Wednesday, June 10, 2015

Why it's hard to reform Harvard Trained Economists or Crack Whores - The Case for Epistemological Modesty

Something I wrote back in 2014.

Okay I'll admit that writing a piece on epistemological modesty demonstrates a certain lack of.....epistemological modesty. But Emodesty (oh please God, don't make me write epistemological again) is an essential ingredient in any constitutional republic or indeed any well functioning society.

So what is Emodesty you ask? It's the recognition that we fallible human beings know rather less than we think we do. There is a natural tendency to explain things that happen in causal "just so" terms: that glass full of vodka smashed on the floor because you were smashed on your feet. You see: cause and effect. Yet that example is flawed: the glass didn't smash just because of your BAC level was north of .3, it smashed because of a whole expanding tree of causality: the glass was glass, the temperature was south of 1500C so the glass was brittle rather than molten, the glass was held in your hand at a height sufficient to gain breakage velocity, the floor was tile rather than rubber, your hands were slippery from the soap you failed to rinse off and so when your smashed mind caused your normally modest gesticulation to explode into a grandiose gesture the friction between your fingers, the soap on your fingers and the condensation on the glass and the glass fell below some critical level that some poindexter could explain but I can't, leading the glass to first go up and then because of the Earth's gravity come flying back down above critical smashage velocity whereupon you were sent home in a cab to sleep it off. See what I mean? It was the damned floor, not six vodka tonics that caused your girlfriend to break up with you.

And I haven't even gotten to the most important reason we need Emodesty: human agency. When you throw humans into a causal tree - when they're part of the sordid "just so" story, so to speak - their dynamic responses to changes in their environment make prediction vastly more difficult.

This is why we have gotten pretty good at predicting tomorrow's weather but can't predict whether the stock market will go up in the next minute, hour or day much less what your date is going to do if you 'accidentally' touch her boob. Natural systems lack any sentient control or feedback loop (well, except when God wants to smite someone) and therefore in theory with enough information and processing power the causal, deterministic tree can be mapped forward from a known state with some level of statistical accuracy. Not so with us humans: we respond to changes in the environment which cause other humans to respond which causes us to respond to their response and so on and so forth. Thus, human mediated processes are far more dynamic and far less predictable or manageable than natural or synthetic systems.

There are several critical drivers of this 'random walk' result:

The information problem - Fred Hayek, a Nobel Laureate from Chicago (my fair school) won his Swedish Gong for explaining why central planners couldn't plan their way out of a wet paper bag. It turns out that there is an immense amount of information that each of us process to make economic and social choices every day. This information is distributed and particular to us and as such, is unavailable to any central authority that seeks to 'optimize' a process to achieve a 'desired' outcome 'for' us. For example, the simple task of figuring out how many T Shirts of which colors, styles, sizes and assortments will be demanded in which locations at what price at what time of the year is beyond any central planner's ability to compute. However, a market composed of many distribution, manufacturing and selling points interacting with millions of customers every day figures it out, even if there's always some ridiculous end of year blowout sale to move the chartreuse XXXL women's half tees with the bare midriff that buyers were sure was going to take the big 'n hefty market by storm last winter.

The just so fairy tale problem - when we seek to manipulate others, we put ourselves in their shoes without being them. So a Harvard trained economist (God help us all) pretends that he's a crack whore on the mean streets of Scarsdale...wait, Hell's Kitchen and then explains how his pet theory or project or more likely manipulation will cause this crack whore cum Ivy League honors graduate living in Scars...Hell's Kitchen to choose the sunny uplands of purity, chastity and rare single malt scotches rather than the crack pipe, Antoine and his list of nice 'friends' who "just want to spend a little time with you". Which works just about as well as a crack whore modelling Biff and Muffy's marital reconciliation which isn't going well because Biff's a real whiffenpoof and Muffy's preference for small batch Gin turns her into a martini whore. We all like fairy tales, particularly ones where we manipulate others to do what we think they ought to do. And like all good fairy tales, we (the prince, or princess or frog who becomes one) always end up getting what we want the way we want it, even though it's not us doing any of it, even the wanting.

The dynamic response problem. Even if we could get all of the information that the crack whore had at a given point in time and even if we were a crack whore who understood crack whores and knows all about Antoine and his so called 'friends' we would still struggle to predict how crack whore #1 would respond to interventions by (Harvard trained) crack whore #2. This is because in human relations every action has a reaction that is driven by so many unknowable factors that it's as if in our smashed glass example the floor kept flipping from tile to cotton wool while gravity fluxed from none to Jupiter's and we cycled between the alcohol consumption of Mormons, backsliding Baptists and Welsh poets.

This is why people who try to use the state and it's immense power to take, hurt and humiliate others into being 'good' or at least 'better' shouldn't. Because inevitably they end up doing things that lead to outcomes that were not predicted in places that they didn't expect that have consequences that weren't in the spreadsheet that Biff typed up to hide his list of glee club 'friends' from his drunk whore of a wife. Frederic Bastiat - the last good French economist (yeah, I'm lookin' at to you, Piketty) called this 'Things seen and things unseen' - the tendency of of interventions into human interactions (which is all an economy or glee club or S&M bar is, really) to generate unpredictable outcomes elsewhere that swamp the so called benefits of the intervention and make the intervener look stupid to the people he is trying to manipulate even if he gets pelted with praise from the other Harvard trained economists (God help us all).

And tragically, our fair Government in the Imperial Capital of Washington DC is just one big fricking manipulation. Indeed a better name for the place would be 'Manipulations R Us' because toying with us is what they specialize in. And sadly, most people, drowning in their own 'just so' logic and towering self righteousness lack the epistemological modesty - hell, they can't even spell it or know what it is, much less have it - to restrain the pack of thugs, pugs, mugs, looters, shooters and Senators that infest the capital like tapeworms.

Their motto, indeed our glorious Federal government's motto is "make them pay". Or maybe it's "gimme, gimme, gimme" - either way it's bad.

In future installments I shall give those willing to undertake the arduous process of recovering their EModesty some rather juicy examples. Recovering one's EModesty is sort of like going to AA to recover your sobriety: it's a lot of sitting around and saying "Hi, I'm Bill and I'm a self righteous, blinkered, manipulative bastard" and then the crowd goes "Hi, Bill". Sadly, there are virtually no EM groups active in Washington DC or Cambridge MA at the present time. God help us all.

No comments:

Post a Comment