Re: On uplifting

Wesley Schwein (schwein@pegasus.montclair.edu)
Tue, 18 Nov 1997 14:47:24 -0500 (EST)


On 18 Nov 1997, Anders Sandberg wrote:

>[Twink:]
> > I also believe intelligence to be best measured by problem solving abilities.
> > Language would be a good way but not the only way to implement this
> > measurement. (There are nonlinguistic tests we can do to test uplifted
> > organisms, such as some which are already done with mice.:)
>
> Yes, but language is important in order to make the uplifted animals
> able to communicate with each other, and hence spread their own
> experiences and memes. It would likely accelerate development just as
> it appears to have done in humans.
>
> It strikes me that we really need to learn more about how we became
> intelligent in order to be good at uplifting. But uplifting also gives
> us valuable information on how we may have evolved.

How nice, I'm just doing some reading on that right now.

Modern human intelligence may result from combining different modules,
each with a different function and different learning mechanisms. For
example, we and other primates seem to have, to a greater or lesser degree
depending upon individual and species, innate knowledge of (1) social
interactions within the species, (2) an understanding of biology (folk
taxonomy, paying attention to life cycles, etc), and (3) a technical
intelligence (talent for building tools). All these focused modules exist
alongside the less powerful learning mechanisms of "general intelligence."

Chimps make tools, but do not act as though they have as good a grasp of
what they're doing as humans --they are using general learning techniques,
not skills specific to the medium. Perhaps consequentally, chimps must
struggle just to maintain what tool-making knowledge they have.

When analyses from different modules can be applied at the same time, we
make analogies based on similarities in pattern. If we apply social
knowledge to animals or landscape (or mythical entities), that's
anthropomorphism. If we apply natural history knowledge to humans, that's
totemism. Thus, early religion is a product of cognitive fluidity.

A literary theorist who has followed the progress of cognitive science
(and whose name and books I forget) proposes that our memory operates in
terms of literature -we remember a story, the detail of which are
'hyperlinked' to other skills and knowledges. For example, when I think
about living in Austria, I remember practicing German, traveling, learning
Habsburg history, and so forth, as I remember specific events --the
weekend Sasha, Oliver and I went to Budapest, what we saw, what we said.
Thus, we have narative memory with oral history and pedagogy, and
considering how long humanity spent as foragers and huter-gatherers, it
should be no surprise what a hold mythology has on us. The practice of
critical thinking, OTOH, is very much younger.

When we combine knowledge of dream lives with the practice of building
tools to alter our living condition, that's the beginning of art and
artisanry. Hunting weapons are designed with specific animals involved,
clothing is made in response to weather conditions, designs are added
according to correspondences with mythology.

Consciousness, under this perview, is paying attention to what you're
doing --each tool, each story, each social exchange is important as it is
used or as it occurs. Meaning has selective advantage.

Where _exactly_ does language fall into all this? We're not sure.
Certainly humans have language, and other primates don't. All chimp
language studies, though in the past interpreted differently, are now
taken as evidence that, while apes can use words, they do not have a
capacity for syntax. Using words is great, but without syntax language
remains at the level of a 2-year-old human. And certainly humans make
great use of technology while chimps do not. Can we safely adduce
language as the primary memetic vector? I certainly think so, but more
tool-using species would be great for comparison.

I think we can say that a species or genre of entities (AIs, Powers) needs
language or a functional equivilent in order to have memetic evolution.
Certainly a species that relied more on the individual's use of, say,
spatial or emotional understanding over inter-individual communication
would use very different approaches to memetic evolution.

These are the main sources I've using.

*** moderately technical, easy read
** moderately technical
* technical, in-depth

* Barkow, J.H., Cosmides, L., & Tooby, J., editors. (1992) _The Adapted
Mind: Evolutionary Psychology and the Generation of Culture_. Oxford:
Oxford University Press.

** Dunbar, R. (1996). _Grooming, Gossip, and the Evolution of
Language_. Cambridge, Massachusetts: Harvard University Press.

* Fodor, J.A. (1983). _Modularity of Mind: An Essay on Faculty
Psychology_. Cambridge, Mass. & London, UK: MIT Press.

** Gibson, K.R., & Ingold, T., editors. (1993). _Tools, Language, and
Cognition in Human Evolution_. Cambridge, England: Cambridge University
Press.

** Lieberman, Ph. (1991). _Uniquely Human: The Evolution of Speech,
Thought and Selfless Behavior_. Cambridge, Massachusetts: Harvard
University Press.

* Hauser, M.D. (1996) _The Evolution of Communication_. Cambridge,
Massachusetts: MIT Press.

*** Mithen, S. (1996). _The Prehistory of the Mind_. London: Thames &
Hudson.

** Nobel, W. & Davidson, I. (1996). _Human Evolution, Language, and
Mind: A Psychological and Archaeological Inquiry_. Hong Kong: Cambridge
University Press.

** Velichkovsky, B.M., & Rumbaugh, D.M., editors. (1996). _Communicating
Meaning: The Evolution and Development of Language_. Mahway, New Jersey:
Lawrence Erlbaum Associates.

--------------------------------------------------------------------------
Wesley Schwein Let's push natural selection out of business.