CARTER, HAROLD, JR.; PHD
STATE UNIVERSITY OF NEW YORK AT BUFFALO, 1986
ANTHROPOLOGY, CULTURAL (0326)
This work reports the results of the comparison to two computer simulation models
of diffusion. These
models are derived from two distinct views of diffusion. The Contagion Model
is based on the
anthropologically ambient view of diffusion as based primarily on proximity
of the exchanging cultures.
This model is similar to the general model of diffusion of innovation in geography,
the Hagerstrand
Model. The Selection Model sees diffusion as a process similar to gene flow
in biology. In this view, what
is passed between cultures is the same as in the contagion model; however, the
selection model differs
by seeing what is maintained in a culture's 'trait inventory' as the result
of selection. This model is drawn
from an evolutionary perspective for the study of cultural change. It includes
ideas similar to Clark
Wissler's, Raoul Naroll's selection interpretation, and Cavalli-Sforza and Feldman's
cultural selection -
natural selection model. The perspective is derived by considering biological
or cultural populations that
undergo evolution as systems. By using the basic notions of system, the main
features of evolutionary
biological theory and applying these to cultural phenomena, a framework for
seeing cultural change as
evolution is developed. The key points gained from the results of the 1600 simulations
and their
comparison with ethnographic data on 12 variables for Africa and North America
are: (1) The features of
the Contagion Model produce a fixation rate of .969 which means that fewer than
4% of the simulation
runs result in a trait dispersion comparable to the ethnographic data. (2) The
features of the Selection
Model produce fixation in less than 14% of the runs. More than 86% of the runs
produce dispersions
similar to those found for the ethnographic data. One of the model's chief features,
consistent with
previous research, are fall off curves of similarity with distance. (3) The
fixation rate measured in the
ethnographic data is between 12% and 17%, which closely matches the rate found
in the selection
model simulations. Within the limits of this study the Model II, Diffusion with
Selection, results better fit
the real data and support the selectionist cultural evolution interpretation
of diffusion.
Social
Systems Simulation Group
P.O. Box 6904 San Diego, CA 92166-0904 Roland Werner, Principal Phone/FAX (619) 660-1603 |