God’s Foreknowledge and the Problem of Evil Essay

In his essay[1] on the possibility of God’s having middle knowledge of the actions of free agents and the relationship of that knowledge, if it exists, to the problem of evil,[2] RM Adams discusses two questions: firstly, whether middle knowledge is possible, even for God, and secondly, whether God could have made free creatures who would always freely do right. These questions highlight the importance of trying to understand how much God knows about the future and the relationship of the answer to that question with the problem of evil. In the present essay I review four major possible views of God’s foreknowledge and highlight their strengths and weaknesses, paying particular attention to Adams’ arguments on Middle Knowledge which lead to his conclusion that there is reason to doubt its possibility. I then review Adams’ arguments about its impact on the problem of evil and, having concluded, as he does, that, middle knowledge being available or not, permitting some evil in order to allow creatures to have free will may contribute to a theodicy but not complete it, I consider how this situation might be improved by accepting that the future is at least partly open. The problem The problem of evil has been the subject of theological dispute for centuries. If God is, as the traditional Christian view would have it, omniscient, omnipotent and perfectly good, how come there is evil in the world? Such a God, the argument goes, would not only wish to dispel evil from the world, but, since he can do anything, he would have done so. Since he clearly has not, either he is not able to do so or he does not care, or perhaps he doesn’t exist. While arguments such as the above call into doubt the possibility of God’s being at once omnipotent and perfectly good, the problem of evil is also closely related to the issue of his omniscience, in particular to his foreknowledge. If God knows everything about the future, including what choices between good and evil I will make, am I really free to make those choices? But does God actually know everything, particularly about contingent future events? There are many views of the God’s foreknowledge; I will consider four principal ones. The Simple Foreknowledge View. This view holds that God knows all truths and believes no falsehoods, or as Hunt puts it ‘God has complete and infallible knowledge of the future’[3], a simple statement and one which is subject to some serious objections. In the context of this essay the most important objection is that it would appear to negate the possibility of human freedom. As Augustine’s interlocutor, Evodius, says, ‘since God foreknew that he [Adam] was going to sin, his sin necessarily had to happen. How then is the will free when such inescapable necessity is found in it? ’[4] Augustine then argues that ‘God’s foreknowledge does not force the future to happen†¦. God foreknows everything that he causes but does not cause everything that he foreknows†¦ sin is committed by the will not coerced by God’s foreknowledge. ’[5] If God did cause or coerce Adam to sin he would be exempt from blame but, Hunt maintains, following Augustine, the simple fact of God’s knowing in advance what Adam (and more generally, we) will do does not constitute coercion. It is true that God’s foreknowing†¦ leaves Adam with no alternatives†¦ But the mere absence of alternatives is irrelevant†¦ simply knowing what the person will do is not an interference of any sort, and its implications for free agency are benign. ’[6] Hunt’s view is that we should ‘trust our intuition’ that Adam is deprived of alternatives but not free will. For me however, this is not my intuition. This and similar arguments elsewhere appear to be doing little more than restating the problem, and do not provide a satisfactory escape route. The problem is one of logic not theology. If it is inevitable, foreknown infallibly, that I will do A then it is not in reality an option for me not to do A. I might think that I am choosing between A and not-A, but if God knows which I will choose then in reality I am deluded: there is no possibility of my choosing not-A and if I don’t have any choice this also seems to remove any possibility of blame or responsibility for my actions. How can I be held responsible for an action which I could not avoid doing? Worse, since I do things which patently are evil and could have been avoided if I really had free will, it is arguable that God himself is responsible for, or at least knows in advance and allows to happen, the evil that I do. In addition to the free-will problem, proponents of the simple foreknowledge view have to explain what we are doing when we pray. Are we asking God to change the future? And if he does graciously agree to change it, would that not mean that he was wrong when he earlier knew, supposedly infallibly, what the future was to include before he changed it? It is an important part of this view of God that he believes no falsehoods, but if our prayers have any effect, that would seem to entail the falsehood of God’s earlier beliefs about that particular aspect of the future. It should be noted at this point that the simple foreknowledge view is fully compatible with the Christian understanding of God’s being outside time. I will return to this later, but sacrificing or compromising this understanding would be a heavy price to pay for many Christian theologians. These objections taken together seem to me to make simple foreknowledge, without some considerable modification, incompatible with an understanding of humans as responsible agents. The other views I discuss below attempt in different ways to make sufficient modifications to deal with this problem while remaining true to scripture. I should of course consider the possibility that, in coming to this conclusion about the difficulties of the simple foreknowledge view, I have not understood the question. Could it be that what I mean by either ‘free ill’ or ‘knowledge’ is somehow different to what generations of theologians have meant? For myself, I maintain that my action is free if I could do otherwise than what I actually decide to do and, crucially, no-one else knows in advance what I will decide to do, not even God. And knowledge in this context can be taken as ‘justified true belief’ which is just the sort of knowledge that God is supposed to have infallibly. It seems that simple foreknowledge is not to be rescued by recourse to a dictionary. The Augustinian-Calvinist View This view, as expounded by Helm,[7] does indeed depend on a careful compatibilist definition of ‘free will’ which enables him to argue that it is not necessary to accept either a modified, reduced account of omniscience, or that human agents are not responsible for their actions. Here ‘compatibilism’ is the view that free will is compatible with causal determinism, a view that Helm maintains was explicitly held by the later Augustine (probably as a result of further thought compared with his earlier writings) and implicitly by Calvin. The latter is evidenced firstly by the distinction he drew between necessity and compulsion, and secondly by his successors’ taking a similar view of free will, calling it the liberty of rational spontaneity while denying the liberty of indifference. [8] Helm distinguishes three concepts of God’s foreknowledge. One is causal in the sense used by Aquinas: God’s knowledge is the cause of things and on this view there is no distinction between what God causes and permits since God foreknows all events and therefore must cause them all. There is an inference from this that God causes future evil but Aquinas is said to have allowed the concept of divine permission whereby God is said to know of it but not cause it. More on that later. The second sense has God’s foreknowledge logically subsequent to his decree and is simply the knowledge of that decree before it takes effect in time, and the third is the reverse of this, with the foreknowledge logically prior to his decree. His arguments entail one or other of the first two senses, but not the third. Based on these starting points Helm raises three arguments in support of the Augustinian position. First there is the role of God’s grace. The argument between those who believe and those who do not believe that God’s foreknowledge is compatible with human incompatibilism, Helm says, is not about the nature of God or of human freedom but about the relationship between God and humankind. Divine grace and free, incompatibilist choice can only be causally necessary for a person’s coming to faith, but not causally sufficient since, given our libertarian will, we could resist such grace and it would not therefore ensure its intended effect. However, scripture tells us that saving grace is irresistible and, when received, liberating: it alone, according to Augustine, ensures true human freedom,[9] and the inference is that such grace is therefore sufficient. The obvious objection here is that some people clearly do resist God’s saving grace, an objection that Helm does not deal with effectively. Secondly there is an argument based on divine perfection as reflected in his omnipotence and omniscience. Helm asks rhetorically how God knows of the causes of evil actions if he is not the cause of them, and quotes Augustine’s answer that God, for the highest reasons (which are at present unknown to us) knowingly permits particular evil actions. 10] In a rather obscure passage, Helm appears to argue as follows: (1) it is theologically desirable that God’s foreknowledge should be as complete as may reasonably be assumed and we should therefore assume that he does foreknow his free creatures freely willed actions; (2) If compatibilism is true then God can foreknow these actions and therefore (3) compatibilism is true. [11] However, as Hunt points out, this is fallacious and Helm should have argued for (2’) If compatibil ism is not true then God cannot foreknow†¦ but he has not done so. Finally Helm argues that God’s omniscience is logically inconsistent with human incompatibilist freedom. He supposes as an example that God foreknew yesterday the truth of the proposition ‘Jones will freely eat a tuna sandwich tomorrow. ’ That foreknowledge is now in the past and is therefore necessary, not logically but accidentally or historically, and therefore it entails the necessity that Jones will eat the tuna sandwich; that putatively free act cannot therefore be free. In that case divine omniscience is inconsistent with incompatibilist freedom. 12] Helm admits that this argument really only works with the assumption that God is in some fashion inside time for ‘yesterday’ and ‘tomorrow’ to have any force. [13] In summary, Helm believes his arguments have made the broadly Augustinian case that divine foreknowledge and human freedom are consistent, but I am hard-pressed to see that any of my objections to the simple foreknowledge argument are any less forceful in response to Helm. My logical worry and the problem of prayer remain, but these are supplemented by the acknowledged need for God to be temporal, at least for part of the argument to be successful. The Middle-Knowledge View This view is that espoused by Luis de Molina, a 16th century Spanish Jesuit theologian, who drew a distinction between three kinds of knowledge that, in his view, God possesses[14]. Firstly, Molina said, God possesses ‘natural knowledge’, that is a knowledge of all necessarily true propositions, such as ‘two plus two equals four’. Since such truths are necessary, nobody, not even God, can make them false. Secondly, God possesses ‘free knowledge’, that is knowledge of all contingent truths that are within his control, but which could have been false under different conditions,. For example ‘I am interested in philosophy’ is a contingently true proposition but God could have brought it about that it was false. Finally, Molina proposes that God possesses ‘middle knowledge’ (so called because it is in-between God’s natural and free knowledge), that is, knowledge of contingent propositions which are true but beyond his control. The most important items of middle knowledge for the purpose of this discussion are the ‘counterfactuals of freedom’ which describe what people would freely do if placed in various possible situations. This is relevant to the problem of evil because ‘it might seem that if God has middle knowledge, He could have secured creatures sinless but free by just creating those that he knew would not sin if allowed to act freely. ’[15] In his discussion of middle knowledge[16] Craig indicates its power and why it is so attractive in the discussion of free will and the problem of evil. If it is true that God has middle knowledge as described above, this not only makes room for human freedom but it gives God scope to choose which free creatures to create and bring about his ultimate purposes through free creaturely decisions. He adduces three lines of argument in support of it – biblical, theological and philosophical. [17] Biblical arguments: Craig uses the example of David and Saul: [18] David is in the Jewish city of Keilah and asks God through an ephod[19] if Saul will attack him there and whether the men of Keilah would give him up to Saul to save their lives. God answers affirmatively to both questions, whereupon Saul heads for the hills, with the result that Saul does not need to besiege the city and the men of Keilah do not need to betray him to Saul. It is clear, says Craig, that the bible passage shows that God has counterfactual knowledge, although he admits that this does not show conclusively that he has middle knowledge. He goes on to accept that biblical exegesis is not enough to settle the matter. [20] Theological arguments: Craig says that ‘the strongest arguments in support of the Molinist perspective are theological’[21] but gives no direct support for this other than to wax lyrical on the power of middle knowledge in theological argument on a range of issues. This may be correct, given the existence of middle knowledge, but that is what we wish to test. Philosophical arguments: Craig asserts that divine foreknowledge and future contingents are compatible ‘for the simple reason that Scripture teaches both’[23] (a theological rather than philosophical statement of course) and goes on to discuss the basis of such foreknowledge. He builds an argument about freedom of action, concluding that ‘from God’s knowledge that I shall do x, it does not follow that I must do x, only that I shall do x. That is in no way incompatible with my doing x freely. ’[24] This is really just a restatement of the problem of free will and Craig does little more here than reassert its truth. Craig’s final conclusion is that ‘philosophically, omniscience†¦ entails knowledge of all truth and, since counterfactuals of creaturely freedom are true logically prior to god’s creative decree, they must therefore be known by God at that logical moment. Therefore we should affirm that God has middle knowledge. ’