From Konrad Lorenz’s The Natural Science of the Human Species (1997: 61-62), the astonishing “Russian Manuscript” written in a Soviet prison camp:
At the age of 5 ½ years, my daughter Agnes was able to recognize any bird belonging to the rail family after becoming familiar with just two representatives, albeit in some depth. Her achievement was remarkable for the following reason: Both of the species with which she was familiar–the moorhen (Gallinula chloropus) and the coot (Fulica atra)–are waterfowl that have adopted a generally ducklike form through adaptation to aquatic life. But the rail family (Raliidae) contains birds with a wide variety of different forms, including small steppe inhabitants with a pronounced chickenlike external appearance, such as the corncrake, and long-legged wading birds, such as Allen’s gallinule, along with many others. Carefully avoiding any suggestive questions or involuntary guidance, I conducted a test on my young daughter using the comprehensive bird collection at the Zoological Garden in Schönbrunn, with which she was quite unfamiliar. She made no mistakes, either with the most chickenlike members of the rail family or with the most rail-like members of the chicken family, and in an enclosure that also contained herons and other wading birds she immediately picked out Allen’s gallinule as a rail, despite the fact that this bird is a very unusual member of the family in external appearance because of its ultramarine blue coloration. In doing this, she solved a problem that had completely defeated Linnaeus, the founding father of zoological systematics. In his “natural system,” the individual members of the rail family are classified into different groups of striding, swimming, or wading birds according to their external adaptations! How was it that this 5-year-old saw something that Linnaeus had failed to see? It was because she was closely familiar with the living bird, whereas Linnaeus had only examined the prepared specimen. In the general behavior and locomotion of the living bird, a wealth of fine details generate a Gestalt. This experiential entity incorporates more quality-determining characters than Linnaeus took into account in his identification key. When I asked my daughter how she could recognize members of the rail family as such, all she could say was that they “are like a moorhen.” The number of characters that determined the quality of my daughter’s “rail Gestalt” for the birds with which she was familiar must have been so great that they completely overwhelmed the aquatic bird features shared by moorhens and coots. In spite of the somewhat distorted basis of her “inference,” she was able to achieve a correct result through intuitive Gestalt perception. The most knowledgeable and eminent modern zoologists concerned with fine systematics exploit this feat of Gestalt perception quite consciously by taking observations of the living animal and of its “physiognomy” as indicators in their systematic investigations.
Realize that Lorenz is not espousing the “supernatural infallibility of intuition” (63); molecular phylogenetics’ dismantling of intuitively cohesive taxonomic units (not just those of folk taxonomy but also, on numerous occasions, the more formally delineated yet ultimately intuition-led groupings of traditional morphologists), are proof enough of his admonition: “intuitively derived certitudes are just as correct as those that are based on more primitive perceptual functions. In other words, in general they are in most cases almost completely correct, but in a few awkward cases they are completely wrong!” (62).
Lorenz, K., & Cranach, A. (1997). The natural science of the human species: An introduction to comparative behavioral research: The “Russian Manuscript” (1944-1948). Cambridge, Mass: MIT Press.
I met some German academics who were Lorenz’s intellectual descendants on a trip to Moscow in 2001. They did remarkable research videoing hundreds of hours of human behavior looking for patterns. For example, here’s an article I wrote on Frank Salter’s study of nightclub bouncers:
http://www.isteve.com/2001_Nightclub_Bouncing.htm
My one suggestion to them was that they could hurry up the hypothesis generating part of the process by asking talented actors, comedians, or impressionists to act out for them different kinds of people in different kinds of situations. Those people are masters of gestalts.