Is faith rational?

I was recently asked my opinion on an argument for the rationality of faith. Here is my response.

You quoted someone who wrote:

So we find that we are forced in almost every deductive argument to accept something in the premises which is either beyond proof (and simply accepted), or which relies on something which can only be offered as a statement anchored in finite (limited) observation. Which in turn, is a good argument for the foundational necessities of both faith and common sense.

As I understand it, the argument goes something like this.

  1. All arguments are either deductive or inductive.
  2. Deductive arguments depend upon at least some assumptions that cannot be proven.
  3. Inductive arguments are fallible and depend upon our limited observations.
  4. So, all human reasoning must assume something which cannot be proven.
  5. Reasoning which depends upon something which cannot be proven involves faith.
  6. Therefore, all human reasoning involves faith.

1-6 is meant to defend the rationality of faith by showing that reason, itself, requires faith. Thus, if faith is generally rational, there cannot be something irrational about Christian faith, or so the argument goes.

I think there are a few problems with 1-6.

First, it is unclear that the notion of faith appealed to is the same as what the Christian means by ‘faith’. For the Christian, the word ‘faith’ means trust. In the New Testament, the word that is translated into English as ‘faith’ is ‘pistis’, which is the Greek word for trust. In Latin — the language employed by Catholic intellectuals — the word for faith is ‘fides’, which is the root of the word ‘fidelity’ (this is why “semper fi” means “always faithful”; this is why Pope John Paul II wrote an encyclical entitled Fides et Ratio, or Faith and Reason).

What is it that the Christian is supposed to trust? Thomas Aquinas provided a decent explanation of what the Christian might mean by ‘faith’. For Thomas, the Christian trusts particular propositions as they are revealed by God. God is a perfect being and information revealed by God cannot be incorrect; God would not lie to us. Notice that this presumes one must first know that God exists in order to have faith and that the knowledge of God’s existence is not arrived at through faith. In order to know that God exists, Thomas says that we must first prove God’s existence as a preamble to the faith (the preambulae fidei); Thomas provides 5 arguments for God’s existence. The lay person, who does not have the ability to prove God’s existence, may trust that others (who are smarter than oneself) can prove God’s existence and in that way have faith that God exists. Either way, one must first know that God exists in order to trust (or have faith in) God’s revelations.

Some Christians interpret Romans 1 as stating that God’s existence is obvious to all humans from the appearance of nature. I doubt that God’s existence really is obvious to all humans — the geographic distribution of beliefs about God seems to be evidence to the contrary — but, even if God’s existence were obvious to all humans, the belief in God’s existence would not be held without evidence. Instead, there would be extremely strong evidence — undeniable evidence! — from the appearance of nature.

If, by ‘faith’, one means trust in God’s revelation, then clearly not all reasoning requires faith. Doubting Thomas (not to be confused with Thomas Aquinas) did not trust in all of God’s revelations because he doubted some of the things that Jesus said; yet Thomas was still able to reason.

Perhaps the word ‘faith’ is meant to refer to a more general kind of trust. After all, as I said, the lay person can have faith that intellectuals can prove God’s existence even if the lay person cannot. But trust does not necessarily involve believing propositions for which there is no evidence. Instead, faith involves trusting those who we have reason to trust (intellectuals have earned our trust by demonstrating themselves to be knowledgeable and that they are not the sort of people who would lie to us). This more general kind of faith — what we might call reasonable trust — cannot be what is referred to in 1-6 as faith, since 1-6 refers to faith as a kind of belief without evidence. One may wonder if ‘faith’, as used in 1-6, is a kind of unreasonable trust.

Perhaps 1-6 is meant to show us that it is not always unreasonable to believe without evidence and that, from this, we are supposed to accept that it is not unreasonable to believe Christianity to be true without evidence. Here, two things can be said in response.

First, suppose that it really is the case that some things should be believed without evidence. Even though it might be acceptable to believe some things without evidence, it does not follow that it is acceptable to believe all things without evidence. For example, we would not say that we should believe a murder suspect to be guilty without evidence nor should we accept a Nazi’s claim that the superiority of the Aryan race should be believed without evidence. Thus, it is left to the proponent of arguments like 1-6 to explain why Christianity is the sort of thing that should be accepted without evidence.

Second, I’m not actually convinced that there are any propositions that should be accepted without evidence. I need to be careful to draw out a subtle distinction here, so bear with me. Consider any belief b. You can ask whether we should accept b. If we should never accept any belief without evidence, we should only accept b if we have some evidence E. But in order to have E, we have to have the belief B2 that E is evidence for b. And in order to have B2, we need evidence E2 that B2 is true. But in order to have E2, we need to have the belief B3 and so on. This proceeds into an infinite regress, and, thus, the notion that we require evidence for all of our beliefs is destroyed. The problem is that some beliefs might be evidence for themselves (they are self-evident). If so, the regress stops at some point. But if some beliefs are self-evident, then it is not necessarily the case that we hold any beliefs without evidence. Perhaps all of our beliefs are either backed up by some external evidence or are evidence for themselves. I don’t know if that is the case, but it seems like a reasonable possibility. Of course, the dilemma for the Christian would become whether Christianity is backed up by some external evidence or is self-evident. If the former, then defending the notion that we should have some beliefs without external evidence is irrelevant to the Christian qua Christianity and the burden of proof is on the Christian to demonstrate that there is evidence for Christianity. If the latter, it becomes incumbent on the Christian to show that Christianity is self-evident.


Chris Hazel has brought another situation to my attention in which we might beliefs without evidence. There are some beliefs which we assume because they are useful to us and because they are indispensable for our successfully navigating the world. Beliefs of this sort might include my belief that other people have minds and my belief in the existence of the external world.

While I am sympathetic to Chris’s position, I’m not totally convinced that we cannot justify these beliefs. For example, it may be that beliefs of this sort have pragmatic justification. Perhaps pragmatic justification and evidence are distinct, but it would be false to say that beliefs of this sort are justificationless. I am not completely convinced that we cannot have evidence of other minds or of the external world. Arguments exist in the literature that our belief in other minds or in the external world may be justified as inferences to the best explanation (for example, John Mackie argues for the latter in his response to Berkeley’s idealism in The Miracle of Theism).

Leave a Reply