The principle of methodological naturalism is a proposed principle that scientists should always exclude any supernatural explanation for anything, and consider only natural explanations. It is claimed by some that this is a “rule of science.” Evoking this principle is very convenient for those who wish to rule out “in one fell swoop” all evidence for any paranormal belief or supernatural belief. The person who evokes the principle of methodological naturalism basically attempts to evoke some supposed “rule of conduct” which prohibits any consideration of the paranormal or supernatural by a scientist.
Below are several reasons why such a principle is very dubious and should not be evoked.
Reason #1: Since we do not know the boundaries of nature, we have no business excluding certain possibilities on the grounds that they are outside the boundaries of nature.
Quite a few of the main things that we might today call supernatural are things that might one day be regarded as natural once we expand our knowledge of nature. Consider an idea such as the soul. We might now consider that supernatural, but once we expand our knowledge of nature we might discover some account according to which the soul is natural. We could say the some thing about apparitions, or alleged human supernatural powers. An expanded understanding of nature might include an account of such things. The diagram below illustrates the idea.
So not knowing the boundaries of nature, we have no business saying, “It is forbidden to believe in X, because that is outside of the boundaries of nature.”
Reason #2: A principle of methodological naturalism conflicts with another older and more widely advanced principle of scientific conduct, the principle of self-correction.
An advocate of the principle of methodological naturalism might try and make it seem simple: well, we scientists have this principle we're supposed to follow, and so we have to follow it; simple as that. But it's not so simple, because scientists have also advanced other principles, and it turns out some of them conflict with the principle of methodological naturalism. One such principle is sometimes called the principle of self-correction, that science must always be willing to revise its assumptions when the evidence warrants. But what if the evidence should warrant the assumption that some supernatural cause was at play? Then scientists would have to make a choice: either follow the principle of self-correction demanding that assumptions be revised when the evidence warrants it, and accept that evidence; or follow a principle of methodological naturalism, and reject it.
Given such a conflict, we might ask: which principle is more in sync with the spirit and essence of science? It is, of course, the principle of self-correction, which has actually been evoked by scientists far more often than the principle of methodological naturalism.
Reason #3: Appeals to the principle of methodological naturalism involve question-begging, the fallacy of petitio principii.
Appealing to a principle of methodological naturalism involves the fallacy of begging the question, of assuming what is to be proven. The reason is as follows: obviously it makes no sense to follow a principle of methodological naturalism unless there are no supernatural effects– for if there were supernatural effects, why would we not want science to learn about them? The principle only seems to make sense if you first assume there are no supernatural effects, and then conclude that science should follow the principle of never admitting the possibility of such effects. But whether or not there are supernatural effects is typically the issue being debated at the time such a principle is evoked, so you can't just start out with the assumption that no such effects exist, because that's the matter being debated. Here we have a fallacy which philosophers call petitio principii, and which is more informally known as begging the question. It's the fallacy of starting out by assuming what you are trying to prove.
Reason #4: The principle of methodological naturalism is supported by appeals to “normative behavior,” but it is very dubious that such a norm actually exists among a great majority of scientists; and even if it did exist, majority support is not a good justification for a principle.
The person arguing for the principle of methodological naturalism typically attempts to argue that it is some kind of rule of conduct for scientists. But it is actually very dubious that such a rule is rigidly followed or has been endorsed by a majority of scientists. Because of all of the vast number of specialties in science, most scientists never become involved with questions of whether we should admit some supernatural effect as a possibility. We certainly do not have a case in which a majority of scientists have signed on to some principle that supernatural effects must never be considered as a possibility. No written declaration of principles endorsed by a majority of scientists supports such a principle of methodological naturalism.
Even if it were a case that most scientists had publicly endorsed such a principle, the fact that the majority held such a principle would not justify it. Opinions held by the majority are very often wrong, and principles followed by the majority are very often wrong.
In the case of the scientific community, we have a cultural subgroup or subculture that may be subject to the sociological effects that influence most subcultures: group taboos, “herd behavior,” groupthink, and so forth. So the fact that some principle may be followed by a majority of scientists does not establish the wisdom of such a principle, and may merely tell us about customs or cultural taboos of a sort that may be mainly cultural or sociological in nature.
Any attempt to establish a principle of methodological naturalism by claiming that most scientists support it (or most scientists have supported it) is reasoning that commits fallacies such as the “appeal to authority” fallacy or the “appeal to tradition” fallacy.
Reason #5: In its most rigid form, the principle of methodological naturalism is a “think like a psychotic” principle with the same general form as principles followed by those clinging to delusions.
In its most rigid form the principle of methodological naturalism holds that whenever we get evidence seeming to suggest the supernatural, we should interpret it as evidence of the natural. Now such a principle has a general form or schema or outline, which is like this:
Whenever you receive evidence that seems to show X, interpret it as evidence that shows non-X or the opposite of X.
There is another group that follows this type of thinking. It is psychotics. When a psychotic person has a delusion, he may interpret all evidence that seems to show his delusion is false as being evidence that is consistent with his delusion or evidence in support of his delusion. For example, a psychotic may believe that he is the king of the world, and may interpret all evidence seeming to contradict that belief as evidence that is consistent with that belief or a confirmation of it.
Not wishing to think like psychotics, we should not follow any principle that has the general form of: whenever you receive evidence that seems to show X, interpret it as evidence that shows non-X or the opposite of X. Since the principle of methodological naturalism is such a principle, we should reject it.
Current evocations of the principle of methodological naturalism are signs of desperation. If no evidence for supernatural effects were accumulating, no one would ever try to evoke such a biased rule, which is so clearly in conflict with the more essential scientific rule of “objectively follow the evidence wherever it leads.” Instead scientists would just say, “We would believe in that if there were any evidence for it; but there isn't any evidence for it.” But when someone is besieged by an accumulation of evidence that he doesn't want to accept, and can no longer plausibly claim that such evidence does not exist, he may in desperation try to evoke some rule that tries to disqualify such evidence by claiming it is a violation of procedural rules.