ibid. 5

reliability of our cognitive faculties 19. For example, intentionality, or aboutness, in our mental states, is a peculiar phenomenon given our mechanistic metaphysics, for it points towards something e.g. my thought about a chair intends or points towards a chair. On naturalism, this is distinctly inexplicable; nothing else physical can be about or point towards anything else. If so, intentionality cannot be accommodated by a physicalist framework, despite its necessity for rationality. Moreover, any attempts to dismiss intentionality become incoherent. Calling intentionality an illusion, for example, uses an intentional term: illusions point towards something! Each argument is compelling, and thus, cumulatively, rationality is severely undermined by naturalism. But if rationality is undermined by naturalism, then knowledge itself is undermined, and so moral knowledge is incompatible with naturalism. Finally, moral reasoning cannot coexist with naturalism. Mark Linville argues that our basic moral orientation 20 is determined by evolutionary psychology, and that all moral reasoning will take place within the arena of this moral orientation. If so, this leaves us with serious difficulties for moral reasoning. Firstly, moral reasoning supposes that our basic moral orientation must line up with moral truth. But, as noted before, evolution selects for survival, not for truth so imposing a tracking relation 21 between moral truth and our moral orientation fails all the usual criteria of scientific adequacy 22. The relation is not parsimonious 23 - it assumes a tracking relation when there need not be a tracking relation, multiplying entities beyond necessity; and it is not clear 24 - it gives no reason for the link between survival and moral truth. The lack of simplicity and clarity here gives us strong reasons to doubt the effectiveness of this as a naturalistic explanation. The problem of evolution and moral truth is further developed by Linville s problem, of Darwinian counterfactuals 25. The idea is that if our moral 19 Reppert, V., 2012, The Argument from Reason, in: Craig, WL and Moreland, JP, ed. 2012, The Blackwell Companion to Natural Theology, Oxford: Blackwell Publishing 20 Linville, The Moral Argument 21 Street, S., 2006, A Darwinian dilemma for realist theories of value, Philosophical Studies, Vol. 127, no. 1, p. 109-166 22 Ibid. 23 Ibid. 24 Ibid. 25 Linville, The Moral Argument 6

orientation is radically contingent 26, then in other possible worlds, we could have radically different moral orientations. For example, if we were raised like bees, then our unmarried females would think it a sacred duty to kill their brother, and mothers would strives to kill their fertile daughters 27 ; or if we were raised like wolves, then we may consider justice [to be] inequality 28 rather than equality, given the fitness- conferring value of these beliefs. This leads us into a dilemma of justice - if this situation is called moral in other possible worlds, then morality seems to be just when it follows reproductive success, making a mockery of our usual belief in morality following certain values of individual rights, happiness, equality and mutual respect. If, alternatively, it is argued that this is not moral elsewhere, then we are left with our previous skepticism- naturalism gives no reason that our moral orientation, given its evolutionary determination, is any closer to moral truth than any other moral orientation. Again, the naturalist is left with skepticism about our moral reasoning, and given that moral reasoning is a precondition for moral knowledge, this removes the possibility of moral knowledge from the naturalist. The theist, in contrast, has explanations for how each of these features could arise. As God is the ultimate standard of goodness, the summum bonum, this means that moral values can be grounded in his nature, and moral obligations can be grounded in his commands, flowing out of his nature, allowing for moral facts 29. Since God is the ultimate mind, we can see how intentionality occurs if we come from God, or are made in the image of God as the Christian theist says. Additionally, our moral beliefs and cognitive faculties can be reliable, if an omnibenevolent, omniscient and omnipotent God guided the evolutionary process so that our moral reasoning and normal reasoning faculties were trustworthy, and created creatures endowed with libertarian free will. The theistic hypothesis has here a far more plausible explanation of these occurrences than the naturalistic hypothesis, suggesting that if we are to accept that we have moral knowledge, then we must choose the theistic hypothesis over the naturalistic hypothesis. However, this requires us to reject naturalism before even considering the evidence from evil, meaning that naturalism is critically undercut here. 26 Ibid. 27 Darwin, C., 1882, The Descent of Man and Selection in Relation to Sex, New York: D. Appleton and Co. 28 Linville, The Moral Argument 29 This formulation of theistic morality avoids the Euthyphro dilemma. 7

3. Attempts to retain the evidence from evil We have seen the naturalist s difficulties in explaining moral knowledge, versus the relative ease of accommodating moral knowledge for the theist. As a result naturalists must, if they wish to press the evidence of apparently gratuitous evil, find another way of doing so. I see two such possibilities for the naturalist. Firstly, the naturalist may rephrase gratuitous evil as gratuitous suffering or gratuitous pain, rephrasing so that the naturalist does not assume moral knowledge in formulation. Draper s reference to biologically useful pain 30 provides an example- arguably this is formulated in morally neutral terms. Nonetheless, this response is weak (and not pursued by Draper), because the naturalist must still contend that a God would prevent gratuitous suffering, as a good God would be obligated to do so. But this is moral knowledge: it assumes knowledge of God s moral character or obligations, and also assumes that suffering or pain is bad. The existence of rocks, for example, is considered no problem for theism, nor are chemical reactions such as photosynthesis- we do not consider God obligated to remove these. If God is then obligated to prevent other natural processes such as suffering, then gratuitous suffering must be in a different moral category to photosynthesis and rocks, being evil, again requiring moral knowledge and returning to this problem. The naturalist s more plausible route is to piggyback onto the theistic hypothesis, and argue that the theologian must adhere and cannot consistently adhere to 31 the goodness of God, the omnipotence of God, and the presence of evil. By arguing that certain events would be gratuitously evil if God existed, and that God would then prevent them, these events can then internally count against belief in God. As John Loftus argues: I m arguing that it s bad to have this amount of suffering from a theistic perspective, and I may be a relativist, a pantheist, or a witchdoctor and still ask about the internal consistency of what a theist believes. The dilemma for the theist is to reconcile senseless suffering in the world with his own beliefs (not mine) that all suffering is for a greater good. It s an internal problem for the theist. 32 We must therefore consider the worth of the evidence as undermining theism internally. 30 Draper, Pain and Pleasure. 31 Mackie, Evil and Omnipotence 32 Loftus, J.W., 2007, Why I Rejected Christianity: A Former Apologist Explains Victoria, British Columbia: Trafford Press 8

The problem with this objection is that it assumes that if naturalism were true, then the theist could possess middle knowledge of what would be gratuitously evil (or plausibly so) and that such evil would be incompatible with God s existence. However, such knowledge is inconsistent with naturalism. If naturalism is true, then if the theist merely believes something is gratuitously evil, they could be mistaken. According to the naturalist here, the theist must know what would be gratuitously evil, in the alternate world where theism is true, so that the evidence from evil contradicts their worldview. Nonetheless, there are multiple reasons to deny this assumption. Firstly, as has been demonstrated, knowledge itself is inexplicable if naturalism is true, so the theist could not possess any knowledge. Secondly, moral facts are plausibly necessarily true. But if moral facts depend on God, as previously argued, then if God does not exist in this world, then he does not exist in any possible world, and so moral facts do not exist any possible world, and cannot be known by the theist in any possible world, including this one. Finally, if theism were possible, but naturalism is true, then the world where theism is true would be a very far possible world, and it would be beyond our epistemological capabilities to determine the nature of that world. As an analogy, Hume (in an entirely different context) considers the example of a missing shade of blue in the spectrum. If we suppose that this shade of blue is possible, but does not exist in our world, then it would be beyond our epistemological capability to determine what this shade would look like in alternate worlds. Similarly, if moral facts were possible, we still could not know what they would be if God did in fact exist- the world is too different, and there is also no causal connection between our world and the other possible world for the information to carry across, a plausible requirement for knowledge. This middle knowledge is arguably more problematic than moral knowledge itself on naturalism. Consequently, the theist could not know what would be moral if God did exist, and no internal inconsistency would remain in theism. If so, the evidence of apparently gratuitous evil continues to hold no weight against belief in God. 9 4. Final considerations We have seen the problem of moral knowledge for the naturalist, and that it renders the evidential problem of evil toothless. We are then left to some final considerations. Firstly, the naturalist may simply deny, given the argument laid out, that we have moral knowledge. The theist should grant this - but if the naturalist does so, then the evidence from evil is not available as disconfirming

the theistic hypothesis. This, however, is commonly considered the strongest objection to theism (along with divine hiddenness, which plausibly lands in a similar position) - and so the naturalist must abandon the sharpest arrow in their quiver. This is a sub- optimal outcome for the naturalist. Secondly, there is clearly a plausible intuition in the evidence of gratuitous evil- if there is moral knowledge and a God then why does this apparently gratuitous evil exist? There are several plausible theodicies, but I lean towards a skeptical theist approach. William Alston lists six cognitive limitations which undermine the reliability of our judgments as to what constitutes a gratuitous evil 33, and I affirm with Stephen Wykstra that If we have realized the magnitude of the theistic proposal, cognizance of suffering should not in the least reduce our confidence that it is true. 34 This is plausibly the message of the book of Job, where the reasons for permitting Job s horrendous suffering remain beyond him, and a reference to the book of Isaiah is appropriate - For my thoughts are not your thoughts, neither are your ways my ways, declares the LORD. As the heavens are higher than the earth, so are my ways higher than your ways and my thoughts than your thoughts 35. Given the difference of such a being, it is not surprising that events would occur that are incomprehensible to limited human minds. Regardless of its explanation, it is clear that apparently gratuitous evil, given the necessary precondition of moral knowledge, does not sink the theistic hypothesis quite the way the naturalist intended it. Either one must accept the theistic hypothesis to make one s argument, or one must reject the evidence as defeating belief in God. Either way, the evidence of apparently gratuitous evil does not count at all against belief in God. Bibliography 33 Alston, W., 1991, The Inductive Argument from Evil and the Human Cognitive Condition, Philosophical Perspectives, Vol. 5, p. 29-67 34 Wykstra, S., 1984, The humean obstacle to evidential arguments from suffering: On avoiding the evils of appearance, International Journal for the Philosophy of Religion, Vol. 16, no. 2, p. 73-93 35 The Bible, Isaiah 55:8-9, New International Version, 2011, London: Hodder and Stoughton Publishing 10

Alston, W., 1991, The Inductive Argument from Evil and the Human Cognitive Condition, Philosophical Perspectives, Vol. 5, p. 29-67 Craig, W.L., The Problem of Evil, Available at: http://www.reasonablefaith.org/the- problem- of- evil [Accessed 30 July 2015] Craig, W.L., 2006, Theistic Critiques of Atheism in: M. Martin, ed. 2006, The Cambridge Companion to Atheism, Cambridge: Cambridge University Press Darwin, C., 1882, The Descent of Man and Selection in Relation to Sex, New York: D. Appleton and Co. Dawkins, R., 1996, River Out of Eden, London: Phoenix Books Draper, P., 1989, Pain and Pleasure: An Evidential Problem for Theists, Noûs, Vol. 23, No. 3, pp. 331-350 Geisler, N., 2012, From Apologist to Atheist: A Critical Review, Available at: http://normangeisler.com/from- apologist- to- atheist- a- critical- review/, [Accessed 30 July 2015] Hume, D., 2008, Dialogues Concerning Natural Religion, Oxford: Oxford University Press. Keller, T., 2009, The Reason For God, London: Hodder and Stoughton Publishing p. 26; Lewis, C.S., 2012, Mere Christianity, London: Collins Publishing Loftus, J.W., 2007, Why I Rejected Christianity: A Former Apologist Explains Victoria, British Columbia: Trafford Press Mackie, J.L., 1955, Evil and Omnipotence, Mind, Vol. 64, No. 254, p. 200-212 Mackie, J.L., 1977, Ethics: Inventing Right And Wrong, London: Pelican Books Mavrodes, G., 1986, Religion and the Queerness of Morality in: R. Audi and W. J. Wainwright, ed. 1986, Rationality, Religious Belief and Moral Commitment, Ithaca: Cornell University Press Plantinga, A., 1993, A Christian Life Partly Lived, in: K. J. Clark, ed. 1993, Philosophers Who Believe, Illinois: InterVarsity Press Plantinga, A., 2011, Where the Conflict Really Lies, New York: Oxford University Press Reppert, V., 2012, The Argument from Reason, in: Craig, WL and Moreland, JP, ed. 2012, The Blackwell Companion to Natural Theology, Oxford: Blackwell Publishing Rowe, W.L., 1979, The Problem of Evil and Some Varieties of Atheism, American Philosophical Quarterly, Vol. 16, No. 4, p. 335-341 Street, S., 2006, A Darwinian dilemma for realist theories of value, Philosophical Studies, Vol. 127, no. 1, p. 109-166 The Bible, Isaiah 55:8-9, New International Version, 2011, London: Hodder and Stoughton Publishing 11

The Veritas Forum, 2013, The Problem of Suffering and the Goodness of God - Ravi Zacharias at Johns Hopkins, [video online], Available at: < https://www.youtube.com/watch?v=t7- gp1gc8gm>, [Accessed 30 July 2015] Turek, F., 2015, Stealing from God, Colorado Springs: NavPress Weisberger, A., 2006, The Argument from Evil in: M. Martin, ed. 2006, The Cambridge Companion to Atheism, Cambridge: Cambridge University Press. Wykstra, S., 1984, The humean obstacle to evidential arguments from suffering: On avoiding the evils of appearance, International Journal for the Philosophy of Religion, Vol. 16, no. 2, p. 73-93 12