A Discourse on Unmet Expectations as an Underminer for Belief in God




"In all sad sincerity I think we must conclude that the attempt to demonstrate by purely intellectual processes the truth of the deliverances of direct religious experience is absolutely hopeless." (William James)


In this paper I attempt to show that there may be an underminer for belief in the existence of God based on a failure of experience to ground conclusions drawn from philosophical and evidential arguments for God’s existence. If one comes to believe that arguments that attempt to ground belief in God’s existence are cogent, that belief should generate certain expectations about how God would act in the world. If those expectations are unmet, the truth value of the beliefs that generate those expectations should be taken to be false, probably false, or irrelevant which seems to entail that the belief in the cogency of the arguments that ground belief in his existence either are (1) false, (2) undermined or (3) irrelevant as a ground for belief in God’s existence.


I offer the following as a kind of discourse that seems to me to establish a strong ground for rejecting belief in God’s existence or at least providing a strong underminer for it. This is not an argument in the strict sense but a narrative by which I intend to create a foundation for an argument. I do not intend to establish a basis for God’s non-existence but to outline a set of scenarios that, if accurate, should undermine belief in God’s existence for someone that holds it. If I were to put the main thesis of this discourse in the form of an argument, it would go like this: if a person comes to believe that God exists based on a set of arguments and evidence (hereafter just arguments), that belief should generate certain expectations about how God will act in the world. If those expectations consistently are not met, this should undermine belief in the cogency of those arguments. Those expectations are not met and this undermines the cogency of the arguments that formed the basis for belief in God.

This argument has a pedigree. It’s nearest relatives can broadly be categorized under the rubric of arguments from evil and the problem of hiddenness. This argument extends arguments in these categories by focusing on the idea of expectation as inference from beliefs someone has to other beliefs that are or should be generated by them. By doing so, I think it bypasses some of the easy avoidance maneuvers that plague arguments from evil (such as the claim that the argument is non-starter unless one assumes a clear distinction between good and evil which seem to rely on a largely religious interpretation of the world) and some of the metaphysical assumptions of hiddenness such as the idea that God can only be hidden if he first exists. This argument also has a certain relation to falsification arguments such as the one proffered by Antony Flew in "Theology and Falsification". Falsification arguments have been roundly criticized mainly on burden of proof issues1. The argument I offer avoids, I hope, some of the pitfalls of that class of argumentation.

An Initial Consideration


There is a fundamental challenge with the notion of expectation as a verification principle as I’ve articulated it here. For any expectation I say would be generated by true belief in God’s existence (and belief that God has all the attendant properties typically associated with a maximally powerful and beneficent being), even if that expectation is met, it’s always possible to attribute the event or events that meet the expectation to something other than God. If this is true, then met expectations provide no real epistemic value and thus can’t function in any scenario that would constitute an underminer. Let’s suppose I expect that if God exists, small children with Leukemia would be healed as a result of prayer. It’s an expectation because, given the properties of an all-loving, all-powerful God, it seems to follow by way of strong inference that such a being would heal suffering innocents particularly if sincere people asked him to. Now let’s suppose that at some point in time something changes. From that point forward, each time a person prays for healing, the prayed-for child’s Leukemia is eliminated immediately and conclusively as a direct result of the prayer. The expectation that God would heal as a result of prayer has been met and this would seem to support belief in God’s existence.

But a skeptic can ask, “Why attribute such a situation to the act of God?” For the skeptic, wouldn’t it be just as explanatorily satisfactory to attribute the relation of the speech act, the prayer, to some yet-to-be-discovered physical process? Isn’t it possible to say in such circumstances that something has changed in the world such that prayers spoken by certain individuals emit some new, undiscovered particles that have a direct effect on cancers in small children? While this would seem very strained to the believer, to the skeptic who is loath to introduce a supernatural entity into an explanatory calculus, attributing the "healing" to such an entity would seem equally as strained and he may be more comfortable categorizing the event as a new scientific research project.

Notice too that the opposite scenario is possible. Any scenario within the bounds of scientific explanation could easily be characterized as products of the hand of God. Suppose you witness an accident and see a person dragged out of a vehicle presumed dead. She’s dead for a period long past that which normal means of resuscitation are expected to have any impact. The doctors perform CPR and other resuscitative activities on her to no avail while onlookers continue to pray. On the ambulance ride to the hospital, suddenly the woman awakens with no sign of brain damage and goes on to live a normal, healthy life. No doubt scientists would have numerous naturalistic explanations for such an event even if those explanations were outliers in terms of what is normal for conditions involving human trauma. But why not attribute the "healing" to God? Even though the scientific explanations might be entirely viable, they are not "normal" and so why not say God, using natural processes, intervened and revived the woman? Her "coming back to life" fits in the realm of the physically possible but the natural occurrence is a result of the direct act of God.4

So the problem is this: the outcome of a scenario sufficiently buffeted by religious acts could be interpreted religiously or non-religiously given certain presuppositions and there doesn’t seem to be a way to adjudicate these interpretation to favor one way or the other. A skeptic will tend to interpret every event as physically explainable or possibly physically explainable such that no event is an act of God. The believer will tend to interpret events as a direct act of God even though there may be a viable naturalistic explanation—miracles are not an either/or situation. If this is true, then expectation may have no epistemic value. That is, if a person expects God to do something given prior beliefs about him, the expectation being met or not met has no implications for truth value regarding prior beliefs about God’s existence. Rather, prior beliefs about God’s existence will inform how one interprets events given one’s expectations.

The situation is, then, that evidence and argument would be irrelevant when it comes to forming  or even sustaining belief in God’s existence since any evidence would be a product of already-formed theistic or non-theistic beliefs. This is exactly the problem existentialists have noticed for centuries and its why they advocate bypassing evidence as an epistemic ground for decision-making about the truth of God’s existence and promote the idea that one should take a leap of faith. Belief in God’s existence, they say, is not a product of having enough evidence but of making a decision to believe.

If this is indeed the situation in which we find ourselves epistemically, notice that the entire notion of expectation is undermined. That is, the believer and non-believer alike would be in the situation where one’s beliefs about God have no implications for how God might interact with human life. But this also would mean that any narrative about God’s supposed activity in the world should have no epistemic implications. If one reads in the Bible that God through Jesus turned water into wine, this narrative cannot be presented as evidence for God’s existence since how one interprets that event solely is a product of what one already believes about God’s existence. Assuming we set aside the debate about the truth of the story and assume it relates a veridical historical event, the skeptic will say that the story relays a very peculiar natural occurrence attributed to a non-existence deity and the believer will say that the story documents God’s activity in the world. But the story itself is not evidence (in the apologetic sense) of that activity that carries any epistemic value.

Because of this, I’m positioning this discourse very specifically to focus on the situation the believer finds himself or herself in with regards to the expectations they have about God’s activity in the world. I’m not suggesting that having one’s expectations met or not has apologetic value. I’m making what I think is a reasonable assumption that believers hold that situations can’t or shouldn’t be epistemically neutral when he or she believes God acts in the world. I’m assuming that believers expect to see God’s hand in the world and that his activity can be distinguished from events in which God does not directly act. That is, one can determine whether or not God acting has a truth value: either he does or he doesn’t and the truth of the matter is discernible by the person who has the expectation. I’m also assuming that God’s activity would be distinguished from natural events. If everything is labeled as a act of God, then expectations become moot and actions like prayer would seem to become irrelevant. If the truth of the matter is not discernible then it seems it wouldn’t much matter whether we attribute a given activity to God or not and the notion that we should expect anything about God acting in the world becomes a moot point. But this also seems to have serious implications not only for the believer but for how we think about the role religion plays in the public square vis-à-vis the individuals who lack those expectations. It seems that under such conditions, religion should play no role.

So this discourse, then, starts with the assumption that believers have certain expectations about how God would act in the world given his existence, that God’s actions are discernable, and that his actions either meet or do not meet the expectations a believer has. 

An Existentialist Approach to Belief Formation

I’ll roughly define an existentialist approach to belief formation as follows: one forms beliefs about what is true based on the deliverances of ones experiences. To stay properly oriented to the spirit of existentialism as I understand it, I’m not claiming that this is how one ought to form belief only that this is, in general, how one does form belief. This also does not rule out beliefs about theoretical truth claims. For example, one could form beliefs about the relations between numbers by studying mathematics even if one never "experiences" a number or forms no further beliefs about the nature of numbers or whether they exist with the properties they are supposed to have inside of the theory. The class of beliefs I’m concerned about are those that directly affect the lived experience of the subject.

Philosophy and Belief Formation

The approach I describe above would seem to undermine the value of philosophy as a discipline that provides a basis for true belief about things in the world (and might even undermine the argument I’m making in this paper). In fact, many existentialists are skeptical to varying degrees of the value philosophy plays here and so this would at least be consistent with existentialism as it’s generally practiced. But I don’t think we need to adopt an either/or position. I think one can acknowledge that some beliefs are in fact formed in just the "existentialist way" I describe above and that a great many others aren’t. Whether theoretical arguments in philosophy or even the hard sciences can establish the truth of the existence of one thing or another can be taken on a case by case basis and I’m going to limit my case to the existence of God in this paper. (The success or failure of the argument will partly depend on the degree to which the reader agrees with me on this.)

What may be more generally true though is that when philosophy attempts to ground the existence (or non-existence) of things that are supposed to exist in the world, it may be inadequate though still relevant. In short, philosophy can’t provide proof here. Rather, it may provide us with a basis for believing that the thing in question is possible but only experience can provide the ground for believing that the thing is actual. Things seem to work this way in the hard sciences as well. Einstein’s theory of gravity proposed what possibly was true when Einstein first developed it but it took experiments to validate the world actually works in the way the theory proposed. If the practical experiments weren’t consistent with the theory, the theory would have to be modified to fit what one encountered in the world (or discarded entirely if the theory could not be made to "fit" the way the world appears to work).

Belief and Truth

I want to call out one other point by way of prolegomena. Holding a belief that some proposition is true is distinct from proposition actually being true. While this should be a fairly obvious distinction, it is one I think is not all that clearly understood outside of philosophy. If one accrues enough evidence and arguments for the truth of a given proposition (say that a human zygote is fully a person in any relevant moral sense and should be afforded all rights and protections ascribed to any other human), one may come to believe the proposition to be true and perhaps be fully justified in that belief. Whether a zygote is in fact fully a person in the sense described above may be an entirely different matter. The "objective" truth value of such things may be outside of our ability to determine so at best we can describe what we believe about such matters but never make claims about what actually is the case. This aligns to the common distinction between rationality and truth—one may be rational in believing p is true while not being in an epistemic position to claim that p is actually true.

In fact Existentialists (and Postmodernists more generally) are inclined to say most, if not all, truth claims are not truth claims at all but belief claims. I think there is something very important in this idea but we don’t need to be committed to this more exhaustive epistemology for my argument to work. All we need to agree on is that the distinction exists and that we need to be aware of it in order to make sense of any epistemic system.

Philosophical Arguments For and Against God’s Existence

I’m ask my reader to grant that arguments for and against God’s existence stand in rough epistemic parity. By this I mean that based purely on philosophical arguments for God’s existence, one could be entirely justified, and therefore rational, in believing that God exists and, based purely on philosophical arguments for atheism, one could be fully justified and rational in believing that there is no God. I do not mean by this that for any given person such a situation is true—if you believe arguments for atheism are conclusive (or more probable than not), then you most likely believe that arguments for God’s existence are faulty and don’t justify that belief for the person that holds it. But I’m taking the approach of Robert McKim here and claiming that, taken across the whole range of believers and non-believers, arguments for and against God’s existence are in rough epistemic parity. This means that on both sides of the question, there are "wise people who think carefully and judiciously, who are intelligent, clever, honest, reflective, and serious, who avoid distortion, exaggeration, and confabulation, who admit ignorance when appropriate, and who have relied on what have seemed to them to be the relevant considerations in the course of acquiring their beliefs." Religious Ambiguity and Religious Diversity (OUP, 2001, p. 129). This seems obviously true but if you find the claim suspect, I’m asking that you at least grant it for the sake of this narrative.

If this is true, then philosophical arguments for or against God’s existence cannot function as proofs. They can only provide a basis for believing either position is more or less probable for a given individual. It also means that one must consider the arguments for the position other than the one they hold to have some merit and this would leave one in an epistemic position of something far less than certainty. This is in fact the situation I think we are in if we stop at what philosophical arguments (and even scientific evidence) can give us with regards to the question of God’s existence.2 

Argument and Expectations

But there may be a way to tip the epistemic balance scales. If we take the philosophical arguments for God’s existence as concluding that its true, or more probable than not, that there actually is a being who exists that is a person who knows all there is to know, has enough power to do anything able to be done, loves human beings "maximally" (to the greatest degree possible without losing some greater good), and who has the capacity to be intimately involved in their lives, then we can use these claims as a basis for experimentation based on experience. In other words, true conclusions to these arguments seem to set up a certain practical expectations about what a person would experience vis-à-vis God existence. If those expectations are not met, it may provide a strong underminer or even a defeater for those arguments.

Of course not having ones expectations met that have been set by theoretical arguments does not necessarily mean the arguments are not sound. It could just be that the expectations the arguments generate are unreasonable or that there are good ways to understand how the arguments can be true and the expectations generated by them can at the same time be unmet. For example, many people who deny Darwin’s theory of common descent claim that the theory generates the expectation that there should be a "universe of transitional intermediates" in the fossil record. Because this expectation is not met, the claim goes, it counts against the theory being true. In response, biologist generally don’t argue for the claim that the fossil record actually is complete but focus on the misaligned expectations on the part of the skeptic.

So for the argument to work, the strength of expectations generated by the arguments would have to follow by way of fairly strong inferences and the most reasonable rejoinders to these expectations and the notion that they should be met fail. In the case of God’s existence, I think we can state fairly obvious expectations that this belief generates and, if we limit ourselves to the Gods described by Christianity, Judaism, and Islam, I think the expectations I note are strongly supported by the key religious texts of those religious respectively. I will not argue for the strength of the expectations here as there are many good sources that can establish the link between the existence of God and the expectations I think would be generated by the belief that such a being exists. But I will note those expectations that I think should obviously follow from the belief that God exists and will leave it to the reader to judge their strength.

Expectation in the way I’m using the term is a type of inference but the basis for the truth of the inference isn’t logical but existential. If we belief proposition x to be true and we can infer y from x then we might also belief proposition y to be true based on the inference from x (where the strength of the belief is directly tied to the strength of the inference). In inference, the truth of inferred proposition is conferred by way of the truth of the argument that generated the proposition. Expectation is a belief or set of beliefs generated by other beliefs a person has. But in expectation, the truth value (or probabilistic truth value) of that belief is not conferred by the truth value of the belief(s) that generated it. Rather, the belief’s truth value is entirely a product of the expectation actually being met or not met in experience. Thus expectation is a certain type of inference such that the belief being inferred is the type of belief whose truth value can only be satisfied by experience.

If I belief that God exists as an actual being in the universe with certain personal qualities, my life should, in some measurable way, be impacted by the existence of that being in ways possible only if that being exists. I come to anticipate certain experiences given my belief in the existence of that being. While other beliefs may be (and hopefully are) generated by my belief in God, expectation involves what I anticipate my lived experience would be like given God’s existence. If those expectations are not satisfied, the beliefs about what should be true turn out to be false (or probably false depending on the type of beliefs we’re considering).

Expectations Generated by True Belief That God Exists

Assuming that philosophical arguments for God’s existence generate belief that God exists or probably exists, what expectations should be generated by such a belief? Admittedly, this is tricky since expectations are inherently personal. Further, it would seem to violate the very spirit of existentialism to proscribe the types of expectations one should have given positive belief in God. Yet, I don’t think it’s beyond the pale to suggest that we should be able to articulate some general expectations beliefs in the class of theistic belief should generate. This seems particularly true for believers of the monotheistic religious traditions. These are not entirely ideological systems and most claim that God is a real being that acts in the world. They also have sacred texts that describe God in ways accessible to anyone who cares to take the time to understand them. The idea here is that the description of who God is and what he is like in the major religions is bounded by a set of properties—God isn’t whoever one wants him to be and doesn’t do whatever one wants him to do or not do. Based on these descriptive boundaries and using them to inform our thinking about God, I think articulating certain expectations generated by positive belief in that God is entirely appropriate.

In order to stick to as clear cases as possible, I will adhere to the following guidelines.

  • I will attempt to keep my list of expectations as general as possible in the hopes that the expectations will apply to the majority of believers instead of a select few at the margins.
  • I will attempt to articulate expectations that I think involve God as an inference to the best explanation for the expectation to be met. This is important in order to rule out other sources for meeting the expectation. For example, let’s suppose I have formed the expectation that Martians will deposit funds into my bank account twice a month. When I look at my bank account on the days I expect to see the money, there is a deposit for exactly the amount I expected. However, in order for the expectation to be met, the deposit could only have been made by Martians and not, say, by my employer or some other source. If another source for the deposit were possible, we should first have to rule out those other sources as explanations before my expectation could be said to be met particularly if those other sources better explain the phenomenon in question. As I stated above, it seems it would always be possible to attribute some outcome to some source other than God (and God is not the only one in this situation—other explanations are always possible for most situations). But we want to look at examples that make it more reasonable than not that God is the source and that other possibilities are far less likely. This will rule out attributing to God things like making headaches going away after taking Advil or finding a parking spot after saying a quick prayer at the mall. There are too many other explanations competing here that even if the believer were to attribute them to God, that explanation would be at the most charitable in rough explanatory parity with other explanations and, I suspect, would fare far worse than other explanations under close examination.
  • The expectations I list would become normalized over time. That is, I will articulate expectations for how God would regularly act in the world, not ways he would act in particular time periods or once every 100 years. The reasons are the same as the previous point. The more infrequently God is said to act, the easier it becomes to attribute the event to a non-theistic explanation.
  • I will confine my list to expectations generated by my understanding of the God of Christianity mainly because this is the religious tradition with which I am most familiar (I hope, however, that my list will be general enough to apply to other monotheistic religious traditions as well). The expectations I list would be generated by the God who has the general properties that Christianity has and does claim He has.

So here is a simple list of expectations I think true belief in God’s existence should generate3. 

  • There would regular events such that the most reasonable explanation would be that God is the source of the event or occurrence. For example:
    • Medical "miracles" such as cancers in small children disappearing before direct observation, severed limbs of innocents being immediately regrown under direct observation, the long-dead coming back to life under direct observation and the like and that are the direct and immediate response to prayer and that involve no medical intervention whatever.
    • Regular, accurate, specific (that involve dates, names, times in extremely complex scenarios) prediction of future events.
    • Individuals that have regular, accurate, and specific knowledge of facts about the world that are far beyond the ken of the science of the day (though that could be confirmed by a future science). The specificity would include exact measurements, precise relationships between entities involved, and clear, accurate descriptions of those entities and their properties. This knowledge should come as the direct response to prayer, sacrifice, or some other religious exercise in which the exercise is designed to prompt a response from God.
    • Individuals that can regularly violate known laws of physics with the ability to move through solid objects, move across great distances instantly, turn water into wine, appear in two places at once, change bodies etc. These activities should be the direct result of prayer, would be entirely beneficent in nature, and should not be explainable by any science of the day.
  • The experience of God and God’s will would be clear and personal and align with the general direction of all other believers who are working towards a common goal. If there is one God, I think it’s epistemically appropriate to expect there to be alignment in what God wants for his people, that his will would unify all believers and that believers would know what this is for their individual lives. This will would go beyond a more generalized set of doctrines in a religious text which tend to be ambiguous, general, and open to vast, often contradictory interpretations. It would be clear, personalized, and create unity among all believers.
  • Each believer should expect to have resources for dealing with life’s challenges that exceed that of non-believers who don’t have access to (or choose not to have access to) an all-knowing, all-loving God.  I think we should expect that believers would have access to resources that would help them with addiction, psychological trauma, interpersonal problems, loneliness and isolation, and social problems like poverty, hatred, inequality, violence, health issues, and the like.  In direct answer to prayer or some other religious act, believers should expect that God would provide them with supernatural or some other non-natural means for dealing with the challenges they face and the resources to help those they wish to help.
  • Believers should expect God to reveal to them a clear purpose for their lives and provide them with the opportunities and means to work towards fulfilling that purpose. While it could be argued that it must be up to the believer to act on what God provides, I think it’s a reasonable expectation that the believer would, at the very least, know what purpose they have, have a clear sense of their gifts and abilities, and the be aware of the opportunities to express those gifts and abilities (even if the individual believer chooses not to so act).

With these expectations on the table, we have to judge whether the expectations do in fact reasonably follow from true belief that the God of Christianity exists. The items I list above certainly seem within the scope of the types of activities God would and could do when interacting with humans since the Bible is full of stories about just such activities and these activities are directly attributed to the actions of God. So the expectations can’t be dismissed outright as somehow outrageous or inappropriate. It seems also trivially true that many humans do in fact have these expectations. The Psalms, Lamentations, and Song of Solomon largely are about the grief David felt at not having expectations like these met. And if one attends a  church service on any given Sunday in just about any Christian church that still believes in prayer, many public, and I expect, private prayers are offered for healing—mental and physical, doctrinal clarity and unity, miracles of all kinds ("please help me pick the right lottery numbers" or "please make it rain so the crops will grow"), and more personal and direct feeling of God’s presence. Many believers pray regularly for direction and read books to help them figure out what it is God wants them to do. They ask God to open doors, for the “right” job, whether they should have children or help others in a third-world country. They expect God to have a will and they actively seek to find it.

If believers do not have any expectations about what God could and would do, it seems prayer and supplication are a neutered exercise. There might be psychologically comforting reasons for prayer but believers would not pray because they expect their prayers to be answered. I suspect that this isn’t the case and believers pray because they do expect God to act. Even hope is based on expectation. I’m reminded here of a (probably apocryphal) anecdote I heard while attending Bible school as a young man. D.L. Moody, the famous Chicago revivalist preacher called a prayer meeting during the 1871 drought (which eventually contributed to the great Chicago fire of 1871) to ask God to bring rain. As he began the meeting, Moody sought to enliven those in attendance by asking repeatedly if they expected God to bring the rain as a result of their supplication. With each escalating repetition of the question, the congregants would heartily agree with an amen. With the congregants loudly proclaiming their affirmations to the final iteration of the inquiry, Moody finally asks dryly, “Then why is it that none of you has his umbrella?”

To reiterate, this is meant to support the claim that the expectations are not misaligned with what people in fact do expect, not that these expectations are appropriate theologically though my argument is that one implies the other.

Are These Expectations Met (in General)?

“If it turns out that there is a God, I don’t think that he’s evil. The worst you can say about him is that basically he’s an underachiever.” – Woody Allen

If these expectations are, then, appropriate aligned with what should be generated by positive belief in God’s existence, are they in general met? Do humans, on the whole, experience God and his activity in the world to satisfy the expectations? I think it’s fairly safe to say that, at least in our time and most times in the past, these expectations are not met for the majority of the people that have them. The literature is over-burdened with accounts, arguments, and testimony of this fact and so I will not further weight down this already too-long article repeating those accounts here. Given that these expectations are not met, the "lived experience" of being a believer turns out to be misaligned with the way a believer should expect to live given God’s existence. My claim, and this is the essence of it, is that this misalignment should serve as a rejoinder for the arguments that ground belief in God in the first place. If a given belief generates an expectation and the expectation repeatedly is not met, the veracity of the belief that generates the expectation should, at the very least, be called into question.

An analogy may help. Let’s suppose that Bob, an orphan, comes to believe his father who he has never met is alive and well, lives in the town that Bob lives in, is intimately interested in Bob’s wellbeing, has tremendous resources to care for Bob’s every need and in fact has promised that he would be there for Bob when he needed it, and is available to Bob if Bob would only seek him out. Bob comes to believe this through testimony of his foster parents, through participating in a community of people who believe the same thing Bob believes about his father, and through letters and written documents that claim that Bob’s father desires to do everything Bob has come to believe.  As Bob grows, his lived experience is such that it appears that he is on his own: he has to work hard, suffers mental pain and anguish from being disconnected with his birth parents, has to seek out medical care on his own and has to pay for all his medical expenses when he is ill, had to work his way through college, and the like but for most of his life, he has been comfortable with his circumstances and, while he has the expectation that his birth father would be there for him, he’s never really made an effort to reach out. The expectations are very much a part of his belief system however.

At one point in his life, Bob finds himself facing an unusual set of challenges and is as low as he’s ever been. He finds himself at the end of his rope without resources to dig himself out and he’s without hope. He decides to actively seek out his father at least to find comfort that someone is there on whom he can depend. He is told by friends and family that he just needs to send email to a particular address and his father would listen and respond. Then he’s told that he needs to have faith that his father is out there intensely interested in his well-being and that if he just believes, help will come. Some people give him a phone number to call, others tell him that he needs to step out in faith and his father will reveal himself. Bob does these things and the very acts brings him comfort and gives him hope. His hope is grounded on the expectation that his father will respond. But, as time passes, his life continues as it always has; his lived experience, the way he has to conduct his life, remains unchanged.

Bob returns to the letters from his father in which he reads of his father’s intense love for Bob, of the fact that he has limitless resources to help Bob when he needs it, that he is there to comfort him when he’s down and support him in his victories. He reads the stories others have written about his father, about how brave he was in the war, how he’s helped countless others who called on him, how he started a foundation to help the poor, and of those who celebrated his kindness and love. These continue to support the expectations Bob has about his father’s involvement in his life but he never experiences the interactions from his father that would satisfy those expectations. Over many months, Bob continues to email, phone, ask friends and do all the things his father said he should do to connect with him. Each time he’s comforted by the hope of hearing from his father but his life and his life circumstances remain unchanged. The only activities that make any difference are the activities that Bob himself engages in to change the course of his life.

Given these circumstances, Bob could seek out explanations for why his expectations are not being met. Perhaps his expectations are misaligned. Perhaps he hasn’t waited long enough. Perhaps there are mitigating circumstances that explain why his father has not reached out. But notice that even if the explanations are emotionally and psychologically satisfying, the beliefs that generated the expectations are in trouble. Let’s suppose that Bob came to expect that his father would support him financially when he needed it because Bob believed his father to be (1) alive, (2) deeply interested in Bob’s physical well-being (he does not want Bob to be in physical distress), (3) wealthy enough to care for Bob’s needs and (4) has the ability to leverage his resources to help Bob financially whenever he wants to. Not having that expectation met itself undermines one or more of those grounding beliefs. This is because the inference of propositions 1-4 strongly implies (5) Bob’s father will assist Bob financially when he needs it and thus generates that expectation.

Now lets suppose Bob seeks out an explanation for why no help has come. A friend tells him that Bob’s father is more interested in Bob’s maturation and growth as a person and that helping Bob out of his crisis would rob Bob of his ability to grow. Bob is unclear how all this works but accepts his friend’s explanation. Bob now must come to believe something different about his father. Either he comes to believe a new proposition: (6) my father believes my emotional and psychological growth are more important than addressing my physical distress, or he has to modify one of his other beliefs, say his belief in proposition (2). He may now come to believe proposition (2′) my father wants me to go through financial difficulties so I will grow emotionally and psychologically and this is the aspect of my well-being that my father is most interested in.

With this modification, all of Bob’s expectations have to be altered and his new expectations have epistemic implications for his grounding beliefs. Let’s suppose Bob no longer expects his father will support him financially when he’s in trouble. Rather, he now expects that his father will not support him financially when he’s in trouble and that that he has to dig his way out on his own based on his belief in 2′. This adjustment may seem like an epistemic tweak and nothing more. But I think something significant has happened. With this new expectation, it no longer seems to matter whether Bob believes propositions 1-4. They no longer need, though they may, serve as a grounding for his updated expectation. In fact, Bob can have this new expectation and believe the negation of each proposition (1′) my father is not alive, (2”) my father is not interested in my physical well-being, (3′) my father is not immensely wealthy and (4′) my father does not have the ability to help me financially whenever he wants to. Bob can believe any one or more of these new propositions and these are entirely consistent with Bob’s new expectation.

But lets suppose Bob continues to believe propositions 1-4 while abandoning the original expectation and adopting the new one. Are these existentially compatible? I don’t see how they can be. Given that Bob believes that his father desires that Bob is not in distress physically, that he has the means and the ability to help Bob, his new expectations seems existentially incompatible with the conjunction of these beliefs. Bob expects his father not to help him even though he believes his father wants to and is able to. If we recall that belief is an acceptance that the world is a certain way, I think Bob’s expectation is irrational given what he believes since his beliefs would seem to generate a wholly different expectation in a rational person*. 

The point is this. Because Bob’s original expectation was not met, the unmet expectation should diminish Bob’s belief in the truth value of the original grounding beliefs because those beliefs no longer serve to support the expectation Bob has. If Bob continues to believe propositions 1-4 and continues to have the expectation generated by them unmet, his epistemic state is not aligned with his lived experience and this misalignment could lead to Bob being irrational or, worse, delusional. Something has to give.

My claim is that this analogy holds with belief in God. If one comes to believe that God exists and has certain properties, that belief will generate expectations, which, if not met, affects either the strength or the relevancy of those beliefs. It’s not possible for the expectations to go unmet, for a person to maintain belief in those grounding propositions, and for those propositions to retain their epistemic strength.


"I think the reason we sometimes have the false sense that God is so far away is because that is where we have put him. We have kept him at a distance, and then when we are in need and call on him in prayer, we wonder where he is. He is exactly where we left him." – Ravi Zacharias

The strength of my argument is directly related to the strength of the intuitions that the expectations I claim should be generated by belief in God’s existence are valid expectations. In what follows, I will consider some rejoinders to that and not to the argument itself. I believe if the expectations can be shown to be unreasonable, faulty, irrational, or in some way ungrounded, the argument loses steam and will crumble due to a lack of foundation.

Rejoinder 1: God does meet the expectations you cite for me personally

This response essentially claims that the expectations are sound and God comes through: he does regularly act in the world in ways that meet all the expectations cited and so there is a direct epistemic relation between beliefs about his existence, the expectations generated by those beliefs, and a given person’s experience. This, in my view, is an empirical claim and the evidence that God does act in the world to meet the expectations cited above is, as far as I can tell, paltry to non-existent. The empirical evidence does not seem to be in the favor of the one making this claim. But since my argument is focused on the person holding the belief, I leave it to the individual to evaluate this evidence.

Rejoinder 2: God has acted in ways that meet the expectations in the past even though he no longer does

This response centers around the idea that the Bible (or another sacred text) relates stories of God’s activity in the world that meets all the expectations cited and these are sufficient to establish the fact that the expectations are met even though God no longer interacts with human in just those ways. In response, I will claim that part of the expectations that are generated by belief in God’s existence is that he would act in ways consistent with Biblical descriptions throughout time. This is partly because the Bible partly sets those expectations. The Biblical stories explain who God is, not who God was and the types of prayers uttered by believers today are based on expectations generated by the activities of God based on his character and essence as explained in the sacred texts.

Additionally, with time, the epistemic grounding of stories related in a book like the Bible to present experience of God’s existence weakens. With advances in science, psychological understanding, and technology, the ability of stories from the past to satisfy expectations generated by belief in God’s existence wanes substantially because these advances tend to undermine the veracity of the stories (consider the modern vs. the pre-modern view of things like witches, disease, and psychological dysfunction—it’s a lot harder to believe in a witch today than it was 400 years ago). As time marches on, the epistemic distance between narratives about God’s activity in the world and belief in his existence grows and, I’d claim, in our current age that epistemic gap is so large that the narratives no longer function as a ground at all. They may function as a basis for how believers view who God is and thus how the believer should expect God to act today and I think those expectations largely go unmet.

Rejoinder 3: By changing your perspective, you’d see that what we view as ‘mundane’ are actually acts of God

The problem, this claimant is making, is not that God does not act to satisfy the expectations I cite because meeting such expectations is superfluous and unnecessary. Rather regular events in the world, those we experience everyday, are direct acts of God for those that have eyes to see. The problem with this claim is that it seems to put God in a situation where he’s over-determined. The expectations generated by belief in God’s existence are such, I’m claiming, that God meeting them isn’t the best explanation. In other words, if the expectations can be met my other explanandum that are simpler and explanatorily sufficient, God as the explanans becomes superfluous explanatorily and thus it doesn’t count as a genuine theistic expectation of the kind that I care about. The idea is this. God is said to be distinct from his creation and has properties that no other being or object has. That implies that there are things that only God can do such that no other explanation could reasonably satisfy the expectation. If there are other, viable explanations for why a given event occurred where the occurrence of that event can be subsumed under a general law of nature or under acts caused by other agents, then one would need another reason to attribute it to God and not subsume it under one of these other categories.

Rejoinder 4: The Gideon problem

It could also be argued that, for the skeptic, no explanation that includes God as an explanans will do even for those events which don’t have a ready naturalistic explanation. For any activity that a believer would claim to be of God, the skeptic will just claim that it’s a naturalistic occurrence but we just don’t have the science yet to explain it. Such a skeptical position is certainly question begging if the presupposition the skeptic brings to the argument is that God does not exist. But It is not question begging if naturalistic explanations are more reasonable than the theistic explanation. For the skeptic that is open to the idea that God might serve as the explanans but so might other, non-theistic explanans, the argument misses the point. This also is why the type of expectations I cite are such that God is the best explanans for the explanandum. The burden is on the theist to explain why it is God who met the expectation and why it probably is not attributed to something other than God when other explanations are possible.

Rejoinder 5: Unmet expectations as an anti-nomological necessity

Related to rejoinders 3 and 4, this rejoinder states that God could not meet the expectations I cite as a regular part of his activity because if he did, his activity would come to be viewed in a law like way and could thus be made explanatorily superfluous or underdetermined. For example, if God healed the sick each time a person prayed for healing, it would be possible to subsume the activity and response under a law and that law would become normalized such that it would no longer have the same doxastic impact of generating the belief that God is the one who healed. In other words, in order for a miracle to be a miracle, it has to be unusual.

This rejoinder may undermine my claim only in those conditions where God’s activity would serve as an the best explanation for why the expectation was met. But even under law-like circumstances, if a given explanans is the best explanation, it is less reasonable to deny that the explanans explains the explanandum. If God serves as the best explanation, it is less reasonable to subsume the events under a physical law regardless of how regularly God meets the expectation. For example, if a lamplighter comes by every evening and lights the street lamps in the village, we might be tempted to call it a law: at dusk, the lamps light. But even in this situation, the expectation that the lamp will be lit by the lamplighter is best satisfied by the lamplighter. The regularity does not underdetermine the epistemic force of the explanans. If, all of a sudden, the lamps started lighting without the appearance of the lamplighter, the law would still hold, but the expectation would not be met and we would need a new explanans to explain the explanandum.

Finally, Christian narrative indicates that God did so act to meet these expectations in just the ways I describe at a point in the past and these events retained their status as acts of God. Believers claim that God acted regularly to heal the sick and feed the poor and avoided this nomological threat both because of the context of the activity and the clear distinction between the nature of God’s action and of natural or human action. Nothing was lost by the regularity of Jesus’ divine actions over the three years he was religiously active.

Rejoinder 6: God is not a cosmic jukebox

This response claims that while the expectations I cite may be appropriate given belief in God, we should not expect the expectations to be met because God does not act according to what we expect him to do: God is not a cosmic jukebox where we get an expected output from a particular input (I’m deliberately avoiding the more common phrase "cosmic slot machine" hopefully for obvious reasons).

The only way to make sense of this response, I believe, is by subsuming it under the class of response that claims that the expectations that are generated are inappropriate, improper, unwarranted or somehow or other wrong expectations. If a belief generates an expectation but the believers should not expect the expectation to be met, it ceases to function as an expectation in the sense I mean. If belief in God is that he is all good and powerful and that generates the expectation that he will help suffering innocents, it does not follow to claim that we should not expect God to help suffering innocents even though the expectation that he will follows from belief in his existence. 

Rejoinder 7: God already has met all these expectations

As I mentioned above, part of the rationale for the validity of the expectations I cite is grounded in the idea that God is said to have acted in the ways I claim we should expect him to act in religious texts like the Bible. Many believers understand the Biblical narrative to be literal descriptions of God’s actions in time and space. Throughout the Bible, God is described as being directly involved in many miracles including healings and the raising of the dead, suspension of the regularity of the laws of nature including stopping the rotation of the earth, causing seas to part ways, and tearing down large stone structures. He also is described as giving clear pronouncements of his will both for individuals and mankind, to comfort for the lonely in times of distress, to predict future events, and changing the will of despots. This rejoinder acknowledges these narratives as accurate descriptions of God’s activity in the world and then claims that these are adequate epistemically to ground belief in God’s existence such that God does not need to continue to meet these expectations for modern believers. As I understand this rejoinder, the interlocutor is claiming that that the expectations I cite are appropriate but that it is inappropriate to expect God to meet these expectations at all times in human history and that God, having met them in the past, has provided enough evidence of his existence such that continuing to meet these expectations for modern believers is superfluous.

In one sense, each individual has to judge what is epistemically adequate for them in terms of what God will do or has done. However, I believe this response misses the nature of the types of expectation I describe. The nature of the case as I’m attempting to frame it, entails that we have expectation of how God will act in our lived experience today even if we can acknowledge that God may have met these expectations in the past. The nature of the expectations I believe are generated by true belief in God is that historical distance is an inadequate because the nature of the expectation is that God will be involved in my life and only by being so involved can the expectations be met.

Further, I think it is disingenuous for a person to claim that God’s actions in the first century and prior is adequate epistemically and existentially while that person engages in fervent prayer with the expectation that God will answer.  As I stated above, given that many believers actively pray for all kinds of divine interaction demonstrates that these individuals do not believe that first-century actions by God are adequate to meet the expectations that true belief in God’s existence generates.

Finally, I think this response misses the mark in that the claim that God has acted in specific ways in the past is intended to have apologetic value (provide evidence for his existence) but fails existentially. That is, referring to God’s past action is given as evidence that of God’s activity in the world but the nature of expectation as I’ve articulated it entails that those expectations would be met in the here and now. To refer to God’s activity in the past as a ground for having current expectations is an apologetic answer to an existential question.

Rejoinder 8: God meets these expectation through natural events and human beings

This response is similar to earlier ones in that the claim is that the apparent failure of God to meet expectations I cite assumes that God cannot meet them through other efficient causes; that he somehow has to act directly in the world and must, himself, be the visible and efficient cause of the events that meet the expectations. For example, if one expects God to heal children with leukemia and that child is eventually healed through the hard work of doctors and medical treatments, the expectation has been met but not in the way I assume God must meet it.Because efficient causes other than God’s direction action brought about the expected result does not mean God did not meet the expectation.

Again here, I think the problem with this response is that it runs the risk of removing God from the explanatory picture altogether and God becomes underdetermined—something I don’t believe believers would want to do. I assume that believers will acknowledge that there is a distinction between a human performing some action and God performing some action. That is, God is a distinct being with his own domain of control that can be distinguished from that of humans (for example, replacing “God did” or “God said” in the Bible with the name of some person and attributing actions to persons that the Bible attributes to God cannot be done without losing the meaning of the original narrative). If this distinction holds, it should apply to the way the expectations I cite are met. If all the expectations I cite can be met by humans, then either it doesn’t respect the distinction and conflates human and divine action, or it removes God as the efficient cause of meeting any expectation and grounds the meeting of those expectations entirely on the actions of humans. Either should be unacceptable to the believer.


This discourse has attempted to establish an existential tension between true belief in God’s existence and unmet expectations that are generated by that belief. This existential tension has an epistemic implication for the strength of the arguments that grounds the belief that gave rise to the expectation, namely, that those arguments are undermined.

1. Falsification arguments have also been dismissed, crudely in my opinion, by being called a type of warmed-over positivism. While there is some truth to this, I think this amounts to handwaving and throws out the baby with the bathwater. While positivism as an epistemology has critical flaws, it does have insights worth considering about some aspects of belief formation.

2. It’s worth noting too that McKim holds (and I tend to agree with him) that given that arguments for and against belief in God do stand in epistemic parity, this accrues a particular epistemic responsibility on the part of believers and atheists alike, namely, that both should hold their beliefs tentatively. By this, McKim means that the belief involves (1) A recognition that the belief may need revision and may be mistaken, (2) A concomitant openness to alternative awareness that some of these alternatives may be plausible, and that one or more of them may even be correct, (3)  Belief that p assigns to p a lower probability than strong belief that p assigns, (4) A psychological component, namely that one is less convinced of p, and (5) An awareness that one may very well be wrong and that alternative views may very well be right. While this seems to be important epistemically, this whole picture is not critical for the argument I’m making here.

3. I will deal with specific rejoinders to these expectations below but I expect that an initial response by believers might be along these lines: God simply does not interact with humans in this way and it’s inappropriate, or arrogant, or silly, or unnecessary, or somehow beyond what humans should expect from an almighty God. Faith just is believing in and experiencing God in lieu of his direct involvement with our lives. Blessed are those who believe and have not seen. The initial problem with this response is that God is said to have done each of the things I mentioned in stories related in the Old and New Testaments of the Bible and he did them for the express purpose of bolstering faith and making his presence known. So at least for a few years he thought it important to get directly involved with humanity and the humans at the time saw that as both helpful and necessary.

4. Many who attempt to describe miracles make exactly this claim. They explain that a miracle isn’t necessarily a violation of natural law. Rather, the miraculous event can be seen as entirely consistent with natural law though it may be entirely abnormal. What makes it a miracle is the occurrence of the event within a particular religious schema (someone prays, or a religious person with God-given power gives some command or the like).

* I’m setting aside whether propositions 1-4 and the proposition that constitutes Bob’s expectation are logically compatible since that would take us beyond the scope of this paper. My intuition is that they could be viewed as logically compatible and so would not be undermined on logical grounds. But in order to preserve the logical consistency of these 5 propositions, propositions 1-4 would, at the very least, need to be qualified by further propositions as I explain earlier.