>Brian Harper wrote:
>
>>a) The argument seems to have changed somewhat from that involved
>>with tool making. Tool making would have to be considered an innovation,
>>so to be consistent with the arguments in this post it seems to me that
>>the key is not how long tool making was in a static condition but rather
>>how rapidly tool making appeared. Anyone know what the time-frame was?
>
GM:========
>Tool making appears suddenly at 2.6 myr ago. It is very hard to have a half
>formed tool. The stone was either fashioned or not and so it is a digital
>situation. In actuality we don't know when tool making began because there
>may have been bone and wooden tools prior to this time which didn't survive
>to the present. We also almost assuredly do not have the very oldest tools.
> it would be highly unlikely that we have found the exact site of the first
>stone tool making.
>
>In any event the advent of tool making does not coincide with the advent of
>art or with the different advent of Jim's Shaman art.
>
Glenns comments reminded me of a few things I had been thinking
about. First of all, just from the definition one would expect
true innovations to appear rather suddenly. But "suddeness" could
also be an artifact of our methods. As Glenn says, tool making could
have been around long before 2.6 myr ago. Perhaps what was sudden
was the change in materials.
Also, the answer to the question "how sudden is sudden?" should
depend very much on the situation and the model under consideration.
For example, tool making and art can be learned and passed on to
future generations. Changes can thus occur more rapidly than if
this were a genetic trait.
Further, not all models of evolution require gradualistic changes. Models
involving nonlinear dynamics can involve rapid changes. A recent
article in the journal <Complexity>
Robertson, D.S. and M.C. Grant (1996). "Feedback and Chaos in
Darwinian Evolution," <Complexity>, 2(1):10-14.
has caused me to rethink the role of natural selection itself. Previously I
had been thinking of natural selection as a kind of stabilizing mechanism,
with the real evolution occuring by some other mechanism. The authors
of the article above give a good argument that even classical Darwinism
is a highly nonlinear process which is potentially unstable and even
chaotic. Of course, nonlinear systems of this type are not continuously
unstable. They can have extremely long periods of stability followed by
rapid bursts of change (PE).
Brian Harper
Associate Professor
Applied Mechanics
Ohio State University