Home > Positioning > Persons > Campbell

Donald T. Campbell (1916–1996)

Campbell argued that every gain in knowledge, from an organism’s trial and error to the growth of science, comes about by the same process: variations that are blind to the solution, a selection among them, and a way of keeping what is selected. He stated the scheme for creative thought in 1960, fourteen years before he gave it the name evolutionary epistemology, and ten years before Lewontin’s generalisation of natural selection to any level of organisation. The same career produced much of the working apparatus of social science methodology, the quasi-experiment and its threats to validity, the multitrait-multimethod matrix, and the warning, now known as Campbell’s law, that a social indicator used for decisions corrupts what it measures.

Donald Thomas Campbell was born on 20 November 1916 in Grass Lake, Michigan. He studied psychology at the University of California, Berkeley, taking his BA in 1939 and, after service in the Naval Reserve interrupted his doctoral studies, his PhD in 1947. He taught at Ohio State and the University of Chicago before moving in 1953 to Northwestern University, where he spent twenty-six years; he then held a chair at Syracuse University’s Maxwell School from 1979 to 1982, and finished his career at Lehigh University. He was elected to the National Academy of Sciences and the American Academy of Arts and Sciences in 1973, served as president of the American Psychological Association in 1975, and was elected to the American Philosophical Society in 1993. He died in Bethlehem, Pennsylvania, on 6 May 1996. The Campbell Collaboration, founded in 2000 to produce systematic reviews of social interventions, is named after him.

Key concepts

Blind variation and selective retention. “Blind Variation and Selective Retention in Creative Thought as in Other Knowledge Processes” (Psychological Review, 1960) set out three conditions for any process that produces fit between a system and its environment: a mechanism for introducing variation, a consistent selection process, and a mechanism for preserving and propagating the selected variants. Blind does not mean random. It means that the variations are not produced with foreknowledge of which will succeed, so that where a solution is not already known, trials cannot be aimed at it. The paper applied the scheme to creative thought, where ideas are tried and discarded in the mind before one is kept, and presented it as the same process at work in natural selection.

Evolutionary epistemology. “Evolutionary Epistemology”, Campbell’s contribution to the volume on Popper in the Library of Living Philosophers (1974), gave the programme its name and its fullest form. Knowledge processes form a nested hierarchy of ten levels: non-mnemonic problem solving, vicarious locomotor devices, habit, instinct, visually supported thought, mnemonically supported thought, observational learning and imitation, language, cultural cumulation, and science. Each higher level is a vicarious selector, a shortcut that substitutes for trials at a lower level: an animal that sees an obstacle need not bump into it, and a theory rejected in thought need not be tried in the world. The shortcuts are themselves products of selection, and their knowledge is only as good as the selection that shaped them. Campbell read Popper’s conjectures and refutations as the scheme at the level of science.

Downward causation. In a paper of the same year on hierarchically organised biological systems, Campbell introduced the term “downward causation” for the way selection acting on a higher-level system constrains the distribution of events at the lower levels that compose it, so that a lower-level structure is explained both by the laws that govern its parts and by the selection that acted on the whole. The term has since become part of the vocabulary of the emergence debate.

Social and biological evolution. His presidential address to the American Psychological Association, “On the Conflicts between Biological and Social Evolution and between Psychology and Moral Tradition” (1975), argued that biological evolution in a species whose cooperators are genetic competitors can produce only limited self-sacrificial altruism, and that the complex cooperation of urban societies was built by social evolution, whose moral traditions work against individual selfish tendencies that biological evolution still selects. Traditional moral teachings, on this view, may carry the results of that social selection.

Methodology. With Donald Fiske, Campbell devised the multitrait-multimethod matrix (1959), a way of testing whether a measure tracks what it claims to by comparing several traits measured by several methods. With Julian Stanley he wrote Experimental and Quasi-Experimental Designs for Research (1963, as a handbook chapter; a book in 1966), and with Thomas Cook Quasi-Experimentation (1979): how to draw causal conclusions where full randomised experiments are impossible, by naming the threats to internal and external validity and designing against them. “Assessing the Impact of Planned Social Change” (1976) stated what became Campbell’s law: “The more any quantitative social indicator is used for social decision-making, the more subject it will be to corruption pressures and the more apt it will be to distort and corrupt the social processes it is intended to monitor.” His hope for an “experimenting society”, in which policies would be tried as experiments and kept or dropped on the evidence, applied the same scheme to government.

Where Campbell stops

The weight of the scheme falls on “blind”, and that is where it was pressed. Paul Thagard’s “Against Evolutionary Epistemology” (1980) argued that scientific theories are not generated blindly but by reasoning aimed at the problem in hand, so that the variation in science is guided and the analogy with natural selection fails at the point Campbell needed it. Robert Sternberg made the same argument for creativity in 1998, that variation in human thought is sighted rather than blind. Dean Keith Simonton has defended Campbell’s model through the same decades, and the dispute over how blind creative variation is remains open.

The hierarchy also joins what later philosophers of the field separated. Michael Bradie’s distinction between the evolution of cognitive mechanisms and an evolutionary account of theories marks two programmes that can succeed or fail independently: that the capacities for knowing were shaped by natural selection does not show that theories are selected in the same way.

Key works

See also: Evolution · Science and knowledge · The criterion · Popper · Hull · Lewontin · Emergence: the types