This YouTube video by Yvette d’Entremont, who calls herself the ‘The Science Babe’, is making rounds on the Internet, also on skeptical websites. She tries to convince the viewer that homeopathic sleeping pills don’t work by swallowing a whole packet of 50 pills of Calms Forté. She doesn’t feel sleepy even after waiting one and a half hour. So, homeopathy is bullocks and the producers of these pills are misleading consumers?! That might be true, but I don’t agree that this is a convincing way to debunk homeopathy in general, but I’ll explain that after the video:
As you can see in the comments on YouTube or in the comment areas of the blogs which show this video, a lot of homeopathy supporters come with counter arguments: the Science Babe didn’t take the pills according to the ‘rules’, like don’t touching them with your hands and swallowing them together with the diet Coke. According to them you should drop some pills under your tongue directly from the bottle and let those dissolve there. So this video is just another example of those pesky skeptics who don’t understand homeopathic principles and start shouting that it is nonsense, based on faulty tests. Do they have a point?
For a proper debunk, I think you have to test the exact claims which are made. Does the producer of these homeopathic ‘sleep aid’ pills claim that they contain active ingredients? Does it claim that taking more pills enhances the claimed effect? What does it say about how you should take the pills? Maybe you can check for yourself on their website.
But that’s not my main problem. These type of homeopathy debunking video’s or actions like ‘1023 Homeopathy: there’s nothing in it’ tend to focus on the implausibility of homeopathy, i.e. (most) homeopathic remedies are diluted to such a point that it is highly unlikely to find even a single molecule of the stuff they started with, the mother tincture, in the pills or drops that are sold. Therefore there is no imaginable working mechanism for these remedies to achieve an effect other than a placebo effect. According to science. But there’s where the discussion with homeopaths doesn’t stop. Some might argue that over the counter sold homeopathy can’t be compared to real homeopathy, which requires a completely individualized approach. Others have set their hopes on nanoparticles as a ‘normal’ explanation for the alledged working mechanism. But in fact many don’t dispute that you can’t detect any molecules of the mother tincture in their stuff and that therefore it can’t work according to accepted science. But they just dispute that the current status of sciencific knowledge has the last word in this matter. Either they claim that science hasn’t evolved far enough to understand how homeopathy works, or that it works in ways the scientific method will never be able to figure out in the first place.
So how to deal with these arguments (how silly they may seem)? I think it’s a better strategy to ask homeopaths how they are able to obtain knowledge on how their products work, while that’s impossible by ‘our’ science. How do they figure out which remedy is fit for which symptom? Of course this standard homeopathic theory, they perform so called provings. If they want to find out what a new homeopathic remedy might be good for, they give a couple of healthy volunteers this remedy in the form of pills or drops and let them notate all symptoms they experience. These symptoms can be quite diverse, from a little itch on the left shoulderblade, to dreams about snakes, actually anything which they consider out of the ordinary. In this way the homeopathic ‘knowledge’ base is constructed: probably millions of combinations of remedies and symptoms. These combinations can also be found by dreaming (dream provings) or group effort (trituration provings). One famous homeopath even loves to take a bath in a tub filled with water to which a single droplet of the new remedy is added and then he lets thoughts run freely to find the symptoms by association.
These remedy-symptom combinations don’t look to be obtained in a very controlled way and skeptics will argue that these are just random combinations. And that’s something that could be tested in several ways. You would expect for instance that repeating the provings would more or less give the same results. If not, what are these combinations worth? Of course you should do these replications proparly blinded. You could also test the skills of homeopaths in detecting symptoms. If they fail to discern the ‘established’ symptoms from a remedy if it is given to them without knowing what they are given, that does say something, doesn’t it? But homeopaths are very reluctant to perform these kind of blind replications when challenged. They have tried themselves a few times in history, but you do not hear much about those efforts from themselves and that’s not that surprising.
Homeopaths in the United States in 1879-1880 performed a test in which homeopaths were given ten vials with sugar pellets of which only one contained sugar pellets moistened with Aconite 30c. Simple task: fgure out which of the vials contains the actual homeopathic remedy. The idea was that they would just do a proving for each of the vials, note down the symptons and see which of the proving results looked most like one of Aconite 30c. This turned out to be an utter failure for homeopathy, read Early use of blind assessment in a homeopathic scientific experiment (Kaptchuk, 2004) for this. Similar tests were done by German doctor Fritz Donner, with similar dissapointing results for homeopathy. But these occured during the reign of the Nazi’s in Germany and are therefore under some cloud of obscurity.
A more recent study is one by Michael Teut: Homeopathic drug proving of Okoubaka aubrevillei: a randomised placebo-controlled trial: “Thirty-one subjects were included (19 Okoubaka and 12 placebo). Data for 29 participants could be analysed. No significant differences in number of characteristic symptoms in both groups were observed between Okoubaka.” Quite damning for homeopathy in my opinion.
There is also recent attempt by Jermey Sherr, a homeopath known for his provings but also for his unethical experiments with homeopathic remedies for hiv/aids and malaria in Africa. Sherr gave several experienced homeopaths a list of symptoms obtained in a recent (re)proving of a remedy which had been used in a proving in the past without telling which remedy. So they had to guess what the most likely candidate was from all remedies in the Materia Medica. It was actually Ozone C30, which proving was done in 1993 and had been replicated around 2008, but these last results had not yet been added to the Materia Medica. Sherr manages to present the result as a success for homeopathy, because two of the seven homeopaths had chosen the right remedy amongst their three allowed guesses in the first round of this test. That looks very impressive indeed if you think that they could have chosen from 2372 remedies in the Materia Medica. But if you look closer to how this test was set up, the results don’t look that impressive. It is also remarkable that Sherr’s wife, Camilla, was one of the participants. On the Skepsis-blog is a thorough analysis, only in Dutch at the moment.
Of course homeopaths find excuses for these failures, but those are not convincing. Also simple tests are shunned by homeopaths, like the challenge at the Quackometer (no homeopath has dared to take this up since December 2007). In the Netherlands the Dutch Skeptics foundation Skepsis almost came to agreement with homeopaths in 2004 to perform a test in which homeopaths were to determine from recorded symptoms by volunteers who either took Sulphur C200 or a placebo. But the homeopaths backed out after some starting negotiations by the secretary of their organisation, who seemed quite willing to cooperate. Probably some members of his organisation understood the risk of participating in this test better than he did. And there’s also the test which famous homeopath Vithoulkas would do for the JREF one million dollar challenge, but this didn’t happen because they failed to achieve agreement on the testing procedure.
Many homeopaths with some clue of science and statistics are probably well aware that their knowledge is not replicable and therefore they will never participate in tests that would easily show this. That’s why they prefer randomized controlled trials, often with faulty designs, which give pretty good chances of getting significant results just by chance (or a bit of help). But those don’t mean a thing because of the prior probability of zero that those ultra diluted substances can have an effect.
Summarizing: the ‘experiment’ of the Science Babe is okay enough to show that homeopathic remedies don’t work like normal medicine, but that isn’t what homeopaths claim. To challenge their actual claims you have to challenge them to reproduce their provings. Challenge them to show that their knowledge is actually something other than just random relations of remedies and symptoms.
By the way: this Science Babe ‘experiment’ with homeopathic sleeping pills was a recurring item in talks by James Randi, so it’s not very original either. See his TED Talk from 2010: