The Armchair and the Lab Throne
(by guest blogger Carrie Figdor)
I've noted that the popular science press is both accomplice and victim in the miscommunication between god and mortal. But some philosophers too -- the other main class of professional skeptics -- may be in the same boat. Here a failure to be sufficiently skeptical can appear in the guise of burning the despised Armchair in favor of the Lab Throne, a.k.a rejecting intuitions as evidence in favor of empirical data (even though elicited intuitions are empirical data).
It's often hard to know exactly what is being rejected. Sometimes it seems to be the counterexample tennis (e.g. the Gettier-problem cottage industry) in which intuition gets used. Sometimes it seems to be the scope of an intuition-based claim (I am always reminded of the joke in which Tonto says to the Lone Ranger: "What do you mean "we", white man?"). Sometimes it seems to be the use of far-fetched hypothetical scenarios, although trolley problems are pretty far-fetched and what we are to make of the moral judgments (a.k.a. intuitions) they elicit is a lively subject of debate.
However, if burning the Armchair means not using intuitions as evidence, the Lab Throne will go up in flames along with it. Science depends heavily on intuition in the form of reasons that would be offered for background assumptions if such assumptions were made explicit and explicitly questioned. Why is memory cashed out in terms of encoding-storage-retrieval? Well, what else could memory be but a kind of storage? The fact that this intuitively plausible answer is only now being questioned by memory researchers seeking a more dynamical conception underlines how difficult it can be to root out the role of intuition in science.
Another way in which intuition is deeply embedded in science is in judgments about which aspects of a phenomenon (a stimulus, a task, a performance) are essential to its being the kind of phenomenon it is. One such case occurred to me during a recent talk by Stephen Stich, in which he argued that intuitive judgments exhibit wide variability based on apparently irrelevant factors (and so presumably cannot be relied on to track the truth). One example was the so-called Lady Macbeth Effect, in which moral judgments are affected by bodily cleansing. In one study, moral judgments of those who used an alcohol-based cleaning gel were significantly less severe than those of subjects who used a non-cleansing hand cream (Schnall et al. 2008).
But why think cleaning one's hands is irrelevant to moral judgment? Hopefully not because it seems intuitively plausible. My own off-the-cuff reason for doubt stemmed from generalizing from Goldin-Meadow's experiments with gesture and counting. If moral judgments are embodied, changing embodied aspects would be changing relevant features. (I should note Steve was wholly amenable to this interpretation; I also later found a similar view defended by two staff writers for the Association of Psychological Science.)
In this particular case, the moral is that in at least some cases the problem with Armchair intuition is that it is disembodied, not that it is a priori. Moreover, some critics of intuition may simply presuppose a disembodied (Cartesian?) view of intuition. (Why?) In general, the problem with intuition seems to be that different disciplines have different Armchairs, and the Lab Throne is just another style of Armchair -- and philosophers who prefer Lab Throne style should be especially critical of that particular Armchair.