How can we test alleged healing miracles?

Healings? What healings?

There are millions of accounts of healings. Most are not well documented, but a few are – I have summarised a few of these in Healing miracles and God. But how can we assess these documented stories objectively, to determine whether it is possible they could be true?

Approaches to assessing miracles

Extreme gullibility

Some people seem to uncritically accept any miracle story they hear. But while this may show admirable ‘faith’, we all know that many stories are urban myths, there are charlatans and gullible people out there, and coincidences occur.

Extreme scepticism

At the opposite extreme are the sceptics who say miracles are contrary to science, and so couldn’t possibly occur. This view is often based on the writings of 18th century Scottish philosopher David Hume, who argued that miracles were contrary to the known laws of nature, which are known from experience to be reliable, and therefore there could never be sufficient evidence to believe in them.

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 endeavours to establish….

Hume’s view was influential for two centuries, but is now criticised for a number of reasons. It is claimed that the argument is circular (i.e. it assumes what it wants to prove) and that Hume ignored how the probability a miracle occurred might increase with many miracle reports by many reliable witnesses. (Some of those critical of Hume use Bayesian probability to show that many independent reports can lead to a justified conclusion that a miracle probably occurred – see references below.)

So extreme scepticism often begs the question. We only know miracles are contrary to experience if we assume from the start that they have never happened. Sceptics claim their view is based on evidence, but their argument has the bizarre result that it makes evidence irrelevant, because it will never be believed.

Scientific approaches

Many attempts have been made to evaluate miracle claims scientifically. For example, rigorous and statistically valid studies have been undertaken of the effects of prayer on healing (see Intercessory prayer and healing), with a wide range of results. Most often, the studies find a small improvement in the group prayed for, but sometimes prayer makes no difference and at least once it seemed to have a small negative effect.

The problem with these studies is that they are not testing apparent miracles, but small improvements, which is a different thing. And this points out a major difficulty with the strict scientific approach. A miracle is (by most definitions) an unusual event, so it is difficult to set up an experimental procedure with a control group in advance.

It is a question of aims and methodology appropriate to those aims. If the aim is to test how often miracles occur, and if enough time and money is available, a long study with appropriate statistical controls is necessary. But if the question is simply whether sometimes unexplained recoveries occur after prayer, a simpler design is both possible and appropriate.

Professor Candy Gunther Brown has tried to develop a reasonable approach to testing prayer as it is used by christians. She suggests four different types of studies may usefully combine to give some answers:

1. Medical records

Medical records can often be used to establish that someone was sick or injured and has recovered, and eliminate clear frauds. They don’t settle the question of the cause of the recovery, but they are a necessary first step.

2. Surveys

Surveys can establish how sufferers perceive healing prayer – whether they tried prayer, or medicine, or both; whether their condition improved and to what they attributed this improvement.

3. Clinical trials

While it is difficult, practically, to achieve the level of control we would like to see, clinical trials can give an in dication of the level of improvements (or otherwise) after prayer.

4. Follow up

Observations and interviews over several years can assess whether healing was temporary or lasting.

A reasonable person

A lawyer friend tells me that some matters of law (e.g. defamation) are decided on the basis of what a reasonable person would conclude. This may be less rigorous than scientific testing, but should still be as rigorous as possible, and thus may be suitable for this type of test. So how would a reasonable person assess apparent miracles?

A promising approach is to set up criteria for what could plausibly be a miraculous healing. A useful set of criteria, based on suggestions by magician and sceptic James Randi, and criteria used by the Catholic church to assess miracles, is:

1. The account of the story comes from a reputable source which provides names, time and place, and there is no reason to believe the story is a fraud, or that anyone had anything to gain by inventing it.
2. The disease had little possibility of natural recovery.
3. The recovery must have been complete, or at least very significant, and not what might be expected from any treatment being received.
4. There must have been prayer for healing not long before the healing occurred.
5. There must be good independent medical opinion (backed up by documentation) that the disease was present before the prayer and not present afterwards.

Obviously none of these criteria can be assessed absolutely and certainly, but if an apparent healing fitted generally within these criteria, a reasonable person would likely conclude that something unusual had occurred, possibly a miracle. If a number of such cases meet these criteria, the cumulative strength of the case increases.

Assessing whether the unusual event could plausibly be believed to be a miracle requires alternative explanations to be evaluated. The only three seem to be (i) it didn’t really happen, (ii) an extremely unusual natural cause, or (iii) miraculous healing. The better the above criteria are fulfilled, the more likely it is that it was a miracle. However a person who was resolutely committed to the non-existence of God could always choose one of the other options, although whether this would be the response of a reasonable person is a matter of opinion.

In his book God and other spirits, philosopher Phillip Wiebe suggests an approach similar to this, which he terms ‘abduction’, for the analysis of events purporting to be caused by non-material or supernatural beings. Abduction takes otherwise unexplained phenomena and posits an explanatory hypothesis which can then be evaluated be obtaining data relevant to that hypothesis. Wiebe argues that this approach could lead to empirical testing of supernatural phenomena (which would include miraculous healings).

Assessing actual miracle claims

Sceptics often focus on miracle claims that should not be believed – due to lack of documentation, doubts about the integrity of the ‘healer’ or the healing process – whether they be ‘quack’ medicines or faith healers. But use of the above criteria should eliminate all these claims, and leave only plausible miracle claims.

I have considered a number of miracle claims using the principles outlined above – a summary of the cases is in Healing miracles and God. I conclude that there is strong evidence that something very unusual has happened in many of these cases, and that miraculous healings have plausibly occurred.

That is about as far as any studies can take us. Whether the “something unusual” is a miracle or a spontaneous remission is a matter of personal judgment. But I feel anyone looking for evidence of God should certainly examine healing case studies.