¹ For example, see Myrdal, Gunnar, An American Dilemma: The Negro Problem and Modern Democracy, 2 vols. (New York, 1972);Google ScholarGreene, Lorenzo and Woodson, Carter, The Negro Wage Earner (Washington, D.C., 1930);Google ScholarJohnson, Charles, The Negro in American Civilization (New York, 1930);Google Scholar and Spero, Sterling and Harris, Abram, The Black Worker (New York, 1930). More recently researchers have provided some revisions to these beliefs, using census data showing that blacks increasingly obtained skilled jobs over time.Google Scholar See Becker, Gary, The Economics of Discrimination; 2d ed. (Chicago, 1971), pp. 135–52,CrossRefGoogle Scholar and Meeker, Edward and Kau, James, “Racial Discrimination and Occupational Attainment at the Turn of the Century,” Explorations in Economic History, 14 (07 1977), 250–76.CrossRefGoogle Scholar

² There are some exceptions. Spero and Harris in The Black Worker and Myrdal's appendices in an American Dilemma go into greater detail on specific industries. See also Northrup, Herbert and Rowan, Richard, eds., Negro Employment in Southern Industry: A Study of Racial Policies in Five Industries (Philadelphia, 1970).CrossRefGoogle Scholar

³ Fishback, Price, “Employment Conditions of Blacks in the Coal Industry, 1900–1930” (Ph.D. dissertation, University of Washington, 1983), pp. 91–96.Google Scholar

⁴ These impressions come from a series of interviews from the West Virginia Oral History Project at the West Virginia Sound Archives, West Virginia Regional and History Collection, West Virginia University Library. For more detail about the interviews see Fishback, “Employment Conditions,” pp. 238–39.Google Scholar

⁵ Laing, James T., “The Negro Miner in West Virginia” (Ph.D. diss., Ohio State University, 1933), pp. 193–97.Google Scholar

⁶ Myrdal, An American Dilemma, pp. 1113–14;Google ScholarBarnum, Donald, “The Negro in the Bituminous Coal Industry” in Rowan and Northrup, eds., Negro Employment, p. 29;Google ScholarNorthrup, Herbert, Organized Labor and the Negro (New York, 1944), pp. 161, 169.Google Scholar

⁷ Myrdal, An American Dilemma, pp. 259–61.Google Scholar

⁸ Becker, Economics of Discrimination, pp. 135–52.Google Scholar

⁹ Dewey, Donald, “Negro Employment in Southern Industry,” Journal of Political Economy, 60 (08 1952), 283.CrossRefGoogle Scholar

¹⁰ Contemporary employment researchers considered hand loading and most day jobs like timbermen and trackman to be semiskilled. Edwards, Alba, “Social Economic Groups of the United States,” Journal of the American Statistical Association (06 1917), 647.Google Scholar For a more extended discussion of the various jobs and skills in coal mining, see Fishback, “Employment Conditions,” pp. 241–53;Google ScholarDix, Keith, Work Relations in the Coal Industry: The Hand-Loading Era, 1880–1930 (Morgantown, west Virginia, 1978);Google ScholarGoodrich, Carter, The Miner's Freedom (Boston, 1925; reprinted New York, 1977);Google Scholar and Hunt, Edward, Tryon, F. G., and Willits, Joseph, eds., What the Coal Commission Found (Baltimore, 1925)Google Scholar

¹¹ Despite attempts by the United Mine Workers (UMW) to organize the region, paid-up membership in the UMW accounted for less than 10 percent of the coal miners in West Virginia until World War 1. The figure then jumped to 30 percent, but fell to 22 percent by 1923 and still lower thereafter (U.S. Coal Commission, Report, p. 1052). The unions were most successful in northern West Virginia and also Kanawha County in southern West Virginia.

¹² Fishback, Price, “Discrimination on Nonwage Margins: Safety in the West Virginia Coal Industry, 1906–1925,” University of Georgia College of Business Administration Working Paper, 83–146, p. 27. These figures give only a rough idea of the relative experience of these ethnic groups, underestimating the actual experience levels. The paper listed above uses the entire experience distribution to discuss the interaction of safety discrimination and experience.Google Scholar

¹³ Dewey also suggested the concept of distaste for proximity in his second law, which states that whites generally did not work alongside blacks except in the most menial jobs (Dewey, “Negro Employment,” p. 283).Google Scholar

¹⁴ In his theoretical work Becker assumes one job. In his empirical work he indicates that the concentration of blacks in less skilled jobs is consistent with his discussion of wage discrimination (Becker, Economics of Discrimination, pp. 135–52).Google Scholar

¹⁵ The contingency tables can also be described in the context of a multinominal distribution (Hogg, Robert and Tanis, Elliot, Probability and Statistical Inference [New York, 1977], pp. 352–57).Google Scholar

¹⁶ See also McGouldrick, Paul and Tannen, Michael, “Did American Manufacturers Discriminate Against Immigrants before 1914?”, this Journal 37 (09 1977), 723–46;Google ScholarHandlin, Oscar, The Uprooted (Boston, 1952);Google ScholarHandlin, Oscar, Race and Nationality in American Life (Boston, 1948);Google ScholarNovak, Michael, The Rise of the Unmeltable Ethnic (New York, 1971).Google Scholar

¹⁷ Ransom and Sutch created a segregation measure for blacks and whites in the cotton South based on the binomial distribution. I choose not to use their measure for two reasons. First, there are three or four relevant ethnic groups involved in the coal industry; therefore, a statistic based on a multinomial distribution is more appropriate. Second, Ransom and Sutch's measure is not sensitive to sample size (Ransom, Roger and Sutch, Richard, One Kind of Freedom [Cambridge, 1977], pp. 220–31).Google Scholar

¹⁸ The statistic measured here is not a perfectly accurate description of the job assignment process. For each job it is assumed that the employer makes the choice between that job and all other jobs first and then makes assignments of workers to the jobs included among all other jobs. Since this assumption is used to describe the assignment process for each job, we miss some of the interdependence of job assignments. A similar caveat is relevant to the comparisons for blacks (southern Europeans) and native whites and northern Europeans that follow in the text. A chisquare that would capture all of the interdependencies is one for a contingency table with all four ethnic groups and all fifteen major jobs, but it provides information only about overall segregation and little on segregation in each job.Google Scholar

¹⁹ The counties he surveyed were McDowell, Mercer, Raleigh, Fayette, Logan, and Kanawha (Laing, “The Negro Miner,” p. 12).Google Scholar

²⁰ Including counties where few blacks were employed adds noise to tests for differences in the job hierarchy, especially since the job hierarchy was similar in all parts of the state. Prior to 1910 blacks were concentrated in southern West Virginia, but black employment began to spread throughout the state. This is reflected in the following list of counties and the years in which the counties are included: McDowell in 1906–1925; Fayette in 1906–1925; Raleigh in 1906–1925; Mercer in 1906–1925; Kanawha in 1906–1925; Harrison in 1906–1925; Mingo in 1906–1925; Logan in 1906–1925; Marion in 1906–1925; Putnam in 1906–1909, 1912–1916; Preston in 1906; Wyoming in 1916–1925; Ohio in 1920, 1922–1925; Brooke in 1920–1925; Barbour in 1916, 1918–1925; Greenbrier in 1922, 1924, 1925; Monongalia in 1920–1925; and Clay 1922–1924. I made exceptions to the 100- black-minimum rule for Harrison in 1910, 1911, 1913, and 1914 and Marion in 1910. Since these counties exceeded the minimum in all other years listed, I felt that they should be treated as part of the labor market where blacks were employed throughout. This selection of counties generally included 97 percent of the black miners, 85 percent of native white miners, and 83 percent of the Southern European miners in the state. Workers killed in disastrous accidents killing four or more people, primarily explosions, also are excluded from the samples. There are few such accidents, but each killed a large group of workers. A sample including such an accident is affected to a large degree by the policies and actions of a single mine. If this mine were not representative of the rest of mines, false conclusions about the general conditions in the state would be drawn.Google Scholar

²¹ The division between Southern Europeans and Northern Europeans is similar to that used by McGouldrick and Tannen in “Did American Manufacturers Discriminate?” Northern Europeans included Canadians, Danish, Dutch, English, Finnish, French, German, Norwegian, Scotch, Swedish, Swiss, and Welsh. Any other nationality is included under the “catchall” group, Southern European. For more details see Fishback, “Employment Conditions,” p. 232.Google Scholar

²² Fishback, “Employment Conditions,” pp. 44–50.Google Scholar

²³ Laing, “Negro Miner,” p. 94.Google Scholar

²⁴ Unlike most trade unions, the UMW accepted black workers as equals in its national constitution, but the union lacked sterght where most blacks were located and there were imstances of attempts at discrimination by union locals. See Fishback, “Employment Conditions,” pp. 56–60 and sources cited there.Google Scholar

²⁵ Very rough calculations of accident rate indices for inside jobs suggest that machine runners (10.0), motormen (12.2), pick miners and loaders (11.5), drivers (12.0), and bratticemen and timbermen (9.8) faced similar risks. The brakeman's job (21.8) was most dangerous, whereas trackmen (6.8), pumpmen (4.8), and inside laborers (2.6) faced the least risk. The problems with bias toward unsafe occupations inside the mines are minimized by examining each job sample individually as in the job-specific chi-square measures rather than examining the entire job structure. Details on these accident indices are available in my Working Paper 84–147, cited above.Google Scholar

²⁶ Fishback, “Discrimination on Nonwage Margins,” A fatality-experience (FE) profile is the distribution of workers killed across experience levels.Google Scholar

²⁷ Laing, “Negro Miner,” pp. 198–201;Google ScholarFishback, , “Employment Conditions,” pp. 295–98.Google Scholar

²⁸ U. S. Coal Commission, Report, Part, 5: Atlas of Statistical Tables, 68th Congress, 2nd Session (Washington, D.C., 1925), pp. 218–30.Google Scholar

²⁹ For an extended discussion of this point see Fishback, “Segregation in the Job Hierachies,” Working Paper 84–147 in the University of Georgia, College of Business Administration Series.Google Scholar

³⁰ These figures were compiled from tables titled “Nationalities of Mine Employees Summarized,” West Virginia Department of Mines, Annual Reports for the years from 1907 to 1925. For information on the percentage for specific years see Fishback, “Employment Conditions,” p. 280.Google Scholar

³¹ Fishback, “Employment Conditions,” p. 315.Google Scholar

³² Laing, “The Negro Miner,” pp. 194, 202–07.Google Scholar

³³ The miner category in the sample probably includes both pick miners and loaders, since loaders were sometimes called machine miners. There is no reason to believe that the loader category contains pick miners; therefore, the loader samples, although smaller, should still be useful in determining the distribution of ethnic groups in loader jobs.Google Scholar

³⁴ Much of the uncertainty about the first period appears to be caused by geographical separation of blacks and machine mining rather than by intrafirm segregation along machine lines. In 1908 the correlation for counties in the sample between the percent black of all inside workers and the number of pick miners as a percentage of all pick miners and loaders was 0.82. Such geographical separation later became less pronounced as the correlation fell to 0.303 in 1913, 0.178 in 1918, and –0.043 in 1923. Intrafirm segregation along machine lines does not appear to have been widespread because the loader job was also nonblack during this period. Much of this geographical separation of machine cutting and blacks in 1906–1910 is attributable to the location of earlier black migration to the West Virginia coal regions prior to the widespread use of cutting machines. Laing (pp. 131–32) noted the existence of migratory pipelines between areas of origin and the coal fields. Their importance is suggested by the positive correlation, 0.588, between the percent black of inside workers in sample counties in 1908 and the percent black of the population in those counties in 1900. See footnote 20 for the counties included in the calculation of correlations in each year. The county data are from the West Virginia Department of Mines Annual Report and the census. Details on sources are available in UGACBA Working Paper 84–147.Google Scholar

³⁵ A summary of black employment patterns in West Virgina coal mining is found in Laing, “Negro Miner,” p. 94. These figures, based on the West Virginia Department of Mines Annual Reports, also show that black employment rose generally until 1927, when it peaked and began declining. The discussion of wages is based on the daily wages reported for daymen in West Virginia mines from 1912 to 1923 in U.S. Coal Commission, Report, Part V, Statistical Atlas, pp. 208–26. In general, daily wages remained relatively constant in nominal and real terms until 1916, when they began rising markedly. They fell sharply in 1922 and then recovered in 1923.Google Scholar

³⁶ Spero and Harris, The Black Worker, pp. 224–27.Google Scholar

³⁷ Laing, “Negro Miner,” pp. 241–47.Google Scholar

³⁸ p. 242.

³⁹ p. 182.

⁴⁰ West Virginia Bureau of Negro Welfare and Statistics, Report, 1921–1922, p. 58; 1927–1928, p. 16. There were several black owners of small mines in the late 1800s.Google ScholarSimmons, Charles, Rankin, John, and Carter, U. G., “Negro Coal Miners in West Virginia, 1875–1925, Midwest Journal 6 (Spring 1954), pp. 6–7.Google Scholar

⁴¹ United States Bureau of Mines, prepared for the Department of Labor, Descriptions of Occupations: Mines and Mining (Washington, D.C., 1918), p. 20.Google Scholar

⁴² West Virginia Mining districts may have been an exception. Educational expenditures per pupil for blacks and whites were similar in the mining towns in the early 1900s, Fishback, “Employment Conditions,” p. 402;Google ScholarCorbin, David, Life, Work, and Rebellion in the Coal Fields: The Southern West Virginia Miners 1880–1922 (Champaign-Urbana, Illinois, 1981), pp. 71–72;Google ScholarLaing, “Negro Miner,” p. 377. This, however, primarily aided children of miners. The relative benefits of equal education in West Virginia were felt with a lag. Better-educated blacks tended to avoid staying in the mines and probably became teachers if they did stay in the area.Google Scholar

⁴³ West Virginia Department of Mines, Annual Report, 1913, p. 11.Google Scholar

⁴⁴ Interview conducted by Keith Dix with U. G. Carter, West Virginia Oral History Project.Google Scholar

⁴⁵ Getting foreman papers was no guarantee that blacks would become foremen. After U. G. Carter established the mining extension course for blacks at West Virginia State College in 1937, he gathered statistics on blacks who passed the exam prior to 1942. Only 23 (or 46 percent) were in positions of competency. “Report to the West Virginia State College Mining Extension Course for 1942–43,” U. G. Carter Collection at the West Virginia Regional and History Collection at the West Virginia University Library.

⁴⁶ Laing, “Negro Miner,” pp. 212–18.Google Scholar

⁴⁷ p. 182. An ex-maintenance man for one company in the 1920s told me that the Raleigh Coal and Coke Company had once set up its No. 7 mine as an all-black mine to reward its harder working blacks with positions of authority.

⁴⁸ Interview with U. G. Carter, West Virginia Oral History Project.Google Scholar