Burden of Proof, Part 2:
What Constitutes Evidence for God?

by Mano Singham

If a religious person is asked for evidence of god's existence, the type of evidence presented usually consist of religious texts, events that are inexplicable according to scientific laws (i.e., miracles), or personal testimonies of direct experience of god. Actually, this can be reduced to just two categories (miracles and personal testimonies) since religious texts can be considered either as miraculously created (in the case of the Koran or those who believe in Biblical inerrancy) or as the testimonies of the writers of the texts, who in turn recorded their own or the testimonies of other people or report on miraculous events. If one wants to be a thoroughgoing reductionist, one might even reduce it to one category by arguing that reports of miracles are also essentially testimonies.

Just being a testimony does not mean that the evidence is invalid. 'Anecdotal evidence' often takes the form of testimony and can be the precursor to investigations that produce other kinds of evidence. Even in the hard sciences, personal testimony does play a role. After all, when a scientist discovers something and publishes a paper, that is kind of like a personal testimony since the very definition of a research publication is that it incorporates results nobody else has yet published. But in science those 'testimonies' are just the starting point for further investigation by others who try to recreate the conditions and see if the results are replicated. In some cases (neutrinos), they are and in others (N-rays) they are not. So in science, testimonies cease to be considered as such once independent researchers start reproducing results under fairly well controlled conditions.

But with religious testimonies, there is no such promise of such replicability. I recently had a discussion with a woman who described to me her experiences of god and described something she experienced while on a hilltop in California. I have no reason to doubt her story, but even she would have thought I was strange if I asked her exactly where the hilltop was and what she did there so that I could try and replicate her experience. Religious testimonies are believed to be intensely personal and unique and idiosyncratic, while in science, personal testimony is the precursor to shared, similar, consistently reproducible experiences, under similar conditions, by an ever-increasing number of people.

The other kind of experience (miracles) again typically consists of unique events that cannot be recreated at will. All attempts at finding either a consistent pattern of god's intervention in the world (such as the recent prayer study) or unambiguous violations of natural laws have singularly failed. All we really have are the stories in religious texts purporting to report on miraculous events long ago or the personal testimonies of people asserting a miraculous event in their lives.

How one defines a miracle is also difficult. It has to be more than just a highly improbable event. Suppose someone is seriously ill with cancer and the physicians have given up hope. Suppose that person's family and friends pray to god and the patient suffers a remarkable remission in the disease. Is that a miracle? Believers would say yes, but unbelievers would say not necessarily, asserting that the body has all kinds of mechanisms for fighting disease that we do not know of. So what would constitute an event that everyone would consider a miracle?

Again, it seems to me that it would have to have the quality of replicability to satisfy everyone. If for a certain kind of terminal disease, a certain kind of prayer done under certain conditions invariably produced a cure where medicine could not, then that would constitute a good case for a miracle, because that would be hard to debunk, at least initially. As philosopher David Hume said: "No testimony is sufficient to establish a miracle unless the testimony be of such a kind that its falsehood would be more miraculous than the fact which it endeavors to establish..." (On Miracles)

But even this is problematical, especially for believers who usually do not believe in a god who acts so mechanically and can be summoned at will. Such predictable behavior is more symptomatic of the workings of as-yet-unknown natural laws than of god. The whole allure of belief in god is that god can act in unpredictable ways, to cause the dead to come back to life and the Earth to stop spinning.

So both kinds of evidence (miracles and testimonies) used to support belief in a god are inadequate for what science requires as evidentiary support.

The divide between atheists and religious believers ultimately comes down to whether an individual feels that all beliefs should meet the same standards that we accept for good science or whether we have one set of standards for science or law, and another for religious beliefs. There is nothing that compels anyone to choose either way.

I personally could not justify to myself why I should use different standards. Doing so seemed to me to indicate that I was deciding to believe in god first and then deciding on how to rationalize my belief later. Once I decided to use the yardstick of science uniformly across all areas of knowledge and see where that leads, I found myself agreeing with Laplace that I do not need the god hypothesis.

Top of page