If a natural phenomenon is claimed to exist and be observable, then it is reasonable to expect a theoretical framework which allows others to formulate experime

If a natural phenomenon is claimed to exist and be observable, then it is reasonable to expect a theoretical framework which allows others to formulate experiments to test it in novel ways. It does not have to be couched in mathematics, but once the description is transmitted it should permit thinking along the lines of "Aha, you're saying that this <well-defined-thing> causes that <well-defined-thing>! If that is the case then this <new-experiment> should have that <surprising-outcome>". That framework should also be clear about under what circumstances the phenomenon occurs, so that an outsider can in principle understand its limits and be able to perform tests without the advocate claiming that their experimental conditions were "not conducive" to the phenomenon. Another example of a "hypothesis" which has no mechanism of action (and I therefore do not regard as a hypothesis) is the "Stoned Ape Hypothesis". It claims that the ingestion of psylocibin explains the sudden increase in size of the brain at one point in our evolutionary history. But it does not actually give a mechanism for how your or my genes today would be affected, it just says that hunter-gatherer culture would have changed and there would be more group sexual activity. If I were to be presented with this "explanation", I would not expect a dramatic effect on brain size, because there would be less sexual selection going on. Homeopathy also strikes me as a claimed effect without plausible mechanism.