How is a meme born? Thought or Information? (Retransmission due to low channel fidelity!)

Mick Ashby (
Tue, 03 Mar 1998 11:37:31 -0500

Date: Tue, 03 Mar 1998 11:37:31 -0500
From: Mick Ashby <>
Subject: How is a meme born? Thought or Information? (Retransmission due to low channel fidelity!)

-----------This is a retransmission--------------------
The original channel had low fidelity!
I don't know why "3D" got appended after each "=" sign.
Hopefully this doesn't suffer the same fate!

thanks Josip, you got me thinking by writing:

>...I say that memes or information can not "survive"
>with-out or out-of the dynamic system.

I wondered what you mean by "survive" - the quotes imply a particular
interpretation - I'm not sure which! On what dynamic system(s), for
example, does the information in a library depend for "survival"? Does
that mean that a meme is only a meme when it is in a host? I hope
not! Certainly, the same meme will be slightly different inside two
different hosts' heads, but what if the meme is defined (and replicated)
in a book or an email?

Let 's consider how is a meme born:

1. Person A has a thought T. The thought may be completely original or
may be a mutation of other people's thoughts.
2. A encodes the thought in a form which can be transmitted to other
people, E(T).
Depending on the thought, the encoding process may or may not be a
loss-free transformation
e.g. mathematical concepts can be encoded with more precision than
subjective feelings.
3. A places E(T) in a communication channel C.
The channel may be verbal, email, letter, book, TV show, or whatever.
4. Depending on the nature of channel C, E(T) will be:
4a) delayed by time D, where (0<D<=infinity)
4b) mutated by fidelity F, where (0<=F<=1),
0 = deleted, 1=perfect reproduction, 0<F<1 = corrupted/modified
4c) directly received by N people, where (0<=N)
Assuming N>0, then for each person Bn, where (n=1..N)
5. Person Bn receives F(E(T))
6. Bn interprets the message (another potentially noisy transformation,
I) and a version of thought T is reconstructed in person Bn's mind as
When Interpretation I, Fidelity F, and Encoding E are "perfect" then
I(F(E(T)))=T, otherwise Bn has a slightly mutated version of T, let's
call it Tn
7. If Bn accepts (consciously or unconsciously) Tn as a belief then Bn
may be said to be a host for meme T (or mutant version Tn)
8. Person Bn may now spread T or (Tn) to other hosts by the same

OK that's how thought T became a meme - is this complete? I hope
not! What's missing? Sure, point 7 hides the details of infection,
resistance, and competition, but it will suffice here to have a binary
function determining the potential host's acceptance or rejection of the

So is a meme a thought or information? Is the meme really T, or E(T)?
The two are not the same. Sure, if the encoding E(T) is *very*
hi-fidelity then we can just talk about meme T. But if T is difficult
to express precisely, then E(T) is possibly the most authoritative,
communicable description of thought T, and E(T) is what *really*
spreads. For example, Richard Dawkins' original "meme" thought is
possibly still in his head - the encoded form (in his book) has spread
very well, but the "meme" meme which we are all discussing is actually
the one defined in his book rather than the definition in Richard's
head. I believe the recent call for papers - CFP: Symposium on Memetics
(Namur, August 1998) - supports this view:

> The initial description of 'meme' by Dawkins is rather vague, which is a
> possible reason for current diverging views on what a meme really is, and
> how the memetic model can be used. We are confronted with an avalanche of
> books, essays, and publications scattered over different journals and
> disciplines, with dialogue flashing up here and there in an unstructured
> manner. This chaos exists because a general framework is lacking.

Tim Rhodes wrote:
>Rob wrote:
>> However in order for memes to reproduce, there must be some medium
>> by which memes can be transfered/invoked. If a meme is commonly
>> invoked by a specific word or set of instructions, does not the meme
>> exist partially in that external medium?
>Do we have a terminology for this yet? I have heard this called a `memetic
>artifact' before--is that considered common usage in the community?

Do we need a special name for a meme while it is outside a host? Sure,
you can if you want to, but I hope not! If we can view the meme as a
piece of communicable information, rather than a state of mind, it is
not necessary to attribute hosts the special status of somehow bringing
a meme "to life". We can simply view the hosts as communication
channels, each having their own characteristics for delay, fidelity, and
number of listeners. From the point of view of the infection of hosts
with a meme, would it matter if one of the "hosts" was a dumb
listserver? Equally, would it matter if one of the meme's
"communication channels" was an intelligent person?

Perhaps the greatest strength now is that memes can be spread by
*reference* or as Rob suggested - invoked by a set of instructions. To
infect you with a meme I don't even have to communicate it to you with
sufficient fidelity that you can recreate the idea in your mind - it can
be as simple as "Hey I read this amazing book! It's by ABC and is
called XYZ." We sure see a lot of this on the net from people called

Be Happy

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)