Book Read Free

The History of White People

Page 23

by Nell Irvin Painter


  Fig. 15.5. “Scandinavia,” in William Z. Ripley, The Races of Europe (1899).

  JEWS POSE another problem. Having long occupied a separate conceptual space within the races of Europe, they must be discussed as a category. At the same time, they are too varied to fit into one of Ripley’s three European races. Recognizing this shortcoming early, Ripley had added a “supplement” on Jews to his Lowell Institute Lectures and articles published in Popular Science Monthly. The Races of Europe allots Jews and Semites a separate chapter.

  Fig. 15.6. “Jewish Types,” in William Z. Ripley, The Races of Europe (1899).

  Anti-Semitic writers had long posited a permanent Jewish race. But Ripley does not like that. Rather, he calls them a “people,” since Jews conform closely to others among whom they live. His photographs of Jewish faces confirm regional variation.11* (See figure 15.6, Ripley’s “Jewish Types.”)

  Then consider their noses. Ripley’s bizarre discussion of the stereotypical Jewish nose betrays his uneasiness. He draws three figures to demonstrate how easily “the Jew” may be turned into a Roman. (See figure 15.7, Ripley’s “Behold the Transformation!”) A tortured paragraph explains this conceptual nose job:

  The truly Jewish nose is not so often truly convex in profile. Nevertheless, it must be confessed that it gives a hooked impression. This seems to be due to a peculiar “tucking up of the wings,” as Dr. Beddoe expresses it…. Jacobs has ingeniously described this “nostrality,” as he calls it, by the accompanying diagrams: Write, he says, a figure 6 with a long tail

  (Fig. 1); now remove the turn of the twist, and much of the Jewishness disappears; and it vanishes entirely when we draw the lower continuation horizontally, as in Fig. 3. Behold the transformation!

  Throwing up his hands, Ripley also explains that Jewish noses do not prevail among urban Jews; besides, many non-Jews have noses that look Jewish.12

  TODAY’S READERS might find intriguing Ripley’s use of a measure of blackness for people of the British Isles. His tool, the “Index of Nigrescence,” had originated with the respected British anthropologist John Beddoe. Over the course of thirty years, Beddoe measured thousands of British heads. Employing impeccable methodology, he analyzed their cephalic indexes, hair, eye, and skin color. Those measurements grounded his classic Races of Britain (1885), whose countless pages of tables convinced Ripley and his generation of scientists that the Irish were dark. Beddoe’s maps and photographs slipped easily into Ripley’s chapter on Britons in The Races of Europe.13 (See figure 15.8, Ripley’s “Relative Brunetness.”)

  Fig. 15.7. “Behold the Transformation!” in William Z. Ripley, The Races of Europe (1899).

  The text reads, “RELATIVE BRUNETNESS BRITISH ISLES, after Beddoe ’85 [Races of Britain] 13,088 observations.” On the right the scale ranks the “INDEX OF NIGRESCENCE,” with light skin and hair at the top and dark skin and hair at the bottom.

  This map also attempts to define linguistic groups: a line between highland and lowland Scotland traces the boundary of the Gaelic speech of Scotland and Ireland (“Gaelic Celtic”); another line separates English from the Gaelic of Wales and the Channel Islands (“Kymric Celtic”); a line through Ireland demarcates the eastern borderline of Irish Gaelic. These unreliable linguistic boundaries often reappeared as racial boundaries between Briton and Celt. Lowland Scots such as Thomas Carlyle and Robert Knox, you may recall, were delighted to be British Saxons rather than highland Scottish Celts.

  RACES OF EUROPE vaulted to success immediately on publication in 1899. The New York Times devoted two full pages to a glowing review, reproducing several of the book’s photographs. The Times reviewer (identified simply as W.L.) raves about Ripley’s “great work” of “elaborate scope and exhaustive treatment.” Best of all, Ripley demolished the “schoolroom fallacy that there is such a thing as a single European or white race.”14 Addressing general readers, the Times underlines this telling point in an era of alarming European immigration. Scholars also loved Races of Europe. The sheer amount of labor it required delighted the American Anthropologist’s reviewer, who gushed over “the best results of the last twenty years in physical anthropology.”15

  Fig. 15.8. “Relative Brunetness British Isles,” in William Z. Ripley, The Races of Europe (1899).

  The British edition of 1900 prompted the Royal Anthropological Society to give Ripley anthropology’s crowning honors, the Huxley Medal and an invitation to deliver the Huxley Lecture of 1902. Occupied with a move from part-time appointments at MIT and Columbia to a permanent post at Harvard, Ripley was unable to accept for 1902. Later he remarked with youthful hubris that Sir Francis Galton, the world’s most famous statistician—and the father of eugenics—had substituted for him.16 Ripley did finally deliver the 1908 Huxley Lecture, the first American to achieve this singular mark of distinction. The New York Times reported his lecture under the pithy headline of “Future Americans Will Be Swarthy.”17 Like Races of Europe, Ripley’s Huxley Lecture delivers racist notions in scholarly tones.* He predicts a “complete submergence” of Anglo-Saxon Americans to follow the “forcible dislocation and abnormal intermixture” of many races and the low Anglo-Saxon birthrate. He also frets that Jews (“both Russian and Polish”) were dying at too slow a rate, “only one-half that of the native-born American,” despite living in abysmal circumstances. While Anglo-Saxons avoided having children, short, dark, round-headed Jews in the tenements multiplied alarmingly. Echoing his mentor Francis Amasa Walker’s phrasing, Ripley warns that immigration was now tapping “the political sinks of Europe,” bringing a “great horde of Slavs, Huns and Jews, and drawing large numbers of Greeks, Armenians and Syrians. No people is too mean or lowly to seek an asylum on our shores.”

  Reiterating favorite themes in European racial science, Ripley warns that the tendency of city people to be darker “can not but profoundly affect the future complexion of the European and American populations.” Ripley’s surmise that these dark-haired men were more sexually potent than blonds repeats speculations of Emerson and others, adding a note of anxiety. Ending ruefully, Ripley posits that all might not be lost. Even if Anglo-Saxon Americans followed American Indians and the buffalo into extinction, surely their mixture with others would mean a continued, worthy life. After all, there exists a “primary physical brotherhood of all branches of the white race.” The white race?

  Dancing about the hot coals of theory in 1908, Ripley says there might be only one white race, not the three of 1899. Even further, perhaps, “all the races of men” belong to the same human brotherhood, and “it is only in their degree of physical and mental evolution that the races of men are different.”18 If so, nevertheless, some races (the white) are more advanced—more “evolved”—than others (the dark).

  This cloudy perversion of Darwinian evolution and Mendelian inheritance thrived in the early twentieth century. Like most of his scholarly peers, Ripley believed that races of people, like Gregor Mendel’s garden peas, inherited gross “unit” traits, such as intelligence, head shape, pigmentation, and height. Because Ripley believed that the original Europeans—those primitive, Stone Age Celts fated for displacement—were dark, he concludes that the “abnormal intermixture” of peoples in the United States would lead to a “reversion to the original stock.” The hybrids might be even darker than the dark parent because of “the greater divergence” of stocks. If Italian men mated with Irish women, they would produce “the more powerful…the reversionary tendency” toward darkness. There could be no intermediate pigmentation.

  Harvard’s prestige played a large part in the longevity of such nonsense as scientific truth. But mostly The Races of Europe spoke to a race-obsessed nation by delivering the right opinions dressed up as science. Never mind that the book could not survive a careful reading, that it bulged with internal contradiction, or that its tables and maps offered a Babel of conflicting taxonomies. William Z. Ripley’s Races of Europe remained definitive for the next quarter century.

  ALTHOUGH Races
of Europe defined Ripley’s reputation over the long run, it represented a detour in his scholarship, for the young Ripley had first been a promising economist rather than an anthropologist. True, Races of Europe brought him excellent job offers from Cornell, Columbia, Yale, and Harvard. But the position he accepted in 1902 took him into Harvard’s Department of Political Economy, where he remained until his retirement in 1933, when, like his mentor Francis Amasa Walker, he served as president of the American Economic Association.19

  In these many years Ripley’s reputation shone brightly. He appeared regularly in the New York Times. In the mid-1920s he warned President Calvin Coolidge, investors, and politicians of the dangers of unsound railroad financing and appeared as a cartoon character measuring railroad financial soundness. (See figure 15.9, Ripley measuring the railroads.) During the Hoover years of the Great Depression, he advocated government regulation along lines that would become President Franklin D. Roosevelt’s New Deal. In fact, Ripley had taught Roosevelt at Harvard, delivering lectures that apparently inspired the reform of the American economy.20

  Through it all the New York Times covered him diligently, printing his warnings and reviewing his books. It reported Ripley’s auto accident, his nervous breakdowns, his retirement, his death, and, a quarter of a century later, the death of his wife.21 All this attention signified a scholar at the top of the heap. But an immigrant of an original turn of mind was rising to challenge his preeminence.

  Fig. 15.9. The New York Times shows Ripley measuring the railroads.

  16

  FRANZ BOAS, DISSENTER

  Born in Minden in Prussian Westphalia to middle-class parents, Franz Boas (1858–1942) had a sound German education. After attending a Protestant gymnasium, he studied at Heidelberg, Germany’s oldest university.* (See figure 16.1, Franz Boas.) Like many German undergraduates, he joined a fraternity (Burschenschaft) known for dueling and drinking. Interestingly, being Jewish did not stand in his way, for during the 1870s his Burschenschaft Allemannia accepted Jewish students; not until later in the century did Allemannia and other fraternities become exclusive anti-Semitic centers of German nationalism. After a semester at Heidelberg, Boas moved on to Bonn, then completed his graduate work at Kiel University, earning a Ph.D. in physics in 1881. Not that Boas escaped the barbs of German bigotry. In both Heidelberg and Kiel he had encountered Germany’s burgeoning anti-Semitism, from the “damned Jew baiters” who provoked “quarrel and fighting.”1 By the 1880s such harassment was becoming endemic, but Boas was able to avenge these insults, at least in part, through duels that left scars on his face, symbols of honorable, upper-class German manhood.

  In 1881 Boas moved on to postgraduate work at Berlin with the pioneering German anthropologists Adolf Bastian and Rudolf Virchow (the latter considered a father of German anthropology), who taught him the physical anthropology of bodily measurement and alerted him to the possible influence of environment on head shape.2 No finer training could be had in Germany, but with anti-Semitism now closing in, his homeland hardly offered Boas a promising career. The German nationalist Berlin movement, led by its intellectual avatar, Professor Heinrich von Treitschke, had made overt hatred of Jews so widespread and respectable that Boas began to ponder emigration.3

  An opportunity arose in 1883 to pursue his study of psychophysics (the relationship between physical sensation and psychological perception) far from Germany, in Anarniturg in the Arctic Cumberland Sound. There, living with the Inuit, he sought to perceive the environment as they did and to think like them. Those two years of fieldwork were useful and enjoyable. Even the hardships and strange food came to hold an appeal. Raw seal liver, he discovered, “didn’t taste badly once [he] overcame a certain resistance.”4

  Fig. 16.1. Franz Boas.

  Late nineteenth-century European anthropologists were typically both provincial and arrogant. They operated from two basic assumptions: the natural superiority of white peoples and the infallibility of elite modern science. Supposedly, scientific methodology endowed European scholars with universal knowledge. Boas might have accepted such dogma as a student, but during his time with the Inuit his independent streak took him toward diametrically opposite conclusions. Trying, for instance, to record and understand the Inuit language and failing, Boas realized that the fault lay not in the Inuit but in his own limitations. He began to see how hallowed European ways had disabled its scholars. “We [civilized people] have no right to look down on them,” he said of the Inuit, an almost unique, even heretical, thought at the time.

  This breakthrough took Boas well toward the cultural relativism that would dominate twentieth-century anthropology. All knowledge, even Western knowledge, was relative and circumscribed: “our ideas and conceptions are true only so far as our civilization [i.e., our culture] goes.”5 To know everything worth knowing was impossible. And to know anything about other people required immersion in their world, to become, in a sense, one of them.

  In the decade following 1885, Boas made several fieldwork trips to the Pacific Northwest and held a series of temporary positions, including stints at the American Museum of Natural History in New York, the World’s Columbian Exposition in Chicago in 1893, and G. Stanley Hall’s Clark University in Worcester, Massachusetts, then in its heady early years. Such wandering afforded refuge from Germany and sufficient time for research, but no financial security. Finally, in 1896, at the relatively advanced age of thirty-eight, he gained an appointment as lecturer in anthropology at Columbia University, joining a faculty that included the young luminary William Z. Ripley. Not that life had suddenly become cushy for Boas. Whereas Ripley had a choice of jobs and enjoyed generous remuneration, Boas began with temporary employment and a paltry salary at Columbia, evidently made possible only by his rich uncle’s underwriting. American anti-Semitism, then on the increase, doubtless played a role in such disparity.6

  Even so, Boas quickly forged an international reputation as a careful yet innovative scholar. His colleagues soon came to respect his fieldwork and publications—they granted him tenure in 1899, when he was forty-one—even as he questioned their conventional notions of racial superiority and civilization. One such innovation appeared in Boas’s 1894 address to the Anthropology Division of the American Association for the Advancement of Science, called “Human Faculty as Determined by Race.” Its premises—that race, culture, and language are separate, independent variables and should not be confused—undergirded Boas’s classic book, The Mind of Primitive Man, first published in 1911 and revised in the late 1930s. But as statements applying to Americans, these points began to enter popular consciousness only in the 1930s, when anthropologists were becoming the experts on race. Boas’s Anthropology and Modern Life (1932) then emerged as a major scientific declaration.

  One theme of the 1894 lecture questioned the evolutionary view of human races that equated whiteness with development and civilization. Downplaying anatomical differences between races, Boas looked to environment and culture rather than to race as shapers of people’s bodies and psyches. Here was the radical germ of cultural relativism, one fast attracting adherents, among them Paul Topinard, successor to Paul Broca and then the leading French anthropologist. Topinard applauded Boas’s thinking, declaring him “the man, the anthropologist [he] wished for in the United States” and with whom “American anthropology enters into a new phase.” Boas, though he had not found a home in Europe, did retain some European chauvinism. He enjoyed outshining American anthropologists, who lacked his Old World education. “Actually,” he conceded, “it is very easy to be one of the first among anthropologists over here.”7

  For all its originality, Boas’s 1894 address contained a number of dated ideas. One of them was the validity of comparing numbers of “great men,” in the fashion of Sir Francis Galton in England and Henry Cabot Lodge’s “Distribution of Ability in the United States.” And Boas remained tentative, closing with: “Although, as I have tried to show, the distribution of faculty among the races of man is
far from being known, we can say this much: the average faculty of the white race is found to the same degree in a large proportion of individuals of all other races, and [even though] it is probable that some of these races may not produce as large a proportion of great men as our own race.” Counting up great men might be forgiven in Boas, even inferring failure in other races and cultures. After all, such was the tenor of impeccable scholarship at the time. Everyone else was doing it, but he was moving on. Boas soon jettisoned notions of comparative intelligence, as well as designations of “higher” and “lower” races.8

 

‹ Prev