Saturday, March 30, 2019

A Bottom-Up Defense of Christianity: Not-Pascal’s Wager

For the past three or four months, Ron Garret and I have been going back and forth in what we suspect might have set the record for the longest running cordial conversation between a young earth creationist and an evolutionist.  We have, to date, exchanged over 540,000 words, which translated to a 156-page PDF document.

The courtesy and respect coming from Ron throughout this conversation have been outstanding.  For my own part, I have attempted to rise to Ron’s level of decorum, and even if I have fallen under reproach once or twice, I am very glad for the access that Ron has provided me to his perspective.  I very much look forward to continuing this conversation on a platform that will hopefully be easier to search and track, since megathreads like these become cumbersome on Reddit’s platform.

Our conversation began when I challenged Ron’s representation of skepticism. Our conversation took place in Reddit, under the Creation subreddit, in the comments of a post titled “Creation/Evolution debate: A priori vs. a posteriori knowledge”.  From there, we have continued our conversation revolving around a game theoretical approach to belief formation, which leads me to call myself both skeptical and Christian.  Along the way, we have touched on many related topics, and I believe we have both challenged one another, as well as helped us each to better understand the other side.

Ron and I have both addressed ideas held by the other which we find reprehensible.  I have stood in defense of God’s wrath upon the Canaanites, whereby they (and, importantly, their children) were subjected to all sorts of suffering, including death by the sword, dislocation, and slavery.  Ron deplores the ideas that I hold, yet he has treated me with dignity and respect, apparently seeking only to correct what he feels is wrong.  Even if he doesn’t change my mind, it may be that by understanding our reasoning, he might be able to more effectively appeal to others like me.

And, contrarily, Ron has stood in defense of the murder of unborn children, which we are used to calling abortion.  While the pro-abortion position itself is abhorrent to me, I am long cured of the delusion that I am morally superior in my nature than anyone who has come to hold it.

We have both recognized the fact that catastrophic evil can result from the most banal of sins.  A little motivated reasoning, a tiny bit of self-deception, and one can feel justified in tolerating, approving, celebrating, even participating in horrendous acts of evil.  I believe Ron is wrong.  But God help me if I ever suggest that I am in some way better than him.  We should never forget that we are all the same kind of thing.  We are human beings.  God help us.

Allow me to introduce myself and my argument at the center of our discussion.  My name is Jimmy Weiss.  I am 34, live in Oklahoma, and I am a dabbler by nature.  I tried my hand in music production and performance, which was by far the most fun anyone could have while failing completely.  I settled down when my wife and I conceived our first child in 2005.  We now have a son and a daughter, and I currently work as an accountant, but enjoy reading and writing whenever I can.

My testimony can be read in full in the Christian Apologetics subreddit.  My initial conversion from atheism to general theism took a form very similar to what we know as Pascal’s Wager.  It was only a year or two ago that I was giving my testimony, and someone pointed out the similarities between my reasons for believing and Pascal’s Wager.  I was told that Pascal’s Wager was deeply flawed.  When I looked into the matter, I found that they were absolutely right.  But I perceived the flaw of Pascal’s Wager to be in the application of the underlying principles to a false dichotomy, Christianity vs. Atheism.  Those principles come from game theory itself, and are demonstrably sound.  They are employed by the casino and insurance industries on a daily basis with enormous success.

The gist of my argument, which I call “Not-Pascal’s Wager”, is that game theoretical analysis of the “basic problem of religion” returns a sound rationale for treating the existence and goodness of God as a properly default belief.  Skepticism, then, is properly applied to propositions counter to this belief.  This is why I call myself both a skeptic and a Christian, and why I originally challenged Ron’s proposition that skepticism ought to resist theistic beliefs.

Our conversations surrounding this argument have led us to the point where Ron sought to understand how this argument related to both my basic axiomatic starting points, as well as my acceptance of the YEC position.  I provided a nutshell version containing 13 points, and Ron and I have agreed that a more exhaustive explanation of those points would make a good introduction to his audience as we switch to this new platform.

(1) The preconditions of intelligibility are true.

I believe that Ron and I are agreed on all but one of these preconditions.  As I understand them, in order for us to be able to make any sense at all of the universe, we have to accept a few basic axioms.  The first is that the laws of logic are absolutely true.  These laws include:

(A) The Law of Identity, which states that something is what it is, and whatever exists must have a specific nature.
(B) The Law of Non-Contradiction, which states that something cannot be itself and not itself at the same time, in the same way, and in the same sense.
(C) The Law of Excluded Middle, which states that a statement is either true or false, with no third option.

Another prerequisite for intelligibility is the general reliability of our own senses and minds.  Of course, as Ron and I have discussed, it is undeniably the case that our minds and senses are not always reliable.  But I generally find that in the moments when my mind has not been reliable, I was not seeking answers to the mysteries of the universe.  When I set myself to thinking about such things, as now, I feel very comfortable with the assumption that my mind and senses are giving me true information about reality.

The third precondition for intelligibility is one that Ron has contested.  Uniformity in nature (not to be confused with “Uniformitarianism”) means that nature will reliably produce exactly one possible outcome for any given starting state of affairs, when all drivers are properly accounted for.  In other words, the laws of nature don’t change.  It has been my position that the practice of science properly depends on this assumption.  Ron has taken the position that uniformity is a conclusion of science.  I am afraid that we will find this divergence at the root of many of our disagreements.

If these first principles are not true at all times and in all places, then we could know nothing about anything.  We cannot prove or disprove them, but we see clearly that accepting them is useful, and rejecting them is harmful.  In fact, one might be able to apply a game theoretical analysis to support the acceptance of these prerequisites as properly default beliefs.  We ought to be most skeptical of any claims contrary to these points.

(2) Basic facts about reality suggest that a “Creator God” exists

The “God” concept that I employ here and in the following game theoretical analysis consists only of the most obvious and necessary qualities which are suggested by these basic facts.  It is important to understand that I am not yet, at this point in the total argument, even remotely interested in the God of the Bible, or any of His named competitors.

Within this point would fall the classical “proofs” of theism.  I have spent a good deal of time studying “cosmological” style arguments, and I am even convinced that certain forms of the argument are valid proofs for the existence of Creator God.  Nevertheless, for the purposes of this total apologetic argument, my only goal here is to show that these basic facts suggest the existence of Creator God.

The basic fact of reality behind cosmological arguments is causality.  Everything we actually observe occurs or comes into being due to some cause external to itself.  My chair came into being because its parts were assembled by a chair maker.  I came into being as a result of my parents’ copulation.  My understanding of multiplication was brought about by my understanding of addition.  And each of these causes have their own, external causes.  The chair maker, like me, came about as a result of his parents’ copulation.  And our parents were the result of their parents’, and so on.

So we find in the actual world an apparently inviolate principle of causation.  And also, we find an actual causal regression, which raises the question of origins.  If we imagine a world without religion, this question of origins would present itself almost immediately to intelligent creatures, having the leisure to ponder such things.  And such a creature would be able to perceive three, and only three basic possibilities, considered with respect to causation.

These three possibilities arise through a nested, double dichotomy.  First, it would be clear that there must either be or not be a First Cause.  The absence of a First Cause means that there must exist an infinite regression of causality, with every cause having its own cause before it.

I believe that Ron and I are in agreement on the point that the concept of the infinite regression violates the law of non-contradiction, and is therefore properly deemed impossible.  Nothing that is actual can be infinite, because infinitudes are purely potential.  Such is the definition of infinity.  Time and space can be infinite, precisely because they are potentialities for actual things to come into relation to each other.  God’s strength can be infinite, because it is His potential for actualization.

So whatever is infinite, must be potential, rather than actual.  But the causal regression that we are wondering about is derived from the actual causal progression we find in progress.  Since it is actual, it cannot be infinite.  Thus, the infinite causal regression is out.

That only leaves the other side of the initial dichotomy.  It is impossible for there to be no First Cause; thus, there simply must be a First Cause.

Now, given that there must be a First Cause, we still must consider whether this First Cause does, or does not have a cause, itself.  Yes, we have already ruled out the possibility that the First Cause might have an external cause.  And while every cause we have ever observed has been external to the thing caused, we have not been able to rule out the possibility that a thing might have its cause in and of itself.  This is certainly exotic to our experience, but is it more or less so than the alternative?  Let’s take a look.

The alternative to a First Cause having its cause in and of itself is for the First Cause to have no cause at all.  This is what I call “null causation”.  Ron believes that there is evidence for null causation in our observation of the most fundamental particles.  This is a point of contention between us, which I believe stems from our disagreement about the metaphysical presuppositions of science.  In fact, it may be the case that the most fundamental particles behave without any natural cause, but that does not mean that they have no cause at all.  I believe that Ron is simply begging the very same question we have to face in this cosmological argument.

There is, as it turns out, a very serious logical problem with the affirmation of null causation.  As with infinite regressions, null causation turns out to entail a violation of non-contradiction.  The contradiction arises because “nothing” is being treated as though it were “something”.  If it were really true that nothing could cause anything, then it must follow that anything could be caused by nothing (making the practice of science impossible).  For nothing can have no properties which constrains what, where, or when it should produce effects.  To say that nothing would produce effects only in certain, rarified situations is to presume that nothing has certain properties which so constrain it.  And if you ascribe properties to nothing, then you are treating it as though it was something, and thus contradicting yourself.

This contradiction satisfies me regarding the conclusion of this cosmological argument.  The impossibility of infinite regression and null causation leaves only one general category of possibility: the existence of a First Cause, which has its cause in and of itself.  For the intents and purposes of these arguments, this First and Ultimate Cause is what I am referring to when I use the title “God” or “Creator God”.

As a side note, this quality of self-causedness, or self-existence, implies certain other qualities.  I have allowed Ron to hold that the quantum wave field or some other aspect of the natural universe could satisfy the requirement for a First and Ultimate Cause.  I believe that this is mistaken, and that this mistake arises out of a lack of understanding about the implications of self-existence.  Nevertheless, the total argument that I am attempting does not require a specifically classical theist concept of God.  I believe it must stand that causality is a basic fact of reality which, at the very least, suggests the existence of a First and Ultimate Cause, which we can very reasonably call “God”.

I will not go into this level of detail for the other “proofs” of theism.  Partly because I have put nowhere near as much effort into understanding them and their implications as I have with the cosmological arguments.  But, so that we might come back to them if needed, here are the arguments, along with the basic facts on which they rely:

The Moral Argument relies on the sensibility and our general agreement about moral truth.

The Teleological Argument relies on the appearance of design in nature.

The Transcendental Argument relies on the intelligibility of the universe, or the absolute truth of the laws of logic.

Again, I cannot go to the mat for these arguments as I do for the cosmological argument.  But at the very least, I would fight for the authentic suggestiveness of these arguments.  This suggestiveness is important because it shows why we are not talking about something comparable to pixies, or flying spaghetti monsters.  Even if one doesn’t think these arguments prove the existence of Creator God, one should at least be ready to admit that Creator God is suggested by basic facts about reality in ways that pixies and flying spaghetti monsters are not.

(3) Game theory leads me to believe for all intents and purposes that God is good.

This is the central point of our conversation: my “Not-Pascal’s Wager”.  This argument will serve to define the standard of evidence which is applied in all following points.  As a result of these conclusions, I believe we should treat as properly default, the belief that Creator God exists and is perfectly good.  In other words, this is the belief that should be held until positive proof, beyond the shadow of any doubt, comes to our attention that God either does not exist, or is not perfectly good.

Not-Pascal’s Wager shares the basic game theorem with Pascal’s Wager:  A wager is favorable when the cost (including opportunity cost and excluding sunk cost) is less than the product of the probability of success and the reward in the case that the wager is successful.  In mathematical terms, a wager W is favorable when true, and unfavorable when false, according to the logical test expressed as:

W = | C < P x R |, where C is cost, P is probability of success, and R is the potential reward

Before I move on to the way that Not-Pascal’s Wager applies this theorem differently than Pascal’s Wager, I want to establish two things.  First, I will detail an example of a simple wager, to which I invariably return in these conversations.  Second, I will establish the connection between wagers and the formation of beliefs.

The simple example that I use is this: You can buy in to a wager for $1.  A coin is flipped three times.  If the coin lands on tails at any time, you have lost the wager, and the house keeps your dollar.  However, if the coin lands on heads three times in a row, you get the reward.

So far, two pieces of this wager have been established.  The direct cost is $1, and the probability of success is 1/8th.  If the direct cost is the only cost, then the reward would have to be greater than $8 for this wager to be favorable.  If there was opportunity cost, then the reward would have to be increased by eight times the value assigned to that opportunity cost, in order for the wager to remain favorable.

This demonstrates an important principle.  The action you take when faced with this wager will change, even when the probability of success remains the same.  You should not take the bet when the reward is less than $8, but you should take the same bet when the reward is increased to, say, $1million.

The force of this argument depends in part on being able to connect these principles to belief formation.  I have usually found it to be the case that people who call themselves “skeptical” seem to believe that belief formation is involuntary.  Once they are presented with evidence, they feel that their formation of a belief occurs spontaneously, completely free of any act of choice or preference.

When they say so, they are describing what is in fact only part of belief formation.  The reality is that we are usually faced with a set of possible explanations, and our involuntary response to the evidence entails a perception of the allocation of probabilities among the set.  I agree that we naturally, involuntarily, experience this shift in perception.

But there is another aspect of belief, which is called “reliance”.  The real test of belief is whether you are willing to rely on that belief; to act as though it was really true.  And this is exactly why probabilism, all by itself, is an insufficient way to describe belief.

Take this back to our simple wager.  Remember that you only have a one in eight chance of winning the wager.  You will probably lose.  If probabilism was the only thing that one should take into account, then you should reject the wager, no matter what the cost or reward are.  Changing the reward should not affect your decision to accept or reject the wager.  And yet, you see that it does.

It should be noted that not every belief is “wager-like”.  The set of possible solutions to a problem may not entail an unequal allocation of cost and/or reward.  If the costs and rewards are evenly distributed across the gamut of possible solutions, then of course you can see that probabilism is all that matters.  By all means, let’s believe and rely on the most probable solution.

But, a problem is wager-like when (1) the set of possible solutions has more than one member (which simply means that the problem hasn’t already been solved), and (2) the costs and/or rewards associated with each possible solution are incongruous with their associated probabilities.  In such a case, it may turn out that the solution to be preferred is not the most probable, but the most favorable, by a game theoretical analysis.

Note that this does not entail any form of self-deception.  Relying on an improbable belief because it is favorable as a wager does not mean that you have to lie to yourself about the probability which you have assigned to it.  If you read my testimony at the link provided above, you will see that my decision to believe in a good God was associated with a very low perceived probability that it would actually turn out to be true.  For me, at that moment, it was either believe this, or kill myself.  Having no hope in this world, I was able to perceive that a slim hope was better than none.  I never pretended that my situation was different, not to myself, and not in my prayers to God.  All that mattered was that I was relying on this being true: in my intent to go on living, and in the purposes to which I directed that new life, I made the decision to rely on the existence and goodness of God.

Two quick notes on that point: (1) I was not then a Christian, and would not be a Christian for over a year; and (2) I am, of course, no longer so uncertain about the truth of the beliefs that I am relying on.  I have allowed in my conversation with Ron a 5% perceived probability that God is not the God of the Bible.

So: when a problem is wager-like, the most preferable solution is not necessarily the most probable, but the most favorable according to the game theorem.  The next step is to properly define the “basic problem of religion”, and see if it might be wager-like.

This is where Not-Pascal’s Wager differs from Pascal’s Wager.  Pascal defined the problem of religion as Christianity vs. Atheism.  In reality, this is only a tiny portion of what I call the “advanced problem of religion”.  Christianity is, in fact, only one of many theistic religions from which we have to choose.  Because Pascal ignored these other members of the set of possible solutions, he failed to properly define the opportunity costs associated with Christianity, which would include the risk that another theistic religion besides Christianity was true.  As St. Paul said, “if Christ has not been raised, your faith is futile...we are of all people most to be pitied.” (1Cor15:17,19).

By throwing our lot in with the God of the Bible, we take a position opposed to, for example, Allah of the Quran.  The two bets are mutually exclusive.  We Christians risk subjection to Allah’s punishment, and forego the possibility of receiving Allah’s reward.  Without establishing any grounds for disregarding these elements of cost in the wager that Christianity is true, the cost and reward sides of the game theorem formula are both infinitudes, and thus the wager is not favorable.  But by first applying the game theorem formula to a more basic version of the problem of religion, we will find grounds for excluding such costs, and thus will find that the wager of Christianity really is a favorable wager.

Now, if the “advanced problem of religion” involves the selection of one out of 4,000 or so specific religions, you will understand that the “basic problem of religion” is more, well, basic.  It is not Christianity vs. Atheism vs. Islam vs. Hinduism vs. etc.  Instead, the basic problem of religion reduces the problem to basic dichotomies.  Either God exists, or He does not exist.  If He exists, then He is either perfectly good, or He is not perfectly good.

One virtue of this construction is that it limits the members of the set of possible solutions to three and only three.  This is a true trilemma, so when we assess the opportunity cost of each member, we only need to define the forgone rewards of the other two members.

The next thing to understand about the virtue of this construction is that it is not arbitrary, because it is useful.  It is possible to dichotomize the problem in other ways, and so one might say that dichotomizing the problem as I have is purely arbitrary.  But it should be easy to see the usefulness of this construction, when opposed to really arbitrary constructions.

For example, we could have nested under the existence of God, the dichotomy of “funny” vs. “not funny”.  Of course you see that the funniness of God has almost no bearing on how we should perceive His costs or rewards.  And, of course, some might prefer gravity to comedy, and thus even perceived skews on those grounds would be purely subjective.  Obviously, we can see that the funny/not funny dichotomy is useless.

Using goodness to dichotomize the problem is useful because goodness is an amalgamation of several relevant qualities, all of which have specific bearing on the potential for God to reward.  Part of goodness is trustworthiness.  Even if God existed, and even if God made a promise of infinite reward, it should have no bearing on our decision to rely on Him, if He was not trustworthy.  Another part of goodness is power.  Why should we rely on Him to deliver a reward if He had no power to do so?  And of course, another part of goodness is justice.  If God does not properly and finally deal with sin, then what hope can we have that His reward would not be marred by sin?  If you found that God was offering you the same reward as Hitler, would you be inclined to put any stock in that offer at all?

But if God is perfectly good, then that would mean He might be willing and able to deliver a reward of ostensibly infinite value; and further, if He promised such a reward, we would be able to rationally put faith in that promise.

For the rest of this exposition, I will at times express this dichotomy as “good vs. evil”.  Of course, everyone recognizes that good and evil are a spectrum.  Some of the things we do are worse than other things, and some are better.  However, there is a good reason that theologians treat good and evil as a true dichotomy, and we better establish that fact right off the bat.

In order to demonstrate the case in which good and evil are dichotomous, I use this analogy comparing righteousness (which is also entailed by goodness) and efficiency:  Righteousness is to society, as efficiency is to machinery.  Efficiency is also a spectrum.  Some machines are more efficient than others.  The nature of our universe is such that no machine can ever be perfectly efficient.  “Perfect efficiency” is, to us, just as much of an ideal as “perfect righteousness” is for society.  But, if it were possible for us to build a perfectly efficient machine, we would observe something very interesting: unless someone interfered with that machine, it would never, ever break down.

You see, efficiency is the measure of how much useful energy is lost in the operation of a machine.  Usually energy is lost due to friction, or as unusable heat energy from chemical reactions.  These forms of energy loss create wear on a machine, causing its parts to eventually break down.  Perfect efficiency means no energy loss, no wear, and no break down.

Now, imagine that you are wanting to build a machine that will run forever.  The spectrum of efficiency is no longer interesting to you.  “More efficient” isn’t good enough.  “Really, really efficient” isn’t good enough.  Only “perfect efficiency” will do.  Your objective is pass/fail on that measure.

So it is with righteousness.  In order for a society to last forever, the parts of that society must be perfectly righteous.  “Not so bad”?  Not good enough.  “Really, really righteous”?  Not good enough.  Only perfect righteousness will do.

Here is another way of illustrating the dichotomization of good vs. evil.  It is possible to express a person’s righteousness by the amount of unjustified suffering they produce in a certain unit of time.  A perfectly righteous person will produce exactly zero unjustified suffering during any given measure of time.  And you see that a person who is not perfectly righteous will produce not-zero unjustified suffering during any given measure of time.  We can express this relationship mathematically:

G x D = E

G equals the amount of unjustified suffering the person produces during one unit of time;
D equals the duration of time for which we watch the person, and
E equals the total amount of unjustified suffering produced by the person during a measure of time equal to D.

Notice something interesting: As the duration, D, approaches infinity, E will also approach infinity for all values of G - except zero!

Now these two illustrations should firmly establish that good and evil, observed on an eternal time-frame, are properly dichotomous.

So, we have defined the basic problem of religion: Either God does not exist; or God exists, but is evil; or God exists, and is good.  We can already see that the problem, stated as such, is properly wager-like.  An infinite reward is at least possible in the case that God is good; no good reward can be rationally hoped for if God is evil; and the rewards available to the atheist are also available to the theist.  There is a skewed allocation of rewards, which means that the most favorable solution may not be the most probable.

Since the problem, at this level, seems like it could be wager-like, the next step is to qualitatively define the rewards for each wager.  We do well to define the rewards first, because our analysis of cost will require us to include the risk of foregone rewards arising from rejecting the successful wager (opportunity cost).  Furthermore, understanding the rewards will reveal why game theory is especially applicable to the problem of religion.

We have already mentioned the obvious fact that belief in a perfectly good God has at least the potential to yield an infinite reward.  It may be that He would be willing and able to deliver a reward of ostensibly infinite value; and further, if He promised such a reward, we would be able to rationally put faith in that promise, since a perfectly good God would, by definition, be trustworthy.

We should note that the whole reason that God’s reward might be of infinite value, is because His reward can be eternally durable.  Again, only that which is potential, like time, can be infinite.  Whatever reward God offers, the goodness that it entails in any given instant of time must be finite, because it will be actual.  But any positive intensity reward which endures forever is legitimately infinite.

Ron raised the right kind of objection to this point: we can call it the problem of boredom.  His argument appealed to our tendency to lose interest in any good thing over time.  And thus, he supposes, there is no possible reward which even has a chance to endure for all eternity.  Eventually, we must find ourselves in desperate, even agonizing boredom.

This is the right kind of objection, because it seeks to make the reward impossible.  That is the standard which a valid objection must achieve.  I will soon be saying more on this subject, but even a very small non-zero, multiplied by infinity, is still infinity.  Thus, it is not enough to object with matters which merely reduce the probability of our eternal reward.  A valid objection must entirely eliminate the probability.

And in this case, I expressed three reasons why the problem of boredom does not eliminate the probability of our eternal reward.  I first stated that, since personal experience attests to the renewal of interest after a sabbatical, then rotation between several activities might maintain positive levels of interest and enjoyment through eternity.  Secondly, I pointed out that any condition impairing one’s ability to perceive and enjoy goodness might be specific to our current fallen nature, and absent in our eternal form.  And finally, it’s possible that the infinite creativity of a perfectly good God might make it so even novelty, at some level, could persist through eternity.

So the reward for betting on the good God is potentially infinite, because only through God can we gain access to eternity; and only a good God could be trusted to deliver an eternal reward worth pursuing.

What are the rewards for betting on the “no God” wager?  Now, we have to be careful.  It is clear that there could be costs associated with believing in God, and the “no God” wager would allow us to forego those costs.  But this should not be accounted for as part of atheism’s reward.  It simply is a cost on the theist side, and is not a cost on the atheist side.  The relevance of this cost differential will become explicit when we come to understand how opportunity cost is accounted for.

Anything else?  Honestly, I have no desire to deprecate the value of life, and the purposes which are undertaken by atheists.  They can strive and succeed in making the world a better place.  They can establish a legacy that can last for thousands of years.  These are all wonderful things that I don’t want to take away from anyone.

But there are two things which exclude them from consideration in this context.  First, as great as they are, they are both finite and temporary.  Come the heat death of this natural universe, nothing any of us ever did will matter in the slightest, if atheism is true.  If we are approaching the problem of religion using an eternal timeframe, then we mustn’t delude ourselves.  Everything in this world will come to nought.

Secondly, these kinds of rewards are not actually exclusive to the “no God” wager.  Theists are just as capable of living good and happy lives, making the world a better place, leaving a legacy which impacts generations of human beings.

The only other reward for the “no God” wager, that I can come up with, is oblivion itself.  And oblivion is certainly not the worst thing we could find outside of this life.  It’s certainly heartbreaking to think that Grandma is just gone, forever.  But if the alternative is hell (and more on that later)?  Well, oblivion is infinitely better, at least, than that.

But here is the thing about oblivion: you don’t have to “bet on” atheism in order to receive this reward.  If there is no God, then all will receive this reward, regardless of their beliefs.  As such, this too is excludable from our game theoretical analysis.

In the end, I can find nothing, nothing at all, that I can credit as a unique reward of atheism.

So now we come to the “evil god” wager.  And here we find something interesting.  If God is evil, then, taking an eternal time-frame, the only thing we should rationally expect as His reward is evil.  In fact, given an understanding of evil, in that its effects ripple and compound, eventually the reward of the evil God must become indistinguishable from hell.

Let’s peek ahead for a moment at the risk of incorrectly rejecting an evil God.  If God turns out to be evil, and we incorrectly rejected Him, we can imagine all kinds of bad things coming of it.  The punitive risk is essentially infinite in magnitude.  But we have seen here that correctly betting on an evil God also yields an infinitely bad outcome.

You see, when it comes to an evil god, we are damned if we do, and damned if we don’t believe in Him.  Nothing we can do will ultimately change our fate if God is evil, which would be essentially hell, either way.  That being the case, this member of the possible solution set becomes entirely useless, because nothing we could do would change the outcome.  We can therefore completely ignore the evil god!  That means that when we go to look at the opportunity costs of the other wagers, we don’t even have to take into account the punitive risk of incorrectly rejecting the evil god.  It is entirely excluded.

Common ground is sacred.  Let’s make note of what should be common ground between all atheists and theists:  Even if Creator God exists, we should reject Him in the case that He is evil.  By the end of this exposition, one may disagree that the “good God” wager is infinitely favorable over the “no God” wager; however we should all agree that the “evil god” wager is infinitely worse than them both.

Before moving on, let’s take stock.  The “evil god” wager is out.  We are ignoring it because it can have no bearing on our behavior, even if we chose to rely on it.  The “good God” wager has a chance to deliver an infinite reward, defined qualitatively as “an enjoyment of positive intensity enduring through time without end.”  The “no God” wager has no unique reward.  Let’s take a look at how this information can be slotted into the game theorem formula, and make note of how we might find each wager favorable as we venture into each wager’s costs.

We can differentiate each wager with lowercase qualifiers: Wg is the wager of the good God; Cg is the cost of the wager of the good God; Rg is the reward of the wager of the good God, and Pg is the probability that Wg will be successful.  In like manner, the variables in the “no God” wager will be Wn, Cn, Rn, and Pn.  So:

Wg = | Cg < Pg x Rg |


Wn = | Cn < Pn x Rn |

Looking first at Wg, our reward was qualitatively defined as “an enjoyment of positive intensity enduring through time without end.”  We can reasonably quantify this reward as infinity.  Now, in the case that God might exist, and might be good, Pg must be at least greater than zero.  And remember, the product of infinity and any real number greater than zero is still infinity.  So as long as the cost of believing in a good God is less than positive infinity, the wager is favorable.  The cost could be very high, and the probability very low, and the wager would still be favorable.

Next we look at Wn, the “no God” wager.  The reward for Wn does not exist, so Rn equals exactly zero.  As such, the reward side of Wn will equal zero no matter how probable we perceive its success.

Now, we should note that if the probability of success for atheism was exactly one, that would mean that the probability for every other wager must be exactly zero.  In that case, the set of possible solutions for the problem would have been reduced to one member.  That would kick this question out of the realm of game theory, because remember, one of the requirements for a problem to be wager-like was that the set of possible solutions has more than one member.  In fact, a probability of one means that the problem has already been solved.  And if the problem has been solved, then game theory could not apply.

So, given that the problem has not been solved, and we are actually in a position of uncertainty concerning the existence and goodness of God, we find that the reward side of the “no God” wager comes to exactly zero.  In order for Wn to be favorable, Cn will actually have to be less than zero.

Is it possible for the cost of a wager to be less than zero?  Sure, of course.  Go back to the simple wager example.  Three coin flips; tails you lose; else you win.  Only now, instead of buying into the wager for $1, we can imagine that you are being paid $1 to enter the wager.  And even if the potential reward was zero, you can see that this wager is favorable.  Why not take the free dollar?

As we move forward into the analysis of the costs of each wager, we already know this: the “good God” wager will be favorable so long as its cost is less than positive infinity; and the “no God” wager will be favorable so long as its cost is negative.

Remember that the costs of a wager include the direct cost, opportunity cost, and exclude sunk cost.  Thus, we could expand the game theorem formula by splitting C into elements dC (direct cost), oC (opportunity cost) and sC (sunk cost)

W = | dC + oC - sC < P x R |

Before we move on, let’s make sure all are on the same page about what these elements actually mean.

The direct cost of a wager includes all that is required to become eligible for the reward of the wager.  In our simple wager example, the $1 buy-in was the direct cost.  Direct costs obviously don’t have to take the form of currency.  Of course, costs are more easily quantified when they are, but since we have established the general boundaries of favorability for each wager, we will not have to be perfectly accurate.  We can recognize direct costs in the form of effort, suffering, loss of relationships, etc.

Opportunity cost is quantifiable as the net benefit (“NB”) which would be enjoyed by the best alternative to the choice under consideration.  The net benefit of a wager is the product of the probability of success of a wager and the potential reward of the wager, reduced by the direct cost and increased by the sunk cost.

NBa = oCa = ( Pb x Rb ) - dCb + sC

NBa is net benefit of Wager “A”.
oCa is opportunity cost of Wager “A”.
Pb is the probability of success for Wager “B”.
Rb is the reward of Wager “B”.
dCb is direct cost of Wager “B”.
sC is sunk cost (which, by definition is the cost common to all options; thus sunk cost does not need a qualifier to differentiate the sunk cost between different wagers).

Thus, when we expand the game theorem formula for Wager A by expressing oCa in terms of the elements of Wager B, we end up with this formula:

Wa = | dCa + [ ( Pb x Rb ) - dCb + sC ] - sC < Pa x Ra |

Notice that we have both a +sC and a -sC.  This cancellation makes sense, because now you see that we are comparing the two wagers, and thus what we are after is the difference between them.  Since the sunk cost is, by definition, the cost that is common to all options, it makes sense that this element should be eliminated at this level.

We also see that we have +dCa and -dCb.  Again, we are comparing the two options, and this gives us the difference between the costs of the two wagers.

So this formula can be simplified and reordered to the following:

Wa = | ( dCa - dCb ) + ( Pb x Rb ) < Pa x Ra |

Now we have an expanded, comparative formula.  Let’s test to see if our previous conclusions have held true.

Wg = | ( dCg - dCn ) + ( Pn x Rn ) < Pg x Rg |

Remember that Rg is infinity, so if Pg does not equal zero, then the product of Pg and Rg will also be infinity.

Wg = | ( dCg - dCn ) + ( Pn x Rn ) < ∞ |

Next, remember that Rn is zero, so the product of Pn and Rn will also be zero.

Wg = | ( dCg - dCn ) < ∞ |

Observe that our previous observation remains true: The “good God” wager will be favorable so long as its direct cost is finite.  Now, we can add another boundary: If the direct cost of the “no God” wager is negative infinity, then the “good God” wager would also be unfavorable.  That is, unless the direct cost of the “good God” wager happened to also be negative infinity.

Now, let’s do the same for the “no God” wager:

Wn = | ( dCn - dCg ) + ( Pg x Rg ) < Pn x Rn |

And again, the product of Pg and Rg is infinity, and the product of Pn and Rn is zero.  So:

Wn = | ( dCn - dCg ) + ∞ < 0 |

Let’s move the infinity to the other side of the test to make things more clear:

Wn = | ( dCn - dCg ) < -∞ |

What does this mean?  It means that the direct cost of the “good God” wager has to be greater than the direct cost of the “no God” wager by a magnitude greater than infinity.

Now, there are in fact certain situations in which infinitudes could possibly be called greater or less than other infinitudes.  Since that is a hairy subject, let’s find out if we even need to go there.  If it turns out that we can’t make the case for the positive infinitude of the direct cost of the “good God” wager, or the negative infinitude of the direct cost of the “no God” wager, then this will be a moot point.

But let’s not forget this: as of now, the only way that the “no God” wager could be favorable is if the direct cost of the “good God” wager was greater than the direct cost of the “no God” wager by an infinitude which is, itself, greater than the potentially infinite reward of the “good God” wager.

Frankly, it already isn’t looking good for the “no God” wager.  But, let us finish the race, and finally get a grip on the qualitative definitions for the direct costs of each wager.

The cost which must be incurred to become eligible for the “good God” wager can be summed up in one word: repentance.  In essence, when we decide to rely on the existence and goodness of God in the intents and purposes of our lives, we will be forced to lay down, once and for all, any hope of “getting away” with anything.  We shall have to do what we know is right, even if it appears that we would escape human accountability.  We shall have to make amends for the wrongs we have done, and if we find we cannot secure the forgiveness of our victims, we shall have to prepare ourselves to beg mercy.  In other words, the direct cost of the “good God” wager is that we will have to try to be perfectly good, and if (when) we fail, we will need to be willing to be made perfectly good.  The essence of repentance is the desire to change.

Invariably, it will be suggested that the cost of the “good God” wager must include the obligation to follow arbitrary and onerous rules, or the risk that one would waste one’s life believing that which is actually false.  On the first point, we should remember how this problem has been defined: no God, good God, or evil god.  The obligation to follow arbitrary and onerous rules should be properly recognized as a cost of the evil god!  If we are being consistent, then we must recognize that the only rules imposed by a good God will be necessary and good rules.  In fact, such was implied above by the words, “we will have to try to be perfectly good.”

One of the examples of such arbitrary and onerous rules are the obligation to attend church.  This is confused on a couple of levels.  First, church attendance is associated specifically with the Christian religion.  Remember we are not yet concerned with Christianity.  Secondly, even with respect to Christianity, church attendance is not a requirement for the Christian’s salvation!  Fellowship is partly a fruit of devotion: we do it, not because we have to, but because we want to!  Of course, I recognize that such will not be the case for all the children who are being dragged to church by their Christian parents.  But those children are not submitting to their parents in order to be eligible for eternal rewards; they are simply maintaining their eligibility for the rewards which are in their parents’ power to give: a roof over their heads, three squares a day, and, if they’re lucky, some screen time.  Fellowship is partly a fruit of devotion, and it is also partly a path to even greater fruits.  We do it because it gives us practice in loving others in the way that we experience the love of God, with longsuffering kindness and grace.  To say that this does not come naturally is, perhaps, an understatement.  But where the sincere effort is made, the bonds that arise between believers is in many ways profitable even here and now, not to mention in the world to come.

The other cost that is sometimes suggested is the risk of being “taken in”, of believing something false.  Ron has pointed out that false beliefs can lead to undesirable behaviors.  My response is that such is a sunk cost, common to all possible wagers, assuming the problem has not already been solved.  Of course, if the existence and goodness of God were proven false, then we should obviously not cling knowingly to a false belief.  But again, we have taken as granted that such is not the case.  God may exist, and He may not.  He may be perfectly good, and He may not.  Whatever belief we rely on, we will be in some danger of being wrong.

Furthermore, while we have certainly observed theistic beliefs leading to “undesirable behaviors”, the same can be said about atheistic beliefs.  One or the other must be true, and yet these undesirable behaviors are a fact of reality.  Let us accept this principle and treat such cost as sunk cost.

Another question needs to be raised: have I denigrated the moral character of atheists by assuming that “being good” is only a cost of the “good God” wager?  This is a point well raised, but the answer is no, because the premise is false.  I recognize that most atheists recognize some form of “obligation to be good”, and that, as previously mentioned, they can and often do make this world a better place.  The “obligation to be good”, we can say, is equally born by all, and is therefore properly considered a sunk cost.  This obligation, by itself, is an oversimplification of what has been defined as the direct cost of the wager of the “good God”.  That cost specifically entailed an inescapable accountability to divine judgment.  We are forced to lay down the hope of getting away with anything.

The fact that most atheists choose not to avail themselves of this benefit supports a nominal valuation of this cost.  When they do forego this benefit, it is not as a cost required of them to be eligible for the rewards of atheism, but rather simply as the necessary cost of a clean conscience.  And the value of a clean conscience must be proportionate to the duration over which such a benefit is enjoyed; which, to the atheist, must be finite.  And since many atheists do, in fact, consider this benefit as justifying the cost of doing right when no one is watching, we must agree that such a cost is also finite.

Remember that the “good God” wager will be favorable so long as its cost is less than positive infinity.  And so it appears to be.  The “good God” wager is thus favorable!

So what are the direct costs of the “no God” wager?  Remember that the “no God” wager, in order to be favorable, requires a direct cost which is less than the direct cost of the “good God” wager by an infinitude which exceeds the infinitude the the “good God” wager’s reward.  We will not find such to be the case.  In reality, the “no God” wager simply has no direct cost, because there is nothing required of anyone to avail themselves of its nonexistent reward.  A direct cost of zero for the “no God” wager means that this wager is clearly not favorable.

In order to be exhaustive, lets define and solve the wager formulas given the qualitative definitions discussed above.

Wg = | ( dCg - dCn ) + ( Pn x Rn ) < Pg x Rg |

Wn = | ( dCn - dCg ) + ( Pg x Rg ) < Pn x Rn |

0 < Pg < 1 (using 1% in demonstration)
0 < Pn < 1 (using 99% in demonstration)
Rg = ∞
Rn = 0
0 < dCg < ∞ (using 1 in demonstration)
dCn = 0


Wg = | ( 1 - 0 ) + ( 99% x 0 ) < 1% x ∞ |
  = | 1 + 0 <  ∞ |
  = | 1 < ∞ |
  = TRUE

Therefore, the “good God” wager is very favorable!


Wn = | ( 0 - 1 ) + ( 1% x ∞ ) < ( 99% x 0 ) |
  = | -1 + ∞ < 0 |
  = | ∞ < 0 |

Therefore, the “no God” wager is very unfavorable.

Because the wager of the “good God” is clearly favorable, and the “no God” wager clearly isn’t; and further because the “evil God” wager is excludable from consideration: it follows that belief in the existence and goodness of Creator God should be relied upon in all the intents and purposes of our lives, until such a belief can be proven false beyond the shadow of any doubt.

(4) Evil, in the circumstantial sense, exists.

Let’s be real though.  Believing in the existence and goodness of God would be a whole lot easier, if it weren’t for all the suffering we experience and observe.  We can’t escape this basic fact of reality any more than causation or morality.  Suffering exists, and if God is good, there has to be a reason.  God’s full revelation must include that reason, but in the meantime, our standard of evidence, determined by game theoretical analysis, only requires the possibility of a reason.

We can find a plethora of examples of sufferings in our experience.  We rightly desire a world without catastrophic weather, a hardship which we would like to be able to eliminate or guard ourselves against, and we find ourselves impotent to do so.  We observe humiliating exploitation of vulnerable individuals and even entire populations.  We have to work our fingers to the bone just to survive.

Why, God?  Why?!  This cry should rise up in all of us.  If God is good, then He must be able to give a satisfactory answer.  And if we are, in the meantime, to believe that He is good, then we must be able to at least speculatively reconcile the circumstantial evil which we observe in the real world with His supposed goodness.

A speculative reconciliation is all we need, because Not-Pascal’s Wager showed that belief in the existence and goodness of God should be relied upon so long as the probability of its truth is greater than zero.  Thus, we do not need to know or be able to prove exactly what God’s justification really is; we only need to know that a justification might exist.  Thus, any speculation which would reconcile the goodness of God with the circumstantial evil which we observe in the real world will be sufficient to maintain such belief.

(5) The only way for God to be perfectly good, given the existence of circumstantial evil, is if this world is a necessary means towards a good end which justifies the present circumstantial evil.

At bottom, the problem of circumstantial evil is this: If God is good, then He must be both willing and able to create the best possible world.  And yet this world is clearly not the best possible world.  This problem is reconciled by the proposition that this world is a necessary means of creating that best possible world.

The first question which arises when we consider this proposed solution is whether the means ever can be truly justified by the ends.  I have considered this question exhaustively, and have concluded that ends clearly can, in certain circumstances, justify their means.  This analysis is similar to the wager analysis.  We are dealing again with costs and outcomes.  However, this question is a little bit different because God is the one incurring the cost, but we are the ones who bear it.  Now, this is not an automatic disqualifier: relationships like this exist in the real world.  We call them fiduciary relationships.  On one side of that relationship is the “steward”, who directs the affairs of the other, which we call the “beneficiary”.  Thus, in our situation, the first requirement which must be met is that God must have the authority to act in the capacity of a steward, incurring costs which we must bear.  And as our Creator, He certainly can claim such authority.

Next, given that the steward is legitimately authorized to incur cost to the beneficiary, we should expect the steward to follow the Golden Rule, incurring costs towards the good of the beneficiary as though both the costs and the benefits were his own.  Thus, as we do for ourselves, the steward should incur only the necessary costs required to accomplish the best possible outcome.

Since the important thing is that costs should be necessary, and the outcome should be the “best possible”, any actual judgment of whether the ends have justified the means must entail a comparative analysis.  Thus, no scheme could ever be justified or condemned, except by comparing it to another scheme.

Once one actually undertakes the actual form of that comparative analysis, it becomes clear that the position that ends can never justify the means is ridiculous.  Such a position essentially entails the proposition that only schemes which have no cost could ever be justified.  But since every scheme can be expressed in comparison to another scheme, then it turns out that cost is always relative.  In other words, there are no schemes which have no cost.  Thus, this position returns the conclusions that no scheme is ever justified, which is absurd.

Furthermore, we can appeal to real world examples of fiduciary relationships, in which the steward incurs real and even significant costs to their beneficiaries, and yet we do not even think to accuse them of evil.  One example is the parent, who subjects his child to education.  To the child, the process of education entails a devastating loss of freedom, hours of boredom almost every day, and the frequent frustrations of failure.  And while the child may, in the pains of his struggle, accuse his parents of evil for so subjecting him to such grievous suffering, we, who know better than the child the ends to which his education is directed, will stand by the parent.

Consider the comparative analysis to the alternative.  The parent may seek to spare the child of the suffering of education.  Is this a scheme which has no cost?  Of course not.  The end of that scheme is the avoidance of the cost of the other.  But its cost will be felt when the child is grown, when he is unemployable, ungrateful, unhygienic, and poor.

Another obvious example is the doctor-patient relationship.  In this relationship, the doctor often subjects his patient to significant suffering.  But we trust that these sufferings are necessary for the cure, which we rationally desire more greatly than the avoidance of the pains caused by the doctor himself.

Let us lay to rest the nonsensical claim that ends can never justify the means.  The ends certainly can justify the means.  The only question is whether they actually do, in this actual situation.

We aren’t going to jump to that conclusion.  There is still an objection which is often raised, and it has strong intuitive force.  The objection is this: since God’s goodness entails both omniscience and omnipotence, He ought to be able to actualize any desired state of affairs without the need for intermediate steps.

I say this objection has intuitive force, but that is only because omnipotence can easily be misunderstood.  It is easy to suppose that omnipotence includes the ability to do that which is logically contradictory.  As though God had power to suspend the laws of logic.  But this cannot be.  This position is in violation of our beginning axioms, including that the laws of logic are absolutely true.

Here is a general example illustrating how God’s omnipotence does not mean the ability to actualize any desired state of affairs without the need for intermediate steps.  In order to actualize the state of affairs in which “B has followed A”, God must first actualize A.  To propose that God could actualize the state of affairs in which “B has followed A”, without first actualizing A; is to say that B could both follow and not follow A.  This is a plain contradiction.

This has given rise to what is called “the euthyphro dilemma”.  Some have perceived that God must either have invented the laws of logic (in which case He should be able to suspend them), or else that they existed without Him having to create them (in which case He is subordinate to them).  In fact, this is a false dichotomy.  The third option is that the laws of logic are entailed by the very nature of God’s self-existent being.  He simply is logical, and the laws of logic are a description of Him.

So, given that God is logical, we understand that His omnipotence does not entail the ability to suspend the laws of logic.  Thus, if the best possible world must follow some necessary intermediate steps, then God will simply have to first actualize those intermediates.

And in fact, we can imagine how this world could be necessary towards the creation of the best possible world.  Our appreciation and enjoyment of any good thing is enhanced by a genuine experience of being deprived of that good thing.  That water tastes best when we are thirsty, is perhaps one more basic fact of reality that we have to deal with.  We can imagine that in the best possible world, we will have perfectly sound bodies.  And since we have experienced bodily corruption, we shall be able to appreciate those perfectly sound bodies all the more.  Our genuine experience of scarcity and toil will allow us to better appreciate and enjoy abundance and leisure.  A good environment is better appreciated by the one who has known a bad environment.  Those who have been subject to evil authorities will better appreciate and enjoy being subject to a perfectly good God.

One lifetime of relative tribulation can thus produce an eternity of bettered joy.  We should all recognize that such ends could justify such means, considered in and of themselves.  Nevertheless, the question remains whether the suffering entailed are the absolute minimum whereby the same end could be accomplished.  Anything above the absolute minimum is, by definition, unnecessary.

That is a question that I can’t answer.  Remember, no scheme can ever be justified or condemned except in comparison to another scheme.  Answering that question would require the perfect wisdom to establish exactly what is entailed by a “best possible world”, and how that should be accomplished with minimum cost.  The same level of wisdom would be required to disprove the goodness of God based on the existence of circumstantial evil.  Since it remains a matter of mystery to imperfectly wise human beings, it remains possible that the circumstantial evil of this world could be necessary towards the accomplishment of the best possible world.  It remains possible that God exists and is good, and thus we ought to continue to hold to that belief, until someone can prove it to be untrue.

Significantly, this point gives us another line of boundary in the beliefs we have established.  Not only have we decided to believe, for all intents and purposes, that God exists and is good, but, given the existence of circumstantial evil in this world, we also are led to believe that this world is a necessary means towards the best possible world.  We thus cannot accept (without proof) the proposition that this world is merely accidental, or the end to which God is creating.  We suffer.  If God is good, there must be a good reason.

(6) Evil, in the moral sense, exists - of particular concern is the moral evil which exists in me.

This is partly an extension of the problem of circumstantial evil.  Our subjection to morally evil society produces in us the same appreciation as our other sufferings: that is, we will appreciate and enjoy a morally righteous society all the more, having experienced a morally evil society.  For this reason, we can understand why God does not turn bullets into butterflies when they are fired at innocent victims.

More pressing is the problem of my own evil.  I have done wrong, and even when I did what was right, I sometimes did so despite a strong desire to do what is wrong.

Remember the efficiency analogy?  The only way a society can go on forever is if all its parts are perfectly righteous, else there will be a breakdown.  In fact, the infinitude of God’s potential reward depends not only on His perfect righteousness, but on my perfect righteousness.  Unless my moral nature is changed, I could never set foot in the best possible world, for the simple reason that the moment I set foot in it, it would no longer be the best possible world.

This point, once again, narrows the field of possibilities entailed by our default belief in the existence and goodness of God.  Previously, we saw that we can only accept propositions which include or at least allow for this world to be a means towards the goal of a best possible world.  And now we see that allowed propositions must include or at least allow for our transformation into morally perfect beings.

(7) Not only am I evil, I am also guilty.  My guilt has resulted in debts that I cannot satisfy.  Thus, I require mercy.

Evil and guilt are not interchangeable terms.  Evil is a description of a person’s nature.  A person who is evil is one who would commit evil in certain circumstances.  Thus, it is possible for a person to be evil, without being guilty, because guilt requires an actual violation of some moral law to which the sinner is accountable.  A person can be evil in that they would do evil, but innocent in that they have not done evil; or, in the case that the person has done evil, but was not then accountable to the moral law which was violated; or, in the case that the person has done evil, was accountable to the moral law which was violated, but has since secured the forgiveness of his victim(s).

I am guilty.  I am convicted under the law of my own conscience.  I have done wrong, knowing full well that I was doing wrong.  I have sinned against family, friends, enemies, and God, Himself.  I can neither entice nor compel anyone to forgive me for certain of my sins.

It has not yet been established that “I require mercy.”  As of this point, it remains possible that I may be able to bear the punishment which is due for my actions.  If I could bear my punishment, and then enter into God’s reward, then I could maintain hope towards that reward without requiring mercy.

So how exactly does a punishment relate to a sin?  In other words, how does a sin relate to the debt that it produces?  I’m afraid we will have to accept that sin debts arise by the same kind of process whereby any other debt arises: through a transaction.

You see, sin is a kind of transaction in which the sinner has forced the victim to confer his benefit, without having completed a negotiation determining how the victim should be repaid.  In a bona fide transaction, both parties have the option to walk away from negotiations, which is the source of their leverage.  But when sin has occurred, the victim has been forced to confer his benefit, and thus the sinner loses his ability to walk away from negotiations, which now must take place after the sin.  The victim has all of the leverage, and the sinner has none.  For this reason, even seemingly small sins can produce debts which we could never hope to repay.  Our victims have all the leverage, and our only hope of becoming eligible for an eternal reward is by seeking mercy, which, by definition, is undeserved forgiveness.

Bear in mind that this further narrows the field of possibilities entailed by our default belief.  Mercy cannot be considered an aspect of goodness, as are power, trustworthiness, justice and righteousness.  Because mercy is, by definition, undeserved, then it must also be true that a perfectly good God absolutely could withhold mercy.  We cannot simply hope for God to be good, because now we find a case in which God is perfectly good, and yet, in this case, we could have no hope in Him.  In fact, if we have any hope at all, God must be both perfectly good, and merciful.

(8) The legitimacy of sin debts means that a redemptive payment must be made.

I said God could be good, without being merciful.  It is also possible that God could be merciful, without being good.  This is the error against which we must be on guard at this point.

If the victim of a sin has a legitimate claim against the sinner, then it would be unjust for God to unilaterally forgive the sinner.  This point should have strong intuitive force.  Say we bring a grievance before a judge, and the judge recognizes that our grievance substantiates a legitimate claim.  But, then the judge reveals that the defendant is a close friend of his, and so the judge decides to let him off.  This, obviously, is unjust.

Thus it must be that God’s execution of mercy on our behalf must include a redemptive payment to our victims, whereby they would willingly forgive their claims against us.  God cannot force anyone to forgive, and of course He cannot force anyone into eternal fellowship with someone against whom they legitimately harbor ill will.  This is not the recipe for the best possible world.

The past few major points have all related to the problem of evil.  The problem of evil serves to narrow the field of possibilities, given the default belief that God exists and is good.  We have seen that this world must be a necessary means towards the best possible world.  We have seen that we must be transformed in our moral nature so that we will be perfectly righteous.  We also must be seeking out mercy, and now we see that any such mercy must be accomplished via a redemptive payment on our behalf.

(9) There is only "one name by which we must be saved", only one religion in all the world that offers salvation via a redemptive payment.

Finally we can approach the “advanced problem of religion”.  We can take a look at the 4,000 or so world religions, and because we have laid the groundwork of establishing default beliefs, we can quickly and easily reduce the field of viable religions.  And in fact, we can reduce the field to only one.

Religions which are atheistic in nature are the first to go.  Buddhism, for example, recognizes no God, and in fact teaches that oblivion ought to be our goal.  This is the end to which Buddhists seek to annihilate their worldly desires.  Only by wanting nothing, can they become nothing.

Polytheistic religions are also out, since a polytheistic “god” is not even the sort of god that we were seeking.  Polytheistic gods, like Zeus and Odin, are dependent beings whose cause is external to themselves.  They are not supposed to be “Creator Gods”.  They are just supposed to be extremely powerful beings.

This already reduces the field considerably.  But we can now further narrow the field by applying the filters which came about in our consideration of the problem of evil.  Any religion which fails to solve the problem of evil can be excluded from our consideration; not because this proves them to be false, but it proves them to be unfavorable wagers.  Their gods are evil gods.

This means that any religion in which salvation is earned by works can be excluded.  These religions are like the judge who let off the sinner, just because he was a buddy.  If a sinner is freed from the legitimate claims held against him by his victims, then the judge has committed injustice.  The victim must be satisfied.  Islam is an example of a religion which could be eliminated on these grounds.  Expiation in Islam is accomplished through observance of religious duties, and thus the claims of the victims of Islamic adherents will go unsatisfied.

This also excludes any religions that simply ignore the problem of evil.  Any religion that claims that good and evil are illusory ignore the obvious fact that the suffering which evil produces is certainly not illusory.  And if it is not recognized as such, then of course we could only rationally expect to find more of the same in eternity.  In fact, if good and evil are illusory, then this, right here and right now, is as good as it gets.

But Christianity gets everything right on these points.  Christianity teaches that the purpose of this world is to create a harvest to be brought in to the world to come (Matthew 13:30).  It teaches that we must be perfectly righteous to enter the Kingdom of Heaven (Matthew 5:48), and further that we can so be transformed (Philippians 1:6).  He will not leave the guilty unpunished (Nahum 1:3), yet He offers a redemption through His own body broken, and His own blood spilt (Matthew 26:26-28).  Our forgiveness is conditioned upon our own forgiveness of others (Matthew 6:12), which is how His redemptive payment can legitimately satisfy the debts to which we are subject (since all who are asked to forgive, also need forgiveness themselves).

(10) Thus, my only hope of infinite reward is that Christianity is true.

That is not to say that Christianity has been proven to be true.  It may be false.  But we have no other options, unless God has completely hidden Himself, and kept secret the actual redemption which He has prepared for us, and how we may secure it.  But this would be tantamount to a wager that you didn’t know how to take.  This is another case of a useless wager, just as we found the wager of the “evil god” to be useless.

Now, of course, we have to take very seriously the charges which can be laid against the God of the Bible.  He has subjected the world to all sorts of travails (Romans 8:20), but we have seen that these may be necessary costs required by the best possible world.  Those who say the God of the Bible is evil on account of such things should bear the burden of proof to show what the best possible world really is, and how that world could be accomplished without those costs.

God has subjected people to suffering in the position of a steward, but He has also subjected people to suffering in the position of a judge.  He flooded the whole world.  The people of Canaan were brought to punishment under the sword and subjection of His chosen people.  And in the end, He will forever hold under wrath, in hell, all who refused their own salvation and transformation.  These are weighty considerations, and of course we will want to know that God is just in administering such punishments (see Revelation 16:5-6).  But such matters are to be considered carefully.

(11) The Bible is long, and challenging to read, but it is the only persistent point of contact from God to me.

I have tried other methods of contact.  Before I came to call myself a Christian, I know that God had reached out to me, and responded to several of my prayers.  I cannot deny certain miracles, which He prepared for me.  I have sought out “holy men” and studied their works.

These are not illegitimate ways to seek out God, and we do well to reach out to Him through every available avenue.  But the written word is something special.  The written word has the ability, if it is preserved, to communicate ideas across time and space.  It is a way that God could make His message available and portable.  Taking this form, our belief can be a form of election, and our participation in its preservation and communication can warrant further reward.

(12) Because the God of the Bible is my only hope of infinite reward, I have decided to trust in Him.

There comes a point when you shift from relying on the truth of a certain proposition, and actually trusting in the person of God.  He has dealt with me, and I with Him.  It is no longer an idea that I contend for, but a person.  I have come to love God, because His dealings with me have made it more and more probable that He really is perfectly good.  For that reason, I am determined to err, where error is possible, in the direction of trust.

(13) Because I trust in God, I give the Bible the benefit of the doubt when it challenges my understanding.

Because of the Bible and those who preached it, I came to know the things which would one day become so clearly my only avenue of eternal hope.  Because the Bible had everything right that I would eventually come to realize needed to be right; and because no one else did, I studied the Bible with deference and persistence, even when it challenged my understanding.  In every instance that I could search out, things that appeared to be problems (apparent contradictions, seemingly unjust punishments, etc.), eventually became reconciled in further study.

If the Bible is wrong, then God has failed to preserve and deliver His message.  It is possible (at least, I haven’t been able to show that it would be impossible) that God could be perfectly good, even in the case that His message has become corrupt.  Nevertheless, the fidelity of His message is within His power to preserve, and not in my power to prove.  Thus, if His message contains errors, that will be His doing: my error will have arisen through trust in Him.

Yes, there were human beings involved in recording, transmitting, and curating the Bible.  But it is supposed that God was also involved, and there is no doubt that God can overcome the weaknesses of human beings.  Furthermore, if the involvement of humans is a point against the Bible, then of course the same point must be taken away from propositions mutually exclusive with the claims in the Bible, as they are just as much the products of human beings.  “Let God be true, and every man a liar.” (Romans 3:4)


We can know things.  God might be real, and in fact, His existence and goodness should be relied upon, unless and until either proposition is positively proven false.  Only the God of the Bible solves the problems of evil, and Jesus Christ is the only name given in all the world by which we must be saved.  His existence, goodness, and the trustworthiness of His message are all defensible, given the standard of evidence which properly applies to the problem of religion.

Once again, I would like to thank Ron Garret for his critical, yet respectful encouragement.  Special thanks also to Matt Harkey for his helpful notes and friendship.  I look forward to continuing this conversation with Ron, and can't wait to bring more creative and apologetic content to this platform!

If you have read all the way to the end of this post, just wow!  Thank you!  I promise that future content that I put out through this platform will be much more digestible.