Estimating London's Population

A General Bill of all the Christenings and Burials From 14th of December 1742, to the 13th of December 1743 A General Bill of all the Christenings and Burials From 14th of December 1742, to the 13th of December 1743.

Introduction

Mapping the distribution of crimes, or votes, or tax receipts is of limited use if this information cannot be associated with the distribution of the population. In order to facilitate this kind of comparison, we have developed population figures for each of the parishes included on John Rocque's 1746 map of London, and associated these with a figure reflecting the physical size of each parish to arrive at a measure of population per square kilometre (also displayable as per statute acre). The geographical extent of each parish was generated by measuring the area within known parish boundaries as recorded on a corrected version of Rocque's map (that is 'warped' to make it consistent with a geographically accurate modern map). In the first instance, the boundary for riverside parishes has been taken to be a mid-point between the two banks(though legally this boundary was the low tide mark). To generate a meaningful population density for these riverside parishes, the whole of the river Thames to it high water mark, has been subtracted from area of these parishes. For details of how these maps were developed, see Mapping Methodology.

The population figures were calculated from three different sources to capture the population distribution at different periods. As with all historical population data, however, the resulting figures (both for population alone and for density) have been developed from data that is either partial, problematic, or which was originally collected for a different purpose. The resulting data therefore needs to be used with extreme care.

Three different sources have been used to create populations estimates for the 1690s, 1740s and 1800:

  • The Bills of Mortality (which record burials)
  • The 1695 Marriage Duty Assessment
  • The 1801 census.

Whichever data series you use, please take note that they provide data about relative populations between parishes at a given moment, rather than secure and absolute population figures. These datasets are not designed to allow for the calculation of a London-wide population (each cover a different area and suffer from different gaps and inaccuracies), and in our estimation cannot be used to calculate the overall size of the metropolitan population at any given date. See below for a more detailed discussion of the methodologies applied with each source.

With these caveats, the mapping of population density allows several of the datasets available through Locating London's Past to be observed through a new lens. In particular, population density can be usefully applied to maps reflecting the following datasets:

  • Old Bailey Online (selecting for either 'Crime Location' and 'Defendant Home')
  • Settlement Examinations (a subset of London Lives data)
  • Rating Records for Westminster (available form 1749 onwards in London Lives)
  • Voting Records for Westminster (available from 1749 onwards in London Lives)
  • Hearth Tax Returns (for both occupations and living conditions).

These suggestions represent the datasets that most consistently contain information about addresses, but any dataset that includes locations can be mapped against population by selecting the parish as the unit against which to display data; and, for instance, Population 1740s (from the Bills of Mortality). You can also select Area from the same menu.

Bills of Mortality (1690s & 1740s).

A map of the area covered by the Bills of Mortality in 1831. A map of the area covered by the Bills of Mortality in 1831. Histpop - The Online Historical Population Reports Website.

The Bills of Mortality were originally designed to monitor deaths from plague, and from the late sixteenth century onwards weekly Bills were published giving the total number of deaths of all sorts and also baptisms for each parish in urban London. The cause of death was normally recorded from 1629 onwards, as is the age of death from 1728. Parish clerks were responsible for collecting this data, and their returns continued to be collected up until 1836, but became significantly less reliable from the 1760s onwards.1

As a source, the Bills have the great strength of providing a long term and relatively consistent data series. At the same time they suffer from several serious drawbacks as evidence of population distribution. First, they record burials rather than population itself, and as difficult, record local burials rather than local deaths. This is a problem as in some parishes, such as St Ann Soho, more people were buried than died; while in others, the opposite was true.[^*Jeremy Boulton and Leonard Schwarz, ‘Yet another inquiry into the trustworthiness of eighteenth-century London’s Bills of Mortality’, Local Population Studies, no.85, Autumn 2010, pp.28-45. ^] Burials also have an uncertain relationship to population itself, with peaks and troughs resulting from environmental and lifestyle factors that cannot be fully factored in. The Bills also omit parishes that by the 1740s contained significant areas of urban development. In other words, the Bills provide only an indication of parochial populations, and then, only when aggregated over several years. The weekly Bills were collected into an annual series by Thomas Birch in 1759, and his Collection of the Yearly Bills of Mortality from 1657 to 1758 forms the basis of the data used here.

To generate a population figure from the lists of burials recorded, we have subjected the raw figures to a series of corrections. First, to overcome the variability evident from year to year, caused by disease and environment, parochial deaths have been aggregated across whole decades to establish an average annual death rate. This data has been collected specifically for the 1690s and 1740s. These average death rates have then been modified by using a Correction Factor to address under-registration. The Correction Factors used here are based on the work of John Landers' and published in his Death and the Metropolis. The 1690s and 1740s were chosen because they are decades that correspond with those for which we have large scale datasets that benefit from comparison to population figures; and to tie in directly to Rocque's 1746 map. John Landers’ Correction Factors incorporate the impact of both under-recording of local deaths resulting from infant mortality prior to baptism, and the burial of non-conformists. The relevant multipliers are 1.0128 for the 1690s, and 1.0164 for the 1740s. Landers calculated these correction factors on retrospective parish register data collected as part of the nineteenth century censuses.2

To arrive at a population figure for each parish, the average annual corrected death rate as calculated from the Bills then needed to be turned in to a population figure. In the 1740s, Landers estimates the crude death rate for London as a whole was 46.0 deaths per 1000 of the population. By dividing 1000 by 46.0, a multiplier is arrived at(21.74) that can be applied to the burial statistics from the Bills to generate a population estimate. There is no equivalent estimate of the crude death rate available for the 1690s, and the earliest date for which a figure can be identified is the 1730s. This figure (48.6 deaths per 1000 of the population) has been used to generate a multiplier of 20.58 to arrive at the parish population figures for the 1690s.

The resulting figures are simple estimates. The use of a single crude death rate for the whole of London, for instance, means that the population figures will tend to overstate the population of healthy parts of town, and understate that of relatively unhealthy parishes. In other words, these figures are best thought of as reflecting an order of magnitude, rather than a precise population.

Although reasonably robust, this methodology generated some figures that were clearly wrong for both the 1690s and he 1740s. By identifying parishes where the calculated population density was uncharacteristically high or low, a subset of parish population figures was chosen for moderation using alternative methodologies. In the case of the 1690s, the population figure from the Bills was replaced in several instances with that recorded in the Marriage Duty Assessment. In a few cases for the 1690s, and for several in the 1740s, the population figure derived from the Bills was replaced with one calculated from the 1801 census, using retrospective aggregate population estimates for different decades, based on the registration of deaths in parish records. The results generated by both these methodologies continued to produce anomalous results and in a small number of cases alternative averaging methodologies (based on all three alternative figures) have been used. Details of which parish populations have been calculated by which method can be downloaded as a Word file below.3

Marriage Duty Assessment (1695)

Marriage duty assessments were the product of a form of taxation on baptisms, marriages and burials, and on bachelors aged over 25 and childless widowers. The tax was in force from 1695 to 1706, and imposed a standard charge for vital events and annual payments, and additionally enforced a graduated system of surcharges based on wealth and social status.4 The Marriage Duty Act of 1694 required assessors to obtain ‘the names sirnames estates degrees titles and qualifications of all and every the persons dwelling or residing within the limits’ of an area, with servants and lodgers to be taxed where they lived (see 6 & 7 Wm and M., c. 6: Statutes of the Realm, VI, 1685-94 (1819), 568-89, at 574.) Lists were compiled of all the inhabitants within a specific locale, indicating the duties for which they were liable, or would become liable during the course of the following year (with those in receipt of alms exempt from the financial provisions of the act). The result was a series of parish assessments that resemble census-like lists of inhabitants, which some scholars believe to form the most 'complete' listing available, in terms of capturing the whole population in the period. Coverage was probably not comprehensive, but the records undoubtedly represent the ‘best measure for determining the number of people and households in the city’.5 Gregory King estimated that the marriage duty assessments were subject to under-enumeration in the region of 7%-10%, a belief shared by Glass.

Not many marriage duty assessment lists have survived, and the only extant assessment returns for London are those for 1695. A small number of collectors returns exist for a small number of London parishes for the period 1695-1706, which record the actual baptisms, marriages and burials that took place in the parish during the year, and the payments that were therefore due from the individuals concerned. While there is some variation in the information provided in the assessments, most listings group people together in distinct 'blocks' of names, taken to indicate all those inhabiting the same house, and often describe the nature of the relationship between individuals. In addition, they frequently provide indicators of personal social and marital status, economic standing and, in some instances, age.

The 1695 statistics are undoubtedly more reliable than the equivalent figures derived from the Bills of Mortality, and while some figures appear unexpectedly high or low, this series has been left unmodified.

1801 Census

A sample page of the Enumeration Abstracts for Middlesex and Wesminster from the 1801 census. A sample page of the Enumeration Abstracts for Middlesex and Wesminster from the 1801 census. Histpop - The Online Historical Population Reports Website.

The first national census was undertaken in March of 1801, in accordance with ‘An Act for taking an Account of the Population of Great Britain, and the increase or diminution thereof'.6 The vast majority of detailed returns have not survived. But the enumerators summary reports have, and these detail population figures by parish for London and the rest of the country.

This data was collected by overseers of the poor and ‘other substantial householders’ and the process was managed by local Justices of the Peace, who in turn passed the returns on to the Principal Secretary of State for the Home Department. The forms were then abstracted under the supervision John Rickman. Because few of the detailed house to house returns have survived, the 1801 census has not been substantially used by historians, but there is little reason to doubt the basic reliability of the abstracts created at the time.

The reports published as part of the Parliamentary Papers contain parish by parish figures for total population, and these have been used in an uncorrected form to create the dataset available here. As with figures drawn from the Bills of Mortality and Marriage Duty Allowance, however, these figures should not be used uncritically. The parishes for which we have census data do not map precisely onto those listed in the Bills, so a direct comparison of population densities between the two series is difficult (the Inns of Court, for instance, are listed separately in the Census, but not the Bills). Efforts have been made to aggregate extra-parochial populations in to those as developed from the Bills, but the overlap is not precise. More generally, the census inevitably suffers from some systemic under-reporting.

The Evolving Parish

Regardless of the accuracy of the data series that underlie these statistics, they also suffer from the changing number of parishes that made up London, and from the existence of numerous Extra Parochial places that were sometimes included in the figures for an adjacent parish, and sometimes not. As a result of the passage of the New Churches in London and Westminster Act in 1711, a commission was established which created some 12 new churches in the greater London area, primarily in the 1720s.7 To overcome this issue of new parishes being created from the area of existing ones, each new parish has been treated as if it had existed from the beginning of the period covered by this website. So, for instance, although St George Hanover Square was built between 1720 and 1725, and was carved from the area of St Martin in the Fields, it has been assigned a population figure for the 1690s. This figure was arrived at by taking the population of St Martin's in the Fields in the 1690s, and dividing it proportionally between St Martin's and St George's according to their area. A related problem was created by the tendency of the 1801 census to list population figures for Extra-Parochial places, such as the Inns of Court, separately from the parishes. In these instances, the area and population of the Extra-Parochial Places have been added to the surrounding or adjacent parish. Details of the way in which individual figures and parishes have been treated can be found in Notes on Calculating London's Population (Word).

Using the Population Statistics

A screenshot illustrating the mapping of crime locations in the 1740s, moderated by population. A screenshot illustrating the mapping of crime locations in the 1740s, moderated by population.

To generate a meaningful map using these statistics you need to choose both the right dataset, and the appropriate unit of division; and to then moderate that data against either overall population size, and/or parish area, or the pre-calculated measures of population density.

To take an example you could choose to select for all crime locations specified in the Old Bailey Proceedings for the period between 1740 and 1749. This is done by choosing Old Bailey Proceedings from the list of datasets, and crime location from the pull down menu labelled Place Type, and then using the date restriction lists to select the 1740s. This query generates 1419 hits, which can be either mapped as individual hits, or else Grouped by geographical areas. If you group the results by Parish, and then Divide By: Area, and select Map from the bottom of the page, you will generate a map reflecting the crime rate per 1000 of the population, moderated by area. If you wish to generate a map that reflects crime events per 1000 of the population alone, simply select Population: 1740s.

You can also map these population figures independently of the other datasets by choosing Population and Area Data from the data query page.

Population Statistics and Notes (Downloads)


Bibliography.

  • Histpop - The Online Historical Population Reports Website.
  • 1801 Census, Abstract of the Answers and Returns under the Act for taking Account of the Population of Britain (BPP, 1802, VI).
  • T. Birch, A Collection of the Yearly Bills of Mortality from 1657 to 1758 inclusive… (London, 1759).
  • Jeremy Boulton and Leonard Schwarz, ‘Yet another inquiry into the trustworthiness of eighteenth-century London’s Bills of Mortality’, Local Population Studies, no.85, Autumn 2010, pp.28-45.
  • D. V. Glass, London Inhabitants within the Walls, 1695 (London Record Society volume 2, 1966).
  • Edward Higgs, Making sense of the census: the manuscript returns for England and Wales,1801–1901 (London: Public Record Office, 1989).
  • P. E. Jones and A. V. Judges, ‘London population in the late seventeenth century’, Economic History Review, 6 (1935), 45-63.
  • John Landers, Death and the Metropolis: Studies in the Demographic History of London, 1670-1830 (Cambridge, CUP, 1993).
  • Peter Razzell and Christine Spence, ‘The history of infant, child and adult mortality in London, 1550-1850’, The London Journal, vol.32, no.3, Nov. 2007, pp. 271-292.
  • L. D. Schwarz, London in the Age of Industrialisation (Cambridge, CUP, 1992), ch.5.
  • K. Schurer and T. Arkell eds. Surveying the people: the interpretation and use of document sources for the study of population in the later seventeenth century (Local Population Studies 1992)
  • Richard Wall, Matthew Woollard and Beatrice Moring, Census schedules and listings,1801—1831: an introduction and guide, (Department of History, University of Essex, 2004).


Footnotes

1 M. Dorothy George, London Life in the Eighteenth Century, (London: Peregrine Books, 2nd edn, 1965), p.398.

2 Landers, Death and the Metropolis, p.166.

3 The relevant 1801 figures were taken from Enumeration abstract, 1801, p. 12; and The Abstract of the Answers and Returns to the Population Act, 41, Geo. III. 1800, pp.210-215.

4 For a detailed discussion of the marriage duty assessments and their surcharge categories, see D. V. Glass, London Inhabitants within the Walls, 1695 (London Record Society volume 2, 1966) pp. ix-xviii, available from British History Online; and T. Arkell, ‘An Examination of the Poll Taxes of the Later Seventeenth Century, the Marriage Duty Act and Gregory King’, in Surveying the People: The Interpretation and Use of Document Sources for the Study of Population in the Later Seventeenth Century, edited by Kevin Schürer and Tom Arkell (Local Population Studies 1992) pp.142-177.

5 P. E. Jones and A. V. Judges, ‘London population in the late seventeenth century’, Economic History Review, 6 (1935), pp.45-63; J. Alexander, ‘The City revealed: an analysis of the 1692 Poll Tax and the 1693 4s. Aid in London’, in Schurer and Arkell, Surveying the People, pp.181-200.

6 41 Geo. III, c.15.

7 9 Anne c. 17.