Wednesday, October 22, 2008

Intelligence - Part III

I am not writing this series because I am a world-class expert on the subject of what constitutes intelligence, but because my views are slowly changing. (I might be a county-class expert on intelligence). Had I not read Taleb’s The Black Swan, I doubt I could have accepted the Palin candidacy with grace. My son Ben is tired of the Palin discussions, and I promise not to stay on the topic long. I fully admit that if the Democrats had done something similar, I would be more likely to consider it a stunt and less likely to consider her candidacy seriously.

But the thought was already growing in my mind that defining pure intelligence as g-factor, that patern-recognition, analogy creating, platonicity was a limitation. I had long been suspicious of top-down central planning, which always looks so plausible but has a poor track record. Taleb’s writing, pointing out the besetting dangers of such intelligence took me a step further. “Smart” people create a different type of errors. They become enamored of a pattern or theory and push it forward in spite of contrary evidence.

Well, sometimes that works. Confidence and perseverance can be good things. But it can also be catastrophic.

Some systems are so complex that patterns do not hold very well. There are weather cycles that are fairly consistent and predictive – seasons, El Nino, monsoons – but even these admit of considerable variation. There are too many interactive variables to make solid predictions. We have gotten much better, but still have surprises.

The economy is even more complex. One of the reasons that Taleb has contempt for most economists is that they believe they can reduce complicated, unpredictable events into some manageable form. They see patterns that aren’t there, or more precisely, are there only in certain circumstances or for a limited period of time. Economists overestimate how well the predictors will work. It is not that there are no patterns, or that there is no value in their knowledge, but that acting as if we know when we have overrun the limits of our knowledge in a crisis is dangerous.

A reasonably smart person who doesn’t have a pattern locked in her head, who is simply looking at the situation cold and trying to manage or adjust the crisis, will do better than the smarter person with a Theory they can’t let go of. The former won’t always get it right either, and will make some errors that the Theory person wouldn’t. The Theory people will then articulate what went wrong and how they would never make that mistake, and convince us that we should stop listening to those seat-of-the-pants people and listen to them, the experts. And this will work great until it stops working, at which point we will have…

You know what we will have: a hundred economists choosing up teams and telling us a dozen different reasons why we had a financial crisis. Each of those explanations will sound entirely plausible while you are listening to it. You will know with assurance that if you got into a live argument with any one of these people they would mop the floor with you, because they know this much better than you do. You would have to leave the debate floor in ignominious defeat, revealed as a fool. And yet… and yet I didn’t recently cost the American people $700,000,000,000, did I? No, it took smart people, Nobel prize winners in economics to do that. One of the small comforts of this whole crisis is also one of the most depressing aspects. There was very little criminality, very little fraud, in this whole mess. People stretched rules, but they didn’t break them. It was bad ideas that brought us to this pass.

No, it was good ideas, arrogantly imposed beyond their usefulness, which brought us to this pass. It wasn’t that these subprime mortgages were all bad loans, which anyone could have seen. Most of them have not defaulted, and will turn out to be good loans. Most of those people are still making payments. They will be homeowners and have escaped from the cycle of poverty (if you’re a Democrat) or become part of the ownership society (if you’re a Republican). It’s just that as a group these loans defaulted at a higher rate, and because we’re not sure what the final numbers on that will be – thirty years from now – we don’t know what complex derivatives that include them are worth, so no one wants to buy or sell them anymore.

Off the rant, on to the general discussion.

Pattern recognition is intelligence*. So is pattern elimination, but it’s usually much less fun. There is a certain emotional kick we get from insight, of believing we have understood some new portion of the chaotic world around us. We like that feeling so much we strive for it. We seek it out, latching on to plausible explanations. Hence, new diets. Hence, alternative medicine. Hence, management techniques, Christian living seminars, EMDR, Values Clarification, new math, tough love, and every other one-step solution that we like.

Eliminating a pattern can have the opposite emotional impact. We move from a place of assurance to a place of doubt, exposing ourselves to the sharp winds of reality. There can be compensations, of course. Eliminating someone else’s pattern, or seeing through a common misconception can be very satisfying. Yet this is chancier. We are still then left staring into the chaos, squinting to see what new pattern might be there, now that the old one is gone.

Some professions actively encourage pattern destruction as well as pattern creation in their training and practice. Medicine is one, most types of engineering another. Anthropology is notoriously resistant to abandoning ideas. New ideas are admitted for discussion, but their proponents usually have to wait until the old guard dies off before they get control of the introductory textbooks. People don’t change their minds a lot. They have their pet theory that they are known for or have sided with, and all their effort goes into defending their idea and attacking the others. This is also true of the arts and literature. A professor who reads from a marxist feminist perspective in 1982 will still be reading from a marxist feminist perspective in 2002 and 2022. She will have added and subtracted a few things along the way, but the central premises will be intact. Religion is a mix, showing both faddism and stability.
Foreign policy is an area where different groups or schools of thought have theories they believe should be adhered to. America should never go to war unless…We should be willing to trade with any nation that…Dictators always…Eastern European countries want us to…All of these are great examples of patterns that work great every time until they suddenly don’t. When patterns suddenly don’t work, the most dangerous person in the room is the one who says “But it has to work. Try it again. Again. It will work eventually if we just keep at it.”

I have a rule of thumb that an expert in any field is defined as one who knows when you can force something, and when you absolutely should not force it. Plumbers, salesmen, coaches, therapists, designers, parents, poets – the expert is the one who knows, partly on the basis of principles but equally on the basis of native intelligence, experience, or intuition, that the standard rules don’t cover this situation.

* Thus the standardised test questions Horse: Colt :: Cow: ? These are pattern questions. Even fill-in-the-blank vocabulary or What Is The Best Title questions are pattern questions, illustrating that you pick up the patterns of language use. The tests themselves are patterns of a subtler type, which is why “test-taking ability” is not necessarily an unfair advantage. After taking a few of these things, you should be picking up what the rules of fair questioning are. When they are asking you for the best answer, they are not going to throw you an all-of-the-above or none-of-the-above. It’s too ambiguous. It’s not clean. This is why these tests measure one type of intelligence very well, but fall short in predicting how people will do in ambiguous situations. As everything in life is patterned and predictable until the day that it isn’t, that’s a serious limitation of the tests.

8 comments:

  1. It's actually closer to 1.4 trillion. They added a whole bunch of other stuff, secretly.

    ReplyDelete
  2. Anonymous6:44 AM

    Your rule of thumb is probably what most people live by. Patterns are there in most human affairs and should be looked for and recognized. But, shouldn't discernment be the guiding light? Testing predicts, but discernment tells us Mrs. Palin is needed at home not in politics. Who will care for the newborn with Down Syndrome? Shouldn't the mother? What about her daughter; teenager, pregnant, unmarried, thrust onto the World Stage? Doesn't she need her Mom?

    McCain is running on the pro-life, pro-family, pro-good-guy, political "schtick," yet, he chooses to rip a mom from her family for personal and political reasons. I'll vote for someone else.

    You have a great blog. Keep it coming.

    Mason

    ReplyDelete
  3. McCain is running on the pro-life, pro-family, pro-good-guy, political "schtick," yet, he chooses to rip a mom from her family for personal and political reasons. I'll vote for someone else.

    Is this a serious comment? Or, is this tongue in cheek?

    I hope it's not serious, because then I would get all riled up and say that Palin, as a woman, makes her own choices and is not a passive agent that McCain has some sort of power over...expressed in an ability to "rip" Palin away from her family.

    What a narrative--helpless woman dragged off by uncaring man. ugh.

    Anyway....before I became distracted...I wanted to ask AVI if true intelligence could be determined by having a high problem-solving ability; the ability to turn on a dime and find a solution in a crisis, using what's at hand, in unconventional ways.

    I've met some terribly intelligent people who were terrible, general problem-solvers.

    Can such an ability be measured?

    ReplyDelete
  4. Good post, good series. I've not yet read Black Swan, and admit to ascribing to the "pattern recognition" theory of intelligence. Yet, as you say, false positive pattern recognition is likely to have worse consequences than the drunken walk of an idiot.

    Especially in view of terri's question, What does the book, or other reading, say about the type of person who reads patterns right? History is replete with huge men with multi-year successes (Napoleon comes to mind) who make inexplicably bad decisions (invading Russia) that undo the string of good choices. Is that inevitable? Is it simply impossible to expect anyone to hit .406 again?

    ReplyDelete
  5. terri, as it is so quickly off-topic I took mason's comment as a bit calculated. There is a certain automatic quality to it, of a person listing talking points quickly rather than entering into a discussion.

    carl, Williams himself said that .406 might not be reached because pitchers had perfected the slider, a nasty pitch he seldom had to deal with. He did note that people don't hit even .375 with power, though.

    As to Taleb's more general points, he argues in favor of resisting patterns because they all eventually mislead. He doesn't object to people using them cautiously and judiciously, but prefers to err on the side of thoroughgoing empiricism.

    ReplyDelete
  6. I enjoy your posts and find them thought-provoking. My comment here only concerns the first paragraph of your current post: not to ignore the remainder, but in order to obtain focus.

    I do not understand what you meant by "I fully admit that if the Democrats had done something similar, I would be more likely to consider it a stunt and less likely to consider her candidacy seriously." Done what similar---and how?

    In my own case, Gov. Palin had made my "short list" of desired VP candidates sometime last June. I had become impressed with her dealings with the Alaskan legislature, her oil pipeline negotiation, her declining the "bridge to nowhere" earmark, and several other things. While I think that there may have been others more competent for the job (Sen. McCain was not my first choice for presidential candidate either!), she was definitely within the realm of adequacy, and had good credentials as far as desireability went.

    A very significant criterion for the Vice Presidential candidate choice concerns electability. My first choice in the primaries, Gov. Romney, unfortunately, despite all his experience and history, would harm the ticket due to the prevalence of anti-Mormon bigotry. [Disclosure: I am Christian, not Mormon.]

    I think Gov. Palin does well, and I am at a loss to understand some of the criticism from people like Noonan, as well as AVI. I'd like to understand it---usually I can understand why someone argues a certain way, even if I disagree---but I'm having difficulty here. The "family first" argument is one I especially don't understand; perhaps because I think each family is in the best position to determine things for themselves. [Aside: my wife was a full-time mom until the kids were in high school, and then has always been off work by the time school got out. But that's us.]

    ReplyDelete
  7. One problem with most testing is that it only tests small-scale pattern matching, not large-scale matching. They are certainly related but not tightly coupled -- I suspect that one shows Intellect, the other Common Sense. Both are useful. Holistic thinking is much less common than intellect, almost as rare as creative, "out of the box" thinking really is (note how that ties into your thesis, too -- knowing when to break the pattern)

    The thing is, we need more such thinkers.

    Heinlein referred to them as "Encyclopedic Synthesists". More recently, they've been called Systems Engineers and Operations Analysts.

    But the idea is to construct working paradigms in your head for a complex set of operations. And that's certainly not the same as small-scale pattern matching. Tactics is not Strategy, much less Grand Strategy.

    ReplyDelete
  8. If the Democrats had nominated someone whose appeal was clearly geared toward the common folk instead of the Washington movers and shakers - a feisty liberal mother of three (but pro-choice, of course) who had gotten a community college degree, belonged to a government union, and run for congress - I would have a hard time looking beyond the PR value of that to her qualifications.

    NRO did a piece on Palin last February and I liked what I read, though it was limited. I still like to brag that I was for Sarah Palin before I saw a picture of her.

    ReplyDelete