Figure 4.5. A list of the surnames of all those in the Inquisitions Post Mortem of 1236–99 was formed from Public Record Office 1904, 1906. From these, a sample of names with clearly discernible modern equivalents was selected. In forming this list, a few surnames (Bruce, Preston, and Sutton) were omitted as they were judged so common that their appearance in the Inquisitions Post Mortem was not informative.
Figure 4.6. The Domesday Book surnames were derived from Keats-Rohan 1999. All those with discernible modern equivalents were used, using Reaney and Wilson 2005 as a guide.
Figure 4.7. Photo Austin Osuide / Wikipedia Commons.
Table 4.1. See sources for figure 4.5; Reaney and Wilson 2005.
Table 4.2. See sources for figure 4.6; U.K., Office of National Statistics 2002.
Chapter 5
Figure 5.1. Surname frequencies: U.K., Office of National Statistics 2002. Oxford and Cambridge surnames, 1980–2012: see sources for figure 4.1.
Figure 5.2. Photo Andy Miles.
Figure 5.3. Photo Dennis Novy.
Figure 5.4. Clark and Cummins 2013, table 5.
Figure 5.5. Clark and Cummins 2013, figure 8.
Figure 5.6. Clark and Cummins 2013, figure 3.
Figure 5.7. Oxford and Cambridge surname frequencies: see sources for figure 4.1. Population frequencies by generations of eighteen-year-olds: estimated from the England and Wales Marriage Register, 1837–2005.
Figure 5.8. The set of surnames used here is all those with five hundred or fewer occurrences in the 1881 census (Schurer and Woollard 2000) that appear at Oxford and Cambridge in the years 1800–1829 (see sources for figure 4.1). Population frequencies by generations of eighteen-year-olds were estimated from the England and Wales Marriage Register, 1837–2005, to establish the overall population trends for these surname groups. A surname count for 2002 from U.K., Office of National Statistics 2002 is used as the surname-frequency benchmark.
Figure 5.9. List of MPs for English and Welsh constituencies from Rayment, n.d. Each election was counted in measuring the relative frequency of surnames, even when the same MP was returned to Parliament. Surname frequencies are estimated as for figure 5.8 but adjusted to an assumed average age for MPs of 50. Surname frequencies for marriages for 1810–37 are estimated from parish records of marriages.
Table 5.1. Clark and Cummins 2013, table 2. All the rare surnames used in this estimate of modern social mobility rates are listed in the appendix of this source. Sample A is the rich, B the poor, and C the prosperous.
Table 5.2. Clark and Cummins 2013, tables 6 and 7.
Table 5.3. Clark and Cummins 2013, table 8.
Chapter 6
Figure 6.2. State employee salaries from Sacramento Bee, n.d.
Figure 6.3. Bureau of Labor Statistics 2010; Statistics Sweden 2011b.
Figure 6.4. Rare-surname samples derived as in chapter 5. Average age of death by surname type from United Kingdom, Civil Registration, Death Index 1866–2005.
Figure 6.5. Rare-surname frequencies by generation from same sources as in figure 5.8. Probate rates for 1858–1966 from England and Wales, Index to Wills and Administrations, 1858–2013.
Figure 6.7. Linkages between the twenty-five thousand rare-surname individuals dying between 1858 and 2012, as described in chapter 5, were established using census records, birth and baptism records, marriage records, probate records, passenger ship lists, and university attendance records, as well as genealogies collected from a variety of sources. See Cummins and Clark 2013.
Figure 6.8. See sources for figure 5.8. The surnames were divided into two groups depending on whether the surname appeared at Oxford or Cambridge in the years 1770–99. The outcomes for the years 1770–1829 are not shown because of error introduced for these years by the way the surnames were selected based on their occurrence.
Table 6.1. Pairwise correlations from the following sources. Cognitive ability and education: Husén and Tuijnman 1991; Scarr and Weinberg 1978; Zagorsky 2007. Cognitive ability and occupational status: Cagney and Lauderdale 2002; Griliches and Mason 1972; Hauser 2002. Cognitive ability and earnings: Griliches and Mason 1972; Zagorsky 2007; Zax and Rees 2002. Cognitive ability and wealth: Zagorsky 2007. Education and occupational status: Hauser and Warren 2008; Pfeffer 2011; Scarr 1981. Education and earnings: Cagney and Lauderdale 2002; Griliches and Mason 1972; Pfeffer 2011. Education and wealth: Cagney and Lauderdale 2002; Pfeffer 2011. Occupational status and earnings: Griliches and Mason 1972; Hauser and Warren 2008 (wages). Occupational status and wealth: Pfeffer 2011. Earnings and wealth: Budria et al. 2002; Hendricks 2007.
Table 6.2. See sources for figure 6.4.
Table 6.3. See sources for figure 6.5.
Chapter 7
Figure 7.2. Clark and Cummins 2014, figure 4.
Figure 7.3. Births for 1880–1999 were obtained from England and Wales, Register of Births, 1837–2005.
Figure 7.4. Photograph by Herbert Rose Barraud / Wikimedia Commons.
Figure 7.5. This height study was reported in Galton 1886. The data used here are from Hanley 2004.
Figure 7.6. Data kindly supplied by Simon Boserup, Wojceich Kopczuk, and Claus Kreiner. The nature and construction of the data are detailed in Boserup, Kopczuk, and Kreiner 2013.
Chapter 8
Figure 8.1. University Grants Commission 2008, 105.
Figure 8.2. Doctor totals for each surname group were derived from the listings of Medical Council of India, n.d., for physicians in West Bengal registered between 1950 and 2011. Attorney surname distributions were based on a list of judges sitting in the High Court and district courts of West Bengal, obtained from the Calcutta High Court website (http://calcuttahighcourt.nic.in/).
Figure 8.3. Surname shares among doctors were derived from a database of all registered doctors in Bengal for 1860–1947 and West Bengal for 1948–2011. For the period 1915–2009, we compiled a list of 57,407 doctors registered in Bengal and West Bengal between 1915 and 2009 from Indian Medical Registry, n.d. These listings include doctors who graduated from medical school as early as the 1880s. For 1860–1909, doctor registrations were calculated from four sources: Government of Bengal, Bengal Medical Department 1903, which includes 1,507 doctors in Bengal licensed in 1903 or earlier; Government of Bihar and Orissa 1930; Burma Medical Council 1930; and a list of doctors registered in Bengal in 1915 who graduated from medical school between 1900 and 1914.
Population shares by period were estimated as below. The imperial censuses give Muslim shares of population in Bengal for 1871–1941. Muslims constituted 48 percent of the population for 1871–91, 53 percent for 1891–1921, and 55 percent for 1921–31 (Clark and Landes 2013, population appendix). For 1951–2001 we take the relevant Muslim population share as being that for the 20–29 age group in the censuses of India. This is larger than the overall Muslim population share because of the faster growth rate of the Muslim population. The Muslim shares were thus 21 percent for 1950–80 and 29 percent for 1980–2010 (Clark and Landes, 2013, population appendix).
We take the Hindu population share to be the rest of the population, omitting the small Christian and Buddhist populations.
To estimate the share over time of the seven Kulin Brahmin surnames, we proceed as follows. We start by analyzing the data from imperial censuses, which show the Brahmin share of the Hindu population for all of India. For the censuses conducted from 1871 to 1931, the population shares were 6.79, 7.31, 7.14, 7.19, 6.71, 6.58, and 6.34 percent (Clark and Landes 2013, population appendix). Thus before 1931 the Brahmin share was declining despite the elite status of Brahmins. This trend is consistent with the finding of Kingsley Davis that in 1931 the Brahmins had a ratio of children 0–6 to women 14–43 that was only 88 percent of other Hindu groups on average. This was mainly a consequence of the social taboo on Brahmin widows’ remarrying (Davis 1946, table 3, 248). Presuming that Brahmins, a group with higher incomes than other Hindus, had better child survival rates would explain the only modestly lower net fertility of Brahm
ins. Brahmins in Bengal represented the same share among Hindus as for all of India in 1921–31. We thus assume this same population trend for Bengali Brahmins relative to other Hindus for the period 1871–1931.
Since Independence there has been no formal count of Brahmins. However, electoral surveys for 2004–07 estimated Brahmins as 5 percent of the entire Indian population, or 6.2 percent of the Hindu population (Center for the Study of Developing Societies 2009). This implies a modest decline in the Hindu share of Brahmins between 1931 and 2004. However, the Kolkata electoral register suggests that Brahmins had much greater life expectancy than the Hindu population as a whole (Chief Electoral Office, West Bengal 2010). Whereas the seven Kulin Brahmin surnames constituted 4.1 percent of the Hindu electorate in the 20–29 age group, they constituted 9.9 percent of the Hindu electorate in the 70–79 age group. If this distribution is representative of national population, it would imply that Brahmins accounted for only 5 percent of the Hindu population age 20–29 in 2004. We assume the same to be true for Brahmins in West Bengal in the period 2000–2009.
Not all Kulin Brahmins had one of the seven surnames we track. But a list of prominent Bengali Brahmins consists almost entirely of people with these surnames, so we take the seven Kulin surnames as comprising 5 percent of the West Bengal population age 20–29 in 2001, acknowledging that this method modestly overestimates their population share.
Other high-status Hindu groups are assumed to follow the same population trends as Brahmins. The three other Hindu surname groups—poor, scheduled caste, and mixed—are assumed to follow the population trend of the remainder of the Hindu population in Bengal.
Figure 8.4. Kolkata Police Recruitment Board 2010.
Figures 8.5 and 8.6. See sources for figures 8.2 and 8.3.
Figure 8.7. The fraction of each surname group admitted through the reservation system was estimated from Bankura Medical College 2009; Kar Medical College 2010, 2011.
Figure 8.8. For West Bengal: see the sources for figures 8.2 and 8.3 (counting doctors registered in 1960 and later). For West Bengali doctors in the United States: American Medical Association 2012.
Table 8.1. Totals derived from the posted admissions list of the All India Institute of Medical Sciences, Delhi, 2012, www.aiims.edu/aiims/examsection/MBBS12_RESULT_MERIT_WISE.pdf.
Table 8.2. On population shares in 2010, see Clark and Landes 2013, population appendix.
Table 8.3. See sources for figure 8.7.
Tables 8.4 and 8.5. Chief Electoral Officer, West Bengal 2010.
Chapter 9
Figure 9.1. Photo Li Zhensheng, Contact Press Images.
Figure 9.2. Wikipedia Commons.
Figure 9.3. The population shares of surnames in China come from a database obtained from the China National Identity Information Center (CNIIC) that gives the population, ethnicity, and educational attainment of the 1,500 most common Chinese surnames and the regional distribution of a selected group of surnames. This information comes from China’s system of household registration (hukou), which covers the entire population.
The share of jinshi with each surname for the period 1820–1905 is from Zhu and Xie 1980. The frequency of surnames among the Republican era elite is from two sources: a list of high-ranking civil and military leaders of the Republican era (Liu 1989) and a list of university faculty for the years 1941–44 (Wu 1971). To construct the list of professors in 2012 (26,429 names), we used the faculty lists of Beihang, Beijing Normal, Fudan, Nanjing, Peking, Shanghai Jiaotong, Tsinghua, University of Science and Technology, Zhejiang, and Wuhan universities. The sample of the rich in 2006 is from the 2006 census of 1.4 million enterprises in China, from which we selected 130,000 chairmen of the boards of companies with assets of one hundred million yuan and above. The list of high government officials in 2010 is from China Government Directory 2010. For more details on these sources, see Hao and Clark 2012.
Table 9.1. The frequency of surnames in the population for each set of counties was estimated from the names of “honored fallen soldiers” of the period 1927–53, recorded in the Chronicle of Zhejiang (1985), and the Chronicle of Jiangsu (1993). The elite in the first period, 1870–1905, were those attaining the juren exam pass, as recorded in the chapters on notable local people in these chronicles. The Republican-era elite in these locations were identified from lists of students at the following universities: Central (Nanjing), 1916–36, 1945–47; Datong, 1923–35, 1940–48; Nanyang, 1905–25; Peking, 1905–48; Tsinghua, 1911–37; Wuhan, 1922–35; Yanjing, 1924–28; and Zhejiang, 1918–47. The elites from Zhejiang in the Communist era were derived from the Chronicle of Zhejiang Jiang 2005. The elites from Jiangsu were identified from the Nanjing university entrants from these counties for the period 1952–2011 (http://dawww.nju.edu.cn/pub/?id=1).
Table 9.2. Hao 2013, chapter 2.
Chapter 10
Figure 10.1. Photo Felice Beato / Wikimedia Commons.
Figures 10.2 and 10.3. Medical researchers, 1989–90: Japanese Medical Researchers Directory 1990. Attorneys, 1987: Zenkoku bengoshi taikan 1987. Corporate managers, 1993: Diamond’s Japan Business Directory 1993. University professors, 2005: Daigaku shokuinroku kankokai 2005. Scholarly authors, 1990–2012: Google Scholar search.
Figure 10.4. Photo Chris Gladis.
Figure 10.5. Sources for Japanese data as for figure 10.2. Japanese surnames among doctors registered in the United States from American Medical Association 2012.
Figure 10.6. Names of scholarly authors from Google Scholar search.
Table 10.1. Lebra 1992, 55.
Table 10.2. Amano 1990, 193.
Table 10.3. Amano 1990, 193; Sonoda 1990, 103.
Table 10.4. Harootunian 1959, 260–61.
Table 10.5. The frequency of surnames was estimated from Public Profiler, n.d. The table assumes a population of Japan of 124 million in 1990. The potential rare samurai surnames are those listed in Takayanagi, Okayama, and Saiki 1964. The kazoku surnames were listed in Kasumi kaikan shoka shiryo chosa iinkai 1982–84.
Table 10.6. Japanese Medical Researchers Directory 1966, 1990.
Chapter 11
Figure 11.1. Depositphotos, Inc.
Figure 11.2. HDI by community from Chile, Ministry of Planning and Cooperation 2006. Average wage by occupation and location from Servicio Electoral Republica de Chile 2004; Chile, Ministerio del Trabajo y Prevision Social 2008.
Figure 11.3. OFF / AFP / Getty Images.
Figures 11.4 and 11.5. See sources for figure 11.2.
Table 11.1. Servicio Electoral Republica de Chile 2004. The electoral rolls listed 6,246,198 voters age 18 and above.
Table 11.2. Occupational wages were assigned using Chile, Ministerio del Trabajo y Prevision Social 2008. Locational wages were calculated as the average occupational wage from this source in each comuna in Chile.
Chapter 12
Figures 12.1 and 12.2. Figures created by a simulation of the status paths of five hundred families over one hundred generations. The average status trajectories for all families observed in any period in the top 0.14 percent of the status distribution are plotted for the ten earlier and later generations.
Figure 12.3. The probate surname data were obtained as described for figure 4.1. The frequency of surnames for 1680–1837 was estimated from parish records of marriages in England and Wales, obtained from the FamilySearch website. Surname frequencies for 1837 and later were estimated from marriages as recorded in England and Wales, Register of Marriages, 1837–2005.
Figures 12.4 and 12.5. Elite rare surnames in each of the periods 1710–39, 1740–69, 1770–88, and 1800–1829 were defined as surnames beginning with the letters A–C that appeared at low frequency in the parish records of marriage in the previous thirty years. The frequency cutoff depended on the numbers of marriages recorded in each of those periods: it was three in 1680–1709, four in 1710–39, five in 1740–69, and six in 1770–99. For 1800–29 and 1830–59, rare surnames were defined as those beginning with the letters A–C o
ccurring at low frequency in England and Wales, Register of Marriages, 1837–2005, for 1837–59 and 1860–89, with a cutoff of ten.
Figure 12.6. The rare surnames of the rich are those discussed in chapter 5 that fell into the rich and prosperous groups. The rare surnames of Oxford and Cambridge students for the years 1800–1829 are those that appeared at the universities in these years and had forty or fewer holders in the 1881 census. The population share of these surnames for the years 1530–1837 was estimated from parish records of marriages, and for the years 1837–2005 from England and Wales, Register of Marriages, 1837–2005. The Oxford and Cambridge share of the surnames in each period was estimated from the university database as described above.
Figure 12.7. The four rare-surname groups are surnames appearing at Oxford and Cambridge in the periods 1800–1829, 1830–59, 1860–89, and 1890–1919 that had forty or fewer holders in 1881. The share of these surnames in each period in the population and at Oxford and Cambridge was calculated as for figure 12.6.
Figure 12.8. The share of Pepyses attending Oxford or Cambridge in each century was calculated as their numbers at the universities relative to the estimated number of Pepyses eligible to attend. For the years 1538–1837, the numbers eligible were estimated from parish marriage records by multiplying the estimated numbers of men attaining age 18 in each century by the marriage share of Pepyses. For the period 1837–2012 the estimate was made in the same way, except that the number of university-eligible eighteen-year-olds included women. For 1400–1537, the share of Pepyses in the population was assumed to be the same as the share in the years 1538–99.
The Son Also Rises Page 33