Source / Printed mortality records
Published for London week after week from the plague year 1603 into the nineteenth century, the Bills of Mortality turned parish reports of christenings, burials, plague deaths, and other reported causes into a recurring printed view of the metropolis.
The bills matter because they made comparison over time and place unusually accessible to officials, householders, writers, and later analysts. They were not a census, a register of every death, or a set of physician-certified diagnoses: the people, categories, parishes, and surviving sheets all changed.
Origins and chronology
There is no single secure “first Bill of Mortality.” Historians distinguish royal and civic orders to report plague deaths, weekly manuscript returns, epidemic print experiments, and the later commercial series. Surviving evidence is uneven because these working papers and single sheets were rarely meant to last (Heitman 2018; Robertson 1996).
1510s–1550s: Crown and City authorities sought parish reports during plague. A surviving report probably dating from 1512–32 already lists parishes, total deaths, plague deaths, and parishes clear of infection. In 1555 London’s aldermen compensated the parish clerks for weekly, parish-by-parish reports of deaths and their reported causes even outside plague years. These were administrative manuscript accounts, not yet the familiar public print series (Heitman 2018).
1591–1603: A unique 1591 manuscript return at the Folger records causes for every reporting City parish, total burials, plague columns, and baptisms. Printed plague bills appeared in the 1590s, but historians differ over which epidemic or surviving sheet should mark the beginning. What is firmer is that regular public printing took hold during the severe outbreak of 1603 (Heitman 2018; Robertson 1996).
1604–1636: The printed geography expanded beyond the City. Suburban parishes and precincts were added in 1604, Westminster in 1626, and further eastern, northern, and southern parishes by 1636. The Company of Parish Clerks acquired a press and licence in 1626; two-sided sheets could then combine the parish table with a fuller citywide table of “diseases and casualties” (Robertson 1996; Heitman 2018).
1662–1665: John Graunt published Natural and Political Observations … upon the Bills of Mortality in 1662, reorganising accumulated annual figures to compare causes, seasons, sexes, parishes, and population. The 1665 compilation London’s Dreadful Visitation preserved the weekly bills of the Great Plague year. These two books put serial data to different uses: Graunt rearranged earlier totals into an argument, while the compilation gathered one epidemic year for readers and posterity (Graunt 1662; London’s Dreadful Visitation 1665; Weinreich 2021).
1727–1752 and beyond: Citywide age-at-burial totals appear from December 1727. The year 1752 is a useful endpoint for the Death by Numbers dataset because Britain’s calendar reform removed eleven days, not because the bills ceased. Weekly publication continued into the nineteenth century and was eventually displaced by civil registration and the Registrar General’s returns (Death by Numbers, “Data”; Krylova and Earn 2020).
Production
The typography makes the bills look impersonal, but no number entered print automatically. Death had to become known to parish officers, assigned a cause, entered as a burial, transmitted, combined, printed, and sold.
London parishes appointed women known as searchers of the dead to view corpses and inquire into causes. Many were poor or pensioned parishioners whose work brought small payments, local authority, and dangerous exposure during plague. Describing them only through later male complaints that they were ignorant erases both their practical knowledge and the administrative system that relied on them (Munkhoff 1999; Heitman 2018).
Sextons, burial arrangements, bells, relatives, and neighbours could bring a death to official notice. The searchers’ judgment was reported locally; the parish clerk maintained the burial register and sent a weekly numerical account to the Company of Parish Clerks. Unlike a parish register, the printed bill usually discarded names and household narratives (Heitman 2018; Robertson 1996).
The Company’s clerk combined parish returns for civic and royal readers, while printers and distributors made copies available by subscription and sale. Publication served government but also generated income. By the seventeenth century readers carried, copied, annotated, discussed, and forwarded the weekly numbers; their interpretations were not confined to medicine (Robertson 1996; Weinreich 2021).
Reading one bill
A typical later-seventeenth-century weekly sheet should be read across both sides and against other weeks. Its sections answer different questions, over different geographies, and cannot safely be merged into individual case records (Death by Numbers, “The London Bills”).
One side groups parishes “within the walls,” “without the walls,” in Middlesex and Surrey, and in Westminster. For each parish it normally gives all burials and the subset reported as plague. This can trace an outbreak’s reported movement, but parish size, membership, and inclusion differed; raw counts are not mortality rates.
The table of “diseases and casualties” aggregates reported causes across the covered area. With plague as the major exception, the printed bill does not usually locate each cause by parish. A reader cannot infer, for example, where every death labelled “consumption” occurred from the citywide total (Death by Numbers, “Data”).
Christenings are ecclesiastical events, not a complete count of births; burials are not identical to deaths. Later bills may divide total christenings and burials by sex, state the official weight of bread sold for a fixed price, and, from December 1727, give citywide age groups for those buried. None of these additions creates a person-level dataset linking age, sex, parish, and cause (Death by Numbers, “The London Bills”; Death by Numbers, “Data”).
Plague, government, and readers
Civic intelligence: Before public print, returns moved among the Lord Mayor, aldermen, Crown, and senior officers. Reported plague in a house or parish could help prompt quarantine, watching, and decisions about movement. The figures therefore participated in government; they did not merely describe disease after the fact (Heitman 2018).
Personal and commercial calculation: People read weekly rises and the location of plague to judge danger, flight, travel, trade, and the safety of contacts. Letter writers enclosed bills; diarists copied numbers; preachers and polemicists treated mortality as evidence of providence and moral condition. Calling this only “medical surveillance” would flatten the civic, religious, and commercial meanings that contemporary readers gave the sheets (Robertson 1996; Weinreich 2021).
The 1665 totals: The year’s general bill recorded 97,306 burials, of which 68,596 were labelled plague. Those are the Company’s classified burial totals for the parishes within the bills, not a complete or laboratory-confirmed death toll for metropolitan London. The bound compilation was printed by the Company’s printer for sale and presented its reports as made to the king; it is a primary record of official counting and publication, not independent proof that every cause was correct (London’s Dreadful Visitation 1665; Krylova and Earn 2020).
Medical limits: “Plague” was assigned through bodily signs, household information, local experience, and administrative judgment, not bacteriology. A plague label could bring coercive restrictions and economic loss, creating reasons to contest or conceal it. Conversely, a rise in other cause labels cannot simply be recoded as plague case by case. The bills document reported classification under pressure, while the true distribution of infection remains partly uncertain (Munkhoff 1999; Heitman 2018).
Historical causes of death
A causes table mixes complaints, visible signs, life stages, accidents, and social judgments. Terms such as “aged,” “childbed,” “consumption,” “fever,” “teeth,” and “rising of the lights” did not form a stable modern disease taxonomy (Graunt 1662, chs. II–V; Death by Numbers, “The London Bills”).
“Fever” or “consumption” might cover multiple illnesses; “teeth” attached death to teething-age childhood rather than identifying a single pathology; “childbed” marked an association with childbirth without specifying haemorrhage, infection, or another mechanism. Translating each label into one current diagnosis creates certainty the source does not contain.
New terms appeared, old terms declined, spellings varied, and printers grouped causes differently. A sudden rise may reflect altered reporting or vocabulary as well as mortality. Even Graunt grouped and compared labels to make them analytically useful and argued that exact diagnostic accuracy was less important for some broad patterns than his critics assumed (Graunt 1662, chs. II–V; Robertson 1996).
Households supplied information; parish officers imposed procedures; clerks abstracted; Company staff totalled; and printers reproduced the result. Historians have shown how later statisticians’ dissatisfaction was often displaced onto female searchers even though the reporting system’s purposes, geography, and categories themselves limited the data (Munkhoff 1999; Robertson 1996).
Graunt and statistical reuse
John Graunt, a London citizen and tradesman rather than a physician, reduced decades of annual bills into comparative tables. He examined recurring proportions, seasonal and epidemic variation, male and female totals, the growth of London, and the difference between city and country records; he also estimated a population that the bills did not directly count (Graunt 1662).
Before Graunt: Officials and ordinary readers already compared weeks, parishes, and plague years. Graunt did not invent attention to mortality numbers from nothing. His distinctive contribution was to rearrange a long run of reports, test regularities, expose inconsistencies, and make further inferences from aggregate data (Robertson 1996; Weinreich 2021).
What the book can show: Natural and Political Observations is direct evidence for Graunt’s methods, assumptions, and proposals for better accounting. Its tables also preserve annual figures from bills that may no longer survive. Because it is a selected and transformed analysis, it should not be treated as a facsimile archive of every weekly issue (Graunt 1662, preface and chs. I–XII).
Later reputation: Graunt is frequently called a founder of demography, vital statistics, or epidemiology. Those labels recognise the reach of the Observations, but they can obscure earlier comparison, the Company’s institutional archive, and the labour that generated each number. The history is better understood as reuse of a collective record than as a lone discovery of population data (Robertson 1996; Weinreich 2021).
Coverage and missing lives
The sheets counted reported christenings and burials in participating parishes. Roman Catholics, Jews, Quakers, and other nonconformists could be absent, while stillbirths and infants not christened complicate any use of christenings as births. These omissions grew more serious as religious and burial practices changed (Heitman 2018; Krylova and Earn 2020).
Parishes were added, removed, or regrouped, but the series never expanded in a simple line with the inhabited city. Fast-growing districts and extra-parochial institutions could lie outside the totals, and Londoners buried beyond the covered area disappeared from its burial count. Long-term comparison requires a consistent geography or explicit adjustment (Krylova and Earn 2020; Death by Numbers, “Data”).
The printed bill rarely retains a name, occupation, household, migration history, treatment, duration of illness, or the testimony by which a cause was assigned. Class and gender may shape exposure, care, burial, and recording without appearing as columns. Parish registers, nonconformist records, wills, institutional accounts, and household testimony are needed to recover experiences that totals suppress.
Using the source
A sequence can reveal weekly seasonality, the reported geography of plague, changing cause vocabulary, and shifts in the relation between christenings and burials. State exactly which parishes, dates, and version of the bill are being compared.
A cause total does not identify a person or prove a modern disease mechanism. Do not convert “buried” to “died,” “christened” to “born,” or an early modern cause word to a current diagnosis without separate evidence.
The Death by Numbers team manually transcribes images, preserves archive and call-number provenance, uses flags for missing or illegible material, and has a second reviewer check entries. Its current data coverage is not the same as the estimated publication run, and the project advises checking original bills where readings matter (Death by Numbers, “Data”).
Across the collection
Follow the transformation of recurring counts into rates, comparisons, causal claims, trials, and policy.
Compare public mortality returns with case notes, registers, laboratory records, and later health data.
Place the bills within changing histories of government, cities, institutions, public health, and empire.
References
London: printed and sold by E. Cotes, printer to the Company of Parish Clerks, 1665. Wellcome Collection catalogue record and complete public-domain digitisation. Collects the weekly bills from 27 December 1664 to 19 December 1665 and the general bill. A primary source for printed format and official totals; as a Company publication, it does not independently verify the completeness or causes of the burials reported.
London: Thomas Roycroft for John Martin, James Allestry, and Thomas Dicas, 1662. Wellcome Collection page images of the first edition. Graunt’s primary account of the bills’ development and his comparisons of causes, plague, seasons, christenings, burials, sexes, parishes, and population. It records his assumptions and transformations rather than supplying neutral case data.
Urban History 23, no. 3 (1996): 325–350. DOI: 10.1017/S0963926800016898. Reconstructs metropolitan expansion, official and public reading practices, the Company’s institutional position, and pre-Graunt interpretation; especially useful for separating the reporting system’s original purposes from later statistical demands.
Gender & History 11, no. 1 (1999): 1–29. DOI: 10.1111/1468-0424.00127. Archival study of the women hired to examine and classify bodies, their economic vulnerability, local authority, and place in plague administration.
Church History 90, no. 4 (2021): 799–823. DOI: 10.1017/S0009640721002833. Open-access study of how bills and mortality numbers circulated through sermons, correspondence, print, and religious argument before and alongside political arithmetic.
PLOS Biology 18, no. 12 (2020): e3000506. DOI: 10.1371/journal.pbio.3000506. Peer-reviewed, open-access analysis with a concise source critique: surviving sequence, Anglican burial coverage, changing geography, missing returns, exported burials, cause classification, and the limits of inferring disease patterns.
The Collation, Folger Shakespeare Library, 13 March 2018. Collection-based account of early manuscript evidence, Tudor orders, the 1555 programme, searchers, parish reporting, government circulation, public print, and the 1626 press. It makes clear where survival is unique and where the early chronology remains uncertain.
Roy Rosenzweig Center for History and New Media, George Mason University, 31 January 2022. Illustrated guide to a later-seventeenth-century bill’s recto and verso, parish groupings, burials, plague columns, citywide causes, christenings, and bread information.
Roy Rosenzweig Center for History and New Media, George Mason University. Project documentation for current weekly and annual coverage, archive-level provenance, manual transcription and review, reported-age data from December 1727, changing parish groups, historical-to-dataset category decisions, and missing or illegible records. The site separately estimates that almost 8,000 weekly bills were published between 1603 and 1752; the database does not claim that every issue survives.