What could be less scientific than a homeopathic proving?
I concluded a recent post, discussing claims that a group of Danish schoolgirls had proven that exposure to WiFi radiation was dangerous for plants, by suggesting that they (or whoever carries this project forward) fully apply the scientific method in their future research, so that any final conclusions would leave no doubt as to which factors (exposure to WiFi, temperature, humidity, sunlight, care by the experimenter, etc.) causes changes in the plant growth. As I was pondering the girls’s work and the subsequent media coverage, and doing some background reading, my mind crept to an exercise that claims to be scientific, but actually shows a large degree of hostility to the scientific method – the homeopathic proving, the process by which people (supposedly) figure out what can be treated by a new homeopathic treatment.
I must point out that I’ve written several articles describing how homeopathy is completely implausible from a chemical standpoint, and there is already plenty of information out there explaining why homeopathy doesn’t work from a medical standpoint. The ridiculousness of the proving process is just one more reason to reject the practice. Perhaps if a proving was conducted with some rigour, it would become plainly evident that homeopathy doesn’t work. For now, though, it serves as a great example of how not to apply the scientific method.
According to the principles of homeopathy, if a substance causes a symptom in a healthy person, that substance can be used in smaller doses to treat an ill person exhibiting the same symptoms – the idea of “like cures like”. So someone who wants to find the homeopathic remedial properties of a heretofore unverified substance – perhaps a dead animal, a piece of an inanimate object, or a dangerous chemical – will enlist some healthy volunteers to conduct a proving. Quoting from the New York School of Homeopathy’s explanation of provings (emphasis is mine):
“A proving is the testing of a potentized substance to find out which symptoms that substance is capable of producing, and hence curing. A proving is conducted on volunteers who are in a reasonable state of health (provers), and who do not know what substance it is they are taking. Doses are repeated until provers start to experience symptoms of a change in state. The provers record everything they experience, whether physical, emotional, mental, or even spiritual, as long as the change in state persists. At the end of the proving all the records are compared to find the physical symptoms, states of mind, feelings, and experiences that the provers have had in common, which can reasonably be attributed to the emerging signature resonance of the substance.”
The only part of the procedure that sounds a bit science-y is the underlined clause – the volunteers are not told what substance is being tested. I suppose that withholding this information would remove bias – although, since the whole point of the exercise is to figure out which symptoms are caused by a substance that has never been proven, the volunteer would not know what to expect, even if they were told what they were taking. To expect a homeopathic remedy to cause a specific outcome is said to be a “cardinal sin” (scroll to the second point in Part I of this link), thus invaliding, in the minds of homeopathy advocates, the randomized controlled trial testing approach.
Contrast this to a pharmaceutical trial, where the volunteer does know what the treatment is expected to do, so they can get a sense whether it is working, while also noting any undesired or unexpected effects. Of course, a person’s improvement does not prove that the pharmaceutical treatment is effective; this is why the trial is conducted with a large group of volunteers, with a control group receiving a placebo and the test group receiving the real product. Other variables that could possibly explain positive outcomes are controlled, so that in the end, an objective review of the trial data can (hopefully) lead to the conclusion that significant improvements in the test group can be attributed to the treatment.
The number of volunteers involved in a homeopathic proving can usually be counted on a few hands. Every healthy volunteer is expected to spend some time before the start of a proving writing down everything involving their mental, physical and emotional states, in order to establish a “baseline state”. When proving day arrives, the volunteer consumes a dose of the “potentized” substance – often a 30C preparation, which means the last 18 or so dilutions involve diluting 100% solvent with the same solvent. If a volunteer experiences a “change of state” after taking the dose, they write down everything they experience while in that “change of state”. It is therefore implied that any changes to a volunteer’s physical, emotional or mental state after consumption are caused solely by the substance being consumed. Apparently, it is not possible that the “changes of state” may stem from what they ate and drank earlier in the day, or fatigue from a bad night’s sleep and a hard week’s work, or the exposure to the temperature or humidity of the day, or contemplation of an exciting new opportunity, or the anxiety from having to track every “change of state” after consuming a dose of an unidentified and unproven substance.
As one example, look at this proving of Virginia Opossum. (While one example can never be taken as concrete evidence of anything, the presentation of this proving is typical of what can be found in proving databases.) A long introduction to the biology of the Virginia opossum and details of the capture of the one used in the proving, while interesting to read, seems irrelevant to the question of human health. The proving was conducted by 15 volunteers, although only 10 reported any symptoms after six doses. The long “summary” of the provings attempt to link vague descriptions of symptoms to characteristics of the animal. And to be clear, symptoms are not limited to what is happening in one’s body. Among the listed “symptoms” that are attributed to a volunteer consuming a homeopathic dose of Virginia opossum: a dead fly in a wine glass at a restaurant, a raccoon under a child’s bed that looked like a possum, and an assignment to clean a dirty refrigerator two weeks after dreaming of a dirty refrigerator.
The supervisors of this proving compiled a list of 622 different symptoms from those 10 volunteers, including 153 different dream characteristics. Any symptom observed by even a single volunteer is included in the list, forever enshrined as a potential remedy for that symptom in an ill person. (This seems to contradict the explanation that the supervisors seek common symptoms, which are then attributed to that substance.) It is notable that some volunteers report many more symptoms than others – perhaps some people are just more obsessed with recording every last thing they observe about their body and mind? There is no objective data that is tracked in the same manner from volunteer to volunteer – as David Bradley writes, “good clinical trials use objective outcome measures – those that are binary (like death or survival), quantitative (like a blood level), or are based upon a specific physical finding.” Instead, the proving relies only on a bunch of anecdotes that are very easily fashioned to the story that one wants to tell.
Everything that a volunteer feels in their body after consuming the dose is attributed to the substance being proven, even though he/she has actually consumed exactly none of that substance in the first place. If a dose does not cause any symptoms, he/she is told to try another dose later in the day. Once a volunteer feels that they experienced a “change of state” from a dose, they are instructed not to take any more doses – I suppose it would be confusing and maybe a bit embarrassing if the second dose gave different symptoms from the first dose. Yet, even if homeopathy really did work, there would be no way to attribute symptoms to a particular substance from a single dose. There is no control for any other variables such as mood, diet, weather and exposure to WiFi. (In passing, while homeopathic remedies are supposed to “last a lifetime”, you probably won’t be surprised to learn that they are said to lose effectiveness within 1-3 days when exposed to evil WiFi and evil cell phone radiation. I still want a homeopath to explain how they can determine that a remedy has lost effectiveness before giving it to an unsuspecting patient.)
In a proving, no one is given a placebo to verify if the recorded changes would occur without taking the actual dosage. The School of Homeopathy tries to explain that a placebo is unnecessary in a proving (again, emphasis is mine):
“The proving of a remedy by members of a coherent group appears to cause an amplification of the effect of that remedy. There appears to be a teletherapeutic effect produced by the field generated by the assembled provers, their experiences being in resonance. The whole group is involved and those members who have not taken the remedy may be as affected as those that have.
This means that the use of control provers who are given placebo is not possible as they are also likely to prove the remedy. Because of the group’s field effect It also means there is no need to repeat the dose if symptoms do not occur immediately.”
The emphasized sentence actually makes sense, although probably not in the way the author meant it. Since a 30C dilution has none of the tested substance within it, there would be no difference in the composition of the “remedy” and the composition of its placebo. Therefore, there is no reason why the control provers should exhibit any different symptoms than those consuming the remedy.
Finally, there is a contradiction in full display when considering what is done with the results of a proving: while we are told that like cures like, we are also told that a remedy must be matched to one’s symptoms. (Wait, I thought homeopathy treated the “whole person” and not just symptoms…) It would seem that the entire practice of the homeopathic proving would be pointless to a homeopath, since what works for one person would not necessarily work for another person, and every remedy is supposed to be tailored to the individual.
So remember: while a homeopathic dose of Virginia opossum caused a healthy person to dream about dirty refrigerators, making it a possible cure for an ill person dreaming about dirty refrigerators, it doesn’t mean that it will necessarily work for your particular case of dreaming about dirty refrigerators.
Hey, that makes as much sense as the concept of the homeopathic proving.