RE: What does the replicating?

Aaron Lynch (
Mon, 16 Jun 1997 23:45:42 -0500

Message-Id: <>
Date: Mon, 16 Jun 1997 23:45:42 -0500
From: Aaron Lynch <>
Subject: RE: What does the replicating?
In-Reply-To: <>

Aaron Lynch responding to Bill Benzon:
>Aaron Lynch says:
>>"Coke" has numerous memes associated with it (like the earlier discussion
>>of Demeter). The idea that there is such a beverage is a meme. So is the
>>idea that this cola is the best cola, an idea with a lower host population
>>than the former. The bottles, cans, papers containing ads, etc are
>>phenotypes of such memes, as well as phenotypes of various memes saying, in
>>effect, that "Coke is profitable." The trillions of copies of the word
>>"Coke" that exist in land fills, store shelves, etc. attest to the
>>popularity of the drink, but are not enumerated as memes themselves. Meme
>>prevalence is measured by counting hosts.
>I realize this is just informal chit-chat, but...I got to wondering about
>the relationship between the Coke meme and all those phenotype memes. In
>the case of complex multi-celled organisms there is a complex epigenetic
>process which connects the genes to the phenotype they encode. So what
>connects the Coke memes to bottles, cans, ads, etc.? At the very least,
>it's that part of Coca-Cola Inc. which is concerned with Coke. That's a
>pretty sophisticated social process. And the Coca-Cola taste meme couldn't
>replicate itself without all that machinery, social, mechanical, and
>And all that machinery involves lots and lots and lots of memes of all
>sorts. We have an enormous complex of co-adapted memes -- many of which,
>of course, exist in other such complexes. PepsiInc, for example, shares a
>great deal of its memetic complement with CCInc (just as we share 98%+ of
>our genetic complement with the higher apes). The essential memetic
>difference between PepsiInc and CCInc certainly consists of different
>formulas for creating cola syrup, different memes for add campaigns and
>probably some differences in corporate culture.
>Now, in what sense are those cans and bottles of coke the expression of the
>Coke memes, as opposed to the profit memes, and the memes for stamping out
>aluminum cans, and the "I need a swimming pool in my back yard memes" of
>Coke executives, etc.? Those "coke is it" memes couldn't replicate without
>the help of all those other memes, and, of course, those other memes get
>help from the coke memes. That being the case, why give them all the
>"causal credit" for the complex process in the course of which they are
>replicated? Why not spread the causal credit around?

Robin Wood asked about Coke memes. In my reply, I should have added that I
do not consider it especially useful to recast marketing science into the
language of memetics (recursive evolutionary replicator theory of mental
information in a population). So Coca Cola is a rather poor topic in which
to seek fresh insights that only memetics can supply.

But the question of concentrating "causal credit" can also be asked about
memes that are analyzed better by memetics. In those cases, such as
masturbation taboos, causal credit remains as you say, widely distributed
among many factors ranging from basic language memes to ideas about
clothing, etc. One only focuses attention on the role a meme plays in
causing its own re-transmission because THIS is a recursive operation, and
hence subject to the vast amplifications attainable in recursive functions.

I don't wish to continue in depth with Coke, because the subject is already
handled well by marketing science, diffusion of innovations theory, etc.,
and is not the sort of topic that calls for invoking the memetic
paradigm--though the topic can be forced into that paradigm just as surely
as your commute home can be forced into the relativistic paradigm. I do
agree that the number of ideas involved in this industry is vast.

>Note, however, if we consider CCInc and PepsiInc, we can reasonably give a
>relatively small set of coke memes and pepsi memes credit for the
>differences in output between these two large meme pools. And if we
>compare, for example, Boeing Inc. and CCInc., both a large late 20th C
>corporations and, as such, share a good many memes. But the differences
>between them are greater than the differences between CCInc and PepsiInc &
>so we'd expect greater differences in their meme pools, but just how much
>greater? If CCInc and PepsiInc share 99% of their memetic complement, what
>do CCInc and Boeing share, 70%, 80%? Do we care? Why or why not? If we do,
>how do we go about measuring these meme pools?
>Whatever the value of that previous paragraph, however, that isn't where I
>was going when I started this note. I was going back to a remark made, I
>think by Mario. Namely, that genes don't replicate, only cells do. And, of
>course, cells can only replicate if they are in the right environment. So,
>gene replication is part of a complex causal process. To talk of genes as
>replicators is, in a way, to "give them credit" for all the "causal
>dynamics" in the system. And that is misleading.

I meant to respond to Mario earlier, but will do so now: I had stated
earlier that in absolute terms, nothing in this universe replicates.
Specifically, in *absolute terms*, CELLS DO NOT REPLICATE--EVER. Think they
do? Its only because you have "abstracted out" a set of properties by which
you can call two cells "the same." In reality, such gross features as the
number and 3-dimensional placement of organelles is not exactly the same
between mother cell and daughter cells, nor between 2 sister cells right
after mitosis. Nor are their masses exactly the same. Nor the placement of
deuterium atoms, etc. etc. etc. In fact, the 2 cells generated in a single
mitosis can look far more different in number and placement of organelles
than do 2 randomly chosen humans in the number and placement of appendages
or internal organs. All of this is tacitly ignored by anyone claiming that
2 cells are "the same." Having said all of that, I will point out that
there is no science without abstraction. So go ahead and call 2 cells or 2
quadrillion cells "the same," but remember that they are "the same" with
respect to an abstraction. The scientists job, then, is very largely one of
identifying those systems of abstraction that are useful in their
explanatory and predictive powers. Taxonomies. Gramm stain results.
Resistance to Vancomycin. etc. I am sure that Mario is a fine
microbiologist. But I find that the most talented scientists are often as
unconscious of the underlying principles of science as good athletes are to
the underlying physics of sport.

As to causation of cell replication, it is again not absolutely
"self-replication," not by a long shot. The Sun replicates cells. So does
geochemical energy. So does gravitation. So does water falling from small
comets, etc.

As for genes, they do not absolutely self-replicate either. But when you
hear a scientist or an author say "the gene is the unit of selection," bear
in mind that they are also not speaking explicitly about all sorts of
underlying assumptions in science. You might want to rephrase their
statement as "the 'gene' is the most broadly explanatory and predictive
abstraction for use in the theory of biological evolution by natural
selection." There are reasons for making this assertion, but there are also
reasons for looking at other biological entities including proteins as
replicators, with prions being a case where such analysis seems not merely
conceivable but necessary.

>Similarly, I think it is misleading to give the coke memes all the causal
>credit for the complex social process on which they depend for their
>replication. And I'd like to put this point in context of the discussion
>the determination of human action by memes vs, determination of that action
>by (free) will in an (autonomous) self. If that's the opposition I'm
>given, then I'm going to come down on the side of will and the self,
>eventhough I'm quite content to think of the self as a social construct
>(and have written about the evolution of that construct in history) and I'm
>not at all sure what a free will is. However, I don't think that's a
>meaningful opposition, primarily because I have trouble imagining those
>memes doing anything outside & independently of a complex social system.
>Thus I'm more interested in thinking about the relationship between
>individual desires and actions and social forces which, perhaps, memes
>being the coin in which the relationship is negotiated.
>>replicator. There is, however, the oft-noted recursion principle of the
>>convinced consumer paying the money that gets this central communication
>>done, and it is mainly in this sense that a "Coke is it" belief does its
>Recursion? I have some idea of what the term means in mathematics, computer
>science, and linguistics, but I don't know what it means in this context.

The recursion principle is central to all evolutionary replicator
(reproducer) theory, from Darwin forward. Recursion means that a given
operation is repeated upon the output of the previous run of "the same"
operation. (Dennett uses the term "algorithm," but "recursive algorithm" is
a more exact term.) So if a belief spreads parent to child, recursion means
that it also spreads child to grandchild, etc. Why does recursion matter so
much? Because that is how small advantages in one cycle get amplified into
vast advantages over many cycles. If a masturbation taboo had just 1
adherent 12,000 years ago in the first farming villages, and spread
recursively parent to child at an average rate of 5% per 25-year mean
generation over 12,000 years, then the result would be 15,000,000,000 hosts
in today's world. 5% thus looks like a huge number when you think of it
recursively as 1.5*10^^10 after 12,000 years.

>I've read your masturbation passage and thought about it a little. Do you
>have any evidence that societies with strong masturbation taboos do in fact
>have higher birth rates than societies which don't have such taboos?

Direct evidence of, say, a 5% higher birth rate would need to be discovered
by way of new surveys--surveys that are more likely to happen now that
there is a theory that the results could help test. As for indirect
evidence, religious populations with numerous sexual repression memes
including anti-masturbation memes generally have large family sizes:
Hassidic Jews, Roman Catholics, fundamentalist Protestants, Muslims,
Mormons, etc.

--Aaron Lynch

How Belief Spreads Through Society
The New Science of Memes
Basic Books. Info and free sample:


--Aaron Lynch

THOUGHT CONTAGION: How Belief Spreads Through Society The New Science of Memes Basic Books. Info and free sample:

=============================================================== 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: