_There are marked resemblances between the traits of the Maori of New Zealand and of the Indians of the northwest coast of North America. Among these are sailing ships and houses. (After figures on a map by Covarrubias, 1940.)_]
Only two anthropologists of any standing have favored diffusion. The first of these, Earnest A. Hooton, rejected “the supposition that these various Asiatic invaders brought with them to the New World nothing but a repressed desire to indulge in independent invention, that they came with culturally empty hands, but brains stuffed full of patents to be filed only after arrival.... I have no use at all for the anthropological isolationists who are determined to maintain the incredible dogma that there was no diffusion of inventions and ideas from the Old World to the New, but only of naked human animals.”[8]
The chief modern American proponent of diffusion is Harold S. Gladwin. What is his case? How does he come to his conclusions? He begins, of course, by noting a large number of random resemblances. Some are in simple objects. Some are in complex ones. As he seeks a scientific basis for his argument, he concentrates on the complex things. Complexity seems to rule out coincidence. If a tool has only one or two parts—like a curved throwing stick or a hafted knife—it is not difficult to conceive of two different men inventing it on opposite sides of the world. A bow and arrow with three essential parts presents a little more of a problem, but not too much, for the three parts are dependent on one another. If the bow has a back reinforced by sinew, if the arrow has feathers and a foreshaft, and the foreshaft has a flint arrowhead—making seven elements in all—then one begins to wonder at the mathematical chances of two men exactly duplicating the whole arrangement. Then, consider the vertical loom with nine separate elements, and eleven if it sports a shuttle and a reed fork.
From citing such coincidences, Gladwin turns to the second step of the diffusionist’s argument. This has to do not alone with one complex object, but with unrelated things grouped around it—let us say a vertical loom and bark cloth, painted tripod pottery, and metal casting by the lost wax method. Now if all these disconnected objects can be found in another locality and in use by another people, the suggestion of diffusion becomes far stronger than even in the case of a single complex machine. As Gladwin puts it: