Placebos, Utilitarianism and Truth

I finally got around to reading Predictably Irrational, and the chapter on placebos got me thinking.  The chapter describes how some surgeries were found to be no more effective than “placebo surgery”, when doctors told patients they would do the surgery, gave them anesthetics and made incisions but didn’t actually perform the part of the surgery that was supposed to be effective.  The usual response when a treatment is proven to be no more effective than a placebo is to stop doing it, or to claim the study was flawed.

But if a placebo is effective (and they are often quite effective), perhaps we should continue giving them.  If placebos require false belief on the part of the recipient, to what extent is it ok for the scientific and medical establishment to deceive people, or at least not expend effort discrediting placebos?

I know this isn’t exactly a novel question, but I haven’t put much thought into it and the answer is not obvious to me.  Like many other who think of themselves as “rationalists”, I am mostly a utilitarian but I put a value on truth that is likely out of proportion to that which can be justified on purely utilitarian grounds.  My modus operandi is to be truthful without even making utilitarian calculations, and even if I made them and they pointed to deception I would likely decide to be a single-issue deontologist.

This tension goes back to the beginning of both utilitarianism and classical liberal truthiness, since JS Mill helped come up with both ideas.  He tried to square the circle and argue that there was no conflict.  Today people acknowledge the conflict but I have not read a good solution to it.  I believe Robin Hanson and Eliezer Yudkowsky have said something like “the conflict exists, I take the side of holding truthfulness as a value in itself but I cannot fully defend this position.” (except for mundane dishonesty)

I guess that’s where I am now too.  However, I do wonder if rationalists should spend so much effort trying to convince people that, say, homeopathy is quackery.  If people turn to homeopathic remedies in lieu of modern medicine when there is a real treatment available, that is certainly bad.  However, in the areas where modern medicine does little better than a placebo, homeopathy is likely to provide a much cheaper placebo.

This issue comes up in economics as well.  Some macroeconomic remedies may return the economy to prosperity by fooling people.  Rational expectations argues against this by saying that the government is incapable of fooling markets.  However, provided that they could, economists face a dilemma where telling people the truth about what government policy is doing could make the country poorer.

This conflict comes up in politics all the time.  Is it ok to use dishonest tactics to get better policies adopted?  Like end-justify-the-means problems generally, much of the problem is due to the fact that everyone considers their own ends to be worthy, but for many reasons their ends would not in fact increase total utility.

This is part of why I say err on the side of truth, but I cannot really defend this position.