Re: On 'information transmission'

t (
Mon, 8 Mar 1999 19:55:37 EST

From: <>
Date: Mon, 8 Mar 1999 19:55:37 EST
Subject: Re: On 'information transmission'

In a message dated 3/4/99 7:10:46 AM Central Standard Time,

<< We may have been using the term "transmit" as a kind of specialized jargon
not mentioned in your dictionary. Yet taking the usage above, there is
still an important distinction. Information can be re-created with the help
of someone who already has "the same" information and who plays a crucial
causal role in that re-creation, or it can be re-created without the help
of someone who has "the same" information--in which case common parlance
would call the re-creation "independent." The former case is "replication"
in a very broad sense. "Replication," then, is what we (in our somewhat
specialized jargon) call "transmission." What is important about
"replication" in that very broad sense is that it constitutes a minimum
condition for recursion, (what Dennett calls an "algorithm," is more
exactly a _recursive algorithm_ or _recursive process_), and this is a
minimum condition for evolution by natural selection.>>

Interesting point. It raises the question, "Can we really say that
information is EVER transmitted?" In some respect, when I recieve information
through my modem, my computer already has much of the "same information", and
my software is really just "recreating" the message by following the
instructions as sent via the modem and already contained in my computer's
software. And then my brain further "recreates" meanings because it too
already has much of the same information as the original author.

Perhaps a "transmission" might be thought more as a communincation that that
actually contains a higher information transmitted to message/meaning ratio,
where as a representation contains a much lower information transmitted to
message/meaning ratio.

In transmission, Information/Message(meaning) approaches One. (more like my
computer with a modem). In representation, Information/Message(meaning)
approaches Zero. (more like Aaron and me talking, more "bang for the
information buck")

Both cases would be instances of replication necessary for an evolutionary
algorithm to operate. We can also have transmissions nested within
representations (like this EM list). Artificial Intelligence, once achieved,
would present a continuous transition between transmission and representation.

In terms of cultural evolution, however, I would think that it would be more
important to focus on representations and deal with transmissions mostly in
terms of the role that they play in forming representations. In Gary Taylor's
book, which I talked about in my last post, he provides a very good analysis
of *representation*. This is where I think that a good cultural inclination
like his would help a lot. I also tend to think that representations are what
people are generally thinking about when I find memes discussed.

"Where memes reside" is really a two part question. Memotypically they reside
outside of the mind, Like a picture hanging in a museum, or postcards made of
that same work. Phemotypically they are expressed/recreated, or represented
in the human mind/brain upon experiencing the memotype, and it is in this
phemotypical expression of the memotype where cultural selection occurs,
determining whether and how the memotype will be replicated.

The genotype is the "potentially immortal" biological replicator when we think
in terms of "selfish genes". When we are talking about the genes for sickle
cell anemia, we are really talking about both of its phenotypical expressions
- recessive resistance to malaria AND the dominant expression of the disease
sickle cell anemia. Thus we are really talking about the genotype. It would
make better sense to treat the memotype as well as the understood default when
we talk about "memes", leaving the meanings represented to the discussion of
the "extended phemotype".


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)