Greetings from Seattle

[Carl Lipo]

Hello to all!

Belated greetings from Seattle! I am a graduate student at the Department
of Anthropology at the University of Washington working with two other
students (Mark Madsen and Tim Hunt) modeling cultural transmission. Our
work is focused on examining cultural transmission in populations that are
structured in time and space. Essentially, we want to look at how we might
go about identifing "social groupings" that archaeologists have sort of
intuitively identified through the definition of phases.  We are seeking
to develop better methods for detecting boundaries in settlement patterns
and community interaction pattern.  The use of phases, it turns out,
hopelessly obscures most of the information present in the archaeological
record about relative interaction.  As a result, we desperately need tools
for examining evolving lineages. 

We have found that simulations provide a great way to generate data that
are the products of known configurations of interaction. We can test our
notions about what parameters, build null models, etc.  We began our work
using simulations last spring. This work resulted in a series of papers at
the SAA meetings.  Although there is relativley little discussion about
the simulation, a brief summary this is work can be viewed at: 

http://www.law.washington.edu/~madsen/saa/

This work was based on an object-oriented simulation (a C/C++ thing done
on the UWs DEC Alphas) where individuals transmitted information to one
another according to a set of rules and examined the patterns of that
interaction in time and space. We found we could use seriation (the old
archaeological method for ordering assemblages in time) as a tool for
examining spatial structure and potentially identifying interacting groups
in prehistory (we examined ceramic data from the Central Mississippi River
valley).  We also used the simulation to build a null-hypothesis for the 
migration and continued interaction of populations into the Pacific
(Melanesia-Polynesia) -- we generated expectations about similarity in
linguistic cognate frequencies on the basis of distance and accessibility.
We found that some of the current wisdom about ancestry and relatedness in
the Pacific is probably very mistaken. 

We are really excited about using SWARM as a way of building tools for the
study of interaction -- the key for building archaeological explanations
of cultural change in an evolutionary framework. We need to be able to
delineate lineages. The only way we can building these tools is to model
interacting populations and study their emprical consequencies. 

Once SWARM is running smoothly, our first goal is to take our first
transmission simulation (actually, it is not too different from AxelCult)
and convert it onto the SWARM platform. We will then be able to begin
adding in components that we were unable to implement in the first version
(adding in geographic configurations with boundaries, movement,
hierarchical sets of interactors, etc.). 

Well, back to work ... 

Cheers!

Carl
______________________________________________________________________________
Carl Lipo, Archy GSA    
Department of Anthropology, Box 353100
University of Washington                               
Seattle, WA 98195-3100                  
(206) 543-7712 FAX: (206) 543-3285                      
Email:  address@hidden