Vital statistics is the systematic collection, registration, and analysis of data pertaining to births, deaths, marriages, and related demographic events. As a formal discipline it emerged in the nineteenth century from the convergence of civil registration law, actuarial practice, and the growing conviction that population data could guide public policy. Its central argument is that the health and growth of a population are not matters of fate but of measurable, and therefore modifiable, conditions.
Population Theory and the Critique of Malthus
The intellectual foundations of vital statistics as a policy science required, above all, a credible account of how populations grow. William Farr, whose collected writings in Vital Statistics (1885) constitute the discipline’s primary nineteenth-century synthesis, began from the empirical fact that England’s population had increased in regular geometrical progression.(Farr, William (Humphreys, Noel A., ed.), 1885) This was not a theoretical claim but an observation drawn from four successive census returns: every one hundred persons living in 1801 had increased to 132 by 1821 and to 175 by 1841, implying a mean annual growth rate of approximately 1.41 percent, analogous to compound interest.
The precision of that figure allowed Farr to examine the mechanisms underlying it. He found that roughly 79 women in every 100 who reached marriageable age in England actually married, at a mean age of 24.3 years, and 78 in 100 men at 25.5 years.(Farr, William (Humphreys, Noel A., ed.), 1885) The remaining fifth of each sex never married. Only one woman in seven at childbearing ages gave birth in any given year, an observation Farr used to show how thoroughly the reproductive capacity of the population was restrained by prudence rather than misery.(Farr, William (Humphreys, Noel A., ed.), 1885)
These numbers constituted a direct challenge to the Malthusian framework. Farr judged the claim that subsistence increases only arithmetically while population increases geometrically to be “pure empiricism” — asserted without evidence, since nobody had shown that the rate of improvement in subsistence was necessarily arithmetical rather than geometrical.(Farr, William (Humphreys, Noel A., ed.), 1885) More fundamentally, Farr argued that Darwin’s reading of the population struggle modified the premise that misery alone checked growth: reason gives human beings certain prerogatives, and the family, the clan, and the nation all acknowledge the claims of children and the infirm, offering protections that inferior species do not possess.(Farr, William (Humphreys, Noel A., ed.), 1885)
The empirical record supported a calmer view. Both Richard Price’s earlier alarm about depopulation and Malthus’s geometrical-progression alarm had contributed to the development of population theory while ultimately misreading the facts.(Farr, William (Humphreys, Noel A., ed.), 1885) Population growth was in practice voluntarily repressed — not through increased mortality but through the timing of marriage — and the means for adjusting it were “simple, efficient, and quite compatible with our ideas of the benevolence of the divine government of the world.”(Farr, William (Humphreys, Noel A., ed.), 1885)
Farr showed that a stationary population could be achieved without raising mortality at all: either by raising the proportion of women who never marry from one-fifth to one-half, or by raising the mean age at marriage from 24.3 years to approximately 30 years.(Farr, William (Humphreys, Noel A., ed.), 1885) This calculation inverted the Malthusian premise entirely. It was not disease and death that checked population but the voluntary choices embedded in social institutions, and those institutions could be analyzed statistically.
If mean lifetime increases — through better sanitation and lower mortality — without a corresponding fall in the birth-rate, population expands. Farr demonstrated this arithmetically: with births held constant, extending mean lifetime from 30 to 50 years raises population from 3 million to 5 million.(Farr, William (Humphreys, Noel A., ed.), 1885) Reducing mortality is therefore not neutral with respect to population: “the numbers of the population bear a definite relation to the duration of life.”
Census Data and Population Dynamics
The raw material of the discipline was the decennial census. The United Kingdom population (excluding army, navy, and merchant seamen) stood at 21,272,187 in 1821 and approximately 27,724,849 in 1851; in the same interval, 2,685,747 persons emigrated, so that the survivors and descendants of the 1821 races, counted as a whole, numbered some 30,410,595.(Farr, William (Humphreys, Noel A., ed.), 1885) By the eighth decennial census (1871), England and Wales alone numbered 22,712,266, having grown at 13.19 percent in the previous decade.(Farr, William (Humphreys, Noel A., ed.), 1885) At the rates of increase observed between 1801 and 1851, England’s population would double every 51 years.
An important distinction governed the interpretation of these figures. Reducing mortality does not inevitably increase population because the birth-rate may fall to an equivalent extent.(Farr, William (Humphreys, Noel A., ed.), 1885) England’s mean lifetime in the 1870s stood at 41 years; extending it to 49 years (the level achieved in the healthiest districts) would reduce the annual death-rate from 2.242 to 2.041 per 1,000 living, but whether population grew would depend on whether births also changed.
England maintained what Farr described as a demographic reserve: at the time of the 1871 census, half the women of childbearing ages were unmarried, births were kept to half their possible number, and there remained 1,246,743 maidens aged 15 to 21 still unmarried.(Farr, William (Humphreys, Noel A., ed.), 1885) This reserve meant the birth-rate could expand rapidly to meet any demand created by war losses, epidemic mortality, or colonial emigration. The nation was, in Farr’s phrase, fertile in men, holding an ample reserve against whatever fate might require.
Comparing England and France illuminated the birth-rate’s role independently of the death-rate. France’s death-rate of 2.36 per 1,000 differed little from England’s 2.242, but France’s birth-rate was only 2.63 against England’s 3.51.(Farr, William (Humphreys, Noel A., ed.), 1885) France had no colonial demand for population; its restraint expressed itself through a low birth-rate rather than a high death-rate. Population statistics thus required the analyst to disaggregate birth-rate, death-rate, and migration — the three proximate levers of demographic change.
In the 39.5 years from 1837 to 1876, more than 8 million people emigrated from the UK, coinciding with a prodigious increase of domestic capital and a rise in wages. Far from being a net loss, the emigrants created new markets, supplied wheat, cotton, wool, and gold worth hundreds of millions, and extended the global reach of the Anglo-Saxon race.(Farr, William (Humphreys, Noel A., ed.), 1885) The same logic applied to new settlements. Population growth depended on the excess of births over deaths, which was in turn regulated by the number of marriageable women; in colonies with a permanent demand for labour, men and women should therefore be induced to immigrate in equal numbers, since “Colonies can only be planted by families.”(Farr, William (Humphreys, Noel A., ed.), 1885)
Marriage Statistics and Fecundity
Marriage registration provided vital statistics with one of its most sensitive economic instruments. Marriages by banns (predominantly working class) outnumbered marriages by licence (upper and middle class) five to one; but because the banns-to-licence ratio shifted measurably when wheat prices rose, it functioned as a real-time test of lower-class economic conditions.(Farr, William (Humphreys, Noel A., ed.), 1885) Successive generations exhibited a near-doubling of annual marriages every 33 years: from 52,666 in the mid-eighteenth century to 72,347 in 1791, 104,180 in 1824, and 158,868 in the 1855–59 quinquennium. Each married couple left approximately two married couples in the next generation.
The fecundity of marriage varied sharply by region and by the proportion of women in the married state at childbearing ages. Across all England in 1856–60, 1,000 married women aged 15 to 55 bore 220 children annually, while 1,000 unmarried women bore 16 illegitimate children.(Farr, William (Humphreys, Noel A., ed.), 1885) Mining counties (Durham 59.9 percent of women married, Staffordshire 56.3 percent) had the highest birth-rates; agricultural counties (Devon 45.2 percent) had the lowest. The county illegitimacy rate showed similar geographic variation: Norfolk had 1 illegitimate birth in every 9, with 25 per 1,000 unmarried women bearing children annually, while Devonshire had 1 in 18 and only 11 per 1,000 unmarried women.(Farr, William (Humphreys, Noel A., ed.), 1885)
Scotland’s apparent paradox illustrated how carefully these statistics required interpretation. Scottish wives were individually more prolific than English wives (24.8 children per 100 wives versus 22.0), yet 1,000 Scottish women collectively bore fewer registered children (120) than 1,000 English women (123), because Scotland had a much lower proportion of formally recognized wives.(Farr, William (Humphreys, Noel A., ed.), 1885) Scotland’s higher illegitimacy rate (89 per 1,000 births against England’s 65) resulted in part from a legal regime that allowed a quasi-marriage state before formal solemnization. England’s refusal to grant legitimation to children born before marriage — unique among European nations — had the counterbalancing effect of inducing more women to marry explicitly rather than hovering in a concubinage status.(Farr, William (Humphreys, Noel A., ed.), 1885)
Comparative European data (1876) showed Italy apparently highest in fecundity at 5.15 births per marriage (possibly an artifact of incomplete marriage registration), Prussia at 4.92, Sweden and the Netherlands near 4.83, England at 4.63, Belgium and Spain at 4.48, Denmark 4.24, Austria 3.73, and France lowest at 3.42.(Farr, William (Humphreys, Noel A., ed.), 1885) The practical consequences of France’s low fecundity were stark. If English wives had reproduced at French rates, they would have borne only 175,048 legitimate children annually in 1861–70 against the 701,309 actually registered — fewer than the 479,450 annual deaths.(Farr, William (Humphreys, Noel A., ed.), 1885) Without immigration, England’s population would have declined. Farr judged England’s higher fecundity a “great and wise” national policy, populating colonies and planting the English race broadly, in contrast to the restraint of French peasant households.
The registration of illegitimate births itself required critical scrutiny. The 1830 census illegitimacy returns collected by church ministers totaled only 20,039, but civil registration in 1842 yielded 34,796 — 74 percent more, against a population growth of only 17 percent.(Farr, William (Humphreys, Noel A., ed.), 1885) The difference was attributable to the deficiency of the earlier ecclesiastical returns. Over subsequent decades, illegitimacy declined gradually: from nearly 7 percent of total births around 1851, to 6.5 percent in 1851–60, 6.1 percent in 1861–70, and 5.6 percent in 1871.(Farr, William (Humphreys, Noel A., ed.), 1885) London consistently ran below the national average.
Origins: Bills of Mortality and Graunt’s Foundation
The formal discipline of vital statistics emerged from an older and cruder tradition of urban mortality surveillance. John Simon, then serving as Medical Officer of the Privy Council, traced the ancestry of the discipline directly in English Sanitary Institutions (1890): London’s Bills of Mortality — kept since the early plague years of the sixteenth century and printed regularly from about 1593 — gave, “imperfect as they were, the first systematic record of urban death.”(John Simon, 1890) John Graunt’s analysis of them in 1662, published as Natural and Political Observations on the Bills of Mortality, was, in Simon’s account, the act of intellectual foundation: it “founded the science of vital statistics and showed what could be inferred from mortality records about the health of populations.”(John Simon, 1890) The civil registration system created after 1836, which Farr spent his career perfecting, was the bureaucratic successor to those weekly bills — an attempt to place the insight Graunt had demonstrated on an accurate, comprehensive, and continuous footing.
Death Registration and Certification
The quality of vital statistics depended entirely on the completeness and accuracy of the underlying registration. A comprehensive death schedule, Farr argued, should record not merely the cause and date of death but also birthplace, duration of residence, parents’ names, marital status, and number of children — information needed for population analysis beyond simple mortality counts.(Farr, William (Humphreys, Noel A., ed.), 1885) A specimen London death schedule for William Canty of Poplar was published to demonstrate the reformed format.
In practice, registration was uneven. An analysis of the first quarter of 1858 found that of 125,819 deaths in England, 79 percent were medically certified, 4 percent certified by coroners, 6 percent were not attended medically at all, and 11 percent were uncertified for unknown reasons.(Farr, William (Humphreys, Noel A., ed.), 1885) London reached 92 percent medical certification; the Welsh Division achieved only 61 percent. In a separate annual analysis, 4,630 deaths had no cause specified and 3,506 were inferred sudden deaths; in London specifically, 93 percent were medically certified, 5 percent by coroners, and 2 percent uncertified.(Farr, William (Humphreys, Noel A., ed.), 1885)
Registration quality also varied by gender of informant. In Northampton, 85 percent of death informants were women, of whom 69 percent could only sign with marks rather than their names.(Farr, William (Humphreys, Noel A., ed.), 1885) The French Civil Code, which excluded women as informants of death, stood in instructive contrast. The Compulsory Medical Certification Act 1874 (37 and 38 Victoria cap. 88) eventually mandated medical certification in England — a legislative milestone Farr had advocated for decades — but progress to that point had been slow.
Mortality Analysis
The mean lifetime of the English population stood at 40.858 years, while in the healthiest districts it reached 49 years; eliminating diseases affecting the first year of life alone would raise mean lifetime from 40.9 to 45.0 years.(Farr, William (Humphreys, Noel A., ed.), 1885) Of 335,956 deaths registered in the year ending June 1838, more than a fifth (71,888) were under one year of age. This made the subdivision of the first year of life into months methodologically essential, since the expectation of life “on the day of birth differs greatly from that at six, three or even one month old.”(Farr, William (Humphreys, Noel A., ed.), 1885)
Population density shaped mortality in a regular mathematical relationship. Density affected mortality most severely in early childhood (ages 0–5) and again in middle age (45–65, after reproduction ceased), and least at ages 15–25 when migrants entered towns.(Farr, William (Humphreys, Noel A., ed.), 1885) In the 47 densest districts (1,718 persons per square mile), childhood mortality was doubled compared to rural areas, and mortality at ages 45–65 was raised by about half.
The 1847–48 London influenza epidemic illustrated the demographic weight of infectious disease. Some 11,339 persons died in six weeks. The epidemic raised childhood mortality 83 percent above the normal seasonal level, adult mortality 104 percent, and old-age mortality 247 percent; it killed twice as many in the insalubrious parts of London as in the healthier parts.(Farr, William (Humphreys, Noel A., ed.), 1885) From these observations Farr developed calculations of what disease suppression would theoretically accomplish. Eliminating cancer would have a small effect at birth (raising life expectancy from 39.68 to 39.88 years) but a larger effect at ages 55 and older, because cancer kills predominantly in later life after other diseases have already selected survivors.(Farr, William (Humphreys, Noel A., ed.), 1885)
The mean lifetime of the English population stood at 41 years while in the healthiest districts it reached 49; 242,325 persons died under the age of 25 annually — figures demonstrating both the gap between actual and achievable longevity and the heavy burden of premature mortality.(Farr, William (Humphreys, Noel A., ed.), 1885) Only 392 deaths were attributed to homicide in the annual registration data Farr analyzed — a surprisingly small number that calibrated the relative scale of violent mortality against the tens of thousands dying of infectious disease.(Farr, William (Humphreys, Noel A., ed.), 1885)
Life Tables and Actuarial Science
The life table was the central analytical instrument of vital statistics. Farr described it as a “biometer” — an instrument that gives the exact measure of the duration of life under given circumstances — and argued that a separate table should be constructed for each district and profession to determine their relative salubrity.(Farr, William (Humphreys, Noel A., ed.), 1885) The methodology this required had been obscure in the earlier work. Richard Price never explained the interpolation method he employed in framing the Northampton and Swedish Tables of Mortality; Farr judged the method probably empirical rather than mathematical, representing a methodological opacity in the foundation of early actuarial science.
Price’s Northampton Table carried a serious error with practical consequences. Price had estimated the mean lifetime at 25.18 years; correctly recalculated from census-based data, the true value was 37.57 years — an error of more than 12 years.(Farr, William (Humphreys, Noel A., ed.), 1885) Life insurance companies that had built premium schedules on the Northampton Table had therefore been charging too much for life assurance. The English Life Table No. 3, constructed from 6,470,720 deaths registered in England and Wales during 1838–54, provided a far more reliable foundation — the largest mortality study constructed to that date anywhere in the world.(Farr, William (Humphreys, Noel A., ed.), 1885)
Life insurance originated in England with policies issued in the late seventeenth century, and by the mid-nineteenth century had become a great national institution whose financial stability depended entirely on the accuracy of the mortality tables it employed.(Farr, William (Humphreys, Noel A., ed.), 1885) The life insurance offices practiced selection of lives, rejecting impaired risks and accepting only the healthiest applicants; their experience tables therefore showed lower mortality than the general population, a selection effect that Farr documented and analyzed as an instance of the broader methodological problem of non-representative samples.
The Healthy District Life Table, constructed in 1859 from 63 English and Welsh districts where mean annual death-rates did not exceed 17 per 1,000 during 1849–53, served as a scientific standard for measuring preventable mortality — the gap between what was and what a well-organized state could achieve. Applying the sickness-mortality ratio to England in 1846, Farr estimated 744,600 persons were constantly disabled by disease, reducing productive national power by one-fifteenth.(Farr, William (Humphreys, Noel A., ed.), 1885) Sick-time increased with age in geometrical progression — a law deduced from friendly society records that provided the mathematical basis for actuarially-grounded health insurance systems.(Farr, William (Humphreys, Noel A., ed.), 1885)
The marriage register also embedded within routine civil registration a proxy measure of literacy and social progress. The proportion of persons who signed the marriage register by mark (unable to write) declined steadily throughout the period of civil registration, affording a trustworthy and impartial measure of the progress of elementary education derived incidentally from vital registration.(Farr, William (Humphreys, Noel A., ed.), 1885)
Occupational and Institutional Mortality
Vital statistics extended its analysis from the general population to specific occupational groups and institutional populations, where mortality differentials were most sharply visible. Clergy had a mean after-lifetime at age 25 of 42.1 years, while publicans had a mean after-lifetime of only 31.3 years — a difference of nearly 11 years that Farr attributed to occupational exposure to alcohol.(Farr, William (Humphreys, Noel A., ed.), 1885) Bernardino Ramazzini had described the effects of alcohol vapour on workers in Modena distilleries — an early occupational health account that Farr cited as support for his statistical argument.(Farr, William (Humphreys, Noel A., ed.), 1885)
Workhouse and prison mortality exceeded general population mortality substantially, and Farr systematically documented the lethal effects of institutional confinement.(Farr, William (Humphreys, Noel A., ed.), 1885) Millbank Penitentiary mortality doubled under imprisonment conditions compared to expected rates for men of the same ages in freedom.(Farr, William (Humphreys, Noel A., ed.), 1885) Farr calculated that imprisonment destroyed more than ten times as many lives as capital execution in England, making incarceration a greater cause of death by deliberate state action than the gallows.
Among lunatic asylums, the mortality differential between types of institution was striking. Pauper lunatics in licensed private houses had approximately twice the annual mortality of those at Hanwell Asylum, with licensed house mortality near 19 percent against Hanwell’s 9 to 10 percent.(Farr, William (Humphreys, Noel A., ed.), 1885) Paradoxically, large asylums averaging 265 patients showed 19 percent annual mortality while small licensed houses averaging only 17 patients showed 9 percent — suggesting that size was not the decisive variable but rather conditions of care.(Farr, William (Humphreys, Noel A., ed.), 1885) Of 1,000 admissions to asylums, Farr’s nosometrical tables calculated that 380 would recover and 620 would die.(Farr, William (Humphreys, Noel A., ed.), 1885) Asylum mortality was highest in the first 18 months of admission, running at approximately 18 percent annually, declining to about 8 percent thereafter for long-term residents.(Farr, William (Humphreys, Noel A., ed.), 1885) Bethlem Hospital’s artificially favorable mortality resulted from its strict admission criteria, which selected only recoverable cases — an early documented instance of selection bias in institutional statistics.(Farr, William (Humphreys, Noel A., ed.), 1885)
The legal procedures for committing persons to asylums were inadequate to prevent improper detention, and the statistical evidence of high mortality in certain institutions demanded reform of the legal safeguards around lunacy certification.(Farr, William (Humphreys, Noel A., ed.), 1885) The conjugal condition of persons in the general population — whether married, single, or widowed — was itself a significant variable in mortality analysis, with French data demonstrating that marriage and its absence registered measurably in death rates.(Farr, William (Humphreys, Noel A., ed.), 1885)
Florence Nightingale proposed a uniform system of hospital statistics to allow proper comparison of mortality across different hospitals, recognizing that raw death rates were confounded by case mix and patient selection.(Farr, William (Humphreys, Noel A., ed.), 1885) Without such standardization it was impossible to determine fairly whether one hospital was more or less fatal than another. This proposal represented one of the earliest systematic arguments for what would later be called risk adjustment in comparative medicine.
Violent Death and Suicide
Vital statistics provided a systematic account of violent death that distinguished homicide, accident, and suicide as analytically separate categories requiring different explanatory frameworks. London Bills of Mortality from the seventeenth to the nineteenth century showed violent deaths declining from 6.8 to 5 per 100,000, and the chance of being murdered diminishing nine-fold over that span — a long secular decline accompanying civilizational development.(Farr, William (Humphreys, Noel A., ed.), 1885) In the coroner’s inquest year 1856, England recorded 21,801 inquests with 482 homicide verdicts (205 murder, 271 manslaughter), leading to 109 convictions and 16 executions — documenting the gap between homicide verdicts and criminal convictions.(Farr, William (Humphreys, Noel A., ed.), 1885)
Annual accidental deaths in England during the 1850s included 391 to 548 railway deaths per year, 1,136 mine deaths (985 from coal mines), 3,045 deaths from burns, and 2,566 drownings, alongside 125 opium deaths and 27 arsenic deaths.(Farr, William (Humphreys, Noel A., ed.), 1885) Suicide averaged 1,083 deaths per year; hanging, cut-throat, and drowning together accounted for eight in ten suicides.(Farr, William (Humphreys, Noel A., ed.), 1885) Suicide rates varied by occupation: outdoor workers (masons, carpenters, butchers) had the lowest rates, at 1 in 9,332 annually; indoor sedentary workers (tailors, shoemakers, bakers) had the highest, at 1 in 1,669.(Farr, William (Humphreys, Noel A., ed.), 1885)
Alcohol, Railway Safety, and Social Insurance
Vital statistics touched the political economy of everyday life through its analyses of alcohol mortality, railway accidents, and the design of social insurance. The physiological action of alcohol in small doses had been characterized by experimental work: Dr. Brunton described small doses as increasing gastric secretion, raising heart rate, dilating fine vessels, and producing warmth, while noting the risk of surface chilling in extreme cold.(Farr, William (Humphreys, Noel A., ed.), 1885) Dr. Parkes’s experiments showed that one pint of good claret raised pulse rate from 76.3 to 80.5 but had no measurable effect on internal body temperature, while larger amounts produced torpor.(Farr, William (Humphreys, Noel A., ed.), 1885)
The occupational mortality comparison between clergy (mean after-lifetime of 42.1 years) and publicans (31.3 years) provided the most cited statistical argument for the health costs of the alcohol trade. That argument required a statistical rather than a categorical approach: arsenic, opium, and chloroform were also lethal at high doses but beneficial or neutral at low doses, and a coherent public health position demanded dose-specific evidence.
Railway accidents presented vital statistics with a problem of social insurance design. The existing passenger insurance companies were inadequate because they covered only accidents “to the train” and excluded approximately one-third of actual railway deaths.(Farr, William (Humphreys, Noel A., ed.), 1885) Lord Campbell’s Act (9 and 10 Victoria cap. 93, 1846) had established the legal right to sue for compensation when death was caused by wrongful act, neglect, or default,(Farr, William (Humphreys, Noel A., ed.), 1885) but of 203 actions brought under the Act in 1868, only 122 resulted in plaintiff verdicts, recovering a total of 68,092 pounds with average damages of 549 pounds — a system too expensive and uncertain to serve most claimants.(Farr, William (Humphreys, Noel A., ed.), 1885)
Even when passengers were killed through their own want of caution, Farr argued that railway companies should pay a fine, invoking the principle of the deodand: without financial liability, companies had no incentive to install preventive measures.(Farr, William (Humphreys, Noel A., ed.), 1885) The compensation paid by English companies for personal injury in 1867 totaled 322,985 pounds — more than a systematic actuarial tariff would have required — suggesting that organized insurance could replace uncertain litigation at comparable total cost.(Farr, William (Humphreys, Noel A., ed.), 1885)
Civil registration indexes had by 1854 grown to contain more than 21 million names after 17.5 years of operation, receiving a yearly addition of 1,350,000 names.(Farr, William (Humphreys, Noel A., ed.), 1885) Under the unsafe railway conditions of the period, Farr noted that the deaths among railway servants and contractors amounted to “a battalion killed every year.”(Farr, William (Humphreys, Noel A., ed.), 1885) Analysis of registration indexes also yielded England’s first large-scale empirical study of family nomenclature. Smith was the most common surname (286,037 registration entries in 1838–54) against Jones’s 282,900, approximately one person in 73 being a Smith and one in 76 a Jones.(Farr, William (Humphreys, Noel A., ed.), 1885) Welsh surnames showed conspicuously low diversity: nine-tenths of the Welsh population could be found under fewer than 100 surnames, because hereditary surnames had not been generally established in Wales until after the Tudor period.(Farr, William (Humphreys, Noel A., ed.), 1885)
Disease Suppression and Economic Modeling
Vital statistics reached its most ambitious formulations when it attempted to calculate the economic consequences of disease and the potential gains from disease suppression. The index of Farr’s collected work confirmed coverage of the economic value of population through calculation of the cost and present and future value of a man, including wages and cost-of-maintenance tables for agricultural labourers.(Farr, William (Humphreys, Noel A., ed.), 1885) The influence of marriage on mortality among the French population, and the separately analyzed effect of chastity upon mortality, demonstrated that conjugal condition ranked among the major social determinants of health that the discipline was equipped to measure.(Farr, William (Humphreys, Noel A., ed.), 1885)
Eliminating cancer would raise mean lifetime at birth from 39.68 to only 39.88 years, a small gain reflecting cancer’s concentration at later ages.(Farr, William (Humphreys, Noel A., ed.), 1885) These calculations grounded a broader argument about the priority of sanitary conditions over more targeted interventions. Suppressing one disease, Farr observed, merely allowed other diseases to fill its place, “as the extirpation of one weed makes way for other kinds of weeds in a foul garden.” Sanitary conditions as to food, drink, and cleanliness therefore stood first in importance, with quarantine and vaccination subordinate.
Vital statistics provided the evidentiary foundation on which public health reform rested throughout the nineteenth century. Its methods were not merely technical: they constituted an argument that the duration and quality of human life were subjects of systematic knowledge, and that knowledge could be translated into policy. In Farr’s formulation, extending mean lifetime from the English average of 40.86 years to the 49 years achievable in the healthiest districts would add a fifth to the economic value of the population — equivalent to adding 1,050 million pounds to national wealth through sanitary reform alone. The discipline was, at its core, the application of mathematical regularity to the oldest of questions: why people live and die when and where they do.