Nassim Nicholas Taleb is a statistician and risk analyst. He recently sent out a tweet that defended homeopathy as being harmless and even beneficial since after all people were only taking a placebo and that it may prevent them from over-treatment of marginal symptoms. He followed up with another tweet saying that “Superstitions can be rational if 1) harmless, 2) lower your anxiety, 3) prevent you from listening to forecast by economists & BS “experts””
Leaving aside the obviously facetious third point, the first two are often invoked by homeopathy’s defenders that even if homeopathic treatments don’t do any good, they at least don’t do any harm.
But Cory Doctorow says that such bland assurances are misleading.
In pursuing this line of inquiry, Taleb ignores the great body of peer-reviewed, published evidence about the real harms of homeopathy, which fall into two categories: first, people with real medical problems (e.g. cancer) substitute placebos for effective therapies.
Second, that people who take homeopathic remedies for difficult-to-diagnose or imaginary ailments waste public/insurance money (in healthcare systems that fund “Complimentary/Alternative Medicine”) and are apt to overmedicate with both homeopathic and real medicines — a 2015 paper looked at 45,000 patients and determined that homeopathic treatments “led to more productivity loss, higher outpatient care costs and larger overall cost.”
This has sparked an angry response from Taleb, calling Doctorow “very stupid” and dishonest”.
No post on homeopathy is complete without this clip from That Mitchell and Webb Look.
Reginald Selkirk says
It can be rational for superstition to lower your anxiety? This must be some new definition of “rational” with which I am not familiar.
Marcus Ranum says
That would all be possibly true if homeopathy was free. But it’s not.
Saying “a placebo may not be bad” only makes sense when you compare it to nothing at all. Is a $40 nostrum better than a $0 nothing when they are equally effective? Only if you work for “big homeo”…
Homeopathy is harmful because, amongst other things, it reinforces magical thinking, fooling you into thinking you’re getting better when you could be benefitting from some effective medical treatment.
Marcus, consider Dan Ariely’s (Ig-Noble prize winning!) work work showing that discounted placebos are less effective at pain relief. So here’s the conundrum: a placebo might offer temporary relief while the body heals or while doctors seek to understand the patient’s condition better, but only if it is in fact expensive. (Hmm, that should be an incentive not to cover homeopathic treatment by insurance or by government programs.) How does cheap placebo compare to no treatment?
For some reason I am reminded of this… http://oglaf.com/nostrum/
Placebo Blocker from xkcd
It can be rational in the sense that it’s rational to realise that people are irrational, and to work with that fact. The underlying behaviour itself is irrational, but given that it exists, it’s rational to exploit it if doing so can provide a benefit.
John Morales says
There’s a difference between being irrational and exploiting such irrationality, and I think Reginald refers to the behaviour, not to its exploitation. Different things.
Try rephrasing your claim in a specific case, and see where it gets you: “I believe that relying on superstition to decrease my anxiety is irrational, but since I am irrational (being a member of the set of ‘people’, who are irrational) it is therefore rational to rely on superstition to decrease my anxiety.”
(This is leaving aside considerations of the cost:benefit ratio; yes, there is some benefit, but you are leaving aside the cost)
Taleb’s phrasing (“Superstitions can be rational if…”) isn’t entirely clear here, but since all of the generally understood meanings of “superstition” usually imply irrationality (and, incidentelly, because I don’t think he’s stupid and am applying a reasonably generous interpretation), I’m assuming that he’s refering to the exploitation of the behaviour, rather than to the behaviour itself. Reginald appears to be applying the opposite interpretation, as you say. I was attempting to supply the alternative (and, in my reading, more reasonable) interpretation.
John Morales says
Fair enough, Dunc. Thanks.
So… the fundamental claim is that the ends justify the means, if one only considers the upside. I think Doctorow makes a good rebuttal.
I’m not sure that I’d agree with that. I think it’s more that this particular kink of human psychology exists whether you like it or not, so you may as well try to get some benefit from it.
John Morales says
The benefit (in the context of well-feeling) is clearly superior if one is only psychosomatically distressed, and arguably superior if one is medically distressed and will either recover without actual medicinal intervention or won’t recover regardless of medicinal intervention.
I put it to you that the rest of the time, the cost probably exceeds the benefit.
(Consider Steve Jobs sad death)
Well, there we’re into a rather wider question… You seem to be assuming that people would accept conventional treatment in the absence of alternatives. I’m not sure that that’s necessarily true. In some cases it may well be, in others it’s almost certainly not. I’m not sure that the costs and benefits are actually linked that closely. People are always going to be irrational to some extent or another, so I’m not convinced that you can get rid of the costs simply by eschewing the benefits -- not fully, anyway.
But anyway, that is a rather different (and much more difficult) argument. I certainly have grave reservations about actively encouraging these sorts of beliefs, for exactly those reasons.
John Morales says
Dunc, neither of us disputes that there are benefits, however illusory.
Yes. Though for me, those reasons pale before the epistemic capitulation that activity entails.
FWIW, there are studies that show that placebos work even when the recipient is aware that they are taking a placebo. Quite remarkable, and congruent with your intuition, that is.
Whenever I hear alternative medicines regarded as “harmless,” I simply go to http://www.whatstheharm.net
Some examples from their Homeopathy section: