Theoretical Integration, Cooperation, and Theories as Tracking Devices

The theoretical problem of integrating evolution, heredity, development, and cognition has a long pedigree with a complicated history. Many of these fields were considered the subjects of one science in the 19th century (Maienschein 1987). Leading theoretical biologists of the age wrote large, expansive treatises that biologists now can read only with equal measures of wonder and incredulity: wonder at their wide and deep learning and incredulity at their peculiar visions of biological integration. Think of Herbert Spencer (1900, pt 2: 367): “Evolution is an integration of matter and concomitant dissipation of motion; during which the matter passes from a relatively indefinite, incoherent homogeneity to a relatively definite, coherent heterogeneity and during which the retained motion undergoes a parallel transformation.” Or consider Ernst Haeckel (Gastraea theory), Hans Driesch (entelechies), August Weismann (biophors, ids, idants), and even Charles Darwin (use inheritance, pangenesis). None of these ideas were especially crazy in their time, but they drove integration projects that are quite different from ours. Historically, each of these special subjects has been a contender for the role of primary theory from which an understanding of all biological phenomena would flow. To take but a few examples, Dobzhansky (1973: 125) gave us “Nothing in biology makes sense except in the light of evolution” on behalf of Darwin’s theory. De Vries gave us his Mutationstheorie and Johannsen his pure line theory as alternative explanations of the root cause of hereditary change. Both were intended originally to replace natural selection as theoretically primary. Driesch gave us entelechies of individual development as a fundamental philosophical principle for the biological sciences. James Mark Baldwin, Pierre Teilhard de Chardin, Francisco Varela, Rupert Riedl, and Donald Campbell, in very different ways, put cognition in the center of theoretical concern for general biological science, evolution in particular. Thinking about integration from the point of view of any of these specialties has traditionally depended on a notion of unification by explanatory theory reduction: explanations flow from a fundamental primary theory to phenomena that are organized at the surfaces, so to speak, of those phenomena considered central for the primary theory—for example, trait heredity and organism development in the light of evolution in Darwin’s day, or population change and the development of hybrids in the light of Mendelian heredity circa 1905, recapitulation and phylogeny in the light of ontogeny in Haeckel’s, de Beer’s, and Gould’s days (Haeckel 1866; de Beer 1930; Gould 1977), learning as coequal to inheritance and selection as a force in evolution. In the 20th century, phenomena have tended to be organized hierarchically, in terms of the composition of matter from the smallest to the largest units, so that there is a general expectation and drift of reductionism toward lower levels of compositional organization—to the behavior of smaller and smaller bits of matter—as the primary concern of fundamental theory. In biology, this reductionist drift led theorists from organisms to organs and tissues, to cells, and on to molecules as the foci of theoretical interest. Genes, the “molecules of life,” have been the focus of biological reductionism for nearly all of that century. A different model of integration centers on cooperation and communication among theoretical and phenomenal equals, rather than on imperialism and competition for primacy and fundamentality, which reduces or replaces one theory by another or trivializes one explanandum as epiphenomenal to another. An explanatory domain can become integrated when its bumps, twists, and turns are smoothly traversable, but we need not achieve integration by leveling the domain and making it conceptually homogeneous, just as nation-states can be unified by the smooth flow of goods, services, people, and ideas across their borders rather than by the obliteration of local and regional differences making flow irrelevant.