Unreal Nature

September 22, 2010

Differential Entrenchment

Filed under: Uncategorized — unrealnature @ 7:57 am

… Changes accumulate elsewhere while these deeper features appear relatively “frozen” over evolutionary time.

This is from an essay “Generative Entrenchment and the Developmental Systems Approach to Evolutionary Processes” by William C. Wimsatt (2001):

… Any evolving systems must meet what Lewontin has called “Darwin’s principles.” They must:

1. have descendents that differ in their properties (variation).
2. some of which are heritable (heritable variation), and
3. have varying causal tendencies to have descendants (heritable variation in fitness).

… But there are two further conditions of great generality. These conditions recognize development’s central role in the evolutionary process. I know of no interesting evolutionary process whatsoever (physical or conceptual) that does not meet them. With these DST becomes not just an additional perspective on evolution, but fundamental to it. The same entities which meet the first three conditions must also be:

4. structures which are generated over time so they have a developmental history (generativity), and
5. some elements that have larger or more pervasive effects than others in that production (differential entrenchment).

Then different elements in the structures characteristically have downstream effects of different magnitudes. The generative entrenchment (GE) of an element is the magnitude of those effects in that generation or life cycle. Elements with larger degrees of GE are generators. This is a degree property. The GE of an element in an evolutionary unit has multiple deep consequences for its evolutionary fate and character, and that of systems impinging on it.

… system elements with greater GE tend to be much more evolutionarily conservative. Changes accumulate elsewhere while these deeper features appear relatively “frozen” over evolutionary time.

… History matters in evolution. It is not too far wrong to say that everything interesting about adaptation is a product of selection for improvements in design, or of history, or their interaction. Gould has emphasized the role of contingency in evolutionary processes, arguing that minor unrelated “accidents” or “incidents” can massively change evolutionary history. It seems plausible — indeed (as I will argue) almost inescapable — that a successive layered patchwork of contingencies has affected not only the detailed organic designs we see, and variations among conspecific organisms, but also much deeper things — the very configuration and definition of the possible design space, and the regions in it they occupy. Deep accidents from the distant past not only define the constraints of current (so-called) optimizations, but constraints on these constraints, and so on, moving backward through a history of the deposition of exaptive dependencies which become framing principles for the design of successively acquired and modified adaptations.

… Because of the dominant role of population genetics in evolutionary biology, stochastic genetic events and processes — point mutations, tandem duplications, inversions, segregation events, independent assortments, and other recombinations in inheritance — are the commonly cited sources of contingency in evolution. Equally important are chance ecological events: meetings leading to matings, migrations, symbioses, exclusions, parasitisms, and predations. As George Williams quipped, “To a plankton, a great blue whale is an act of God.” Better design as a plankton will not help if it is in the wrong place at the wrong time.

… The more deeply generatively entrenched something becomes, the more context-independent and the larger is the fitness loss if it is disturbed. Deeply generatively entrenched things become really established — the deepest become functional necessities.

… most evolutionary “contingencies” start small — the luck of the draw for George Williams’s plankton or single-base mutations initiating selective cascades of layered exaptations with divergent consequences. Oxygen production — metabolic byproduct in ancient plants — presumably started small, but hardly any contingency has had broader or greater consequences for evolution, by spreading as these plants succeeded, and becoming a much larger process. As this atmospheric poison rose in concentration, oxygen was initially adapted to, then became utilized almost universally throughout the animal kingdom, driving an energetically richer metabolism which now depends on it.

… taking developmental systems seriously simultaneously explains two of the most striking features of organic life:

First, generative entrenchment explains the frequent deep similarities in organic architecture of varying degrees of generality roughly mapping phylogeny onto causal depth in developmental process in the life cycle of organisms.

… Second, GE explains why some features that seem bizarre, inefficient, or only arbitrary should show so much persistence, so much evolutionary inertia after chance or selection have kept them for a time, and other things have come to depend upon them. It does not explain why they occur in the first place. These marks of contingency are more quickly lost — never visible (as in the quantum transitions through which an ionization might lead to a base substitution), or local and fleeting (the predator from which it zigged instead of zagged and got away), or not so local and not so fleeting but still not saved (changing selective optima).

… People have pointed to the importance of contingency in evolution, argued for it, presupposed it, but not tried to explain it — not as a property of individual cases (left- versus right-coiling shells) or periods (the great loss of Cambrian Bauplans) — but as a generic property of systems that can fix and build up richly textured tapestries of accidents. GE does so. … Unlike the leveling and homogenizing processes attributed to the second law of thermodynamics, evolution seems to leave an ever more complex and filigreed history over time. Generative entrenchment is the primary explanation for this difference. Note that in this sense, evolution can be cumulative without being progressive.



Blog at WordPress.com.

%d bloggers like this: