Generative Entrenchment and the scaffolding of individual development and social institutions

William Wimsatt
Philosophy, University of Chicago

     Full text: Not available
     Last modified: April 21, 2005
     Presentation date: 07/14/2005 11:00 AM in ROZH 105
     (View Schedule)

Abstract
Generative entrenchment is found in any systems showing development and
a life cycle subject to selection, and thus in reproducers threading
through core configurations. Consider an adaptive structure generated
from a smaller number of elements to make a larger system. This
suggests the relation between genes and phenotype or a set of axioms
and the (much larger) range of results generated from them. Either can
be reproduced for their adaptive consequences. Such "generative"
systems will have some parts with a greater and more varied involvement
than others in the production of the larger system. This differential
entrenchment is very robust, indeed "generic". Moreover, selection will
drive away from a totally symmetric state towards increased evolving
asymmetry of contributions-spontaneous symmetry-breaking. A
developmental process will thereby generate differential rates of
evolutionary change among its parts because those with widespread
impacts are far more likely to cause major malfunctions if changed.
Evolution is skewed towards changing things that don't matter much. (In
such systems probabilistic predictions of directions of evolutionary
change are possible without knowledge of the genetics-or, for culture,
without any relevant genetics!)

These predictions do have exceptions. Some structural arrangements can
change expected outcomes. But we can analyze what factors can make
predictions break down, yielding major changes which may be sources of
important innovation. Regulation, canalization, redundancy and excess
capacity can allow deeper and novel experimentation without deleterious
consequences and occur spontaneously in biology (e.g., "tandem
duplications" in genetics) or social systems (bringing more people than
necessary to accomplish a task may give more freedom in how it is
done). These are all found in biological systems, but there are richer
and more powerful cultural means for making deep modifications (with
far reaching implications) and getting away with it. I will discuss
half a dozen features which make major innovations possible for culture
which have no analogues, or weaker ones, for biological systems.

Things that get relatively fixed in cultural systems by entrenchment
tend both to play a larger generative role, and to be protected from
falsification or rejection in other ways. These are properties
associate with norms, conventions, and standards, and with institutions
that scaffold culture and individual development. This gives GE a
potentially powerful edge in explaining features of systems that are
curious hybrids of normative and descriptive. We tend to layer newer
features on older ones (even if creativity and revolutionary change
will occasionally upset apple carts) and to use things we already have
readily to hand for new functions. This will tend to accumulate
contingencies of structure, behavior, procedures, technology, and
symbolic culture which yield arbitrary differences in cultures.
Repetitions of this cycle should generate a kind of fractal order of
contingencies on multiple scales of generality and importance, often
noted as characteristic of historical process. Such fractal richness
also suggests one of the sources of the "thick descriptions"
characteristic of culture.

Multiple Paper Session:
Other papers in this session:
When Does Cultural Selection Explain Cultural Novelty?
Cultural Variation and Epistemic Access
Constraints on Cultural Adaptation
Reproducers and the evolutionary development of culture
Core Configurations that Can Think, Learn and Create