Re: i-memes and m-memes

Bill Spight (
Mon, 30 Aug 1999 10:21:57 -0700

Date: Mon, 30 Aug 1999 10:21:57 -0700
From: Bill Spight <>
Subject: Re: i-memes and m-memes

Dear James,


Bill... They are no longer memes, but they are still potentially memes. If I
might be so arrogant as to quote from my previous post:

<< Some replicators replicate well. Other replicators replicate poorly. But
if a replicator replicates so poorly that it doesn't replicate AT ALL, then
there is no point in continuing to view it as a replicator.


I am inclined to agree.


Besides, without the appropriate decoding system, it becomes meaningless to
talk of an artefact containing information.


Well, it certainly is meaningless to talk about their containing meaning. But we knew that Egyptian hieroglyphs and Mayan stellae contained information before we could decode them.


Speaking of decoding... Let us consider a simple case of memetic
transmission, i1->m->i2. In order for the i-form to count as being
successfully transmitted, the copy (i2) must be sufficiently similar to the
original (i1).

But there is another way of looking at it. The i-meme is successfully
replicated if the process of decoding it is the INVERSE of the process of
encoding it (i.e. the decoding acts to undo the encoding).
In (rather informal) symbolism:

i1 = i2 iff inv(m->i2) = (i1->m)


Def inv(mx->iy) = iy->mx , right?

Well, I think if (i1, c) -> m and (i2, c) -> m, where c is the context for m, for all c, then i1 = i2. I don't think that the converse holds, however, assuming that c does not include the person who incorporates the meme. (It depends on how you classify m-forms. A tricky question. There is also a question of novel contexts. If it appears that (i1, c4138) -> m135 and (i2, c4138) -> m154, but i1 and i2 produce equivalent m-forms in all other known contexts, maybe other memes are involved in c4138, and i1 and i2 are still actually identical. Tricky. <s>)


Bill, do you think this sort of thing could evolve into a rigorous logic of
memetic transmission? I started toying with these ideas after seeing you
constantly writing out L-G-L-G-L and so forth.


Yes. <s> I think your idea is more promising than the hope that i-memes can be equated by examination of their physical substrates.

Strictly speaking, however, I think that we are very often talking about i-m-i'-m'-i''-m''-.... The mutation rate for memes is high, since human perception is constructed, as well as human memory. (There are other reasons, as well.)

Best regards,


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)