Re: [Gnu-arch-users] OT: trained dependency

[Andrew Suffield]

On Tue, Nov 09, 2004 at 02:28:49PM +0900, Stephen J. Turnbull wrote:
> >>>>> "Thomas" == Thomas Lord <address@hidden> writes:
> 
>     > From: "Stephen J. Turnbull" <address@hidden>
>     >     Andrew> The Japanese school system is exceptionally good, by
>     >     Andrew> comparison to other countries.
> 
>     > You couldn't prove that by my students, and they're in the top 10%
>     > nationwide.  If this is "exceptionally good", then it's not just
>     > American education that has problems.
> 
>     Thomas> Could you, not necessarily being overly specific, describe
>     Thomas> some general trends that characterize how you think your
>     Thomas> students are underprepared or have a poor approach to
>     Thomas> academics?
> 
> "Can't sing, can't dance, can't act, they're no fun."
> 
> They're not underprepared; they're overprepared.  In fact, they're
> cooked to the point where all the juice has evaporated.  Japanese
> students know how to sit at desks, look at books, and memorize huge
> amounts of material.  They have huge amounts of information stored,
> but no concept of how to wrap that expressive material around original
> thoughts.

This is the normal perception of graduate students by industries
everywhere. Probably deservedly; most people have no more than a
handful of original thoughts in their entire lives.

I have never seen any education system which performed appreciably
better at this. As best I can tell, it's normally left to natural
aptitude. I've never seen any evidence that there is a practical way
to teach this sort of thing.

> The result is a nation whose high school students regularly rank at
> the top of the OECD in science and math competitions (though they get
> the tar beat out of them by the various Chinese and the Koreans), but
> whose rate of Nobel Prizes in science per capita is the bottom of the
> OECD, and 90% of the cited research was done abroad.  Full professors
> at the "best" universities make whole careers out of very mechanical
> efforts: get more data on a different sample, do the same calculations
> as the last paper, publish in house journal or in journal of academic
> association your PhD advisor is president of (or you are president
> of).  Lather, rinse, repeat.  No joy, anywhere.  It's a job....

I would say this is a cultural effect rather than the result of the
education system. It's what Japanese culture expects people to do, so
it's what they do. (I expect you're familiar with what I'm referring
to). There is nothing that the education system could do better to
change this (taking as a given that it can't suspend normal cultural
influences).

In essence, a highly effective education system is spectacularly let
down by cultural constraints that inhibit people from doing anything
useful at the highest levels with their education.

> You know those exam questions about "read this passage and summarize"?
> They don't have those on the mass exams, so typical Japanese students
> simply can't do it.  They're very good at sed'ing literal answers out
> of the text, but they're lost if the task is taking three or four
> sentences from the original and smoothly combining the premise of the
> first with the conclusion of the last.  Answers to such questions
> invariably are selected near quotes from the text, with large gaps,
> not a coherent summary at a higher level of abstraction.

Summarising information is a learned skill, and a fairly mechanical
one. It's not a particularly good measure of anything else (despite a
number of popular testing systems which are based on it).

> Of course, there's wide variability, and there are many people
> (including students) with interesting, thoughtful things to say.  But
> the system provides no place for them to say them, they say them far
> more freely to us gaijin than to other Japanese (and thus they don't
> get systematized and refined by broad discussion), and students have
> few good role models for original thinking.

<snip continuation>

Yes, I think you've summarised Japan's primary failing (in overall
terms, not just relating to this subject) quite effectively.

-- 
  .''`.  ** Debian GNU/Linux ** | Andrew Suffield
 : :' :  http://www.debian.org/ |
 `. `'                          |
   `-             -><-          |

signature.asc
Description: Digital signature