Should a Pragmatist Believe in God? If So, When?

Following from my previous post on the Biological Basis of Attraction that was submitted as part of a project for my Introductory Psychology class, I would like to share another piece of academic work here today.

I wrote this paper as part of the requirements for my Philosophy of Religion class. It will be useful to note that I wrote it from the point of view from a pragmatist. Arguably, the strongest pragmatic argument for a belief in God is that of Pascal’s Wager. Being a non-believer, I attempt to evaluate the strengths and weaknesses of this argument, helping me better understand my own beliefs (or lack thereof).

It is a possibly controversial piece. Here, I attempt to raise logical and substantiated points to Blaise Pascal’s pragmatic argument for belief in God (now popularly known as Pascal’s Wager). While reading it, it will be best to read it with an open mind, and definitely, always read the footnotes when referred to.

Should a Pragmatist Believe in God? If So, When?

In this essay, I will prove that for pragmatists convinced by Pascal’s Wager, they should not immediately believe in God, instead they should be agnostic until the day he or she is to die. In the Pensées, Pascal put forth a very convincing pragmatic argument, known as Pascal’s Wager, for belief in God[1]. Referring to Table 1 below, since we do not know God’s existence conclusively, choosing to believe in God even if God’s existence is unlikely, is the choice that gives me the highest expected net-benefit. For the sake of brevity, Pascal’s Wager will not be discussed in depth as there exists much literature on it. Being a pragmatic argument for belief in God, Pascal seems committed to the assumption that even if our belief in God is motivated purely by self-interest (although this will be uncomfortable for many theists), God will still reward us with the ‘promised’ infinite gain of eternity in heaven.

Let us call this assumption, A.

(A)  Belief in God will lead to the ‘promised’ infinite gain despite the self-interested rationale for believing.

Table 1

The possible contentious aspect of A will not be discussed here as this essay is targeted at pragmatists. It suffices to say that pragmatists are people who would likely not be uncomfortable with the idea of making decisions based on self-interest.

Before further discussion, it is timely to define pragmatism. I define pragmatism to be the guiding principle in which pragmatists consistently apply whenever there are competing alternatives so that the action that maximises their expected net-benefits will always be chosen.

Pascal’s argument is very convincing; indeed, a pragmatist should believe in God no matter how unlikely God’s existence is. So long there exists at least a minute positive probability for God’s existence, a pragmatist should always wager on God existing, given the infinite gain that can result if God exists and the finite loss if God does not exist.

Which God?

However, after being convinced by Pascal’s Wager, nontheistic pragmatists such as myself now face a conundrum.

Suppose there is an atheistic pragmatist named Bob. Bob has lived all his life without believing in God’s existence and he finally dawns upon Pascal’s Wager on the final day of his life. He is absolutely convinced and decides to believe in God. But wait – which God shall he believe in? Clearly, he does not have enough time to gather more information on the various religions that exist in the world. Rationally, he decides that his best bet is the religion that Pascal believed in – Christianity. After all, it was Pascal who convinced him.

The conundrum is now clear. If the God of Christianity existed, Bob would have gained an eternity in heaven. But if the ‘true’ God is not that of Christianity, he would likely have suffered an eternity in hell.

This conundrum is arguably the strongest criticism of Pascal’s Wager. Let us call it the ‘Which God Objection’ (WGO). WGO rightly pointed out that Pascal’s Wager can be applied to argue for any other religion[2], and not just Christianity that Pascal professed. To see this for yourself, replace ‘God’ with ‘Allah’ in Table 1. Now, a pragmatist who does not reflect on WGO would have believed in Allah. But he shouldn’t, at least not yet.

Therefore, Pascal’s Wager represents a set of options that is too narrow and does not reflect the multitude of options that are actually available in our lives (for instance, Singapore has 10 official religions[3] already, and there are many more in other countries). Stated another way, Pascal’s Wager considered only whether God exists or does not exists, but it gives us no clue as to which God to believe in. There exist many other religions and many (or most) of them are exclusive, believing their God to be the only true God. If only one is true, then this necessarily means that the others are false. And we cannot afford to believe in a ‘false’ God! For brevity’s sake, suppose there exists only two religions – Christianity and Islam. And a belief in the wrong God would mean an infinite loss (see Table 2 below).

Table 2

There is no logical reason to think that Christianity (indeed, any particular religion) is definitely correct and every other religion have absolutely zero chance of being true. It is certainly incumbent on anyone who might contest this to explain why only Christianity (or any particular religion) is correct and not others. Granted, we might be able to eliminate some religions, such as Pastafarianism (created to oppose the teaching of intelligent design in American schools[4]) from among the options that are actually available to us. However, it is very difficult (or impossible) to prove false those religions that are asserted to be true and especially difficult if they are asserted to be religions that received revelation from God.

Further, the arguments for the existence of God in the Philosophy of Religion advanced by philosophers have near equal strength even when used for different religions (particularly, monotheistic faiths) so they are of little use in helping us choose one religion over another.

Given that we do not know any information that would make one religion more likely true than others. It is therefore logical to invoke the Principle of Indifference to help us understand WGO:

Table 3

Taking the example of Singapore and considering only official religions, each of them has 0.1 probability of being correct.

It is clear that there is a high chance that we may very well choose the wrong God and the possible infinite gains (eternity in heaven) and risks involved (eternity in hell) necessitates us to be more prudent in our choice of belief.


Suppose there is an atheistic pragmatist named Stan. Stan read Pascal’s Wager on the day he turned 21 years old. Despite his mere two decades of life, Stan has been a very dutiful pragmatist. For every decision he had to make in the past two decades, he consistently chose the most pragmatic action. Unlike Bob, Stan has many more years ahead of him.

Being a pragmatist, Stan is convinced by Pascal’s argument. A belief in God gives him an expected net-benefit that outweighs the expected net-benefit of a disbelief in God. But he also realises that Pascal’s Wager does not require him to make an immediate ‘leap of faith’. After all, the infinite gain (or loss) can only be realised after his death.

Therefore, Stan resolves to believe in God before his death. For now, what should Stan do? Let us call this problem B.

To resolve problem B, Stan has, broadly, 3 options[5].

1.     Become a believer.

2.     Remain an atheist.

3.     Become an agnostic.

Let us analyse his 3 options in detail.

Scenario 1, Stan chooses to become a believer immediately. He immediately dawns upon the WGO. He is unable to choose one religion to believe in as each religion is equally likely to be true (recall the discussion on the appropriate application of the Principle of Indifference). To simplify our discussion, suppose Stan has 10 religions to choose from, similar to a Singaporean. Given that Stan wants to be a believer immediately, he decides that since all 10 options provide equal expected net-benefits, he should randomly choose between the 10 options. Perhaps this is done by drawing lots. Suppose he drew a lot that determined that he profess Christianity (just like Pascal and Bob). As a Christian, Stan experiences some extraordinary events, perhaps he marries the perfect woman, has the perfect career and the perfect family. He might also go through some ‘religious experiences’, such as witnessing the successful treatment of a scientifically incurable disease[6]. Let us term these set of extraordinary events C. He likely attributes C not to his own hard work or good luck, but to his Christian God, further confirming his belief that Christianity is the one true religion. This can be attributed to the confirmation bias, where people subconsciously select and interpret information to affirm their already-held beliefs[7]. At his deathbed, he recalls Pascal’s Wager but he now has confidence that he has chosen the correct religion and placed his trust in the ‘true’ God.

Let us rewind back to his drawing of the lots. There was a 10% probability that he chose to believe in Christianity and an equal 10% probability that he believed in one of the other 9 religions. Even after the many years that has passed, the probability of each religion being true should have remained equal, at 10%. Of course, there is also the probability that there is no God and therefore no true religion at all. But we shall ignore this possibility since we are discussing the next steps for pragmatists convinced by Pascal’s Wager.

It is clear that Stan’s decision to immediately become a believer has caused him to attribute C to his randomly chosen Christian God. In fact, regardless of the religion his random draw of the lots suggested, given that he experiences C and his inevitable (since he is only human) confirmation bias, he would have attributed C to his randomly chosen religion. To test this, suppose Stan drew the lots and believed in Hinduism, and he experiences C. He would likely have attributed C to his Hindu God and not any other God.

Any logical person would agree that the risks involved in choosing the wrong God to believe necessitates Stan to be more prudent. No one should randomly choose a religion to believe in.

Unlike Bob, as Stan has many more years to live, he could gather more information on the different religions and change the probabilities of each religion being correct (at least, for himself).

It seems clear that Stan should not choose to become a believer immediately.

Scenario 2, Stan chooses to remain an atheist. He experiences C. Due to confirmation bias, he attributes C purely to his own hard work and good luck, further affirming his atheistic belief. On the last day of his life, he decides it is time to make a pragmatic switch to believe in God.

Alas, he faces the WGO.

Pragmatically, choosing one religion (no matter how arbitrarily) is better than not choosing at all. Suppose Stan is therefore able to choose one of the 10 religions to believe in, perhaps also by drawing lots. With his atheistic beliefs affirmed throughout his life, it is likely that Stan will find great difficulty in convincing himself that a God exists. Pascal would likely have agreed with this difficulty as he mentioned that after being convinced by Pascal’s Wager, to believe in God, people should behave as if they believe, “taking holy water, having masses said,” and “this will naturally cause you to believe and blunt your cleverness[8]”. It takes time to convince oneself to truly believe in a God. Stan, being unable to truly believe in a God, even if that God exists, this God will likely not reward him with the infinite gain that he desires.

Some might counter that if Stan was an atheist, he might not have experienced C. This is unreasonable, seeing that amongst successful people in all fields, there are both atheists and theists.

It seems that Stan should also not choose to remain an atheist.

Scenario 3, Stan decides to remain agnostic and remain open to information that could help him decide which God to believe in. Agnosticism is defined here to be the belief that does not affirm God’s existence or inexistence and does not affirm one religion to be more correct than others, until evidence suggests so. Given the agnostic imperative[9]:

for all persons S and propositions P, if S believes that P is just as likely as not-P, then it is impermissible for S to believe either P or not-P.

An agnostic will therefore suspend belief whenever the evidence is insufficient – making a decision on what to believe only when there is evidence that suggests one option to be more likely than others. Suppose he experiences C. As an agnostic, he might profess that for C, there is a probability of divine help, but he knows definitively that his hard work and good luck contributed to C (after all, he reasonably knows that he exists and that he has free will). Confirmation bias does not affirm his belief in a particular God (as in Scenario 1), nor does it affirm his belief in atheism (as in Scenario 2). However, Stan can aptly decide what is the best explanation for the particular events that happens to him. Suppose his perfect wife unfortunately suffers from a terminal disease – late stage cancer. Whether he is advised by his close friends to pray to a Christian God, Allah, or seek more expert oncologists, and his wife recovers fully, his beliefs will be altered accordingly. To illustrate this, suppose Stan is advised by his best friend, Gaston, a Christian, to join a church service, where Gaston’s pastor will pray for Stan’s wife. A desperate Stan obliges and Stan’s wife recovers several months later. It is reasonable for Stan to mentally adjust and increase the probability that Christianity is correct.

As expected, at his deathbed, Stan faces the WGO. This problem is likely to be more easily solved now, as he had the opportunity of his entire life to seek information that support one particular belief and undermine others. Unlike in Scenario 2, Stan now would likely face minimal or zero difficulty in believing in God, as he has experienced events that suggests there to be a God.

Therefore, after Stan is persuaded by Pascal’s Wager, he should make the optimal decision to be agnostic, withholding belief until evidence suggests one belief to be more likely correct than others.

Possible Counterarguments

Some theists might counter here that Stan should not wait until the final day of his life to decide which God to believe in. Instead, he should decide immediately once he receives information that can be interpreted as support for a particular God. Recall the example of Stan agreeing to join Gaston for a church service, where Gaston’s pastor prays for Stan’ wife. After his wife’s recovery, the argument suggests that Stan should believe in Christianity immediately.

I disagree. This is not the optimal decision.

The problem with this is that we cannot confidently rule out the possibility of his wife’s recovery from cancer after Stan’s visit to the church as being due to coincidence. The problem with Stan immediately believing in Christianity as a result of just one event is that he could, from now on, be unable to objectively evaluate other events as evidence to support other religions, due to the confirmation bias. Therefore, Stan should remain agnostic until more evidence suggests Christianity to be right.

Also, some atheists might counter that despite using our entire life to seek evidence for and against the various religions, one might still be unable to choose the correct religion and therefore, we should not believe in any God at all. There is always this possibility. However, it is reasonable and justifiable for us to give our very best (indeed, our very best is our entire lives) to search for evidence that can help us best achieve the infinite gain and best avoid the infinite loss. The infinite magnitude of the gains and risks involved necessitates that we give all we have, our entire lives, to help us make the best-possible decision.

It is clear, that the optimal decision for Stan, and all of us in Stan’s shoes is to remain agnostic until the final day of our lives, where we would have maximised the amount of information we could have possibly gathered to help us best choose the correct religion among all the possible religions.


[1] I refer to a copy of Pascal’s Wager translated by John Warrington, accessible at

[2] Philosopher J.L. Mackie stated that “the church within which alone salvation is to be found is not necessarily the Church of Rome, but perhaps that of the Anabaptists or the Mormons or the Muslim Sunnis or the worshippers of Kali or of Odin” in his book, The Miracle of Theism, published by Oxford University Press, in 1982.

[3] Inter-Religious Organisation Singapore, accessible at

[4] James Langton, “In the beginning there was the Flying Spaghetti Monster”,

[5] The 3 options considered are arguably the most common. Therefore, other types of positions that someone can take with respect to religion, are not considered here.

[6] An example of such a miracle is described by medical historian and hematologist, Dr Jacalyn Duffin, accessible at

[7] Raymond S. Nickerson, “Confirmation Bias: A Ubiquitous Phenomenon in Many Guises”, Review of General Psychology, 2, no. 2 (1998): 175-220

[8] Pascal’s Wager

[9] Defined by Jeff Jordan at the Stanford Encyclopedia of Philosophy, accessible at


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s