MEMETICS: Memetic engineering

Eric Watt Forste (
Tue, 31 Dec 1996 12:26:47 -0800

James Rogers raises some excellent questions.

>I may have misrepresented myself here. I have no designs to control people.
>I am trying to ascertain, based on what we know of memetics, to what
>granularity a generalized (as opposed to individually targeted, e.g.
>deliberate brainwashing) meme can be focused. In other words, is it
>practically possible for a meme to propagate only within a very narrowly
>defined segment of the population? Say for example, a meme which only
>propagates amongst white females, ages 34-37, who live in the Pacific

I doubt it. I suspect the phenomena are simply too complex. Mental
activity resembles cyclical turbulence in some ways, I think, and
sociocultural mental activity would be a turbulence composed of
turbulences. But I'm sure that you know a good deal more about
turbulence than I do... what do you think of this metaphor?

I do think that it's a vain hope that we will ever ascend to a
stage in which our actions and words will no longer have any
unintended consequences, and it is the unintended consequences that
make the phrase "memetic engineering" something of a joke.

>There does seem to be instances of fine grained memetic
>propagation occuring.

For instance?

>I am interested in the factors that control
>susceptibility to memetic propagation. What aspect of human cognitive
>function is responsible for memetic susceptibility? Some people seem to be
>more susceptible than others (bandwagoneers? skeptics?).

I suspect that these questions are probably best researched within
psychology, particularly the psychology of leadership, which is a
mighty contentious field, lying as it does at the boundary between
psychology and political philosophy.

>Also, would it be possible for a person to detect a meme carefully targeted
>at ones self?

Now *this* is a good question, except that it begs the question whether
or not it is even possible to "carefully target" a meme at a particular
individual in any kind of way that would allow objective testing of the
results. I suspect that this question reduces to "Are there any methods
by which we can avoid getting hoodwinked by people who are intimately
familiar (either through long acquaintance or keen skills of
psychological observation) with our individual psychological structure?"
And of course that's an interesting question. But I don't know if there
are any easy answers.

>I suspect that it would probably *have* to be done using some type of fuzzy,
>genetic algorithms because I am not so sure humans will ever have a good
>model of this phenomenon. It's not just that; how do you represent
>something as abstract as a meme in the concrete environment of a computer in
>such a way that it could be useably processed?

The problem seems to me to be more fundamental than that. Physically,
in their active (i. e. interesting) lifestage, memes are distributed
information systems comprising multiple human brains. Human brains
are the most sophisticated supercomputers the universe has come up
with so far, and their computational processes seem to be at least
a bit chaotic. I doubt that any simplified model run on any simpler
computer can tell us very *much* about the entire process, although
of course a lot of excellent work is being done in experimental
economics. I just don't think that this work is going to give us
"the whole story" in any interesting sense.

Eric Watt Forste ++ ++