Skeptical Empiricism and Theory Building: The Search for Mechanisms, rough cut

A better term for the hyperrationalism I’ve been talking about would be…well, hyperrationalism, but even better: Platonicity.  Plato said it is irrational that we prefer the use of one hand over another.  Of course it’s “rational” in terms of the universe we’re in (which exhibits ‘handedness’ in chemistry) and how our brains evolved (to be asymmetric for a reason).  I may return to this in more detail.  Taleb defines Platonicity as “the focus on those pure, well-defined, and easily discernible objects like triangles, or more social notions like friendship or love, at the cost of ignoring those objects of seemingly messier and less tractable structures.”  (He doesn’t ever note the irony involved in Platonicity, which postures as being as rational as possible but it’s actually grounded in focusing on the “easily discernible” over the messiness of life as it actually exists 99% of the time.)

Taleb’s book failure is the failure to follow through on what he presents.  He effectively trashes much of academia (though he fails to show the limits of the more directly empirical sciences like physics, where Platonicity manifests more indirectly).  He shows how easily our theories and stories about the world are trash.  Most scarily, he doesn’t just show that for predictions of the future, but our understanding of the past.

In the end though, his answer really boils down to “learn the biases [common errors] and heuristics [shortcuts that create holes] in your cognition and develop practices to avoid them” and he gives some examples, like not giving your money to a mutual fund manager.

But what would he replace the academy with?  What research programs would he devise?  What theories of the world (other than the theory that we have certain tendencies toward certain errors) would he construct and with what method?

He has shown that understanding the past and predicting the future are even harder than most of us realized, and he has even perhaps narrowed the limits of what we could predict and understand even if “perfect.”  However, by identifying the mechanisms that distort our understanding, he has made it possible to be more precise in our theories and models of the past and future.  While what we can predict may be more limited than thought before, using his tools, it can be strengthened.

But he shies away from grasping at the reigns of history.  Wimp!

One thing he taught me about theory building relates to the search for mechanisms.  

A Platonist would reject something whose mechanisms seem impossible according to his theories.  At least an average skeptical empiricist may accept something that seems to work yet conflicts with his models of the world but would also not attempt to figure out why that is (or Taleb seems not to).

Take homeopathy.  The mechanism by which it claims to work (diluting a substance with water so much that it’s impossible that any significant amount of the substance is left in the ‘medicine’, taking which is supposed to help the person) is scientifically impossible.  However, there is some evidence that it “has an effect over a placebo.”  The harder the investigation, the less evidence.

Now turn to antidepressants.  The mechanisms by which they claim to work are scientifically plausible, but only plausible.  They’re actually little understood and disputed.  And a recent review of the studies as to effectiveness show that in a majority of studies for most patients the difference from placebo was minimal (and it should be noted that homeopathy has no direct side effects, whereas antidepressants have many, including suicidal urges).

A skeptical empiricist could say the “superstition” of someone following homeopathy would’ve protected him from the rational idiocy of something like antidepressants.

But I want to propose a mechanism they have in common: the relationship of the client and healer.  I propose that it’s that relationship that produced the positive effects (including from the placebo, especially in the depression studies: not the belief per se that the pill was making them better, but the hope that someone was caring for them).

Now, the empiricist says, “Where’s your data?”  I would point to the qualitative but indirect data of how humans evolved in close-knit communities, how babies can die if given their “physical” needs but are not held, etc.

I think a better approximation is (from the overcoming bias blog): “How does this constrain your expectation?”  In other words, if a statement can equally explain or not explain something, it’s nonsense.  How would the world be different if my proposition is wrong?  One could then design experiments based around my supposition.  I think Taleb’s method fails to show how to get to this point.  How to develop a theory in a way that is aware of the common errors and takes steps to prevent them: a back and forth between empirical (concrete) and theory (abstract).

How does one decide what medicines to take?

The skeptical empiricist will outlive the Platonist.  But the anti-Platonist theorist will save both their lives.

But what that means could be developed more.  Saying a balance or back and forth between concrete and abstract is not enough.  A big step is 1) taking into account the anti-biases of Taleb and 2) passionately striving for the grand theories, for taking up the reigns of history, anyways.  Something like that…perhaps.

Tags: Life philosophy, platonicity, rationalism, science