Marvin Minsky Comment on Schooling

Subject: interesting comment on schooling from Marvin Minsky
Date: Wed, 29 Jun 94 17:40:56 -0400
From: dm@hri.com
Reply-To: home-ed@world.std.com

In the July 1994 _Communications of the ACM_ there is an interview with Marvin Minsky (former director of the MIT AI lab, widely considered to be one of the fathers of artificial intelligence, and an occasional collaborator of Seymour Papert's).

Among other things he says (on page 26):

(Interviewer): ...For an older student in a conservatory, we can imagine having to study Gregorian chants for a few months before getting any highly (positive) feedback. But in the case of a five-year-old child learning piano or composing, we cannot depend only on delayed feedback or abstract feedback.

Minsky: I'm afraid that's true, at least for most young children, but the evidence is that many of our foremost achievers developed under conditions that are not much like those of present-day mass education. Robert Lawler just showed me a paper by Harold Macurdy on the child pattern of genius. Macurdy reviews the early education of many eminent people from the last couple of centuries and concludes (1) that most of them had an enormous amount of attention paid to them by one or both parents and (2) that generally they were relatively isolated from other children. This is very different from what most people today consider an ideal school. It seems to me that much of what we call education is really socialization. Consider what we do to our kids. Is it really a good idea to send your 6-year-old into a room full of 6-year-olds, and then, the next year, to put your 7-year-old in with 7-year-olds, and so on? A simple recursive argument suggests this exposes them to a real danger of all growing up with the minds of 6-year-olds. And, so far as I can see, that's exactly what happens.

Our present culture may be largely shaped by this strange idea of isolating children's thought from adult thought. Perhaps the way our culture educates its children better explains why most of us come out as dumb as they do, than it explains how some of us come out as smart as they do.

- dave mankins (dm@world.std.com, dm@hri.com)