Re: information transmission

Thu, 11 Mar 1999 12:25:07 GMT

Date: Thu, 11 Mar 1999 12:25:07 GMT
From: UEA <>
Subject: Re: information transmission


> For example, I know my home phone number. If it tell someone else the
> number and they have a good memory, then they know it, too. If a
> researcher asks both of us 'what is mark's home phone number,' we will
> both reply correctly. Thus, the researcher has discovered a similar
> behavior in two people that can only be explained via a sort of
> 'transmission' effect. The behavior (speaking my phone number properly)
> has been transmitted.
> This is what I suspect you are saying.
> Am I close?

Well, what I'm really saying is that this scenario only becomes interesting when the number HASN'T been "recreated" in me. There are
a number of evolutionary phenomena which could claim responsibility for this phenomenon. It could be genotypic (I'm an elephant,
and therefore have no concept of phone numbers or language), phenotypic (I'm deaf), memotypic (!?) (I can't speak English) or it
could be a transmission effect (I wasn't listening, an aeroplane went over when you spoke, and so on). My point is, if you don't know
the reason why, then you can't discount the possibility that it was a transmission effect that was responsible for the lack of success of
that meme.

> My perspective is different. I'm arguing that we have added the illusion
> of 'transmission' in this experiment. Actually, both subjects have
> learned a method of 'token' interpretation. In this case, the 'tokens'
> are sound vibrations known as words. I transmitted tokens (words) to the
> other person and he interpreted the token in such a manner that
> appropriate behavior occurred.
> When a receiver senses words, the sensations stimulate a meaning
> generation event. If the words are in a known language, meaning occurs.
> If the language is unknown, the token is dismissed as noise. The outcome
> of this mental activity is a slightly altered world view and new
> behavioral options.
> One might use 'transmit' as a sort of shorthand, but it confuses
> understanding of communication processes.

OK. Do you know of a letter in some Southern African languages which we write as a question mark (!) and is actually pronounced as
a click? Why does this letter not exist in our language - or any other language for that matter? The answer is that the people indigenous
to this area and need to communicate over vast expanses of plains. The ! letter gets transmitted better in this environment (correct me if
I'm wrong but I think it actually signals a following vowel - which doesn't travel so well). Now why does this happen - how did it
EVOLVE? Your theory, I believe, cannot answer that question, because it only talks about reproduction of existing tokens.

> Another example is this attempt to transmit my perspective about
> information to you. All I'm doing is sending word tokens for you to
> interpret. You may convince yourself that I think X, Y and Z. You may
> convince yourself that this knowledge was transmitted via my email. But,
> all I'm hoping for is some level of skill on your part to recreate my
> intentions.

Once again, what if the email was down? Your thinking would not have been recreated, and it would have been nothing to do with
yours or my mind, or any "internal" attempts to recreate knowledge. Your theory does not take into account this kind of phenomenon.

> >The reason that transmission is important, especially for
> >evolutionary theory, is that things can take place in that
> >medium ...
> >Memetics, being an evolutionary theory, is about the
> >replicators AND their environments - WITHOUT THE
> >linchpin of evolutionary theory). To put it simply, in
> >memetics, your concept of recreation is about the
> >replicators, and transmission is about the environments -
> >you need both.
> Again, let me see if I can recreate what your intent is here. I don't
> think I got the transmission of info, if one was intended.
> You've mentioned medium and environment. I suspect you the two are
> interchangeable in your argument.
> You've said that environments are where a survival differential can be
> observed.

No, I said that without environments, there can be no such thing as a survival differential, and thus no evolutionary theory of
information (which is what memetics is)

In other words, there needs to be an environment for natural
> selection to take place.
> Putting it all together, I suspect your perspective starts with a
> definition that memes are transmitted from person to person. By your
> definition, memes are assemblies of 'information' that get transmitted
> from one human to another. Thus, one can say that memes evolve based on
> how they survive transmission and then force their new host to transmit
> them to other humans. One might say memes are like DNA viruses, except
> they are 'information' bits instead of DNA assemblies.
> Since I discount transmission, your evolution scheme comes to a grinding
> halt.
> Am I close?

Yes and no. Yes, without transmission it can't be evolution. But no, I'm a "Dawkins A - 1976" man (see Gatherer's 1998 article in
JoM), so I think memes can be instantiated in artefacts as well as human minds. As Dennett would put it, a wheel of a cart is not only
transmitting some cargo from one place to another, it is also transmitting the notion of a wheel to other minds.

Interestingly, from reading the rest of your mail (which I haven't included here) it seems that your problem lies not with the word
"transmission" but with "information". I'd be quite prepared to concede that the physical/notional nature of the information as
instantiated in the brain is different from its nature whilst it is actually being transmitted. So you can call them different things, and if
you want to call the stuff in the mind "information" and the stuff in the ether (for want of a better word) something else (let's say merely
the token of the information), that's fine, but in my view, they are both instances of the same meme.

so to really open a can of worms . . .

EITHER a concept of both is inescapable if we want to do evolutionary theory, OR the fact of the matter is, that the information you
speak of is also nothing but tokens (I think this is Dennett's line). That is to say, the mother doesn't actually have the "experience" of
the burnt finger in her head, all she has is a token (a memory) of it. She can encode this token into a transmittable form (pictures,
language, acting out the scenario), and depending on how she encodes it (and which way the wind's blowing, so to speak) the baby will
get the idea (literally). It might be a token useful enough for the baby never to get its finger burned, but on this interpretation, it isn't the
experience of the burnt finger the baby has in its head, just a token of it.

cheers, alex rousso.

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)