In England, homicides per capita fell roughly by a factor of thirty from the fourteenth-century to the late twentieth century. This progress of civilization wasn’t associated with a secular reduction in gender inequality in life expectancy. Elite women’s life expectancy in medieval England was perhaps nine years less than elite men’s. Women achieved near equality with men in life expectancy in the seventeenth and eighteenth centuries. But women’s lifespan shortfall subsequently grew to about five years in late-twentieth-century England. These facts of gender difference in life expectancy are largely unknown. Reducing gender inequality that disfavors women has never been of public concern. Whether that anti-women bias continues may determine the future of civilization.
Violence against women in late medieval England made women’s life expectancy much less than men’s. The best available data are for the legitimate offspring of British queens, kings, duchesses, and dukes. For such persons born from 1330 to 1479, women’s and men’s expected additional years of life at age twenty were 21.7 years and 31.1 years, respectively. Women at age twenty thus expected to have 9.4 less additional years of life than men had. The share of violent deaths to all deaths for women ages 15 and older was 46%. If women dying from violence are excluded from the life-expectancy calculation, women and men at age twenty had nearly the same expected additional years of life. Violence against women in medieval England explains why women expected to have much shorter lives than men did.
Women probably had much shorter life expectancy than men did across late medieval Europe. The sparse available evidence indicates that the female/male ratio of homicide victims was 13, 7, and 3, in the thirteenth-to-sixteenth centuries, the seventeenth century, and the eighteenth century, respectively. The leading scholar of long-term historical trends in homicide observed:
Generally, the shift toward lower homicide rates appears to have been primarily — but not exclusively — a drop in female-to-female violent encounters. 
Violence has always been vastly disproportionately directed against women. Homicidal violence was high enough in medieval Europe to be a considerable factor in life expectancy. These facts imply that medieval European women had a considerable life-expectancy shortfall relative to men.
During the European Age of Enlightenment of the seventeenth and eighteenth century, men and women apparently had nearly equal lifespans. Life expectancy calculated from English parish registers indicates that females had roughly a half-year advantage in life expectancy on average across the seventeenth and eighteenth centuries. The experience of the Age of Enlightenment makes clear that there is nothing natural or inevitable about women suffering relatively short life expectancy. Use of reason in pursuing social reform can promote gender equality in the most fundamental dimension: gender equality in life expectancy.
The growth of women’s life expectancy shortfall from the early nineteenth century to the late twentieth century probably reflects women’s historically disproportionate burden of financially supporting families. In England, excluding decades of world wars, women’s life expectancy shortfall peaked at 6.2 years in the 1970s. In the U.S., women’s life expectancy shortfall peaked at 7.7 years about 1970. Subsequent movement toward gender equality in life expectancy is plausibly associated with men’s greater participation in the paid labor force, particularly in highly stressful jobs previously associated with women.
Much work remains to be done to achieve gender equality in life expectancy. Women continue to face enormous gender discrimination in family court decisions. Women continue to be deprived of equal opportunities with men to withdraw temporarily or permanently from the paid workforce and be financially supported by their partners or spouses. The effects of gender inequality can be measured ultimately in life. In England and the U.S., women currently fall about four years short in life expectancy relative to men. The long shadow of medieval chivalry remains in devaluing women’s lives to this day.
 Eisner (2003) p. 96, Fig. 3:
Cf. Pinker (2011) p. 61, Fig. 3-2, “Source: Graph from Eisner, 2003.”
 Hollingsworth (1957) pp. 10, 8. Life expectancy at birth was 24.0 years for females and 32.9 years for males. The full time-series data are available in the life-expectancy gender trend worksheet.
In documenting medieval English mortality, Clark (2007), Table 6.2 p. 122, has an imprecise population description (“English aristocrats”) and an incorrect source citation. Table 6.2’s data on life expectancy at birth are for British queens, kings, duchesses, and dukes. Its source is Hollingsworth (1957) p. 8. The data on “fraction of deaths from violence” appears to be an estimate for female deaths from violence relative to all female deaths, rather than the directly reported figures for the female violent death share for deaths after age 15. Clark’s estimates appear to be made by using the survivors per 100 females born at age 5 and 20 (64 and 54, respectively), id. p. 11, to estimate 57 survivors at age 15. Assuming all violent deaths occurred after age 15 gives Clark’s Table 6.2 violent death share estimates. Pinker (2011) p. 81, Fig. 3-7, is a line graph of Clark’s death share estimates, described as for “English female aristocrats.”
[3 Eisner (2003) p. 118, Table 5.
 Id. p. 119.
 Deuteronomy 20:12-15 describes a general commandment to massacre all the women, but take the men and children as spoils. A mass grave at Schöneck-Kilianstädten (Germany) from about 7000 years ago clearly indicates a massacre of at least nine women ages 20 to 40, no men of those ages, and eleven children ages seven or under. That demographic distribution strongly suggests that men ages 20 to 40 were present, but abducted rather than killed. Meyer et al. (2015). Demographic data from human groups at Sredny Stog and Novodanylovka about 7000 years ago indicate that life expectancy at birth was 7.8 years longer for males than for females. Estimated life expectancy at birth was for males, 43.6 years; for females, 35.8 years.
 Prospects of survival apparently favored aristocratic men relative to aristocratic women in tenth and eleventh century Saxony. Violence against women is a plausible explanation. Leyser (1975) pp. 56-7. Leading medieval biologist Albertus Magnus in her Cologne lectures in 1258 declared that, in exception to an Aristotelian generality, men then outlived women. See Albertus Magnus, Quaestiones super libris de animalibus, Bk. 15, quaestio 8. Herlihy (1975), pp. 11-2, found Albertus’s view consistent with other medieval evidence from the thirteenth to fifteenth centuries. For a detailed, somewhat tendentious review of Albertus’s claim, Biller (2000) pp. 286-95. Here are links to Albertus’s works available online.
Among first marriages in the elite Nesle family in northern France, 1100 to 1300, husbands outlived wives nearly two to one. Hadju (1980) p. 129. The implications of that statistic for gender differences in life expectancy depends on gender differences in age at first marriage. Among person born in British ducal families from 1330 to 1470, women at first marriage were 5.3 years older than men at first marriage. Hollingsworth (1957) p. 14. In the Nesle family data, the average length of widowhood was 19.5 years and 35% of widows remarried. Those facts suggest that widows’ former wives were dying quite young.
 Wrigley (1997) Figure 6.21, Table 6.27, pp. 307-8. These calculations included only married persons. Childbirth created some additional mortality risk for men. If unmarried persons were included, then perhaps men would have had a slight life expectancy advantage. For the data, see the life expectancy gender trend worksheet.
 Based on national vital statistics for England and Wales. Estimates compiled in the Human Mortality Database. See the life expectancy gender trend worksheet for details.
 Chivalry can take subtle forms. Consider the UK Longevity Science Advisory Panel’s conclusions on gender inequality in life expectancy:
The gender gap in human lifespan is profoundly affected by societal and behavioural factors and movement towards greater parity in lifestyle between women and men is a major factor in the recent reduction in gender gap in life expectancy. Nevertheless there is such a significant range of genetic, endocrine, cell and molecular biology differences between women and men with impacts on longevity that we are led to the conclusion that a gender difference in longevity will persist. At age 65 this is probably of the order of 1-2 years.
Finally we believe that raw data exists which could be analysed to eliminate social and environmental factors and provide a more accurate estimate of the underlying gender gap in longevity. We plan to explore this possibility in the near future.
Pattison et al. (2012) p. 44. Humans have never and can not exist apart from “social and environmental factors.” The “underlying gender gap in longevity” is a meaningless concept. Achieving gender equality in the fundamental human capability of being alive is clearly feasible. The remaining important question is whether gender equality is truly a constitutional public value.
[image] Dames killing other dames while men watch and applaud. Illustration from Codex Manesse, Zurich, created between 1305 and 1315. UB Heidelberg, Cod. Pal. germ. 848, fol. 17r. Thanks to University of Hiedelberg and Wikimedia Commons.
Biller, Peter. 2000. The measure of multitude: population in medieval thought. Oxford: Oxford University Press.
Clark, Gregory. 2007. A farewell to alms: a brief economic history of the world. Princeton: Princeton University Press.
Eisner, Manuel. 2003. “Long-Term Historical Trends in Violent Crime.” Crime and Justice. 30: 83-142.
Hajdu, Robert. 1980. “The Position of Noblemen in the Pays Des Coutumes, 1100-1300.” Journal of Family History. 5 (2): 122-144.
Herlihy, David. 1975. “Life Expectancies for Men in Medieval Society.” Pp. 1-22 in Rosmarie Thee Morewedge, ed. 1975. The role of men in the Middle Ages: papers of the sixth annual conference of the Center for Medieval and Early Renaissance Studies, State University of New York at Binghamton 6-7 May 1972. Albany, N.Y.: State University of New York Press.
Hollingsworth, T. H. 1957. “A Demographic Study of the British Ducal Families.” Population Studies. 11 (1): 4.
Leyser, Karl. 1979. Rule and conflict in an early medieval society: Ottonian Saxony. Bloomington: Indiana University Press.
Meyer, Christian, Christian Lohr, Detlef Gronenborn, and Kurt W. Alt. 2015. “The massacre mass grave of Schöneck-Kilianstädten reveals new insights into collective violence in Early Neolithic Central Europe.” PNAS (Proceedings of the National Academy of Sciences, US). Published online before print August 17, 2015, doi: 10.1073/pnas.1504365112 The online supplement includes the demographic data on the bodies in the grave.
Pattison, John, Klim McPHerson, Colin Blakemore, Steven Haberman. 2012. Life expectancy: Past and future variations by gender in England & Wales. LSAP paper 2. Longevity Science Advisory Panel.
Pinker, Steven. 2011. The better angels of our nature: why violence has declined. New York: Viking.
Wrigley, E.A. 1997. English population history from family reconstitution, 1580-1837. Cambridge: Cambridge University Press.