Re: virus: Replication of Memes

Tue, 13 Aug 1996 09:54:08 -0500

David McFadzean wrote:
> KMO (quoting Fred Dretske) wrote:

> > "[T]here is no reason to think that every meaningful signal must carry
> > information or, if it does, that the information it carries must be
> > identical to it's meaning."
> If the meaning of the signal is taken to be its effect, then I agree
> with the 2nd part of this statement but not the first. I wonder what
> he means by 'meaningful'.

The meaning of the signal is just that. Words have assigned meanings,
and those meanings are not dependent on an effect they have in any one

I have no cat and you know that I have no cat. Now suppose I leave a
message on your answering machine saying, "Stephen, my cat is
pregnant." That sentence is meaningful. The nouns refer to things or
types of things with which you are familiar, and the syntax of the
sentece relates the words in a meaningful fashion, yet, since you know I
don't have a cat, the sentence does not convey the information that my
cat is pregnant. It might carry the information that the feds are on to
our little scheme and it's time to lie low for a while, but the
conveyance of that information would require that you know more than
just the content of the signal (sentence). The content of the signal is
what it means. The information it carries is what you can learn from
it. Because I don't have a cat, and so you couldn't possibly learn
(form a true belief) that my cat is pregnant. The signal could convey a
wide range of information, but it can't convey the information that my
cat is pregnant. The sentence is meaningful, but the information it
might convey (if it conveyed any information at all) is not identical
with its meaning.

> What one learns and what one can learn are often radically different.
> Which corresponds to the information?

This is where I'm a bit fuzzy on Dretske's view. I suspect that he
means to say that information corresponds to both what is learned and
what could be learned from a signal. What he fails (as far as I can
tell) to differintiate between is the notion of information that is
contained in a signal and information that is successfully conveyed to a
recipient. If he's right, most signals contain for more information
than they convey to the recient of the signal.
> > "Roughly speaking, information is that commodity capable of yielding
> > knowledge, and what information a signal carries is what we can learn
> > from it."
> Here he seems to identify information with what one can theoretically
> learn from the signal, potential knowledge as opposed to what is
> actually realized.

Again, what is learned and what could be learned both correspond to
> > "[I]nformation is a question of what, and how much can be learned from a
> > particular signal, and there simply is NO LIMIT to what can be learned
> > from a particular signal about another state of affairs."

> Does this imply a signal can potentially contain infinite information?

It would seem so. For Dretske, information is that which facilitates
knowledge. While you can believe something that isn't true, you can't
know something that isn't true. Given this definition of information,
you can't learn something that isn't true, i.e. no signal can give you
knowledge of something that is false. That being the case, despite what
Dretske says in the above quote, there is some limit to what can be
learned from any given signal. Only true things can be learned. Of
course there may be an infinte number of facts, and, if so, a signal
could potentially carry an infinte number of facts. It seems, however,
that the recipient would have to posses infinte processing power and
near omniscience to start with in order to actually learn an infinte
number of true facts from the signal.

> > "How much information a signal carries is NOT a function of how much
> > information the recipient thinks it carries."
> Does this mean the information the signal carries IS a function of how
> much the sender thinks it carries, or is the sender irrelevant too?

Many signals have an unconscious source. I don't think a drill bit at
the bottom of the ocean thinks anything about the signals it sends to
oil company geologists on the platform above.
> > According to Dretske, what is to be expected from any scientific theory
> > is "a more or less complete, precise, and systematic description of
> > those entities and processes underlying the phenomena of interest."
> >
> > If we accept Dretske's definition of information and his criteria for
> > scientific theories, and if we take memes to be the information in
> > cultural transmission as opposed to the content of cultural signals,
> > then it seems unlikely that we'll be able to formulate a science of
> > memetics.
> My own view on this is if memetics has the possibility of being a
> science, it will be based more on computational metaphors (treating
> memes as programs) than information theory (which is concerned more
> with the process of transmission).

Hmmm... Seeing memes as historical entities which are necessarily
classified by their lineage is central to the concept of memetics.
Information theory (or 'signal theory' as Dretske would call it) seems
to be an indispensible element. If there's an intractable problem with
identifying a meme at the stage of it's lifecycle when it is transmitted
from one host to the next, then there seems to be an intractible problem
for a science of memetics.

The concept of the memes as ideas seen through a Darwinian lens is not
an indespensible concept if you limit your inquiry to what goes in
inside the skull and set aside issues dealing with the transmission of
information between individuals. If we're not concerned with
transmission, we don't need to talk about memes.
> > Of course, if we use Dretske's notion of information and if a honey
> > bee's dance or a termite's pheremones are to convey information, then
> > bees and termites must be capable of KNOWING things.
> >
> > I have a problem with this, as I think knowledge requires justified true
> > belief, and I'm not convinced that bees and termites have beliefs.
> Let me ask you this then: Do bees and termites behave *as if* they
> have beliefs?

They most certainly do, but so do alot of things I wouldn't be willing
have beliefs, e.g. industrial robots, coffee makers, and house plants.

Take care. -KMO
"The least questioned assumptions are often the most questionable."

-Paul Broca

"All our science, measured against reality, is primitive and childlike-
and yet it is the most precious thing we have."

-Albert Einstein