The Seven Sins of Evolutionary Psychology - Konrad Lorenz Institute

Scott Atranpalm lines. From the fact of a bigger body, whatcould one possibly deduce about hearts, livers, kidneys,hands, faces, placentas and so on? And nevermind exaptations, such as language, or spandrels,such as religion. From the fact of a bigger (or denser,or more folded, or grayer) brain, what could one possiblydeduce about perception, emotion, categorization,inference or any of the other capabilities humansshare with apes but in more vastly elaboratedform? Probably nothing at all.It may well be true that little insight is to be gainedinto higher-order human cognitive structures byconsidering possible evolutionary origins and functions.Adopting the hypothesis of GOULD andLEWONTIN or P&P, which assumes this truth, is practicallyguaranteed to block insight, whether or not itis true. By contrast, adopting evolutionary psychology’srequirement that candidates for exaptationsand spandrels be described, as far as possible, withreference to evolved adaptations, then it might bepossible to find out if the hypothesis is true or not.If it is true, then evolutionary psychology wouldhave provided the empirical evidence that shows itto be a significant and surprising scientific insight,and not one that depends entirely on intuition, analogy,eloquence or wishful thinking. If it is not true,then evolutionary psychology will have helped todiscover something new about human nature.Three Examples of Modularity: NaiveMechanics, Theory of Mind, FolkbiologyEver since CHOMSKY jump-started the ‘cognitive revolution,’successors to the behaviorists whobelieved in an all-powerful general thinking devicehave tried to reconcile CHOMSKY’s insights with faithin flexible intelligence by reluctantly granting somespecificity to language, and language alone. But cognitivepsychology today concentrates more on discoveryand exploration of domain-specificmechanisms than on general-purpose computation.Each such device has a particular ‘content-bias’ inthat it targets some particular domain of stimuli inthe world (‘set of inputs’): for example, the edgesand trajectories of rigid three-dimensional bodiesthat move by physical contact between them(mechanics), the contingent motion a self-propelledactors that can coordinate interactions withouthaving physical contact (agency), or thebehaviors and appearances of nonhuman livingcreatures (species relations). The particular inferentialstructure of each domain-specific processor thentakes the isolated exemplars (or relatively poor samples)of the stimulus-set actually encountered in aperson’s life, and spontaneously projects these relativelyfragmentary instances onto richly-structuredcategories (‘classes of output’) of general relevanceto our species: for example, the objects and kinds offolkphysics (naive mechanics), folkpsychology(ToM) and folkbiology. Much work on domainspecificityhas developed, and now develops, independentlyof sociobiology or evolutionary psychology(ATRAN 1989; HIRSCHFELD 1996; KEIL 1989; LESLIE1994; CAREY/SPELKE 1994; SPERBER 1985).Within the emerging paradigm of cognitive domain-specificity,there is much speculation and controversy—again,as might be expected in any newlyemerging science. For example, there are competingaccounts of how human beings acquire basic knowledgeof the everyday biological world, including thecategorical limits of the biological domain and thecausal nature of its fundamental constituents. Oneinfluential view of conceptual development in folkbiologyhas been articulated by Susan CAREY and hercollaborators (CAREY 1985; CAREY/SPELKE 1994;JOHNSON/CAREY 1998). On this view, young children’sunderstanding of living things is embedded in afolkpsychological, rather than folkbiological, explanatoryframework. Only by age 7 do children begin toelaborate a specifically biological framework of theliving world, and only by age 10 does an autonomoustheory of biological causality emerge that is not basedon children’s understanding of how humans thinkand behave. A competing view is that folkbiology andfolkpsychology emerge early in childhood as largelyindependent domains of cognition that are clearlyevident by ages 4 or 5, and which may be innatelydifferentiated (ATRAN 1987; KEIL 1989; GELMAN/WELL-MAN 1991; HATANO/INAGAKI 1999).To address this issue, my colleagues and I carriedout a series of cross-cultural experiments (LÓPEZ et al1997; MEDIN et al. 1997; ATRAN et al. 2001). One setof experiments shows that by the age of 4–5 years(the earliest age tested in this regard) urban Americanand Yukatek Maya children employ a concept ofinnate species potential, or underlying essence, as aninferential framework for understanding the affiliationof an organism to a biological species, and forprojecting known and unknown biological propertiesto organisms in the face of uncertainty. Anotherset of experiments shows that the youngest Mayachildren do not have an anthropocentric understandingof the biological world. Children do notinitially need to reason about nonhuman livingkinds by analogy to human kinds. The fact thatAmerican children show anthropocentric bias ap-Evolution and Cognition ❘ 50 ❘ 2001, Vol. 7, No. 1