From: "Richard Brodie" <richard@brodietech.com>
To: <memetics@mmu.ac.uk>
Subject: RE: Memetics not tautological or circular
Date: Fri, 4 Jun 1999 07:20:50 -0700
Derek,
I am overjoyed this morning because I finally think I've spotted the
difference in our models.
Derek wrote:
<<What memes do you infer from an achievement test? Suppose the test asks:
"What is the capital of Turkey?" You get the right answer straight away,
but I struggle to remember and fail the test. Half an hour later it comes
to me in a flash "Of course, it's Ankara!". Now what does that tell us
about our memes? Did you have the meme and I didn't? Did I have it later
when I remembered but not at the time of the test? Did I always have it in
some way but somehow couldn't access it? These are the kind of problems we
run smack into as soon as we try to start quantifying any kind of mental or
memory entity.>>
That's the reason achievement tests have more than one question on them.
Certainly there are differences in "test-taking ability" among individuals,
and anyone using the results of tests has to realize that there will be such
problems. But statistically, over many questions on a test and many
individuals taking a test, there are correlations between the knowledge
tested and the future ability of someone to perform.
<<You see what I'm driving at here, Richard, is not that mental content
doesn't exist (no Skinnerian behaviorism from me), nor that the idea that
there are replicating mental entities is in any way particularly implausible
(I mean I held to Dawkins B for many a year myself....) but rather that it
doesn't stand up to close scrutiny. The difficulty is methodological.>>
The methodology needed is statistics. To use Jack Cohen & Ian Stewart's
metaphor, there is no predicting the behavior of one fan at the end of a
football game. We don't know if he will go straight out to the train
station, wander around the parking lot, or go use the men's room. But
sitting up in a helicopter, the swirling patterns of bodies moving out the
exits after the final gun is eminently predictable.
Similarly, the replication of memes is predictable. If 100 people go through
a cleverly designed viral seminar, about half of them will enroll in the
next course, 55% of them will bring a guest to the introductory talk for the
next seminar, and so on. When you get too close to any one mind, the memes
slip away. But when you examine the whole population the effect is obvious.
<<To what extent is the ability to drive a car memetic? Car driving, once
one
has learned is cerebellar. While you're learning it's cerebral. Different
parts of the brain are involved depending on one's degree of expertise. A
driving test tests behavioral ability, not anything concrete inside the
head. That seems clear to me.>>
Now I'm lost again. In what way is a learned behavior, taught from father to
son and in schools around the world, NOT memetic? Are you claiming that the
methods of driving are not replicated? Perhaps it's a difference in culture.
Here, drivers take a written exam showing understanding of the rules of the
road, in addition to their driving exam. That may be a clearer example, or
maybe not.
Richard Brodie richard@brodietech.com
Author, "Virus of the Mind: The New Science of the Meme"
Free newsletter! http://www.brodietech.com/rbrodie/meme.htm
===============================================================
This was distributed via the memetics list associated with the
Journal of Memetics - Evolutionary Models of Information Transmission
For information about the journal and the list (e.g. unsubscribing)
see: http://www.cpm.mmu.ac.uk/jom-emit