The World of SPLectrum
Home > Positioning > Subjects > Evolution > The criterion
The Criterion
When a field asks whether something evolves in the Darwinian sense, it reaches for a criterion that biology wrote for itself and then found it could state without mentioning biology.
Lewontin’s three conditions
Richard Lewontin’s “The Units of Selection” (Annual Review of Ecology and Systematics, 1970) gave the Darwinian mechanism its most compact statement. Evolution by natural selection occurs in any population where three things hold: individuals vary in their form, physiology and behaviour; different variants have different rates of survival and reproduction in a given environment; and the variation is heritable, so that parents and offspring are correlated in their contribution to future generations. Lewontin drew the consequence that mattered for every field outside biology: any level of organisation that can be grouped into a population of units meeting the three conditions has the potential to evolve by natural selection. Genes, organisms, groups and species are candidates on equal terms, and so is anything else that varies, differs in success and passes the difference on.
Before Lewontin, Donald T. Campbell had made the same generalisation from the other direction. “Blind Variation and Selective Retention in Creative Thought as in Other Knowledge Processes” (Psychological Review, 1960) proposed blind variation with selective retention as the mechanism behind any growth of fit between a system and its environment, from trial-and-error learning to the growth of scientific knowledge: a criterion of the same shape, stated for knowledge rather than organisms.
Universal and generalised Darwinism
Richard Dawkins coined “universal Darwinism” in a 1983 essay, arguing that wherever life is found it will have arisen by Darwinian selection, no other mechanism being known that produces adaptive complexity. Daniel Dennett widened the claim in Darwin’s Dangerous Idea (1995): natural selection is a substrate-neutral algorithm, variation, selection and retention with no dependence on what the varying things are made of, and it is a “universal acid” that, in his words, “eats through just about every traditional concept and leaves in its wake a revolutionized world-view”. David Hull supplied the vocabulary the programme mostly uses, replicators and interactors, and applied it to science itself.
In economics the same move is called generalised Darwinism. Geoffrey Hodgson and Thorbjørn Knudsen’s Darwin’s Conjecture (2010) holds that the core principles of variation, inheritance and selection apply to social and economic evolution as a general framework, whatever the underlying mechanisms, and claims the framework for business practice, law, technology, language, custom and science. The standing objection is Ulrich Witt’s continuity hypothesis: economic evolution is continuous with biological evolution, shaped and constrained by its outcomes, rather than an analogue of it, so the abstract principles are the wrong level of description. The two contest the field.
Godfrey-Smith’s Darwinian populations
Peter Godfrey-Smith’s Darwinian Populations and Natural Selection (2009) turns the recipe into a space. A Darwinian population is any collection of particular things with the capacity to undergo evolution by natural selection, and a Darwinian individual is any member of one. Populations differ in degree along several parameters, three of which carry most of the weight: the fidelity of heredity, H, how closely offspring resemble parents; the continuity or smoothness of the fitness landscape, C, whether small changes in character make small changes in fitness; and S, how far differences in reproductive success depend on the intrinsic character of the individuals rather than on their position or surroundings. A fourth, V, is the abundance of variation. Paradigm populations sit high on all of these; marginal ones sit low on one or more, and “does it evolve?” gets a graded answer rather than a yes or a no.
Godfrey-Smith uses the space to place cases the field argues about. Cultural evolution he sets toward the marginal side, and memes lower still, on the ground that cultural transmission is low in fidelity and that the asymmetric power relations of human societies make reproductive success depend on position rather than intrinsic character, which is low S. His conclusion is that “Darwinism is not likely to unify and transform the social sciences.” The book also argues that the replicator, the unit universal Darwinism had built on, is not what selection requires: what it requires is reproduction with heredity, and copying is only one way of getting that.
What the criterion does not measure
The criterion is biology’s instrument for grading one sense of the word, descent with modification. It has nothing to say about the other sense, in which a single thing or system changes with no population, no heredity and no selection: the evolution of a star, of the Earth’s minerals, or of a physical state under a law of motion. That sense is older than the criterion, and those uses are not attempts to meet it; Hazen and his co-authors say so of mineral evolution in their own words, minerals do not mutate or compete, and “quartz does not out-compete olivine”.
Where a field has placed its own process in the space, the placement is the field’s and this page reports it. Nelson and Winter call their evolutionary economics “unabashedly Lamarckian”, since routines change under deliberate search and pass on what they acquire. Brian Arthur says outright that the base mechanism of technological evolution “is not Darwinian” but combinatorial. William Croft’s theory of language change is built to carry the full apparatus: the lingueme, a linguistic unit in an utterance, is the replicator, replication happens in the production of utterances, and the speaker is the interactor. Roger Lass holds that languages are not biological systems in any deep sense. The arguments behind these placements are on the Darwinism bundle’s cultural extensions page and on the lines of this subject.
See also: Evolution (the subject landing) · Darwinism · Cultural extensions · Lewontin · Godfrey-Smith · Dawkins · Dennett · Hull