Re: Subject: Re: Meme Extinction

Aaron Lynch (
Thu, 29 May 1997 14:16:31 -0500

Date: Thu, 29 May 1997 14:16:31 -0500
From: Aaron Lynch <>
Subject: Re: Subject: Re: Meme Extinction

Hello, Timothy.

Nice to converse with you after having read your book SEX SIGNALS: THE
BIOLOGY OF LOVE years ago!


> >This harkens back to the thrust of my original point. If a meme is truly
> >extinct, it cannot be studied, by definition. If we try to re-create it
> >from surrounding and circumstantial evidence, we cannot (by definition)
> >know if we have done so with complete accuracy.

The extinct meme can be studied indirectly by the evidence it left, much
as even the Big Bang can be studied.

> TP: This summarizes the paradox very neatly. Extinction of a meme *means*
> it is gone from its host -- the mind -- and is not thought by anyone.
> -----
> Aaron Lynch brings his usual sharp and acute analysis to his comments on
> meme extinction. I agree with most of what he says, but feel that in a few
> areas some further observations are needed.
> >Aaron Lynch on 28 May, 1997 responding to Timothy Perper/Martha Cornog:
> >>
> TP/MC>>
> >> Bloch's paradox centers on the idea of extinction itself. To give a
> >> concrete example, imagine religion in ancient Greece, for instance the
> >> rites and rituals associated with the goddess Demeter. We may assume that
> >> much has been lost of those rites (e.g., what words were actually said in a
> >> household ritual). But people *did* speak those words, and the rituals
> >> *did* exist -- or so we assume -- and they were memes. But those memes
> >> became extinct as the worship of Demeter vanished. But, historians working
> >> today might reconstruct such a ritual, and -- abracadabra! -- those memes
> >> are brought back to life.
> >
> AL:
> >Actually, reconstructing the ritual on a theater stage, for instance,
> >does not inherently reconstruct the meme that I would call "Demeterism."
> >How a particular meme is defined has a profound effect on whether it has
> >hosts or not, and hence whether it is extinct or not, as well as on what
> >constitutes a replication event.
> >
> >We might define a meme for "Demeter-definitionism" for which
> >instantiation happens merely by knowing the word and some core elements
> >of its definition. You might find that this meme never went extinct. The
> >other meme, "Demeterism," would be instantiated only in those who
> >actually believe that there exists a goddess named Demeter who meets the
> >specifications of the core elements of the word's definition. Further,
> >you can even define a "full-blown-Demeterism" meme, whose hosts are all
> >Demeterists and who actually hold some extensive list of ancillary
> >beliefs, ritual instructions, learned desires, etc. Finally, you can
> >define another meme "Demeter-expertise" which combines the word
> >definition with an extensive list of other beliefs about what
> >full-blown-Demeterists believed or did. Notice that, if archeologists
> >and scholars have done a good enough job and the memeticists define
> >their memes carefully, then their definition of "Demeter-expertise"
> >would be such that a substantial number of full-blown-Demeterists in
> >ancient Greece were also "Demeter-experts," and all of the ancient Greek
> >"Demeterists" were also "Demeter-definitionists." The only thing that
> >archeologists and other scholars might bring back from extinction, then,
> >would be Demeter-expertise and perhaps Demeter-definitionism if the
> >latter ever went extinct.
> TP: These are valid distinctions and very useful.
> >
> >snip<
> >
> TP>>
> >> Aaron Lynch's distinction between a meme and the idea of a meme -- the
> >> original worshipers had the memes, the scholars have ideas of the meme --
> >> doesn't seem to solve this problem. It merely says that scholarly ideas
> >> about Demeter worship are different from Demeter worship itself, a notion
> >> we can surely accept. But it leaves the original memes extinct and
> >> therefore incapable of being thought. Therefore they are no longer memes
> >> because they cannot be transmitted. Well, what are they, then? (And to
> >> say "They're _extinct_ memes" begs the question.)
> >
> >
> AL>
> >I do not say that the original memes are "INCAPABLE of being thought."
> >If the words of an ancient scribe somehow convince a person in the 24th
> >century that Demeter exists, then this would be a replication event that
> >took millennia to unfold, and which would cause the host population to
> >fluctuate up from zero.
> TP: Hmmm. I think there's a distinction to be made here. The original
> memes, held by real worshipers of Demeter, in their culture and time, were
> thinkable. IF -- note: IF -- Demeter worship had not died out, in the
> sense that if the *authentic* worship of the goddess had continued down to
> today in unbroken lineage of worshipers, teachers, and acolytes, then those
> memes (or their close relatives) would still be thinkable. But because the
> religion of Demeter did die out, modern versions of Demeterism must belong
> to the list Aaron provided above -- Demeter-expertism,
> Demeter-definitionism, and so on. I'd add another -- neo-Demeterism -- for
> the hypothetical rejuvenated religion of modern people who worship Demeter
> (if any such people exist).

I do think neo-Demeterism is a good term to use after so long an

> But once the authentically ancient religion has died, and with it memories
> of at least some of its beliefs, those memes are "extinct" in a curiously
> paradoxical way. It isn't a matter that they were thinkable, but concerns
> how the act of thinking *creates* the meme itself. If we think them, we
> create them -- and by definition they are no longer *extinct.* Now, it
> may be that we only believe that we are thinking the original memes, and
> perhaps we delude ourselves where in fact we've gotten the details of
> Demeter worship all wrong. if so, then we are not thinking the original
> memes, and they are still extinct.
> For example, we reconstruct Demeter worship as involving a certain
> liturgical ceremonial, but if observers with video recorders had
> materialized in the ancient days, they would have recorded a different
> liturgy. Then our reconstruction is wrong, and we have what might be
> called the memes of "false Demeterism." The memes of authentic ancient
> Demeterism remain unreconstructed, and therefore unthought by us -- and
> therefore extinct. But the moment we *do* think them, they are no longer
> extinct!

Actually, the whole notion of ever saying that two entities are in some
way "the same" is a tricky matter that is often taken for granted in
science and elsewhere. It is the axiom of abstraction, (also called the
axiom of comprehension), and we can only say that two entities are "the
same" with respect to an abstraction. But then we try to make our
abstraction system have a good correspondence to external reality, and
biological evolution has favored brains capable of this to some extent.

But here we have abstractions about the mental or neural information of
animals,--in this case humans. Even in an ancient Demeterism no two
people had EXACTLY the same neural information, but only the same with
respect to some abstraction. Hence in Demeterism retransmission, the
resultant new host's neural was never exactly the same as anything in
her instructor's nervous systems. But we are especially tempted to think
of them as EXACTLY the same because of their close proximity and
intimate communication. But what if communication goes from an ancient
scribe to a slightly less ancient reader several generations later? Or
several millennia later? Surely the "copying fidelity" of such
transmission events will be lower than for the typical mother to
daughter communication. Whether you count the 20th century believer as
having "the same" belief depends on how strictly you define Demeterism.
Embedded in this is how well any modern abstraction system (including
the definition of Demeterism) represents a real ancient phenomenon, for
only by an abstraction that corresponds well to reality can you be
CORRECT in saying that someone's ancient belief was "the same" as
someone else's modern one.

> ---------
> AL>
> >Do you want to define the word "extinct" so as
> >to be absolutely irreversible? What about actually bringing back a
> >recently extinct bird by using sophisticated biochemistry? Does that
> >prove the bird was not really extinct? A similar definitional problem
> >can happen with the word "life" when it comes to freezing then thawing
> >embryos, or even adults, etc.
> TP: I posted something like this on an earlier note, and Paul Marsden on
> 5/27/97 was quick to point out that I had missed the central issue. On
> thinking about his comments, I think he was right in saying that
> reconstructing a biological system does not have what he called the
> "self-referential" quality of the meme. Yes, we reconstruct the animal
> around the fossil, but thinking does not cause the animal to exist. By
> contrast, thinking and communicating *does* create memes. Thus, the fossil
> existed as a living organism in the past, but no human being knew about it
> -- so we had no memes dealing with this creature. Today, when we find the
> fossil and reconstruct it, we invent memes, shall I say, about the
> organism, its bones, its habits, or whatever else. In principle, there's
> nothing paradoxical about this at all -- it's just innovation and
> discovery. The reason is that the existence of original but now fossilized
> organism did not in any way depend on *our* thinking it. But that is not
> true of memes. Their existence does depend on thought. And so the notion
> of an "extinct meme" seems to be a contradiction in terms, even though the
> now lost forms of Demeter worship seem to instantiate it.
> ----------------
> AL>
> >Defining the word "extinct" is not really critical to the theory as long
> >as there still is a mathematical way of discussing host populations of zero.
> >snip<
> To be sure, mathematical formalisms may permit certain parameters to have
> the value zero, but the question about memes is what that actually means.
> Bloch's paradox suggests that the moment we examine the actual memetic
> content of such zero-frequency memes, their frequency is no longer zero.
> Observing the meme brings it into existence, much like a quantum mechanical
> observation. I agree that such things might not particularly important for
> scholars reconstructing Demeter worship, but the existence of a paradox --
> any paradox -- is always a pointer that we need to consider the arguments
> and logic *very* carefully.

Well, to examine ANYONE else's belief system requires us to become host
to at least PART of that person's neural information, even if merely in
a representational way. But with a well-defined "sameness" criterion,
this does not generally cause a new instantiation of the meme under
study, and hence neither an incrementation of that meme's host
population. If it did, then the question of whether the other person
lives today next door or last millennium next continent does not matter.
When I study the beliefs of a Muslim next door, I do not thereby
increment the host population of Islam by 1. By the same token, if I
indirectly study the ancient Demeterist, I do not thereby increment the
host population of Demeterism by 1--and Demeterism remains extinct even
under retrospective scrutiny. It's a matter of being careful (and not
getting confused) about what you count as "sameness."

> I'm not saying that Aaron's analysis is wrong, but that it somehow doesn't
> quite seem to eliminate Bloch's paradox.


--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: