Consider Google Books Search results for “the races of Europe” and the wild-card “* races of Europe”. The dramatic rise in 1911-1915 is arrested in 1916. Until 1928, things look like they had stabilized.
Then it collapses in 1929 and enters into secular decline. Even so, a 1980s cycle is detectable; as is a truncated cycle that begins in 2003; Google censors books from 2008.
A Search for “the races of Europe” and “the races of man” is interesting. It is only during the interwar period that the frequency of “the races of Europe” becomes comparable to the that of “the races of man”. There is a massive cycle in the latter during 1867-1879 that peaks in 1870. The collapse of the 1870s takes place long before the professionalization of anthropology begins at the turn of the century. Did the violence of the 1870s boom-bust cycle interact causally in either direction with the diachronic historical logic underlying the Ngram cycle for “the races of man”? Is this the cooling of relations on the continent quarterbacked by Bismark? Or perhaps the historical resolution of the national question?
For “the races of man” we find three regimes of the discourse after 1880: the late-19th century, the short twentieth century, and the contemporary period after the anti-systemic turn triggered by the Tet Offensive in 1968.
Zooming into the mid-century passage, 1930-1970, we can detect a certain stability, even rigidity, in the fluctuations. The precise timing of the three cycles is of some historical interest. The sharp rise in the first cycle is 1933, and the fall is 1940; in the second, the big escalation is 1946, the fall is 1954; in the third, the escalation is 1962, and the fall is 1969. The discourse of “the races of man” seems locked into some sort of geopolitical cycle during the mid-century passage, 1930-1970.
We have opened up a Pandora’s box. Is the cycle after the anti-systemic turn driven by the rise of Boasian antiracism and the subsequent counter-movement to it? Why did the discourse of ‘the races of Europe’ collapse in 1929? Can we read these graphs micro-locally in time? Is it possible to interrogate the correlation matrices of a few well chosen sound bites to arbitrate historical arguments? What is really going on with the regimes and the cycles? How are they related to escalation cycles? What is of historical interest here is precisely the diachronic pattern of the discourse. What we need is an emplotment of these discursive fluctuations. That is, we need to write a history of high racialism.