re:on imitation(longish)

singa (
Sun, 29 Nov 1998 23:17:11 +1100

Date: Sun, 29 Nov 1998 23:17:11 +1100
From: singa <>
Subject: re:on imitation(longish)

From:Alex Brown (
Date: Wednesday 29th November 1998

In response to Paul Marsden's e-mail of Tue, 24 Nov 1998 with apologies
for the delay:

To my comment that: Imitation = replication. This is a very useful
clarification. However, I would ask what it is in cultural terms that is
being replicated. Behaviour?

PAUL: >No Instructions, which (back to Cloak) ARE embedded in contexts,
they are the functional relation of cues and behaviour. Instructions
are not internal rules.

ALEX: Okay, I’m guessing to some extent (not having read Cloak), but
these instructions are like clues (or cues) which tell the recipient
what behaviour is expected in a particular context. If this was it, I
could certainly agree with it, but the use of the term ‘instruction’
seems to suggest that the environment (other people or situations)
demand a single ‘correct’ response from the recipient, or that
individual behaviour is totally determined. The concept of modelling in
my view is a more appropriate process since it accepts that the model
can never totally describe or express the event. (Reduction being what
it is).

PAUL: >Depends which school of thought you come from [how we get
variation from replication] some would have irreducible individual free
will is the source of creativity, others would > say errors or mutations
in the replication process (both are saying the same thing, but
ascribing different processes to this), Errors perhaps ultimately to due
to a nested hierarchy of the evolutionary loop operating internally
(i.e. Campbell, Dennett, Cziko), or simply errors in embedding an act in
a suitable context. (how many times have I put my wallet in the

ALEX: In my opinion we don’t need the idea of ‘errors’ in the system at
all to explain variation. It is already built in to the evolutionary
process: The state of the environment varies over time. Cultural systems
model their environment by continuously adapting their behaviour. They
do so by selecting appropriate responses to new environmental conditions
from a repertoire of typical routines filtered out of previous
behavioural acts. These discrete elements are re-combined into a model
(a set of elements and relations) which act as a behavioural metaphor of
the environment. Variety or diversity is therefore a result of a system
tracking changes in its environment. Who needs errors? While they will
inevitably occur during the selection-combination process, they cannot
be the mechanism of variation since the system would get its behaviour
wrong (relative to the demands of its environment) more often than it
got it right, and we wouldn’t be here to talk about it.

PAUL: >Durham (Rindos, and to a lesser extent Sperber) says selection,
i.e. the selective environment is based on the complex of memes already
culturally acquired, Lumsden and Wilson and the EP mob generally think
selection is based on epigenetic rules, (AKA Darwinian algorithms AKA, >
evolved structure of the brain - e.g. incest avoidance) that are
genetically > inherited. I think both are important.

ALEX: So do I. The "complex of memes already culturally acquired" is
equivalent I think, to my ‘repertoire of typical elements’ described
above and more generally to the collective cultural phenomena such as
styles, canons, genres, theories, paradigms, customs (and, if you wish,
meme-plexes) which I have mentioned in a previous post and which act as
constraints on new behaviour. The epigenetic rules would define the
biologically-derived procedures (algorithms) or processes of selection
and re-combination from which new models of the environment are
produced. Equally, these algorithms would provide the classificatory
mechanisms for assimilating diverse behaviours into typological groups
(the styles, paradigms, etc mentioned above) by matching similarities
and differences. In other words, these rules define what and how it is
possible to think.

PAUL: [In cultural terms we have diversity and uniformity]....which can
be explained in terms of descent (uniformity) with modification

ALEX: Yes. In cultural terms, ‘descent’ is cumulative past behaviour
(memory) written out in the form of prevailing behavioural styles while
‘modification’ occurs with the individual selection and combination of
these stylistic elements to model new and variable environments. The
assimilation of these many individual adaptations into coherent sets of
behaviours occurs through communication and exchange between the
individuals who act within a cultural system. (Assimilation = similar
responses are merged into Types and circumstantial or improbable
responses are excluded. There is no conscious or teleological intention
involved in this. It is purely the result of exchanges between
individuals in a confined environment - the cultural system itself).
Memory is the key issue in biological and cultural systems. In biology,
memory is in the form of the genetic code. In culture, memory is stored
or written out in the behaviour or artefacts produced by the system. But
this memory is not simply a record of all that has happened. It is
filtered to identify the most probable, characteristic or typical
elements in previous experience, (note Freud’s ‘condensation’ principle
and the production of discrete elements of communication (memes), and
biologically, a product of the epigentic rules mentioned above). Memory
(meme-ory) provides a vocabulary - not a program for future action. In
my opinion it is just plain silliness to suggest that behaviour is
totally determined by memes as it would seem from the latest ‘popular’
books and theories on memetics. Do we really believe that memory alone
determines behaviour? Why do they do it? Because its easier than also
trying to factor in context or environment, and after all, memes are
‘things’ and easy to think about. Events emerging from complex
interactions are not. To reverse the flow from genetics to memetics we
could state that genes are just the memory of the human system and, as a
vocabulary thus incapable of determining behaviour on their own. How did
we ever think they could? Two reasons: one, because (like memes, above)
its easy to conceptualize (and reify) genes as the answer; two because
we have been told so by authority figures.

PAUL: >Yes, I agree with most of this. I prefer to think of them
[styles, theories, paradigms, etc] in terms of sets of instructions,
rather than brute behaviours, recipe knowledge if you will, but I agree
with you. But how do you define a cultural system, what is it composed
of? and what is the stuff it is made of? Uniformity of what?

ALEX: The boundaries of a cultural system are ultimately defined by the
nature of the medium which is used by the agents within the system.
Communication and exchange between those agents takes place through a
process of mutual selection and combination of experience. It is out of
this network of mutual exchanges taking place within a limited
communicational space (defined by the number and connectivity of agents
within it) that cultural paradigms emerge as representative models of
collective experience and behaviour. The paradigm acts as the collective
memory of the system and is written out and retained in the in the
uniformity of behaviour and material artefact which characterise the
products of the cultural system. What is it made of? Information, bound
or organized into behavioural codes.

PAUL: >….I don't think these regularities are due to genetically
acquired instructions, rather they are socially acquired, that is they
have been culturally replicated through communication. The next stage
is to account for why this instruction was replicated, instead of
another, i.e kicking me back.

ALEX: Variable environments = variable responses.

PAUL: .....It depends how you define power, the capacity to control the
behaviour of others can be integrated into evolutionary culture theory.
(Ownership (in a non-homuncular sense - i.e. structural access) of the
means of replication and selection.

ALEX: Yes it can, but it is not necessary. It simply inserts conscious,
intentional motives into the otherwise neutral and non teleological
operations of a system. In my view it is a confusion of logical levels,
between the individual, conscious and motivated and the collective level
where there are no such motivations or intentions. Its a bit like
introducing God into human history to explain good and evil.
Evolutionary theory can generate uniformity and similarity of behaviour
without resort to conspiracy theory. As above, communication and
exchange between individuals in a confined communicational space will
generate increasing similarity of behaviour. Nobody needs to direct it
and there is ‘nobody’ at the collective level to do it anyway. (except
of course God and he seems to be on holiday since the 19th century).

PAUL: >If you take account of the context, they [cues/signals from the
environment] are predictable, most people are VERY predictable.

ALEX: Environments are variable. People are only ‘very predictable’
because they are faced with the same boring environments every day. In
fact they are incredibly creative. Look at the diversity of human
cultures on this planet or within a single society. Whose predictable?

PAUL: >They [cultural repertoires] are all transmitted culturally so the
media is the same, the >vector is different - mass media, and the code
might be different (language) but it is all culture. If you prefer
modelling to replication or imitation then fine, but I think we are
talking about the same process.

ALEX: There is a confusion here between ‘reality’ and the modelling of
that reality. There is no single cultural code which operates
throughout a society which in reality is a constellation of different
cultural systems. However, to model society requires us to reduce this
diversity to a kind of fictitious single system. Thats okay provided we
realize that it is a model - "its all culture" ( and not the territory,
so to speak). The media are not the same. Each system has its own medium
of expression, its own language, (music, scientific theories, words,
etc). For the model on the other hand, this veritable Babel of languages
can be reduced to the idea of communication and exchange. For me at
least, :"transmitted culturally" doesn’t mean anything in terms of a
specific mechanism and not in this context does "mass media".

PAUL: > Has it, by whom and how? [available theories describing
processes of cultural evolution]

ALEX: Too many, but my favourite remains Anthony Wilden: ‘System and
Structure’ , Essays in Communication and Exchange. But there is Bateson,
Waddington, the whole Complexity group, Nikolas Luman, Levi-Strauss,
Dennett, Freud, Foucault, Chomsky, and so on and on….. All these
relevant ideas ready and waiting to be integrated into a coherent theory
of cultural evolution. Meanwhile back at memetics we have the
biochemistry of individual brain structure. Ho hum!

PAUL: >If you can show me a way of using symbols in a way that doesn't
invoke a ghost in the machine, then I might agree with you. Until
somebody comes along with a more productive standard scientific
epistemology, then I'll stick to what we've got - because it works.

ALEX: This sounds like the mystique surrounding the process of
‘emergence’ and just as easily resolved. Symbols are the currency which
is used to exchange information between systems or between different
levels of the same system. Since we can’t exchange actual experience, we
have to represent (report) these experiences or events in another
‘language’. RELATIVE to the actual or initial environment this
description is symbolic. IT is not the ‘thing’, it represents the thing.
Its the map, not the territory. Its the model, not the environment. The
language of the model is the symbol. The symbolic function - the
capacity to translate actual events into reports about reports about
reports about those events is the sine qua non of human culture.
"Standard scientific epistemology" is a classic symbolic system which
tries to model the environment with culturally-derived symbols. What is
mathematics if not the purest form of symbolic system? Mathematics,
neolithic cave paintings, literature, art or the traces of particles in
an accelerator. What are these other than symbols? So where is the
ghost? Within each system, these are not symbols, but behaviours. In
other words, they are ‘real’.

PAUL: >Good, but what is the first level from which all the other system
levels emerge? [with reference to meta-systems}. Our stab in the dark is
assess the usefulness of the evolutionary loop of replication, variation
and selection, or in the Causal logic of natural selection,
"self-embedding" (Cloak) Emergence is magic unless you start by
describing the intitial conditions of the first system.

ALEX: I have already given an operational definition of ‘emergence’ as
the reorganization of the behaviours in an antecedent system by a
process of assimilation of their most typical or recurrent
characteristics into a new state called the meta-system. The "initial
conditions" simply describe the plurality of behaviours which existed
before the process of assimilation. Magic? I don’t think so. As to the
"first level from which all the other system levels emerge", I’m not
sure if this is a meaning of life question or not. If not, then this
"first level" inevitably involves diversity - different ways of doing
the same thing. Communication and exchange inevitably reduce this
diversity of behaviour to a single behavioural template or routine.
"Self-embedding" (recursion) I have found to be an extremely powerful
concept in cultural terms and worth discussing on its own at another

Regards and apologies for the length of the e-mail

Alex Brown
Lefke University
North Cyprus

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)