So Malinowski reshuffled the two packs of cards traditionally displayed in Eastern Europe, and took what suited him from each pack. He did ethnography, as it was done in the Carpathians, on the Danube, in the Balkans or the Caucasus, but he did it in the Trobriands, and justified it, not by love of homeland, but by science and a radical empiricism. This enabled him to revolutionise anthropology. He transformed it from a time-machine into a history-exterminator. In a sense he created a new subject, and he certainly created a new profession, with its very severe rules – fieldwork as the sine qua non initiation rite – and with its clearly demarcated membership. At the very same time, he had solved his personal-political problem. The key device, the re-assemblage of romantic and positivist elements in the package, he then christened ‘functionalism’. It helped make sense of his subsequent life, his career, and his professional-political orientation. One can only admire the elegance with which one set of ideas could be used for so many purposes, and endowed with so much coherence.

The discipline initiated by Malinowski continued to exhibit the interesting mix of elements which he had bestowed on it. It was, at least until very recently, a firmly empirical science: its practitioners were expected to gather their data strenuously and carefully, and the level of ethnographic accuracy in this tradition was very high. At the same time, its members were expected to have a sense of the social interconnectedness of things; and, although it was not compulsory to come out with a functionalist conclusion demonstrating ‘seamless unity’, nevertheless, they were expected to look carefully into institutional interconnections.