From: Keith Henson (email@example.com)
Date: Sun 30 Mar 2003 - 23:47:43 GMT
At 07:33 AM 31/03/03 +1000, you wrote:
>I hate being the "nay" sayer, but predictive is out of reach, almost by
>definition. I say "almost" because I do accept the "egg already falling"
>short term possibility, but still with some reservations.
Oh, I agree with you, and said so way back in 1989.
>SPACE Digest Volume 9 : Issue 438
>Date: 12 May 89 19:02:53 GMT
>From: firstname.lastname@example.org (H Keith Henson)
>Subject: Re: memes
>John Roberts (email@example.com) ends a rather thoughful post with:
>>My question is not whether the concept of memes can be used to form a
>>model of the spread and application of ideas, but whether such a model
>>would be of any actual use, for predicting future events or for some
>>other application, which it can handle better than conventional methods.
>"Conventional methods" is a term which I do not understand, but as to
>predicting future events, I have argued that the mutation/communication
>complexity of the memetic ecosystem is subject to chaos, that is long
>term projections are *impossible* for the same reason weather is. Short
>term projections, on the order of the popularity of a song or book can
>be made (and are by conventional guesses). As to the use of memetics,
>I think its most useful application might be to teach people to be more
>thoughtful about which ideas they accept into their minds.
While I am still arguing that chaos will make certain kind of predictions
impossible, we *can* say a lot about even the classical chaos example,
weather. We have excellent reasons to expect it to be a good deal colder
in winter than in summer, and right now we are getting to the point of
understanding what flips the earth into an ice age. (Probably shutdown of
the North Atlantic heat conveyer by too much fresh water.)
>The one good predictive tool we have is demographical statistics. But
>psychohistory is still science fiction.
>If I have time over the next few days I will expand my comments,
>as I think that the sooner we stop trying to turn memetics into some sort
>of magical social engineering tool, and concentrate on the benefits it
>delivers to ethnographic and cultural anthropology work, the better.
I think memetics and evolutionary psychology along with demographic and
economic data might generate models of great interest, ones that could be
calibrated by historical data and be useful for future predictions,
particularly about situations leading up to wars. I don't think you can
get detailed predictions, but knowing what long term climate is going to be
like (as opposed to knowing what the weather is going to do next month) is
still darned useful. (Do you build flood control dams or desalinization
I expect this might cause a small shift in our view of population
growth. Population growth in and of itself is not so bad, but if the
resources per capita does not keep pace you are in for a hell of a
ride. Since resources don't grow every fast, especially with limited land
or limited investment capital, this leads to the conclusion that slow
population growth, much slower than in most of the poorer parts of the
world, is the only kind we can have without wars and functionally similar
Likewise, a part of the world that has fast rising population and falling
per capita income might be properly seen as a keg of powder with a lit fuse.
Is any of the above a new observation? I don't think so. It's just
fitting what we have long felt or suspected into a model--ultimately based
on the way our psychological traits were shaped by millions of years
evolution in small groups.
PS. You might be able to predict characteristics of memes that arise in
certain areas. For example, socially significant religion type memes
arising in XXX areas are very likely to have many characteristics of XXX,
where XXX can be Christian, Islam or traditional Chinese.
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)
This archive was generated by hypermail 2.1.5 : Mon 31 Mar 2003 - 12:37:44 GMT