McShea, Daniel W. “Bernd Rosslenbroich: on the origin of autonomy: a new look at the major transitions in evolution.” (2015): 439-446.
URL1 URL2
What would a Grand Unified Theory of big-scale evolution look like? Here is one answer. It would unify the various trends that have been documented and suspected, the features of life that have been said to increase over its history—body size, fitness, intelligence, versatility, evolvability, energy intensiveness, energy rate density, and complexity-in-the-sense-of-part-types, and complexity-in-the-sense-of-hierarchy. It would show us how these putative trends are related to each other, how they are all the product of some single simple principle or some small set of principles. It would identify and connect all of the variables that are expected to increase over the long haul in evolution, perhaps in a single equation, or if not yet quantifiable, perhaps in a single breath. The obvious unifying candidate is fitness. Natural selection acts to adapt organisms to local circumstances, of course, but it might also act over the long term to produce adaptation on a bigger scale, to produce organisms that are adapted not only to the environment of the moment but to average environments, on timescales of tens or hundreds of millions of years. A modern snail species might not only be better suited to its present environment than the sister species with which it now competes. It might also be better adapted to the present era—to the average environment of that era—than its ancient ancestors were to theirs. More concretely, a modern snail might be better adapted to the now-540-million-year-old Phanerozoic Eon than a snail early in the Phanerozoic. Indeed, there are reasons in theory to think that selection could act cumulatively to produce adaptation on very long timespans (Darwin 1859), to produce a fitful trend in what has been called “absolute fitness” (Van Valen 1973).