Nov 15 2007

Sloppy Thinking about Homeopathy from The Guardian

Jeanette Winterson recently wrote a piece for The Guardian titled In Defense of Homeopathy. It’s always fascinating to read or hear people defending nonsense – in doing so they have no choice but to use bad evidence and bad logic. Homeopathy has no scientific plausibility and the clinical evidence shows that homeopathy does not work for any specific indication. So homeopathy’s defenders have a real task on their hands – thankfully they are armed with numerous logical fallacies and every manner of sloppy thinking, so they are up to the task.

Homeopathy has been under the gun recently, especially in the UK where physicians and scientists guided by reason and evidence are trying to remove it from the national health service. The defenders of homeopathy, in response, have really been turning up the heat, desperately trying to find cover wherever they can. Winterson’s article is one such attempt, and she pulls out all the techniques of bad science and bad thinking that are the homeopath’s bread and butter.

She starts out with the bedrock of medical pseudocience – the anecdote. She informs the reader that she knows homeopathy works because it has worked for her. She writes:

Picture this. I am staying in a remote cottage in Cornwall without a car. I have a temperature of 102, spots on my throat, delirium, and a book to finish writing. My desperate publisher suggests I call Hilary Fairclough, a homeopath who has practices in London and Penzance. She sends round a remedy called Lachesis, made from snake venom. Four hours later I have no symptoms whatsoever.

She concludes from this personal experience that homeopathy must have something to it, but this conclusion is not justified by such uncontrolled observations. There are many other possibilities. It is quite possible that her cold was just running its course and would have gotten better anyway. I have certainly awakened with a cold and felt better hours later after some food and hydration. Perhaps she felt better because of the ibuprofen she took earlier (she doesn’t say she did, but she doesn’t say she didn’t take anything else, and perhaps in her “delerium” she forgot). We also don’t know what was in the homeopathic remedy she took – perhaps it was spiked with a real drug.

I have no way of knowing if any of these alternative explanations are true – and neither does anyone else. That’s the point. Uncontrolled observations are uncontrolled, we cannot conclude that any one of the many variables are what caused the symptoms to improve. Therefore Winterson is make the post hoc ergo propter hoc logical fallacy of assuming cause and effect.

Later in the article Winterson gives us the standard rationalization for why controlled clinical trials have shown that homeopathy does not work. She says:

It is not enough to say Disease A is caused by B and can be cured by C. Homeopathy, in common with other holistic approaches, asks that we look at the whole picture – the person, and not just his illness. Specifically, in the case of homeopathy, the remedy picture, which is carefully drawn up after full consultation with the patient,…

This is just a desperate attempt to dismiss negative evidence, and worse is an example of the logical fallacy of inconsistency. Winterson and other homeopathy defenders will cite anecdotal evidence of homeopathy’s efficacy, even when a remedy is chosen entirely based upon the symptom. Of course, any over the counter homeopathic remedy can only be based upon symptoms – not a “full consultation” with a homeopath. Also, does Winterson wish us to believe that her phone conversation was a “full consultation” and the choice of homeopathic remedy was based upon her holistic individual illness, rather than just her symptoms?

Therefore, homeopathy defenders will accept anecdotal evidence that favors homeopathy, even if the treatment were based upon symptoms, but then refuse to accept controlled clinical trials of homeopathy that are negative because the treatments were based only upon symptoms and not a holistic full consultation.

While we are on the topic of anecdotal evidence, Winterson writes:

I am sure that there is a placebo effect in homeopathy, but it is a fact that many of the people who end up visiting a homeopath do so as a last resort, when nothing else is working. That such people often see an improvement suggests that the remedies themselves are contributing to the wellness of the individual.

Wrong. This is just more sloppy thinking from Winterson. There is a well-known statistical effect (more generally known as the rainmaker effect) that people with chronic symptoms will continue to seek treatments and that when the symptoms spontaneously remit or wane they will credit the most recent treatment with the success. They will think it is amazing that after a long period of no relief the last treatment was followed by improvement. They fail to recognize (because people have generally poor intuition about statistics) that it is not statistically significant because they were most of the time trying some new treatment, so if and when they had a spontaneous improvement (which most ailments will do) they were very likely to be taking something at the time. From some people we hear that they tried chiropractic, then acupuncture, then homeopathy and finally the homeopathy worked. From other people we hear that they tried homeopathy then chiropractic then acupuncture and the acupuncture finally worked. In most cases people remember the hits and forget the misses – they go around saying that “homeopathy worked” not that “chiropractic and acupuncture didn’t work.”

Another phenomenon exists whereby when people seek out and engage in a novel therapy this is likely to generate a new placebo effect, even though they were previously being treated. Novelty has an effects. What’s more, some people may have just decided that they were finally going to take their health seriously, so they improved their diet, exercised, were more compliant with their meds, and went to a homeopath. Guess who gets the credit for their symptomatic improvement?

To say that the anecdotal experience of many people is and should be convincing is naive in the extreme, and flies in the face of a couple of centuries of hard won scientific and medical wisdom. It represents another logical fallacy, the argument ad populi, or argument from popularity.

Winterson also makes what I will call the “holistic fallacy.” This is to argue that effective treatment takes into account the whole person, unlike scientific medicine which treats diseases. First, mainstream scientific medicine considers the entire biological organism, including psychological and sociological factors, in its evaluations. This is not incompatible with basing treatments on appropriate reductionism – knowing that the influenza virus causes the flu, for example. The holistic fallacy is a false holism – it ignores the actual mechanisms of disease, and the actual biological responses to disease, and pretends that there is a magical mystical deep connection and we must therefore submit ourselves to the mysterious ministrations of the guru in order to be holistic.

Addressing the crushing problem of homeopathy’s utter lack of scientific plausibility, Winterson desperately writes:

Objections to homeopathy begin with what are viewed as the impossible dilutions of the remedies, so that only nano amounts of the original active substance remain, and in some cases are only an imprint, or memory. Yet our recent discoveries in the world of the very small point to a whole new set of rules for the behaviour of nano-quantities. Thundering around in our Gulliver world, we were first shocked to find that splitting the atom allowed inconceivable amounts of energy to be released. Now, we are discovering that the properties of materials change as their size reaches the nano-scale. Bulk material should have constant physical properties, regardless of its size, but at the nano-scale this is not the case. In a solvent, such as water, nano particles can remain suspended, neither floating nor sinking, but permeating the solution. Such particles are also able to pass through cell walls, and they can cause biochemical change.

She doesn’t use the word “quantum mechanics” but the implication is there. The world behaves differently at the “nano” scale. There are numerous problems with her lame defense. The first is that most homeopathic remedies do not have “nano” scale active ingredient – they have absolutely no active ingredient. Also, dilution has absolutely nothing to do with the scale of the particles. A substance diluted in a solvent will typically be comprised of individual molecules. A molecule is the smallest amount of something that retains its chemical properties, and this is independent of the amount or the “scale” (which doesn’t really have any scientific meaning in this context). Her entire nano discussion is vacuous scientific nonsense. It is meant to achieve one thing only – to sound scientific and offer a “truthiness” to homeopathy (with apologies to Steven Colbert).

But, to hedge her nano bets she also cites the “water memory” defense. I wrote about this previously – there is no biologically relevant, persistent, and information carrying water memory. Water has transient structure, but only transient – it is a liquid after all – and homeopathy defenders are simply engaged in anomaly hunting to provide more truthiness cover for homeopathy. What they have absolutely not found is a possible mechanism for homeopathic remedies, nor can they demonstrate that under controlled conditions they can tell the difference between homeopathic water and regular water. They are just using pseudoscience to cover up pseudoscience.

Winterson (as all ideologues do) must find some reason to explain the scientific criticism of homeopathy. That we have a legitimate point is not even considered. She writes:

This homeophobia is, I think, a genuine terror of what homeopathy is suggesting; which is that we think differently about the relationship between the cure and the disease.

Interestingly, when I debated Rustom Roy he used the exact same term, “homeophobia.” We hear this from the UFOlogists, the ESP enthusiasts, the conspiracy theorists, and every crank, quack, and charlatan alive: Scientists – those who have dedicated their lives to investigating the truth – are secretly afraid of the truth. We are just scared little children of the awesome implications of homeopathy (or substitute any pseudoscience for homeopathy).

This is just an ad hominem logical fallacy. If homeopathy worked, we would be all over it. If we could prove it works and could not explain how, then that would open up a new research avenue – a window to discover something new about nature. That is what scientists live for, what they base their careers on. Most scientists, however, don’t want to waste their careers on blind alleys or utter nonsense.

What Winterson is dismissing as terror is really just respect for all the medical knowledge that we have accumulated to date. If we defend scientific medicine it is because this approach has been fabulously successful. Meanwhile, after two hundred years, homeopathy as been a complete and total failure. Homeopathy does not have one single success story – meaning not one proven treatment that stands up to scrutiny. Not one after two hundred years. The assumptions of homeopathy have all been disproven. Two hundred years of scientific advancement has not confirmed a single premise of homeopathy. I humbly suggest that it is no we scientists who are afraid of the truth.

Winterson concludes:

I would like to see homeopathy better regulated. I would like to see the Society of Homeopaths engaging with its critics, as well as initiating more research. There will always be rogue homeopaths and bad homeopaths, but that is true of any profession. Above all we should be careful of dismissing the testimony of millions who say the remedies have worked for them.

Again, absolutely typical of the pseudoscientist mold. I would like to see homeopathy regulated also – by which I mean I would like to see it reduced to a footnote in the history of bad science, along with N-rays and Mesmerism. But was promoters of nonsense want is to regulate – meaning to validate with degrees, licenses, and certifications. You cannot regulate nonsense, you just have to call it what it is and get rid of it.

The call for more research is also ubiquitous. UFOlogists, ESP believers, 9/11 truthers, HIV denialists, and antivaccination kooks all want more research. They endlessly want more research, because they never like the results of existing research. But more research is not the answer, because pseudoscientists do not base their beliefs on evidence. Research is all about cover – maintaining the air of plausibility, sincerity, and legitimacy for as long as possible. It’s a sham.

If homeopaths were serious about research the first thing they would do is stop all homeopathic practices that are not supported by existing scientific evidence, by which I mean all homeopathic practices period. Then engage in proper clinical studies, of the kind they say are necessary, with proper informed consent, and live by the results. But that will never happen. The studies they are now asking for (after the ones they previously asked for came back negative) will be done, and they will change nothing.

She then dismisses the many sins of homeopathy, such as prescribing homeopathic remedies to treat serious illnesses like malaria, as the occasional anomalous “bad homeopath” just as exists in all professions. Right – just as there are some fake psychics, but they shouldn’t detract from the “real” psychics. First you have to prove that there is such a thing as a good homeopath before you can condemn the bad homeopaths.

And finally she makes one more appeal to populism and anecdotes. She has apparently never heard the adage that the plural of anecdote is anecdotes, not data.

Winterson wants homeoapthy defenders to engage with its critics. Bring it on. Bring on the logical fallacies, bad science, and sloppy thinking. We can take it. But I suggest that the homeopaths need to be intellectually honest and put their money where their metaphorical mouth is. There need to be stakes – actual implications. If they can make their case with reliable evidence and valid logic, I will start prescribing homeopathy myself. If logic and evidence is against them, they need to abandon their claims.

Rather, they want to play “heads I win, tales you lose.” Sorry. I’d rather not play. I would rather engage in actual science. If they would only give it a try we could end the controversy quite easily.

16 responses so far

16 thoughts on “Sloppy Thinking about Homeopathy from The Guardian”

  1. TheBlackCat says:

    I think there is another key issue. The burden of proof is always on the person making a claim that something happens or something exists, never on the person claiming it doesn’t. It is the homeopaths’ job to do the proper scientific research to show that homeopathy works. They think they are right, therefor they have a responsibility to do the scientific research to show they are right. But for the most part they don’t. They leave it to real scientific researchers to do the work.

    This, of course, leaves them a perfect “out”. The people doing the research weren’t homeopaths, they don’t understand how homeopathy works, the methodology was all wrong. The homeopaths should just get together, figure out a good way to do a valid test of homeopathy, and then do the tests themselves (with proper controls to avoid cheating). This is how real researchers who come up with a new idea do things. They don’t sit around all day demanding that others do their work for them. They go out and prove their case themselves, or disprove it and accept their mistake. Relying on real researchers to do the test instead of doing it themselves is just another excuse to allow them to avoid having to accept they are wrong.

    A standard pseudoscience tactic is to avoid doing research themselves. They tend to demand that someone else do it for them. If that doesn’t happen they whine about being ignored or suppressed. If it does happen and the results are negative they whine about how the tests were flawed (despite the fact that the researchers were doing the very experiment they demanded) and demand a new experiment. When a real scientist has a new idea, they show that idea works or doesn’t themselves.

  2. Jim Shaver says:

    Great post, Dr. Novella! You’re getting rather good at this blogging thing. (Sorry if I seem to be grovelling; I’m still on a high from watching Judgement Day.)

  3. delaneypa says:

    As a practicing MD myself, I would LOVE it if homeopathy worked, so I could make promises of cures without the worry of pesky side effects. Instead, I have to spend a significant fraction of my day informing people about the possibility of diarrhea with antibiotics, erectile dysfunction from blood pressure medications, liver damage from cholesterol medications, etc. Show me a medication without side effects, I’ll show you one with no effects.

    Most MDs will first try non-pharmaceutical methods to treat chronic conditions when safe….weight loss for diabetes, restricting salt for blood pressure, quitting smoking for coronary artery disease, etc. But it is the rare individual who will change their lifestyle when they could simply take a pill instead. So much for the “holistic” approach.

    Homeopathy would certainly make my professional life easier. Thus, to say that because I reject homeopathy that I am a “homeophobe” boggles the mind.

    -P Delaney, MD

  4. Bowdinium says:

    Great article Dr Novella but you have competition. In the Guardian today, Ben Goldacre published an excellent piece both refuting homeopathy itself and tackling the wider issue of magical nonscientific thinking.

    I think Goldacre would be an excellent subject for an interview on the podcast to educate us about more specifically British problems with the propagaters of alternative medicine and bad science.

  5. Richard says:

    “Winterson wants homeopathy defenders to engage with its critics. Bring it on. Bring on the logical fallacies, bad science, and sloppy thinking. We can take it.”

    OK, here is some bad science:

    Shang A, Huwiler-Muntener K, Nartey L, Juni P, Dorig S, Sterne JA, Pewsner D, Egger M. Are the clinical effects of homoeopathy placebo effects? Comparative study of placebo-controlled trials of homoeopathy and allopathy. Lancet. 2005; 366 (9487):726-32.

    This is the study whose abstract claims to compare 110 allopathic studies and 110 homeopathic studies. But when you look at the paper itself, the authors actually compare only 6 allopathic studies and 8 homeopathic studies, without even saying which ones were used.

    This is a pretty small sample, and it violates the transparency rules of normal allopathic research. Yet not only did Lancet publish it, but the editors even went so far as to say that it heralds “the end of homeopathy.”

    Have you even read this paper, Steve? And if so, can you say with a straight face that you would give it a passing grade in an undergraduate medical research seminar?

    If the homeopathy skeptics really have science on their side, why do they need to uphold such sloppy work?

  6. Richard,

    Your point is not valid. You have to interpret the entire literature and when you do homeopathy fails to show any effect. It also cannot propose a viable mechanism.

    Regarding this particular study, you make it sound like they are being deceptive. But in the abstract they say how they selected trials and the number they selected. There is no deception. This is a standard method for a systematic review – they gave their search methods, the number of trials used, and out of those trials how many met specific criteria – in this case large and high quality studies with a design adequate to minimize bias. Eight large high quality studies is not a small amount of data – and it shows that the best studies are negative. This is a very significant pattern in any research. In mainstream evidence-based medicine such a treatment would not be used – it would be considered ineffective. What evidence-based justification do you have to consider homeopathy effective for anything?

  7. Richard says:


    This is a pretty small selection, as I mentioned.

    Given the sweeping nature of their claims and the finality of the Lancet editorial, you would think that this “anecdotal” meta-study would at least specify the input data clearly enough for interested parties to review it.

    Or are you telling me that the authors didn’t need to disclose which six allopathic studies and which eight homeopathic studies they based their analysis on because this … is evidence-based medicine?

  8. jasonb says:

    I think debunking homeopathy is important and worthwhile. There is, however, a more pressing matter at hand.

    Following on the heels of recent news that many drugs prescribed by doctors (and sold by pharmacies) lack an FDA approval, how many people are aware that LESS THAN FIFTY PERCENT (50%) of western medical practice is based on valid scientific evidence and some experts estimate that number is low as FIFTEEN PERCENT (15%).


    At 15%, I have to imagine the whole of Alternative Medicine has more valid science behind it than that.

    It would be a tremendous service if you could identify the top dozen or so western practices that would utterly astonish people to know have as much scientific validity as homeopathy. There has to be some real doozies at the top of the list.

  9. eiskrystal says:

    She had a temperature of 102, spots on her throat, delirium :
    and these symptoms were completely cleared up in a mere 4 hours?

    I’m suspicious. Very suspicious. That’s a bit quick dontcha think

  10. Stephen_Bain says:

    It looks like the sloppy thinking is not quite over yet.

    After one of the Guardian’s Comment is Free bloggers posted an article rebutting Ms. Winterson’s, a former lecturer in Islamic Studies (No, I don’t see a link here either…), Denis MacEoin, has added yet another pro-homeopathy article.

    Once again it seems that all those scientists who were trying to test homeopathy were doing it wrong, because gosh-darn-it homeopathy is just untestable.

    I see ad hominem piled on top of the argument from ignorance and a bunch of other exciting fallacies.

Leave a Reply