This post is a follow-up to my post on Hyper-Rationalism and Segulos. In the comments thread, a number of questions and challenges were raised about my categorizing Rashba as non-rationalist in his approach to these topics. In order to explain this further, it is useful to discuss the general idea of empirical testing.
How do we know if a remedy works? There is a broad spectrum of possibilities.
1. The most basic stage is to accept that it works based on one person's say-so. As far as some are concerned - and this approach was prevalent in antiquity - if that one person is an authority figure, or dead, then his/her words carry much greater weight. (See Shabbos 66b for Abaye's description of remedies that he learned from his nursemaid. This is literally an "old wife's tale" - the word "wife" in that aphorism is from the Old English wif and refers to any woman, rather than specifically to a married woman. The aphorism refers to unverified welfare-related beliefs, often superstitious in nature, that are passed down from the women of an older generation to a younger generation. However, the aphorism today has derogatory overtones, especially with those who lack an appreciation for how epistemologies change over time. Needless to say, not all old wive's tales are false!)
2. The next stage up is to require more than one person's attestation. The Gemara rates an amulet as being effective if it has worked on three occasions.
3. The next stage is to require a much larger number of attestations - ten, a hundred, a thousand.
4. The next stage is to realize that one also needs to assess if all the people who recovered would have done so anyway - and so one needs to also look at those people with the same condition who did not use the remedy.
5. The next stage is to realize that the placebo effect is very powerful, and to counteract it by having a control group - a group who is receiving something that looks exactly like the remedy, but has inactive ingredients.
6. The final stage is to realize that those administering the remedy/ test may themselves be consciously or subconsciously influencing the results. It is thus necessary to have double-blind testing - where neither the subject nor the experimenter knows which is the remedy and which is the placebo, until the survey is complete.
Now, there is also another spectrum of people, ranging from those who strongly believe in all kinds of supernatural events, to those who are methodological naturalists (denying any kind of supernatural event). There is a strong, although not absolute, correlation between the people on this spectrum, and the people on the first spectrum that we discussed. Generally, those who freely believe in supernatural phenomena have lower standards for accepting that a given remedy works.
Rashba was towards the non-rationalist end of the spectrum. He did not see the lack of a naturalistic explanation as being any reason to doubt the validity of a remedy, since he believed that supernatural processes are just as common. And he had low requirements for accepting the efficacy of a remedy - a claim of it working a few times was sufficient empirical confirmation for him.
Rambam, on the other hand, was towards the other end of the spectrum. He was extremely reluctant to accept the existence of supernatural phenomena. And he was skeptical of remedies based on supernatural explanations, even if there were a few claims of their having worked. The reason for this skepticism is that he was well aware of the power of the placebo effect (see Hilchos Avodah Zarah 11:11).
Rashba writes that he is puzzled by Rambam's stance. On the one hand, Rambam says that all allegedly paranormal phenomena are false and therefore forbidden. But on the other hand, Rambam permits one to go out on Shabbos carrying a nail from a gallows, and a fox's tooth, because they are believed to have proven beneficial as remedial items!
The explanation of Rambam, put forth by Dr. Marc Shapiro in Studies in Maimonides and his Interpreters, is that Rambam believed these items to be beneficial as placebos. He did not think that they could genuinely work, since there is no naturalistic explanation for them, and he did not consider the "thrice tested" standard to be sufficient. In Rambam's view, one may carry these items on Shabbos because they are believed to be therapeutic.