I'm not sure if what's going wrong with, say, the Scocca piece is just bad analogical reasoning. I trust you that The Genetic Lottery isn't similar enough to ye-olde race science for that to be plausible. Rather, Scocca's "reasoning" seems to be abductive -- a bit of ye-olde leftist "hermeneutics of suspicion." For the suspicious reader, the text is treated as a symptom for some deeper, hidden syndrome. If you're Marx-inclined, that syndrome is capitalist ideology: a naturalization of the injustices birthed by our economic order. So we see Scocca write.
"Whatever higher purposes an individual researcher may have in mind, there is only one question the phrenology business has ever sought to answer: Isn't it right that things are the way they are?"
In its most lazy expression, such a hermeneutics of suspicion becomes an excuse to ignore the surface of the text in favor of its imagined depth. Scocca doesn't feel the need to offer a close reading of The Genetic Lottery: it is enough to place the author and the work amid a constellation of known suspects, as a conspiracy theorist might pin a photograph to a board lousy with string. At its most all-consuming expression, such paranoid reading becomes a replacement for politics. As Scocca closes his piece.
"The question the neo-phrenologists are really incapable of answering about their chosen work isn't even "How?" It's "Why?" or, more bluntly, "Who gives a shit?" We know—objectively, factually, beyond the shadow of a doubt—that our educational system, like our society at large, is unequal. We know that poor children and nonwhite children are sent to worse schools. We know that they live under conditions of greater stress and deprivation, which interfere with learning. We know they receive less individualized attention, less support, and more hostility when they struggle."
Never mind that good answers can be given to these questions. (For instance, if we knew that little Timmy wasn't going to grok multivariate calculus no matter what, we could put our money to better use than endless tutoring sessions.) If you're truly paranoid, it is scarcely worth investigating the world as you already know ("objectively, factually, beyond the shadow of a doubt") what the problem is.
***
If something like this is what's going on with Scocca, then I doubt that Bayes can help. This is because Bayesians either lack a good answer to or even reject the question "What should my priors be?" The theory only tells us, given our priors, what our posteriors should be in light of this evidence. But it is fully possible to have paranoid priors. If I believe, with near certainty, that everyone is out to get me, then the fact that it doesn't look like anyone is out to get me will do little to shake this certainty. Rather I'll just believe that everyone is pretending to act normal so that I'll let my guard down (the better to get me).
"Rather, Scocca's "reasoning" seems to be abductive -- a bit of ye-olde leftist "hermeneutics of suspicion." For the suspicious reader, the text is treated as a symptom for some deeper, hidden syndrome."
That's probably true, but I'm responding more to this "genre" than this piece in particular. I see a lot of "similar too" sort of reasoning out there, and I think it's wrong.
As for the hermeneutics of suspicion itself, I guess I just can't communicate with people who use that. If you're starting with a set of axioms that I don't agree with, I don't really see what I can do.
I agree that you are not likely to have much luck reaching people with substantially different priors. I might be more pessimistic in that I don't think we're likely to have much luck in reaching people who reason badly: pointing out that an argument is bad will, if you're lucky, undermine the argument, but it quite probably won't undermine the conclusion (especially if the argument is just a bit of motivated reasoning rationalizing the conclusion); more generally, getting better at constructing / recognizing decent arguments might well make you not anymore rational, but just better at rationalizing.
I really enjoyed this!(Cricket, “baseball” haha) And the point is well taken! All the “phrenology” stuff in the linked book review was maddening. The link to the Slate Star Codex post about teams and outgroups was interesting too!!
The Scocca piece is maddening, and it's just one example of the archetype. Scientific American published a piece about EO Wilson that made the same error.
I like SSC, but whole rationalist community seemes focused on Bayes and biases. I'm really surprised there's no popular writer that I know of (except maybe Singer) who focuses on basic analytical philosophy, ie what are the premises and conclusions. Maybe because it's boring, or maybe because many topics are too complicated to be broken down like that.
Well, analyzing premises and conclusions can be very useful indeed to point out flaws and show what something isn’t; but off the top of my head (so, this could be wrong) it seems like many problems are too complex to use premises / conclusions to show what something _is_.
I'm not sure if what's going wrong with, say, the Scocca piece is just bad analogical reasoning. I trust you that The Genetic Lottery isn't similar enough to ye-olde race science for that to be plausible. Rather, Scocca's "reasoning" seems to be abductive -- a bit of ye-olde leftist "hermeneutics of suspicion." For the suspicious reader, the text is treated as a symptom for some deeper, hidden syndrome. If you're Marx-inclined, that syndrome is capitalist ideology: a naturalization of the injustices birthed by our economic order. So we see Scocca write.
"Whatever higher purposes an individual researcher may have in mind, there is only one question the phrenology business has ever sought to answer: Isn't it right that things are the way they are?"
In its most lazy expression, such a hermeneutics of suspicion becomes an excuse to ignore the surface of the text in favor of its imagined depth. Scocca doesn't feel the need to offer a close reading of The Genetic Lottery: it is enough to place the author and the work amid a constellation of known suspects, as a conspiracy theorist might pin a photograph to a board lousy with string. At its most all-consuming expression, such paranoid reading becomes a replacement for politics. As Scocca closes his piece.
"The question the neo-phrenologists are really incapable of answering about their chosen work isn't even "How?" It's "Why?" or, more bluntly, "Who gives a shit?" We know—objectively, factually, beyond the shadow of a doubt—that our educational system, like our society at large, is unequal. We know that poor children and nonwhite children are sent to worse schools. We know that they live under conditions of greater stress and deprivation, which interfere with learning. We know they receive less individualized attention, less support, and more hostility when they struggle."
Never mind that good answers can be given to these questions. (For instance, if we knew that little Timmy wasn't going to grok multivariate calculus no matter what, we could put our money to better use than endless tutoring sessions.) If you're truly paranoid, it is scarcely worth investigating the world as you already know ("objectively, factually, beyond the shadow of a doubt") what the problem is.
***
If something like this is what's going on with Scocca, then I doubt that Bayes can help. This is because Bayesians either lack a good answer to or even reject the question "What should my priors be?" The theory only tells us, given our priors, what our posteriors should be in light of this evidence. But it is fully possible to have paranoid priors. If I believe, with near certainty, that everyone is out to get me, then the fact that it doesn't look like anyone is out to get me will do little to shake this certainty. Rather I'll just believe that everyone is pretending to act normal so that I'll let my guard down (the better to get me).
"Rather, Scocca's "reasoning" seems to be abductive -- a bit of ye-olde leftist "hermeneutics of suspicion." For the suspicious reader, the text is treated as a symptom for some deeper, hidden syndrome."
That's probably true, but I'm responding more to this "genre" than this piece in particular. I see a lot of "similar too" sort of reasoning out there, and I think it's wrong.
As for the hermeneutics of suspicion itself, I guess I just can't communicate with people who use that. If you're starting with a set of axioms that I don't agree with, I don't really see what I can do.
I agree that you are not likely to have much luck reaching people with substantially different priors. I might be more pessimistic in that I don't think we're likely to have much luck in reaching people who reason badly: pointing out that an argument is bad will, if you're lucky, undermine the argument, but it quite probably won't undermine the conclusion (especially if the argument is just a bit of motivated reasoning rationalizing the conclusion); more generally, getting better at constructing / recognizing decent arguments might well make you not anymore rational, but just better at rationalizing.
I really enjoyed this!(Cricket, “baseball” haha) And the point is well taken! All the “phrenology” stuff in the linked book review was maddening. The link to the Slate Star Codex post about teams and outgroups was interesting too!!
The Scocca piece is maddening, and it's just one example of the archetype. Scientific American published a piece about EO Wilson that made the same error.
I like SSC, but whole rationalist community seemes focused on Bayes and biases. I'm really surprised there's no popular writer that I know of (except maybe Singer) who focuses on basic analytical philosophy, ie what are the premises and conclusions. Maybe because it's boring, or maybe because many topics are too complicated to be broken down like that.
Well, analyzing premises and conclusions can be very useful indeed to point out flaws and show what something isn’t; but off the top of my head (so, this could be wrong) it seems like many problems are too complex to use premises / conclusions to show what something _is_.