Dan Moerman kicked off and pushed a particular minimal line. Placebos are inert substances. So although they seem sometimes to have effects, it isn’t because of the placebo because they are, by definition, inert. So why? He suggested that this might be a matter of regression to the mean because of homeostatic mechanisms. Or natural history: the resolution of viral colds in about a week. But these are some cases which do not reduce to those. Since in his phrase ‘there is never nothing going on’ in such cases he suggests that there is a ‘meaning reaction’.
He then proceeded to outline his favourite dozen or so published studies of meaning responses. These included cases where cases where patients were given doses of pain killers, some knowing and some not. In such a case, there was no placebo and hence could be no placebo effect. But there was a meaning response. In other cases, a drug that blocks opiates also blocked meaning-response reductions in pain, too. Perhaps most helpful for me was a study in which stomach ulcers were cured through a meaning response. In other words meaning responses (or placebo effects if one speaks that way) need not be merely mental.
Catherine Blease extracted some ideas of what placebos and placebo effects might be from the recent philosophy of science literature. She favoured an idea based, I think, on an expectation response and then outlined an evolutionary theoretical account of why there might be such a response based on the idea that an evolutionarily selected mechanism connected to status detection and requests for help. I wasn’t sure whether this second element of the talk was supposed to be an independent explanation of an assumed concept or whether, additionally, it served as a condition of adequacy of any definition of placebo that such an account must be possible: an expression of a prior metaphysical commitment. Perhaps it didn’t matter.
David Teira asked whether the relevant contributions of different elements of the placebo effect could be measured, contrasting merely being on a waiting list, receiving a sham treatment, and receiving such a treatment with additional interpersonal factors. The gist of the presentation was that this could only be researched by a RCT in which social factors could be standardised. But from any perspective on meaning, there are reasons to hold this cannot be done. I was left with a kind of confusion about how surprising this should seem. It seemed to me that it relied on an implicit comparison between placebos and drugs and hence deploying the methods for assessing the latter to the former. But one might think of placebo effects merely negatively: as what disrupts the drug treatment RCT not as a something with a general pattern to be tracked. (I don’t think that there could be a RCT about the reasons why relationships break up. No talk of ‘the reasons’ would imply that they should generalise.)
Phil Hutchinson gave a paper which connected the placebo response to emotions. Both of which were meaning responses (he cited Moerman) pitched at a level higher than merely brute causation but lower than fully propositionally thought. One motivation was to avoid the worry that that latter position ruled out animal emotion. His positive suggestion was that there could be a middle ground in which humans or animals ‘take in’ the world but short of propositional thought. This had conceptual structure and involved taking it in under aspects. I had all the usual worries. What kind of concepts are involved given that various intensional differences won’t be available? And what is the aspect under which animals and people see the world if not ordinary concepts? All of this was more puzzling because Phil denied that thoughts have propositional structure. So I was less clear as to what analytic role talk of ‘concepts’ played.
An enjoyable day wrapped up with pleasant conversation over a beer in a micro pub, sadly a sweaty dash from the station afterwards.
