[Squeakland] Panel discussion: Can the American Mind be Opened?

Alan Kay alan.kay at vpri.org
Wed Nov 21 16:23:26 PST 2007

Hi David --

I think "constructivism" (like "object-oriented") has been so 
appropriated and redefined as to be a useless term today (I certainly 
don't know what it means -- and am not sure I ever did).

I think it's much better to simply try to puzzle out the nature of 
the desired learning, and then to find workable pathways for the 
different kinds of learners. This has been done quite well in 
(admittedly simpler) areas like sports and music, where the learner 
has the distinct advantage of being able to watch practitioners and 
gain some idea of what the subject area is all about and what might 
be fun and rewarding about it.

Two of the biggest barriers to math and science learner are (a) the 
prospective learner has very little idea what the activities actually 
are (though they still might think they could be "cool" because of 
"rocket ships", etc.), and (b) there are so few real practitioners 
around (available) to help give them a sense of how to get started. 
Schools (and many adults) introduce another barrier, which is a 
profound misunderstanding of what it means to be fluent in math and 
science (the misunderstanding is usually in the form of thinking that 
math and science are fact and pattern based, and that learning the 
facts and the patterns is what is required). I've used the " 'music 
appreciation' instead of 'music' " analogy for this misunderstanding.

Once we get some sense of what "the doing of math and science" is all 
about, the main question remaining is "for learners, what is the best 
balance between doing and being advised that should be set up?". This 
is pretty well understood for sports and music for both children and 
adults. As Tim Gallwey once said about teaching tennis: "the main 
problem with traditional tennis teaching is that the parts of your 
mind that learn to play tennis don't understand English!" (Of course 
he meant that a little English goes a long way, and a lot of English 
simply can't be translated into tennis action.)

It is almost certainly the case that different subject matters (and 
different learners) can tolerate more or less of direct instruction 
in English, so it is worthwhile to get a rough assessment of this 
when trying to invent a curriculum. However, I don't think it is 
controversial to state that learning to play music or tennis is 
really about lots of actual guided and coached (and uncoached) doing 
of the activities. Most mathematicians would agree about math 
learning, and most scientists would agree about science learning.

If we look at human history, we can see that "pure discovery" 
learning by children or adults usually results in weak ideas. On the 
other hand, rote learning usually doesn't work very well for any 
subject that has some art to it. (Playing lots of scales or 
memorizing chord progressions does not a musician make.)

So there has to be discovery and creativity of a sort, and this is 
done by good teachers and writers as a kind of "guided discovery" 
(sometimes by great environmental design as in classic Montessori 
education). Perhaps the most wonderful thing about human learning is 
that something that required a genius to invent or discover (like 
calculus) can very often be learned by non-geniuses if given help. 
One of the best accomplishments of the Etoys work over the years (and 
reaching back to Seymour) is that, while no 10 year old has ever 
invented calculus, we now know how to help most 10 year olds get 
fluent in a number of the most important ideas in calculus. This is 
real progress.

I think science is the most difficult of the "new thinking" to teach 
and learn because it is the farthest from normal commonsense 
perception and thinking. It is also the most critical of human reason 
because the nice crisp logic of math is only approximately mapped to 
considerations of the actual universe (it doesn't have to work like 
our current math or brains). So just what "doing science" should mean 
for children is not nearly as clear as for sports, music or math. I 
think that the "Galilean Gravity" project that is done so well by 5th 
graders is an excellent example of one of the "real science" 
activities children should be doing. But I would be surprised if it 
and projects like it are comprehensive enough to cover all that is 
needed. Part of the internalizing of the epistemology of science 
seems to come from so many examples from so many parts of science 
that show "the world is not as it seems", but also allow some pretty 
powerful generalizations to be drawn about many of the non-intuitive 
workings of the universe.

One of the paradoxes about many kinds of learning is that you can 
learn a lot about a subject by reading after you have learned the 
subject pretty well by lots of doing. But the subjects we've been 
discussing are not often (if ever) learned above threshold without 
lots of doing to provide a foundation of deep understanding for later 
listening and reading.



At 09:32 AM 11/21/2007, David Corking wrote:
>Mark wrote:
> > Re: attempts with constructivism
> >
> > I hope you're right. I have heard criticisms of constructivism, based on
> > anecdotes, but I've always wondered whether what's been evaluated is
> > actually constructivism or just some group's ideological interpretation of
> > it (the group that says they're implementing the pedagogy, that is). I
> > haven't studied it in detail, but the ideas behind it, as presented by Kay,
> > make sense to me.
>I think it is worth studying in detail, but I am not sure where to
>start.  First I think we need to learn to distinguish among
>1. constructivism the psychological hypothesis - as proposed by Piaget
>as I understand
>2. constructivism the pedagogy
>3. constructionism - another pedagogy - and a word coined by Seymour
>Papert.  Note the 3rd syllable.
>(There is also constructivism the epistemology, which I can't even
>spell, that also originates with Piaget.)
>I recently read this unsympathetic 2003 article on the US history of
>constructivist pedagogy in maths
>But it is largely anecdotal (which is fine for a historian, but not
>when we are responsible for the education of the next generation.)
>However, beyond such material, I get thoroughly confused by an
>inability to distinguish proven knowledge, accepted wisdom, and pure
>pseudo-science.    It seems that a lot of educational research is done
>by anecdote rather than by controlled blind large group studies.  Any
>pointers to the good stuff?  Or tips to help a natural scientist to
>understand the research methods of the social sciences?
>Squeakland mailing list
>Squeakland at squeakland.org

More information about the Squeakland mailing list