1 Antonovsky, Aaron, ‘Social class, life expectancy and overall mortality’, Milbank Memorial Fund Quarterly 45 (2), Part 1 (1967), 31–73.CrossRefGoogle ScholarPubMed

2 Woods, Robert and Woodward, John eds., Urban disease and mortality in nineteenth-century England (London, 1984), p. 21, sets the scene from an environmental health perspective.Google Scholar

3 See, for example, Fox, A. J., Goldblatt, P. O. and Jones, D. R., ‘Social class mortality differentials: artefact, selection or life circumstances?’, Journal of Epidemiology and Community Health 39 (1985), 1–8CrossRefGoogle ScholarPubMed; Wilkinson, R. G. ed., Class and health. Research and longitudinal data (London, 1986)Google Scholar; Hart, Nicky, ‘Inequalities in health: the individual versus the environment’, Journal of the Royal Statistical Society 149 (3) (1986), 228–46CrossRefGoogle Scholar; and Goldblatt, Peter, ‘Mortality by social class, 1971–85’, Population Trends 56 (1989), 6–15.Google Scholar

4 Pamuk, Elsie R., ‘Social class inequality in mortality from 1921 to 1972 in England and Wales’, Population Studies 39 (1985), 17–31, quoted from p. 27.CrossRefGoogle ScholarPubMed

5 The questions, ‘What is social class?’ and ‘How may social classes be defined?’ are central to the whole debate on long- or short-term variations in mortality inequalities. We shall return to them later, but for the time being let us assume, along with Antonovsky, that distinct and meaningful ‘social classes’ may be identified empirically in every age. We return to this problem in Section V below and in our presentation of Figure 9 and Table 2.

6 Collins, Selwyn D., Economic status and health. A review and study of the relevant morbidity and mortality data, Public Health Bulletin No. 165 (Washington, 1927)Google Scholar; Titmuss, Richard M., Birth, poverty and wealth. A study of infant mortality (London, 1943)Google Scholar; Morris, J. N., Uses of epidemiology (Edinburgh, 1964)Google Scholar; Glass, D. V. and Eversley, D. E. C. eds., Population in history. Essays in historical demography (London, 1965).Google Scholar

7 The most important of these nineteenth-century sources are George Farren, Observations on the mortalities among members of the British peerage (London, 1832)Google Scholar; Edmonds, T. R., ‘Duration of life in the English peerage’, Lancet (10 02 1838)CrossRefGoogle Scholar and ‘Lineage of English peers’, Lancet (9 03 1839)Google Scholar; Farr, William, ‘Comment on duration of life among British peers’, in Humphreys, N. A. ed., Vital statistics. A memorial volume of selections from the reports and writings of William Farr (London, 1885), 393–4Google Scholar; Bailey, Arthur and Day, Archibald, ‘On the rate of mortality prevailing amongst the families of the peerage in the 19th century’, Assurance Magazine and Journal of the Institute of Actuaries 9 (6) (07 1861), 305–26CrossRefGoogle Scholar (life expectancy at age 20 in years (e ₂₀) males given as 41.46 and IMR males as 78 (infant deaths per 1,000 live births): both compare well with Hollingsworth's figures); Ansell, Charles, On the rate of mortality at early periods of life, the age at marriage, the number of children to a marriage, and the length of a generation and other statistics of families in the upper and professional classes (London, 1874)Google Scholar. Also Sigismund Peller, ‘Births and deaths among Europe's ruling families since 1500’, and Hollingsworth, T. H., ‘A demographic study of the British ducal families‘, both in Glass, and Eversley, eds., Population in history 87–100 and 354–78Google Scholar.

8 Villermé, Louis René, Tableau de l'état physique et moral des ouvriers, Volume 2 (Paris, 1840)Google Scholar; Gavin, H., Sanitary ramblings (London, 1848)Google Scholar; Clay, J., Report of the Commission for Enquiry into the State of Large Towns and Populous Districts, Appendix (London, 1844)Google Scholar; Humphreys, Noel A., ‘Class mortality statistics’, Journal of the Royal Statistical Society 50 (2) (1887), 255–92CrossRefGoogle Scholar; Rowntree, Seebholm B., Poverty and progress: a second social survey of York (London, 1941).Google Scholar

9 Milne, Joshua, Treatise on the valuation of annuities and assurances of life, etc. (London, 1815)Google Scholar; Farr, William, Vital statistics (London, 1885)Google Scholar; Stevenson, T. H. C., ‘The social distribution of mortality from different causes in England and Wales, 1910–1912’, Biometrika 15 (1923), 384–8Google Scholar; Henry, Louis, ‘The population of Francein the eighteenth century’, in Glass, and Eversley, eds., Population in history, 434–56.Google Scholar

10 Titmuss, , Birth, poverty and wealth and Poverty and population: a factual study of contemporary social waste (London, 1938Google Scholar). See Oakley, Ann, ‘Eugenics, social medicine and the career of Richard Titmuss in Britain 1935–50’, British Journal of Sociology 42 (1991), 165–94CrossRefGoogle Scholar, for a critical interpretation of Titmuss's work, his role in the eugenics movement and his pioneering of the importance of social class divisions in fertility and mortality.

11 We are especially grateful to Jim Oeppen for bringing some of these points to our attention.

12 The sources used for Figures 1 and 2 are as those for Table 1 with the addition of a series for the Benedictine monks of Priory, Christ Church, Canterbury, , from Hatcher, John, ‘Mortality in the fifteenth century: some new evidence’, Economic History Review, 2nd series 39 (1986), Table 2, p. 28. See also note 19, below.Google Scholar

13 Wrigley, E. A. and Schofield, R. S., The population history of England, 1541–1871: a reconstruction (London, 1981; Cambridge, 1989)Google Scholar. See also Woods, Robert, On the historical relationship between infant and adult mortality’, Population Studies 47 (1993), 195–220.CrossRefGoogle ScholarPubMed

14 Oeppen, Jim, ‘Back projection and inverse projection: members of a wider class of constrained projection models’, Population Studies 47 (1993), 245–67.CrossRefGoogle Scholar

15 This factor was derived from the mean of the ratio of e ₂₀ for men to e ₂₀ for the total population found in the first six official English life tables which cover the period from 1841 to 1900.

16 In Figure 3 the series for France and the French bourgeoisie come from Houdaille, Jacques, ‘La bourgeoisie ancienne selon le dictionnaire généalogique d'André Delavenne’, Population 43 (1988), Table 11, p. 323CrossRefGoogle Scholar, and for the Benedictine monks of St Maur from Hervé le Bras and Dominique Dinet, ‘Mortalité des laïcs et mortalité des religieux: les bénédictins de St-Maur aux XVIIe et XVIIIe siècles‘, Population 35 (1980), Table 1, p. 354Google Scholar. Birth cohorts have been converted to approximate period measures and e ₂₀ has been estimated from reported e ₂₅ as in note 13, above.

17 Perrenoud, Alfred, ‘L'Inégalité sociale devant la mort à Genève au XVIIe siècle’, Population 30 (numéro spécial) (1975), Table 11, p. 236.CrossRefGoogle Scholar

18 Blum, Alain, Houdaille, Jacques and Lamouche, Marc, ‘Mortality differentials in France during the late 18th and early 19th centuries’, Population (English selection) 2 (1990), Table 4, p. 175Google ScholarPubMed. Other French data are summarized by Vedrenne-Villeneuve, Edmonde, ‘L'Inégalité sociale devant la mort dans la première moitié du XIXe siècle’, Population 16 (1961), 664–98CrossRefGoogle Scholar, and Houdaille, Jacques, ‘La mortalité de la noblesse de robe à Paris aux XVIIe et XVIIIe siècles’, Population 25 (1970), 637–41Google Scholar, and ‘Mortalité dans divers groupes de notables du XVIIe au XIXe siècles’, Population 35 (1980), 966–8.CrossRefGoogle Scholar

19 See Table 1 and note 12, above. The series for Halesowen peasants comes from Razi, Zvi, Life, marriage and death in a medieval parish: economy, society and demography in Halesowen, 1270–1400 (Cambridge, 1980), 130Google Scholar, and the series for medieval England from Russell, Josiah Cox, British medieval population (Albuquerque, 1948), Table 8.14, p. 190Google Scholar, converted from birth cohorts. Rosenthal, J. T., ‘Medieval longevity: the secular peerage, 1350–1500’ (Population Studies 27 (1973), 287–93Google ScholarPubMed) gives an e ₂₀ for males of approximately 30 for the birth cohorts from 1325 to 1451, but see also Hollingsworth, T. H., ‘A note on the medieval longevity of the secular peerage’, Population Studies 29 (1975), 155–9Google ScholarPubMed. Loschky, David and Childers, Ben D. (‘Early English mortality’, Journal of Interdisciplinary History 24 (1993), 85–97CrossRefGoogle Scholar) have recently attempted to put some of these series together. Figure 4 also shows a series for the Benedictine monks of Westminster Abbey from Harvey, Barbara, Living and dying in England, 1100–1540: the monastic experience (Oxford, 1993), Figure IV.3, p. 128Google Scholar. Both Hatcher's series for the Benedictine monks of Christ Church Priory, Canterbury, and Harvey's series are based on data for the same 25-year overlapping entry cohorts; see especially Oeppen, Jim, ‘Appendix IV. Estimating the life-expectancy of the monks of Westminster’, 236–8 in Harvey, Living and dying, together with 127–9.Google Scholar

20 See Antonovsky, Aaron and Bernstein, Judith, 'Social class and infant mortality’, Social Science and Medicine 11 (1977), 453–70CrossRefGoogle ScholarPubMed, and Pamuk, Elsie R., 'Social-class inequality in infant mortality in England and Wales from 1921 to 1980’, European Journal of Population 4 (1988), 1–21.CrossRefGoogle Scholar

21 Hollingsworth, T. H., ‘Mortality in the British peerage families’, Population 32 (numéro spécial) (1977), Table 2, p. 327. The infant mortality rate among baby girls was 104 in 1550–1574 and 135 in 1575–1599.Google Scholar

22 See Woods (1993) cited in note 13.

23 Woods, Robert, Williams, Naomi and Galley, Chris, ‘Infant mortality in England, 1550–1950: problems in the identification of long-term trends, geographical and social variations’, in Carlo, Corsini and Viazzo, Pier Paolo eds., Historical perspectives on infant mortality in Europe (Florence, 1993).Google Scholar

24 Kunitz, Stephen J., ‘Making a long story short: a note on men's height and mortality in England from the first through the nineteenth centuries’, Medical History 31 (1987), 269–80 (esp. pp. 272–3 and Figure 2, p. 274).CrossRefGoogle ScholarPubMed

25 Kunitz, Stephen J., ‘Speculations on the European mortality decline‘, Economic History Review, 2nd. series 36 (1983), 349–64, quoted from pp. 354–5.CrossRefGoogle ScholarPubMed

26 Fildes, Valerie, ‘Neonatal feeding practices and infant mortality during the eighteenth century’, Journal of Biosocial Science 12 (1980), 313–24CrossRefGoogle Scholar; Breasts, bottles and babies: a history of infant feeding (Edinburgh, 1986)Google Scholar; ‘The English wet nurse and her role in infant care, 1538–1800’, Medical History 32 (1988), 142–73CrossRefGoogle Scholar; Wet nursing: a history from antiquity to the present day (Oxford, 1988)Google Scholar, especially Chapter 6 on the seventeenth century (‘In this period, women with any status in society rarely breastfed their own children’, p. 83) and Chapter 8 on the eighteenth century (‘Although maternal breastfeeding had been urged on mothers for hundreds of years, especially in Protestant countries, this campaign was not effective until the mid-eighteenth century. In England it only became fashionable and acceptable from the 1750s, following a period of, perhaps, 80 years during which much experimentation with artificial feeding had taken place’, p. 116).

27 Thomas, David (‘The social origins of marriage partners of the British peerage in the eighteenth and nineteenth centuries’, Population Studies 26 (1972), 99–111) demon strates just how complicated this marrying-out factor was.CrossRefGoogle ScholarPubMed

28 Eccles, Audrey, Obstetrics and gynaecology in Tudor and Stuart England (London, 1982), p. 120Google Scholar. See also Loudon, Irvine, Death in childbirth: an international study of maternal care and maternal mortality, 1800–1950 (Oxford, 1992), 161.CrossRefGoogle Scholar

29 Senama, Simon, The embarrassment of riches (London, 1987), 531Google Scholar. Schama also remarks: ‘There is still a great deal of uncertainty about infant mortality rates at delivery or in the first few days postpartum. There is no question that mortality rates in the first year varied from 15 to 30 percent, but much of this was certainly due to environmental circumstances - malnutrition, dysenteric and gastric diseases - rather than to obstetric malpractice. At any rate, the assumption that the stillborn rate in preindustrial Europe was higher than for industrial Europe should certainly not be taken for granted’ (p. 652, note 63 to Chapter 7, ‘In the Republic of Children’). Marland, Hilary, Lieburg, M. J. van and Kloosterman, G. J. (“Mother and child were saved”: The memoirs (1693–1740) of the Frisian midwife Catharina Schroder (Amsterdam, 1987)Google Scholar) provide a fuller account and translation. The stillbirth rate (late foetal deaths at or over 28 weeks gestation) was about 40 per thousand of all birth events in England and Wales in the late 1920s when official records first began.

30 Schofield, Roger, ‘Did mothers really die? Three centuries of maternal mortality in “The world we have lost”’, in Lloyd, Bonfield, Smith, Richard M. and Wrightson, Keith eds., The world we have gained (Oxford, 1986), Table 9.5, p. 248.Google Scholar

31 Schofield, ‘Did mothers really die?’; Hollingsworth, ‘Mortality in the British peerage families’, cited in Table 1, Table 2, p. 327. Birth cohorts have been converted to approximate periods as in Figure 1.

32 Loudon argues that in the nineteenth and twentieth centuries ‘there were not in general close links between maternal and infant mortality rates’ (Death in childbirth, 516). But it would be reasonable to expect infant and maternal mortality to be linked in a positive fashion when, in earlier centuries, some decline in neonatal mortality was induced by improved delivery practices. The question is still open.

33 Landers, John (Death and the metropolis: studies in the demographic history of London, 1670–1830 (Cambridge, 1993), 192)CrossRefGoogle Scholar has estimated the IMR for London in the 1730s and 1740s at ‘somewhere above 300’ while by the 1840s it was in the 160s. His study of the London Quakers (Table 4.3, p. 136) suggests the following IMRs: 1650–1699, 260; 1700–1749, 342; 1750–1799, 276. Quite why the general and Quaker IMRs for London declined in this way during the eighteenth century remains a mystery.

34 Watterson, Patricia A., ‘Role of the environment in the decline of infant mortality: an analysis of the 1911 Census of England and Wales’, Journal of Biosocial Science 18 (1986). 457–70CrossRefGoogle ScholarPubMed, and ‘Infant mortality by father's occupation from the 1911 Census of England and Wales’, Demography 25 (1988), 289–306.CrossRefGoogle Scholar

35 A detailed account of the background to and problems associated with using this material is provided in Naomi Williams, ‘The occupational mortality statistics of the General Register Office, 1861–1911’, Liverpool Papers in Human Geography, New Series, No. 2, Department of Geography, University of Liverpool, 1990. Antonovsky (‘s ocial class, life expectancy and overall mortality’ p. 61) regarded The Annual Reports of the Registrar General as ‘the outstanding source of information on the relationship of social class and mortality’, but did not use the data himself.

36 Szreter, S. R. S., ‘The genesis of the Registrar-General's social classification of occupations’, British Journal of Sociology 35 (1984), 522–46.CrossRefGoogle Scholar

37 The series for England and Wales 1895–1897 and 1910 in Figure 7 are from Woods, R. I., Watterson, P. A. and Woodward, J. H., ‘The causes of rapid infant mortality decline in England and Wales, 1861–1921. Part I’, Population Studies 42 (1988), Table 5, p. 364CrossRefGoogle ScholarPubMed; the remainder are from Samuel Preston, H. and Haines, Michael R., Fatal years: child mortality in nineteenth-century America (Princeton, 1991)CrossRefGoogle Scholar, reconverted from index numbers reported in Table 5.4, p. 185, using q(5)s (the probability of dying before age 5) from Table 2.3, p. 58.

38 Compare Haines, Michael R., ‘Inequality and childhood mortality: a comparison of England and Wales, 1911, and the United States, 1900’, Journal of Economic History 45 (1985), 885–912CrossRefGoogle ScholarPubMed (esp. the concluding statement on p. 912), and Ewbank, Douglas C. and Preston, Samuel H., ‘Personal health behaviour and the decline in infant and child mortality: the United States, 1900–1930’, in Caldwell, John C. et al. eds., What do we know about health transition? The cultural, social and behavioural determinants of health, Volume 1 (Canberra, 1990), 116–49Google Scholar (esp. p. 143: ‘child mortality differentials [in the USA] widened by social class during the period 1895–1925 in a manner consistent with the faster adoption of behavioural innovations by upper-class groups’).

39 Williams, Naomi (‘Death in its season: class, environment and the mortality of infants in nineteenth-century Sheffield’, Social History of Medicine 5 (1992), 71–94) attempts a solution to this problem.CrossRefGoogle ScholarPubMed

40 The SMR represents the number of deaths (in thousands) that a particular occupational group would generate if its age-specific mortality profile were applied to a standard population, in this case all males in England and Wales in 1911. Because age exerts such a strong influence on occupational mortality, it is preferable to use all four age categories (25–34, 35–44, 45–54 and 55–64), when given, to calculate the SMRs. For this reason the material published for 1880–1882, which distinguishes only two age groups, has not been used. The material for 1890–1892 has not been incorporated here either for similar reasons of consistency since it only relates to occupied males. We have followed the General Register Office practice in combining the data for 1860–1861 and 1871 to ensure that numbers are sufficiently large and therefore reliable.

41 The 71 occupations represent 46, 51 and 53 per cent of the total male population aged 25–64 years in England and Wales in 1860–1861 & 1871, 1900–1902 and 1910–1912, respectively.

42 Goldthorpe, J. H. and Hope, K., The social grading of occupations: a new approach and scale (Oxford, 1974)Google Scholar; see also Erikson, R. and Goldthorpe, J. H., The constant flux: a study of class mobility in industrial societies (Oxford, 1992), 35.Google Scholar

43 For contemporary accounts see Arlidge, J. T., The hygiene and mortality of occupations (London, 1892)Google Scholar; Thomas Oliver ed., Dangerous trades (London, 1902)Google Scholar; and Legge, Thomas, Industrial maladies (Oxford, 1934)Google Scholar. More recent reports are provided by Lee, W. R., ‘Emergence of occupational medicine in Victorian times’, British Journal of Industrial Medicine 30 (1973), 118–24Google ScholarPubMed, and Weindling, Paul ed., The social history of occupational health (London, 1985).Google Scholar

44 See Crossick, G. ed., The lower middle class in Britain, 1870–1914 (London, 1977).Google Scholar

45 Stone, Lawrence, 'Social mobility in England, 1500–1700’, Past and Present 33 (1966), 16–55, esp. pp. 17–22.CrossRefGoogle Scholar

46 Dyer, Christopher (Standards of living in the later Middle Ages (Cambridge, 1989), chapter 1) develops this point further.CrossRefGoogle Scholar

47 Antonovsky, , 'Social class, life expectancy and overall mortality’, 66.Google Scholar

48 Ibid.

49 Schofield, Roger, Reher, David and Bideau, Alain eds., The decline of mortality in Europe (Oxford, 1991).Google Scholar

50 Preston, and Haines, , Fatal years, 209, cited in note 37, above.Google Scholar

51 Cooper, Jacqui and Bolting, Beverley (‘Analysing fertility and infant mortality by mother's social class as defined by occupation’, Population Trends 70 (1992), 15–21) demonstrate how difficult this problem is.Google Scholar