Data

The data for the Death by Numbers project has been manually transcribed out of photographs and digitized microfiche images of the original bills of mortality. To ensure that scholars are able to trace our transcriptions back to the original, archival sources, we have divided our transcribed bills into multiple datasets that are named and described according to the archive and call numbers where we found collections of bills. This is most apparent in the downloadable CSV files and can also be traced in the unique IDs for each transcription. For more on our project workflow, see our methodologies essays.

Data can be accessed through:

The London Bills of Mortality dataset currently consists of weekly mortality bills from 1636 to 1752 and annual bills of mortality from 1656 to 1750. It includes data from both the city of London and the greater London metropolitan area.

All weekly and annual data which are currently known to be extant from these two periods are part of the dataset. To avoid disrupting the time series, the dataset ends when England reformed its calendar by removing 11 days.

The bills consist of burial data collected by London's parishes, which are part of the administrative structure of the Church of England. Therefore, the data does not necessarily include all deaths, particularly in the eighteenth century, as people increasingly opted out of membership in the church.

Parish-level burial data is subtotaled and totaled in the bills. The urban subset of parishes "within the walls" remain most consistent throughout this time period. Due to urban sprawl, additional parishes were regularly added to the bills and thus the suburban subset of parishes "without the walls," in Middlesex/Surrey, and in Westminster. Users should be alert to parishes being temporarily moved from one subset of parishes to another, such as occurred in 1673-4 when Christ Church was moved from the parishes without the walls to the out-parishes in Middlesex and Surrey. The parish associated with the Tower of London—St. Peter ad Vincula—rarely appears in the bills but can be seen, for example, in 1729 when it is considered to be a parish "within the walls".

Cause of death data was collected at the parish level but—aside from plague—was only printed at the city-wide level. Some location data is recorded for accidental deaths, drownings, etc. and has been transcribed when available.

The project uses the categories suicide, killed, and other in ways that are distinct from early modern usage of those terms. Suicide includes all self-murders, including those excused by distraction or lunacy. Killed includes all non-suicide deaths that involve human agency, whether or not intentional. Other includes all causes of death not encompassed by another category in the dataset. Users should be particularly mindful that the boundary between deaths with and without human agency is not entirely clear and may want to make their own categorical judgements based on the information transcribed about suicide, killed, and other deaths.

Christening data was collected at the parish level but only printed at the subset-of-parishes level and city-wide level. It is probable that children involved in abortive, miscarriage, and stillborn burials would not have been christened; users interested in birth data should consider adding those burial numbers to the christening numbers to better approximate the number of births.

From December 1727 on, the bills include data at the city-wide level on the reported age of burials. This data has not yet been fully transcribed.

Additional data from the bills on the price of bread and other foodstuffs has not yet been fully transcribed.

All data transcriptions were reviewed by a second person, to increase the accuracy of transcriptions. The "is illegible" flag has been used both to mark numbers that were impossible to transcribe and numbers that both the transcriber and reviewer were uncertain of. The "is missing" and "is illegible" flag are used in combination to indicate when a page or torn or data has been otherwise rendered illegible by damage to the original document. Given the poor quality of surviving sources and/or reproductions of them, users are advised to consult the original bills if there is any concern as to the accuracy of the transcription.

Some of the bills have data that has been overwritten or otherwise corrected by previous owners. Due to the research aims of this project, the original printed data has been transcribed rather than the corrections.