Re: Meaning generation

Mark Mills (mmmills@OnRamp.NET)
Wed, 23 Jul 97 23:19:02 -0000

Message-Id: <>
Subject: Re: Meaning generation
Date: Wed, 23 Jul 97 23:19:02 -0000
From: Mark Mills <mmmills@OnRamp.NET>
To: memetics list <>

Scott DeLancy,

>Now, here's where I have a real problem, but I can't tell for sure
>if our disagreement is substantive or just terminological. Let us
>stipulate that every human being (absent color blindness or other
>pathology) has a visual system that responds strongly to a particular
>wavelength/saturation of light, the one at the center of the field
>that English labels _red_, Chinese _hung_, etc. I don't see anyplace
>here where the concept of "meme" is relevant. If all humans respond
>equivalently to a particular stimulus because that's how they are
>built, this is still squarely in the realm of genetics, not memetics.
>Only when people start *naming* this color, subdividing the color
>space into labelled subfields, into a system which other individuals
>can then acquire, is there anything memetic going on.

Where to start... This is a longer than I'd like, but hopefully it
answers you question about the utility of memetics.

Let me begin at the beginning. I start with the idea that there is an
analogy between genes and a new construct called 'memes.' Genes are
involved in biological reproduction, 'memes' in cultural reproduction.
Everything is built on this foundation.

Genes are generally thought to be chunks of DNA code, so I'm assuming
'memes' are physical chunks of code, too. There are two places for
physical chunks of code to reside, upon the neural network within the
brain and out in the physical environment. To evaluate these two
locations for 'meme holding,' I've gone back to the genetic analogy.
Since code requires a 'processor' to translate 'input' into 'output'
(otherwise code has no meaning and is not code), a hunk of memetic code
requires a processor. Until the arrival of computers, all decoding
(processing) took place in the neural system. Thus, I've concluded
'memes' are code in the brain. To give form to this, I'll point to
recent brain research which has found appropriate proteins which lodge in
synapse connections. These can easily be seen to act as weighting
factors for neural processing. Thus, I've defined memes to be the code
set upon the invariant substrate of neural networks in the brain.

With this done, I've got a physical analogy to genes. It's a starting

Next, we have to ask 'how we identify or decode the meme?'

I suggest going back to the analogy with genes. Genes involve DNA codes
upon a sugar-phosphate substrate. DNA is not a blue print, it simply
serves to as a template for protein production. DNA, though inert itself,
has a number of behaviors due to the activities of enzymes. These
behaviors leave no record. We can't really study them. We know what DNA
does because of it can be linked to proteins (artifacts of the process)
which stay around long enough to be studied. Thus, researchers generally
decode DNA and 'locate' genes by linking DNA chunks to specific proteins.

By analogy, we end up decoding memes by isolating behaviors and

Note that this is a process oriented perspective. There is input
(minerals and energy), processing and output (proteins). I haven't
defined memes in terms of 'effect' or 'human activity.' I haven't even
said that DNA 'controls' anything. DNA and proteins are simply the
easiest elements of the process to identify. Without a host of transient
enzymes, RNA, mitocondria and substrate flexures, nothing works and the
process comes to a halt.

By analogy, there is a 'memetic' process, one where the neural network is
decoded as part of organism behavior and artifact production. Artifact
production can be as simple as footprints in the mud or as complex as a
human computer.

Thus, memes are not defined by 'intent' or 'use.' Memes are defined by
their process and substrate. It doesn't matter what the code is use for
or how it was placed on the substrate. Genes are still genes regardless
of human engineering.

Memes are a level of organization and processing built upon genetics.
Memes involve a pattern of proteins (code) lodged in the synapses of
neural networks (substrate). Based on the this, memes have been around
as long as brains. In fact, I suspect they are a side-effect of neural
organization. Thus, culture is as old as brains, too. (Bonner has a good
book on this.)

With a physical definition for the construct and a suggested path for
investigation, we can turn to the question of utility. As you point out,
what does it do for us?

origin of human language, but the real purpose of memetics is
evolutionary instruction. The construct memes offers yet another way to
become unattached to Platonic idealism. For example, the value of
genetics is not their explicit definition, they remain elusive and vauge.
Genes are often called 'units of inheritance' rather than DNA chunks
because all of what happens in reproduction cannot be ascribed to DNA.
The value of genetics is the notion that we need not be attached to
Platonic ideals for starting points in research. We can reasonably
continue to work in the vast and wildly different disciples of biology
confident that it all fits together as a united whole. Memetics offers
an opportunity to unite the social sciences in a similar way, ie
detaching them from a variety of antiquated Platonic notions and
superseded epistemologies.

With unattachment in mind, let's look at memes (as I've defined them) and
the origin of language. Your comment that it is only when people start
'naming' things that there is anything memetic going on reflects an undo
focus on a recent evolutionary stage in cultural evolution. The
transition from 'pre-naming' human culture to 'naming' human culture was
probably extremely gradual, perhaps a million years in duration. It
seems odd to suggest this phase was a non-memetic human culture. Surely
our pre-naming ancestors shared our ability to memorize and mimic. Why
define them out of the memetic history?

If one can gasp the memetic definition I've outlined, it seems clear that
it offers a way to explictly chart the evolution of 'naming' humans from
'pre-naming' primates. One could start with the study of non-human
primates for a notion of 'pre-naming' memes in primate culture. These
memes could be compared with 'naming' memes. With the two sets in hand,
one can start building evolutionary models to bridge the two.

Without going into detail, let me suggest that I think the evolutionary
perspective will recommend a review of how mimicry serves memetics as sex
serves genetics. Mimicry is the blending process leading to memetic
replication, regardless of how conscious the mimicry. It blends the
mimic's experiential mind with a physical re-enactment of another
individual's activity. Sometimes the mimic focuses on the wrong thing or
is unable to skillfully act out the behavior. Poor mimicry is still
replication and blending. It is analogous to mutation and is definately
a feature of pre-conscious cultural activity.

Mimicry is held in low esteem due to Cartesian and Platonic notions, but
evolutionary perspectives will revive its importance.

Language acquisition by children is a highly mimetic. Children can
identify differences in expression which will disappear at older ages.
This allows them to be master mimics.

Language and the mimicry that fosters it did not evolve 'of itself.'
Culture is the translation of memetic code, a 'mirror' image. Language
is an aspect of culture, one of many cultural artifacts. Recent
discoveries in Africa suggest stone tool goes back 2.5 million years,
clearly pre-dating 'naming' skills. Our modern 'naming' language is
thus an enhancement or expansion on pre-human mimicry, a dynamic and
powerful means of memetic replication as compared to both genetics and
pre-conscious mimicry.

One shouldn't forget the symbolic opportunity tools presented for
enhanced mimicry (what we now call ritual). These tools would have been
mostly plant fiber. If you take a look at your 'Theme - Loc' linguistic
framework, you will see it naturally fits into a system for communicating
ritualized tool skills, a useful activity for 'pre-naming' human

Sorry to go on for so long. I haven't really defined specific memes.
Instead, I've tried to provide a process orientation and evolutionary
perspective. I hope this helps.

See for reference
on stone tools


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)