Coercion and the Hiddenness of God

Michael J. Murray


But if I go to the east, he is not there; if I go to the west I do not find him.

When he is at work in the north, I do not see him;

when he turns to the south, I catch no glimpse of him.

Job 23:8-9


    The sentiments expressed by Job in the above quotation are ones that have been expressed by the sophisticated atheist as well as the typical church-goer. Most of us, in fact, have wondered at one time or another why it is that God does not reveal Himself in some dramatic fashion if He actually exists. Yet, while this question is widely entertained, it has received surprisingly little attention in the philosophical literature. In addition to puzzling many theists, the fact of divine hiddenness has prompted some non-theists to challenge the theist to provide some explanation for God's apparent silence. The problem they have raised can be roughly stated as follows: If, as most theists claim, belief in God is essential to ultimate human fulfillment, one would expect that God would provide us with unambiguous evidence for His existence. However, such evidence is not forthcoming. Therefore, it is unlikely that the theist's God exists.

    The atheist Norwood Hanson makes this case against the theist as follows in his essay "What I Do Not Believe":

. . . 'God exists' could in principle be established for all factually-it just happens not to be, certainly not for everyone! Suppose, however, that next Tuesday morning, just after breakfast, all of us in this one world are knocked to our knees by a percussive and ear-shattering thunderclap. Snow swirls; leaves drop from the trees; the earth heaves and buckles; buildings topple and towers tumble; the sky is ablaze with an eerie, silvery light. Just then, as all the people of this world look up, the heavens open-the clouds pull apart-revealing an unbelievably immense and radiant-like Zeus figure, towering above us like a hundred Everests. He frowns darkly as lightening plays across the features of his Michelangeloid face. He then points down- at me!-and explains, for every man and child to hear: 'I have had quite enough of your too-clever logic-chopping and word-watching in matters of theology. Be assured, N.R. Hanson, that I most certainly do exist.'

. . . . Please do not dismiss this as a playful, irreverent Disney-oid contrivance. The conceptual point here is that if such a remarkable event were to occur, I for one should certainly be convinced that God does exist. That matter of fact would have been settled once and for all time. . . .That God exists would, through this encounter, have been confirmed for me and for everyone else in a manner every bit as direct as that involved in any non-controversial factual claim.1

    Hanson's point, of course, is that since God has not produced such a theophany, we not only lack good evidence that such a God exists, but that this heavenly silence actually inveighs against God's existence. The argument is made even more forcefully by Neitzsche in the following selection:

A god who is all-knowing and all powerful and who does not even make sure his creatures understand his intention-could that be a god of goodness? Who allows countless doubts and dubities to persist, for thousands of years, as though the salvation of mankind were unaffected by them, and who on the other hand holds out the prospect of frightful consequences if any mistake is made as to the nature of truth? Would he not be a cruel god if he possessed the truth and could behold mankind miserably tormenting itself over the truth? - But perhaps he is a god of goodness notwithstanding-and merely could express himself more clearly! Did he perhaps lack the intelligence to do so? Or the eloquence? So much the worse! For then he was perhaps also in error as to that which he calls his 'truth', and is himself not so very far from being the 'poor deluded devil'! Must he not then endure almost the torments of Hell to have to see his creatures suffer so, and go on suffering even more through all eternity, for the sake of knowledge of him, and not be able to help and counsel them, except in the manner of a deaf and dumb man making all kinds of ambiguous signs when the most fearful danger is about to befall on his child or his dog? - A believer who reaches this oppressive conclusion ought truly to be forgiven if he feels more pity for this suffering god than he does for his 'neighbors' - for they are no longer his neighbors if that most solitary and most primeval being is also the most suffering being of all and the most in need of comfort. - All religions exhibit traces of the fact that they owe their origin to an early, immature intellectuality in man - they all take astonishingly lightly the duty to tell the truth: they as yet know nothing of a Duty of God to be truthful towards mankind and clear in the manner of his communications. - On the 'hidden God', and on the reasons for thus keeping himself hidden and never going more than halfway into the light of speech, no one has been more eloquent than Pascal - a sign that he was never able to calm his mind on this matter: but his voice rings as confidently as if he had at one time sat behind the curtain with this hidden god. He sensed a piece of immortality in this 'deus absconditus' and was very fearful and ashamed of admitting it to himself: and thus, the one who is afraid, he talked as loudly as he could.2

    The challenge to the theist is to explain this heavenly silence.


    In order to understand the nature of the problem of divine hiddenness it is important to ask exactly what the objector to theism finds problematic here. The real problem, as I see it, is the fact that the hiddenness of God seems to be closely tied to disbelief. For most Christian theists, disbelief is a form of sin, possibly the most damaging form. As a result, the problem appears to reduce to the fact that God's self-imposed obscurity seems to be indirectly, or possibly directly, responsible for an important form of evil.3 The atheists challenge, then, amounts to this: why has God established conditions, or at least allowed conditions to prevail, which seem to lead to the occurrence of a significant amount of evil, especially evil of such a grave sort? Seen in this way, the problem is similar to a number of others which fall under the traditional problem of evil. One might thus be led to consider, first, whether or not the hiddenness of God might simply be treated as a species of the problem of evil and thus be resolved by appealing to certain traditional theodicies regarding this problem. What I intend to show here is that certain traditional theodicies do seem to provide some interesting resolutions to the problem of God's hiddenness. I will begin in this section with a discussion of the traditional free-will defense and show how it can be brought to bear on this vexing problem. I will begin by applying one theodicy in detail to the problem at hand and go on to mention another which may be equally valuable in resolving this difficulty. I do not mean to imply, however, that I believe these are the only two relevant theodicies. I choose these two, first, because I feel that they reflect accounts which are adopted (or are similar to those adopted) by certain Biblical authors who address this issue. Second, these two theodicies seem to provide two accounts which, while inadequate on their own, complement one another well. The first theodicy I will examine is a free-will theodicy. With respect to this theodicy I will try to show that God's hiddenness is a necessary condition for beings endowed with free-will to be able to make morally significant free choices. I will also argue that the existence of free-will allows beings so endowed to put themselves into epistemic circumstances where even evidence which might be readily available for God's existence becomes distorted. Second, I will look at the punishment theodicy. This theodicy also seems to play a key role in traditional religious thought, in that many Biblical texts speak of God withdrawing his presence as punishment for particular moral failings.

    Briefly, a free-will theodicy claims that the existence of free-will causes, allows, or presupposes the possibility of certain evils. However, there are two distinct species of free-will theodicies, both of which I will make use of in the course of this discussion. The first type of of free-will theodicy argues that one of the consequences of endowing creatures with free-will is that these beings have the option to choose evil over good. As a result, it is impossible that God actualize a world such that there are both free beings and also no possibility of these beings undertaking evil actions. I call theodicies of this type consequent free-will theodicies. They are "consequent" in the sense that evil is to be accounted for in terms of conditions that arise as a consequence of the existence of free-will in our world. It is this sort of theodicy that is most often invoked by theists in order to account for the existence of moral evil in the world.

    However, the type of free-will theodicy I am going to be concerned with first is somewhat different. The theodicy that is important here argues that there are certain antecedent conditions that must necessarily hold or fail to hold if beings endowed with freedom are to be able to exercise this freedom in a morally significant manner. For example, Swinburne, and others, have argued that any world which is such that free beings can exercise their freedom in a morally significant manner must also be a world in which there are stable natural regularities of some sort. If this were not the case, it is argued, free creatures could never come to understand that there are regular connections between their undertakings and the consequence of their undertakings. So, for example, if there were no stable natural regularities, firing a gun at another person's head at point-blank range may, on one occasion, give them a haircut, whereas on another occasion it may kill them. But it seems clear that one could not be said to be morally responsible for their actions if they had no way of knowing that their undertaking, in this case firing the gun, would have the undesirable consequence of taking another life. As a result, free creatures must be created in a world in which such stable connections between undertakings and the consequences of undertakings obtain. And it seems plausible to suppose that such a world requires a set of stable natural regularities to insure the stability of this very connection. It is only when we can be assured that, for example, gun- firings result in certain predictable consequences, that we can be responsible for the outcomes of such actions.

    However, the argument continues, the existence of stable natural laws may also lead to other events which result in natural evil, for example, hurricanes, earthquakes, and so on. Thus, if one can argue that there is some overriding reason why God should create a world with beings that are free and also able to exercise that freedom in a morally significant fashion, then the existence of these laws which give rise to natural evil are justified.

    This argument strategy thus contends that certain antecedent conditions must obtain if free creatures are to be able to exercise their freedom in the most robust sense. And since there is good reason for creating creatures who can exercise their freedom in this fashion, there is good reason to create the necessary antecedent conditions which would allow for such exercising of freedom. One can then argue that even though certain evil states of affairs might result from these antecedent conditions obtaining, such is necessary if God is going to be able to bring about the greater good of actualizing a world in which free creatures can exercise their freedom in a thoroughly robust manner.

    Clearly, theodicies of this sort differ from theodicies of the consequent type in that they argue that there are certain antecedent conditions which are requisite for free beings to be able to exercise their freedom and that such conditions may incidentally lead to certain other evil states of affairs. However, it is argued, the circumstances for which these antecedent conditions are necessary are sufficiently good to justify the evil which arises as a result of their obtaining. I will refer to this class of theodicies as antecedent free-will theodicies.

    In addition to arguing that certain conditions must obtain for free creatures to be able to exercise their freedom, it can also be shown that certain conditions must fail to obtain if free beings are to be able to exercise their freedom in a morally significant manner. Specifically, it appears that one cannot act freely when one is in the condition of compulsion by another in the context of a threat. Under conditions that I will specify below, it seems clear that fully robust and morally significant free-will cannot be exercised by someone who is compelled by another in the context of a threat. Further, I will argue that if God does not remain "hidden" to a certain extent, at least some of the free creatures He creates would be in the condition of being compelled in the context of a threat and, as a result, such creatures could not exercise their freedom in this robust, morally significant manner.

    It seems at least prima facie plausible to claim that morally significant freedom cannot be exercised by an individual who is being told to perform a certain action in the context of a significant threat, say, hand over his money to one holding a gun to his head and threatening to shoot. The threatened individual is compelled by another in such a way that morally significant free-will cannot be exercised. This claim, however, is certainly not uncontroversial. Adequately defending this position would require a separate treatment on the nature of coercion and its relation to freedom, a task too great given the limitations of this essay. However, a few things need to be said here about the relationship between a significant threat and freedom. First of all, I am certainly not claiming that freely willed acts are metaphysically impossible in the context of a significant threat. There is even no physical impossibility involved in the case of one refusing to comply with the demands of the robber described above. But if this is so, what is going on in cases where we are threatened?

    There are at least three alternative answers. The first possibility is that we are free in such cases but that under the circumstances no rational person would choose to act contrary to the demands of the threatener. We might say, further, that as a result of the threat, our ability to rationally deliberate about alternatives is blocked because the threatener has brought it about that there can only be one rational choice. If we look at the matter in this way what is surrendered in such cases is the deliberative or reasoned exercise of freedom. On this account the external threat of an intentional agent has limited the rational possibilities of action to just one. And in such circumstances, one may hold, we are not able to fully exercise our freedom.4

    The second thing one might say is that we are not free at all. It is a common view amongst libertarians that a free action requires that neither metaphysically necessary truths, the history of the world, nor the laws of nature prevent us from choosing between more than one option. One might hold, however, that there are cases in which certain operative psychological laws make only one alternative psychologically possible. When such laws are operative we are in a state where we cannot view physically possible alternative courses of action as legitimate possibilities for action for us at that time. In such cases, these psychological laws make it the case that one psychologically could not choose to fail to act in accordance with the threateners demands. As a result, in cases such as these the libertarian might legitimately hold that we are not free in the most fundamental sense.

    Finally, one might argue that what is not possible in these threat contexts, is free actions that are morally significant. One adopting this approach might argue that freedom has not been lost here because the threatened can consider the alternatives and choose what, in that instance, is the most rational course of action. What the threat does provide, however, are excuses for the behavior-excuses which suffice to relieve the threatened of moral responsibility for the action committed.5 Traditional Christian theists often argue that it is not only freedom, but morally significant freedom which is desirable for free creatures. Plantinga argues, for example, that the moral significance of the free actions is important because God desires to increase the diversity and amount of good in His creation. One way to accomplish this is to create free beings who can exercise this freedom in a morally significant manner, thus create a world containing moral good in addition to just, say, metaphysical goodness, the sort of goodness that attaches to somethings mere existence or being. If this is correct, one might argue that praise and blame are not justified in the context of significant threats because such threats provide adequate moral excuses for the behavior performed. Suppose that the individual being robbed in the case described above, call him Barney, had been sent by his friend, Fred, to make a deposit to Fred's savings account. On his way to drop off the deposit, Barney is stopped by our robber and promptly hands over the cash. In such a case it seems clear that we would not feel that Barney is morally culpable for giving up the money as we would if he were simply to hand it to some passerby. The fact that there was a significant threat provides an excuse which is sufficient to make Barney no longer morally culpable for an act he would have been responsible for had the significant threat not been present. Because praise or blame are not justifiably imputed in such cases of compulsion it would appear that although freedom simpliciter is not eliminated, the moral significance of the action performed is.

    Whichever interpretation one wishes to place on such cases of compulsion, the fact remains that the prospect of being in such circumstances is quite unsavory. If one thinks that such threats make the threatened unable to act freely, then threatening destroys freedom. But even if one concludes that the threatened is still free, there is still something defective about the activity of the threatened in such a case. Whatever this defect might be explains why we legislate against such coercive behavior and do not allow robbers to excuse their action by claiming that they do nothing wrong since the victim "gave his money over freely." For the purposes of this essay I will not argue for any of these three views about the relationship between threats and freedom. Instead I will simply note that such cases do put the threatened in an unsavory position, one which in some way interferes with their exercising morally-significant freedom in a fully robust manner.


    There is, however, an ambiguity regarding exactly what constitutes a "significant threat". Not just any threat counts as a compelling one since, for example, one would not feel compelled to hand their money over to a robber who simply threatened to call them a dirty word. What would be helpful is a list of necessary and sufficient conditions which would suffice to clarify exactly what constitutes a threat significant enough to eliminate the possibility of morally significant, rational freedom. Unfortunately, the subject matter here does not allow for such precision. However, there are certain factors which jointly determine "threat significance." Below I will discuss these factors in an effort to provide a clearer picture of how threats give rise to compulsion and how this this compulsion affects the exercise of morally significant free-will.

    The three factors that are important for my analysis are what I will call threat strength, threat imminence, and wantonness of the threatened. By threat strength I mean the degree to which the threatened person feels the consequences of the threat to be harmful to him. By threat imminence I mean the degree to which the threatened perceives that the threat will inevitably follow given that the conditions for the threatened consequences being enacted are met. The notion of "inevitably follows" is ambiguous here and intentionally so. Below I will explain that this notion must be carefully unpacked since the notion of threat imminence is multi-faceted. Finally, by wantonness of the threatened I have in mind a characteristic of the individual threatened to disregard personal well-being in the face of threats to his freedom. My claim is that the degree of compulsion is directly proportional to threat strength and imminence and inversely proportional to wantonness. I will now discuss these conditions in more careful detail.

    It should be obvious that the degree of compulsion is directly proportional to the degree of threat strength. The degree to which I feel compelled to do an act that I would not otherwise do (say, to give all my money to a stranger) would be much greater if the threatener held a gun to my back than if he threatened to call me a dirty word if I failed to comply with his wish.

    It is more difficult to see exactly how threat imminence relates to compulsion simply because it is less easy to characterize. By examining a few cases I think it will become clear that the notion of the consequences of a threat "inevitably following" when the threatened fails to satisfy the conditions of a threat must be cashed out in more than one way. There are, in fact, at least three distinct senses of threat imminence which must be distinguished for my purposes.

    The first type of threat imminence is what I will call probabilistic threat imminence. Consider the standard robber case above in which I am threatened with being shot if I fail to hand over my wallet to the thief. In this case I would consider it highly probable that the thief would shoot me if I failed to comply with the conditions of the threat. As a result, the probabilistic threat imminence would be high in this case. However, consider another case in which certain prisoners are allowed to spend recreation time in an enclosed prison yard. Surrounding the yard are high barbed wire fences which are periodically punctuated by guard towers. The prisoners have been told that the guards have orders to shoot if any of the prisoners attempt to escape. As a result we have a case which, in important respects, is similar to the standard robber case. Most importantly, in both cases the threatened individuals are under a threat of the same strength, namely, being shot if the conditions of the threat are not satisfied. However, in the prison-yard case, a prisoner might be more tempted to attempt to escape because he might feel that there is some significant probability that the threat would not be successfully carried out because, for example, the guards might miss him at that distance, or because they may fail to see him since they are so busy watching the other prisoners. Thus, in this case the degree of compulsion is somewhat lower than in the standard robber case because the probability that the threat will be carried out is somewhat less even though the threat strength is identical.

    The second type of threat imminence is what I call temporal threat imminence. With this type of threat imminence, compulsion is greater in those cases in which the threat will be carried out with more temporal immediacy, once the conditions of the threat have not been met. To show this consider the standard robber case once again. In such a case the temporal threat imminence is high since I know that if I fail to comply with the robbers demands I will be shot on the spot. Compare this, however, to a case in which the robber tells me that he has a blow gun with darts which he will shoot at me if I fail to hand over my money. Furthermore, the robber tells me that these darts contain a poison which has no antidote and will lead to my certain death in fifty years. In the former case, compulsion is higher because the temporal imminence of the threat is greater.6 Differing degrees of temporal threat imminence may also explain phenomena such as the fact that some individuals choose to eat high fat foods which they know, in the long run, are very likely to cause, say, fatal arteriosclerosis, while these same individuals would not ingest antifreeze, which although quite sweet tasting, is very likely to be immediately fatal. Ingesting both types of substances makes death likely; but ingesting high fat foods makes death likely sometime in the future, whereas ingesting ethylene glycol makes death immediately likely.

    Finally, there is epistemic threat imminence. This type of imminence is also quite difficult to characterize but it is one with which I believe that we are all familiar. It is this third type of imminence that explains why we believe that massive advertising campaigns are effective in reducing the incidence of smoking or drinking and driving. In both of these cases it seems that few engaging in the behaviors really believes that it is not bad for them; they are usually quite well aware that they are so. Clearly, then, the purpose of such advertising campaigns is not to inform the individuals engaging in these behaviors that they are bad for them. What then is their purpose? It can only be to make the fact that these behaviors are dangerous more epistemically forceful. Somehow, by repeating the message over and over we become more powerfully aware of just how harmful such behaviors potentially are. As a result, the more epistemically forceful the danger is, the more likely we are to not act in such a way. Likewise, when we are discussing compulsion, the more epistemically imminent a threat, the more compelled the threatened individual will feel.

    In the case above the threat imminence is high where the thief who wants my money has the gun to my back. However, one can easily imagine cases where the same threat applies but that threat imminence is much lower thus leading to a lesser degree of compulsion. For example, a prison inmate may know that there is a standing threat of being shot if he tries to escape. However, it is not the case that each inmate has a guard who is holding a gun to his head at all times awaiting his attempted escape. Instead, the threat is more "distant" (i.e. less imminent). There will be guards watching who are somewhat likely to spot the prisoner and carry out the threat. But the certainty of these guards being able to carry out the threat, viz. shooting the prisoner, on the condition that the prisoner fulfill the conditions of the threat , viz. the prisoner's trying to escape, is much less than it is in the case of the robber holding the gun to my back. In typical prison-yard cases it seems that what I call the threat imminence is less than in the standard case of my being robbed, and in such cases of decreased threat imminence one is less compelled to obey the demands of the threat (or, alternatively, will be more likely to feel free to disobey the threat). Consider, for example, the case of a would-be robber who demands my money and, rather than pulling a gun, assures me that if I do not comply he has a gun at home he will go and get.

    Threat imminence, however, is a two-sided coin. There are at least two aspects of threat imminence and both must be understood as essential elements in how threat imminence relates to compulsion. The aspect is objective. That is, threat imminence is objectively built into the case. Merely specifying the conditions of the threat makes it clear how imminent the threat is in an objective sense, i.e., regardless of what the person threatened may think about threat imminence in the case. For example, by a mere examination of the conditions in the standard robbery case and the prison-yard case one can see that threat imminence is greater in the former case than in the latter. This determination can be made without any consideration being given to the belief state of the individual who stands under the threat. However, this is not always the case which leads one to the second or subjective aspect of threat imminence, i.e., threat imminence may also be determined simply by what the threatened individual believes about the threat. For example, imagine that when I am being robbed I wrongly believe that I am wearing a bullet-proof vest. This would be incorrect but I could certainly believe it to be so. If I were to believe this, the threat imminence would be greatly decreased for me and I would be much less compelled to obey the terms of the threat as a result. Thus, threat imminence can be understood as a function of either or both of these factors of threat imminence/distance.

    However, these two factors of strength and imminence alone are not sufficient to explain compulsion by another in the context of a threat completely. This is evident when we look again at our prison-yard case. Why, one might wonder, do certain prisoners try to escape, while others in similar circumstances do not, even though threat strength and imminence are the same for all prisoners? Assuming that none of them wishes to remain in prison, why do they not all try to escape? This question points to the need for a the third factor, and this factor is the wantonness of the threatened. Again, this factor is difficult to define precisely. However, it does seem clear that different individuals under the same threat and with the same degree of threat imminence can feel compelled to different degrees depending on a certain internal character trait which can be described as incorrigibility or threat indifference. This trait can be roughly characterized as a feeling of indifference for one's well-being in cases where that well-being is threatened should there be a refusal to submit to the terms of some restriction on one's freedom.

    These, then, are the factors which must be taken into account when we consider the the degree to which a threat prevents the exercise of robust morally significant freedom. While it is surely impossible to quantify these characteristics in order to define exactly what constitutes a threat which overwhelms freedom, it can be said that the degree to which freedom is compromised is directly proportional to threat strength and imminence and inversely proportional to wantonness.


    One feature of the major Western theistic traditions is that they seem to involve the issuing of both temporal and eternal threats for disobedience to the divine will. Passages from, for example, the Hebrew and Christian scriptures, such as the following, represent both aspects of this threatened punishment:

A man who remains stiff-necked after many rebukes will suddenly be destroyed -- without remedy  (Prov.29:1)


But because of your stubbornness and your unrepentant heart you are storing up wrath against yourself for the day of God's wrath. (Rom.2:5).

    As a result, those who are aware of such threats and are convinced of their veracity are in a state where their freedom is at risk. What this creates, simply, is some degree of compulsion by another in the context of the threat. Specifically, it is compulsion by God in the context of a threat of both temporal and eternal punishment. Consequently, on the picture painted by these traditions, God has issued threats, both temporal and eternal, which will be carried out if one fails to submit to Him, in action or belief, in certain ways.7 Here I will focus particularly on the Christian tradition and the notion of a threat contained therein.

    Since these appear to be quite significant threats, the theist must provide some explanation for how this threat can be mitigated so as to prevent the compromising of human freedom. To do this, one of the three factors of compulsion must be mitigated in some way. I will now look at each one to see where the force of compulsion could be averted.

    Certainly, with regard to the factor of threat strength, the threat posed by the prospect of eternal damnation is equal to the strongest imaginable threat. One, of course, might wonder why God does not simply eliminate the threat of hell for disobedience and in doing so eliminate or severely limit the threat strength and thus the compulsion. This is an interesting question but not one I will address here. My goal here is to determine whether the traditional, orthodox Christian position can be reconciled with the fact that God does not reveal himself in the manner Hanson might wish. Since the existence of hell is, I take it, a presumption of the traditional Christian view, I will take it for granted at this point. By doing so, however, we also preclude the possibility of mitigating compulsion by attenuating threat strength.

    As a result, unless one of the other two factors can be appropriately controlled, it would seem that morally significant exercise of human free-will would be precluded. What about wantonness? It is unlikely that this factor will provide what is required to avoid the consequence of compulsion which eliminates free-will. The reason for this is that it seems likely that the development and functioning of traits such as wantonness is something which falls within the domain of the freedom of the individual. To attempt to argue for this claim in any complete way would lead into the complex psychological question of whether such personality traits in general are acquired by heredity, environment or elements of individual free choice. Another area that would need to be addressed is how we develop character traits relating to wantonness. Aristotelian views on the development of virtues by the willful cultivation of habits of right-acting, for example, would support the view I hold above in my claim that wantonness is a factor that God cannot manipulate if He desires to preserve free-will. As a result, my claim is that if God were to preserve human free-will, manipulating this element of the picture would not be an option.

    This leaves us with the possibility of controlling the degree of threat imminence. Let's begin by looking at probabilistic threat imminence. This condition seems to provide little help since, on the Christian story, it is nothing less than certain that the threat will be carried out if the conditions of the threat are not met.

    What about temporal threat imminence? Clearly this condition has some relevance to our case since carrying out the threat does not follow immediately upon failure to obey the conditions of the threat. There is no trap door to hell that opens upon one's first sin or willful failure to assent to the Christian plan of redemption. Yet merely reducing the temporal imminence of the threat does not appear to be sufficient guarantee that creature's freedom is not compromised by divine compulsion. Given the strength of the threat involved it does not seem that merely delaying the carrying out of the threat temporally is sufficient to mitigate compulsion. If it were, it appears that we should be content to say that God could appear in the sky, a la Hanson, issuing the relevant temporal and eternal threats, and yet not have the actions of free creatures be compelled by the issuing of such threats. Yet, it seems that the actions of such free creatures clearly would be compelled if they were to be confronted by such obvious threats.

    We are left then with epistemic threat imminence as the final factor which can be attenuated if God desires to preserve the exercise morally significant freedom by creatures. My claim is that the hiddenness of God is required in order for free beings to be able to exercise their freedom in a morally significant manner given the strength of the threat implied by knowledge of the threat implicit in the traditional Christian story. If God revealed his existence in a more perspicuous fashion we would be in a situation very much like the one in the standard robbery case, i.e. strong threat strength and strong threat imminence such that the level of wantonness of most, if not all, individuals would not significantly diminish their feeling compelled to act in accordance with the demand of the threatener. However, if God desires that there be individuals with free-will who can use it in morally significant ways, then He must decrease the threat imminence of eternal and temporal punishment and He, in fact, does so by making the existence of the threat epistemically ambiguous. It is this epistemic ambiguity that we call the problem of the hiddenness of God.

    This may make it clear why God does not, say, open the sky and give a world-wide, unambiguous proclamation of his existence. However, it does not seem to explain why there is the particular degree of divine hiddenness that there is. An objector may reply here that God may not be able to "open the sky" without the loss of morally significant freedom on the part of humans; yet, must that also mean that merely one more unit of divine manifestation in the world would cause the fabric of significant moral freedom to collapse? The answer is no. What this argument is intended to provide is a response to the question of why God does not provide a grand, universal display of general revelation.8 But why then does God provide the fairly low general level of revelation that he does? Since God is concerned with preserving the freedom of each individual, the level of general revelation must be such as not to preclude the possibility of anyone's exercising his or her free-will in a morally significant fashion. Since threat strength is constant, God must tailor the degree of general revelation to the individual most likely to be compelled by a threat, namely, the least wanton individual. If this is correct, the degree of threat imminence, and consequently the degree of divine manifestation in the world, must be appropriately moderated. And, the degree of moderation here is likely to be great, with the result that the amount of unambiguous general revelation that God can provide is likely to be fairly minimal.


    It will be helpful here to consolidate the ground that has been covered up to this point. What has been shown is that with respect to the general revelation that God gives for his existence, there are tight constraints on the amount which can be provided, given God's desire to preserve the morally significant exercise of human freedom. Showing this in itself would alone be of quite some interest, if indeed I have achieved it, simply because it answers a number of theistic critics who hold that God should "part the heavens and show Himself." However, the solution offered above does not answer the whole question with respect to the hiddenness of God because it does not address the possibility that God could supplement his general revelation by individual revelation that would be such that each individual is maximally aware of God's existence to the extent possible for that individual without such revelation impugning the possibility of the exercise of free-will. If, for example, as Calvin taught, there is a sense of the divine, or sensus divinitatus, in each of us, making us aware of the divine presence, why is God not able to make up the lack of general revelation at an individual level and thereby avoid the difficulty of having to cater to the least common denominator with respect to wantonness?

    It is this question which emphasizes the limitations of the antecedent free-will theodicy I have chosen and shows why it is unable to handle the entire problem of the hiddenness of God. What this theodicy lacks, however, can be supplemented by a consequence free-will theodicy combined with a second theodicy, i.e., an Augustinian-style punishment theodicy. In what remains I will explain how such theodicies would apply in this case although I will not develop them in any detail since others have reviewed such theodicies and their application in this context.9

    Let us assume that it is correct that God does supplement general revelation through a sensus divinitatus which provides each individual with knowledge of God yielding a maximal threat imminence without thereby eliminating the possibility for the morally significant exercise of that individual's free will. How could one then account for the hiddenness of God? The consequence free-will theodicy claims that part of the hiddenness of God can be accounted for as a result of some act which prevents the individual from interpreting this revelation properly or giving the properly interpreted evidence the appropriate epistemic weight.10 As a result, one might argue that one source of God's hiddenness is that free individuals can turn away from the less ambiguous internal evidence that God has provided for his existence. These theories are sometimes characterized as "human-defectiveness" approaches to the hiddenness of God.

    Not only does the corruptness of human nature contribute to the hiddenness of God, but it also seems clear that the the Jewish and Christian traditions represents God, in some cases, as veiling His existence and "hardening" the epistemic capacity that is normally used to understand the revelation He gives. This hardening is usually a punishment which results from some form of moral disobedience. As such, this explanation for the hiddenness of God can be subsumed under the punishment theodicy which claims that some evil (in this case, some divine hiddenness) is the result of justified punishment for sin.11 Passages such as the following indicate instances where such hardening occurs:

[Because the Israelites worshipped the golden calf] God turned away and gave them over to the worship of heavenly bodies. . . . (Acts 7:42)


But my people would not listen to me; Israel would not submit to me. So I gave them over to their stubborn hearts to follow their own devices.(Psalm 81:11-12)

    The view that humans can orient themselves in such a way as to make divine revelation less readily understood, either by a direct act or as a result of cultivating a sinful character, as well as the claim that God sometimes withdraws revelation or the ability to properly interpret revelation, combine to give a potential solution to the problem of "individual" divine hiddenness. These supplement the earlier argument as to why God's general revelation is as ambiguous as it is by showing that even if individual revelation were originally intended to be at its maximum, while still allowing morally significant employment of free-will, it may still be the case that divine hardening as punishment for sin and/or a self-induced human blindness with regard to divine revelation may cause one to find divine exposure non-compelling.


    In a recent essay on the subject of divine hiddenness, Robert McKim argues that even if the theist can give an account which successfully explains why a theistic God would remain hidden, the fact of divine hiddenness still yields some lessons which militate against the claims of many of these particular traditions.12 In particular, McKim argues that divine hiddenness shows us 1) that "the fact of divine hiddenness gives us reason to doubt that it is very important that we should believe."13 and 2) that whatever theistic beliefs we do hold should not be held dogmatically.14 Thus even if the the theist can explain the hiddenness of God, she cannot explain why belief, and afortiori dogmatic belief, is so highly valued by the theistic traditions.

    McKim defends the first claim by arguing that if belief were important, our circumstances would surely have been more conducive to it. However, he adds the qualification that if divine hiddenness is explained in terms of human defectiveness, then it is still consistent for the theist to hold that belief is extremely important. This is so since hiddenness, in such a case, is explained not by the fact that God fails to make Himself evident (in virtue of the fact that faith is not of a high relative importance to God), but simply that free creatures have done something to impede their ability to appropriately understand or interpret the divine manifestation that has been provided. However, these human defectiveness theories would also have to explain how these defects account for the extent of divine hiddenness. If they do not adequately do so, we can assume that God could have been more evident than He is by, say, putting forth a better effort, and in fact would do so if belief were of high relative importance. But since human defectiveness theories cannot explain the degree of hiddenness in this way, the theist ought to remain sceptical about the paramount importance of belief.

    What if the theist were to respond as follows. What God values is not only belief but a number of other things as well. Amongst these is the existence of a world which has on balance the greatest amount of moral good with respect to moral evil. In creating such a world, God must create beings who are capable of bringing about moral good, i.e., creatures that have significant moral freedom. Of course, as I have argued above, achieving this goal requires that God remain hidden in a substantial way. As a result, while theistic belief (and possibly also acting in a manner obedient to divine commands) is important, in addition to being the prime requisite for the fulfillment of a free and rational creature, there are other things God values, such as the goodness of the world he actualizes. Thus, belief in a world where God attempts to maximize this value will not be as easy (or better, belief will not be forced) as it might be if God did not hold these other values in creating.

    McKim might respond to this in a manner similar to his response above. We can imagine him saying something on the order of, "Of course one can make that claim, but in doing so you once again demonstrate that belief is not that important. What is important is freedom, both regarding action and belief, while the actual attainment of belief takes the back seat to these values." But this response fails show that belief is of secondary importance. What it shows is that belief or action in accordance with the divine will simpliciter is not what God values. The theist may hold that what God values is belief and action which contribute to the moral goodness of the world. So, what is ultimately valued is a world containing the greatest balance of moral good with respect to moral evil, and in such a world both believing and acting in a certain manner (namely, a free manner) are what suffice to yield this goal. As a result, belief and right action are not subservient to freedom, it is just that they lack value when had in the absence of freedom.

    But what about McKim's second claim that the theist should not be overly confident about specific claims concerning the nature of God? Does divine hiddenness entail such a consequence? McKim argues that divine hiddenness entails that our knowledge of the divine nature will be obscure at best. If we lack access to unambiguous evidence on the question of whether or not God exists, is it not also reasonable to assume that we would have no better access to the particular nature of that divine being? And if we answer 'yes' to this question should we not also be cautious about being dogmatic concerning the claims of particular traditions about the divine nature?

    In proposing this move it appears that McKim has simply become confused about exactly what divine hiddenness amounts to. Nothing in the account which I have elucidated above entails that there are not demonstrably certain means for acquiring knowledge about the divine nature. While many of the Western theistic traditions could endorse the account of divine hiddenness I offer above, many of these traditions also contend that the deliverances of natural theology can provide one with a substantial amount of even certain knowledge about the divine nature. What God cannot do is make this knowledge, and its supporting evidence, so epistemically forceful that they compel belief or behavior.

    What McKim seems to have errantly assumed in making his case is that divine hiddenness amounts to the claim that there is no way for one to come to discover, in any demonstrative way, answers to questions about the nature and existence of God. But God need not be hidden in such a profound manner. All that is required is that the demonstrative evidence for God's existence be sufficiently obscured so that it does not compel belief or behavior in accordance with the perceived divine standard. Consequently, it does not appear that either of the lessons McKim takes from the fact of divine hiddenness actually follow.


    What I have shown then is that it is possible to view the problem of the hiddenness of God as a species of the problem of evil and to apply relevant theodicies for evil to this problem. In particular, it seems that to preserve the exercise of robust, morally significant free-will, God cannot provide grand-scale, firework displays in an effort to make His existence known. Further, free creatures may hinder attempts by God to reveal Himself on a more individual basis by human defectiveness understood in terms of willful rejection of revelatory evidence. Finally, the fact that God sometimes utilizes justice requires punishment for sins committed results, at least in some cases, in God "hardening of hearts" as a means of punishment for sins provides another explanation for the fact that evidence for God's existence appears to be ambiguous at best. While the solution that I have sketched here is open to amendment as other theodicies may be developed or adapted to apply to this problem, I believe that invoking these three theodicies succeeds in providing a coherent response to the problem of the hiddenness of God.15


1 Norwood Russell Hanson, What I Do Not Believe And Other Essays, New York: Humanities Press, 1971, pp.313-314.

2 Freidrich Nietzsche, ,Daybreak, trans. R.J. Hollingdale, Cambridge:Cambridge University Press, 1982, pp. 89-90. I thank Leon Galis for this quote.

3 In fact, disbelief may be the worst form of evil since its presence can, according to traditional Christian theism, lead to one's being eternally damned.

4 A couple of qualifications are worthy of note here. First, I say "the external threat of an intentional agent" by which I hope to exclude a) cases where the threat is self-imposed ('If I eat one more piece of cake, I will starve myself for a week') or b) cases where I am limited in my choices merely by the physical circumstances surrounding my choice. As an example of this latter situation consider a case in which I, dressed in shorts, t-shirt, and with bare feet head to the door to go play basketball. Upon opening the door I find that there is a blizzard outside.To strengthen the case imagine that I also have a disposition towards frostbite of which I am aware and about which I am concerned. In such a case it seems that I am limited to just one rational option--turn around and head back inside. But in such cases we do not feel that our freedom has been inhibited as it is when the robber holds a gun to our back. This means one of two things: either i) that what makes the robber case a case of coercion is the fact that an intentional agent knowingly limits our rational options to just one or b) that I am in error here and that our freedom is equally restricted in both cases. Regardless of which is correct, I do not believe it detracts from the central thesis of the essay.

5 However, this point itself is not at all free from controversy. Some have surely defended this alternative way of looking at action in contexts of a threat. For example, Ian Tipton agrees that acts performed under compulsion are ones for which the agent cannot be held responsible (see The Handbook of Western Philosophy, Macmillan Publishing Company, N.Y., 1988, p.500). However, the issue is not quite as simple as I make it out here. Aristotle, in Book III Chapter 1 of the Ethics, recognizes the difficulties involved in such cases. Whether or not the action should be considered to have moral import depends on both the strength of the threat and on what the threatened is being asked to do. If the threat is sufficiently strong we do not hold the threatened morally responsible for the action, generally. However, there are some cases where one is held morally responsible no matter how strong the threat is; Aristotle gives the example in which one is threatened with death if he does not kill his mother. Under these circumstances, says Aristotle, the threatened is still morally culpable if he gives in to the threat and commits matricide. So there is some principle here that requires us to balance off the strength of the threat on the one hand with the gravity of what the threatened is being asked to do on the other hand. When the severity of the action being demanded of the one threatened reaches a certain threshold, no threat will excuse the performance of that action by the threatened. While all this seems true, however, it is not, I believe, of much import for the thesis of this essay.

6 Some have objected to this illustration of the role of temporal imminence since, they argue, the difference between the standard case and the blowgun case is really a matter of threat strength. In the robber case I stand to lose the remainder of my life if I fail to comply whereas in the blowgun case I will still be able to live most of the rest of my life, missing only little of it. Thus, what I stand to lose in the robber case is much greater and that is why I feel more compelled in that case. This may be so, but I believe that other cases can be constructed which make the temporal imminence feature much more salient. To show this consider the following two cases. In the first case a would-be robber tells me that if I fail to hand over my money he has an extremely powerful cattle-prod which will deliver a shock so severe that, while it will not kill me, will cause me extreme pain as well as a short but painful two week hospital stay. In this case I suspect that failure to comply with the threat will lead to an immediate shock-experience. Compare this to a case in which the would-be robber tells me that he has a delayed-action cattle prod which delivers the same shock but the shock that it delivers is such that I will not actually experience the shock sensation and two week disability for fifty years. It seems to me that the result is the same here, namely, that the degree to which I feel compelled is greater in the case with the higher temporal imminence.

7 I put the matter this way so as not to take a stand on the relation between performing good works, faith, and salvation.

8 By general revelation I have in mind revelation given to all, or a very large number of, individuals.

9 Robert McKim summarizes these approaches in his essay, "The Hiddenness of God", Religious Studies, vol.26, pp.141-161. There he cites the arguments of Marylin Adams, John Hick, Terence Penelhum, and William Alston.

10 This type of explanation seems to be employed by Paul and Jesus in the following passages: "The wrath of God is being revealed against all the godlessness and wickedness of men who suppress the truth in unrighteousness. . . . ." (Romans 1:18), and, "For the people's heart has become calloused; they hardly hear with their ears, and they have closed their eyes. Otherwise they might see with their eyes, hear with their ears, understand with their hearts and turn, and I would heal them."(Matthew 13:15) In such passages it seems that humans are represented as being able to dispose themselves in such a way as to depreciate the epistemic value of divine revelation whether internal or external.

11 Passages such as the following indicate instances when such hardening occurred: "[Because the Israelites worshipped the golden calf] God turned away and gave them over to the worship of heavenly bodies. . . ." (Acts 7:42), and, "But my people would not listen to me; Israel would not submit to me. So I gave them over to their stubborn hearts to follow their own devices." (Psalm 81:11-12)

12 Robert McKim, "The Hiddenness of God," Religious Studies, Vol.26, pp.141-161.

13 Ibid., p.157.

14 Ibid., pp.159-160.

15 An earlier version of this paper was read at the Society of Christian Philosophers, Midwest Division Regional Meeting held at Central College in Pella, Iowa. I am indebted to A.A. Howsepian, and Trenton Merricks for helpful and insightful comments and criticism on earlier drafts of this essay. I especially thank Thomas V. Morris for encouraging me to pursue this topic further and for voluminous comments on countless earlier drafts.

*This essay appeared in American Philosophical Quarterly, 30:1 (1993), pp. 27-35.