Eva Jablonka and Marion J. Lamb have created their own genre of science writing, marked by visits to imaginary worlds, end-of-chapter dialogs with an engaged lay person, and points made in the text being fancifully illustrated by Anna Zeligowski. In this book the playfulness masks an imaginative integration of disparate concepts into elaboration of a single meta-concept, facilitated variation. The four dimensions of the book's title that the authors represent as almost-equivalent agents in the creation of variation are genetic, epigenetic, behavioral and symbolic, "symbolic" standing for human language and culture. By identifying intermediates between the 4 agents the authors imply they actually form a continuous spectrum of agency, from random genetic variation at one end, entirely non-facilitated, to highly facilitated variation in the form of human planning at the other. Very nice. Natural selection originally failed for want of a source of continuous variation. Now suddenly we've knee deep in variation of all kinds.
For me, all this does is bring into sharp focus the inadequacy of the other pillar of modern evolutionary theory: selection. In this book it's treated as magic. While variation undergoes the most minute scrutiny, facilitation is entirely taken for granted. Some quotes:
"The laboratory experiments we described have shown that through selection a phenotype that has been revealed by an inducing substance or stress can be genetically assimilated. . .
"After rearing the larvae on salty food for many generations, the modified anal papilla was assimilated, being retained even when the larvae were raised on normal food. Thus, in this case, an acquired character became an inherited one in the traditional sense: the trait was adaptive and genetic assimilation was through natural, rather than through artificial, selection.
"Of course, their subsequent evolutionary elaboration required further selection, but there is no reason to doubt that such selection would have taken place.
"This is certainly an adaptive response, and it is not too difficult to think how it may have evolved. Selecting genetic changes that link the existing mechanisms that turn genes on an off to the error-prone mechanisms that repair DNA would do the trick.
"One possible answer is that EISs were selected because they enabled early cells to survive conditions that were constantly changing."
These authors seem uniquely qualified to question established wisdom, so how can they titrate "facilitation" from one end of their spectrum to the other and not wonder if perhaps all variation requires facilitation of some kind, and wonder what form it takes at the genetic end? Why is "selection" the only term they use without a little tire-kicking?
My impression of this book is of a lot of separated dots being brought together into the form of a triangle, when what I wanted was a square.