Country Survey: England and Wales

(A) THE GEOGRAPHICAL AREA TO BE COVERED

A. England and Wales

A. England and Wales are currently two of the component nations of the United Kingdom, although Wales has limited devolved powers as compared to Scotland. Even prior to the recent devolution of powers, Scottish local government differed from that for England, there was a separate census office in Edinburgh and so on; none of this was true for Wales.

--

England existed as a nation-state since before the Norman Conquest of 1066. During the medieval period its borders with Wales were an area of considerable conflict, but the London government has now controlled Wales for many centuries. This means that the only land boundary is with Scotland, and this too has seen few changes for several centuries. While the country has been politically stable over a very long period, industrialisation and urbanisation were major forces for change from the start of the 19^th century onwards. The growth of large towns necessitated major changes to local government from the 1830s onwards.

(B) ADMINISTRATIVE GEOGRAPHY

Despite or perhaps because of its political stability, England has repeatedly revised its administrative geography: the system was completely changed in 1974, and further major revisions were made in the mid-1990s. The geography used for the last census in 1991 was as follows:

NB the enumeration district is a purely statistical unit with no administrative function, and the main role of wards is as electoral units for county and district elections. In rural areas, wards are identical to 'Civil Parishes', which used to have significant administrative functions but are now mostly inactive. Since 1991, some of the 55 counties have been replaced by 'Unitary Authorities' which combine the powers of counties and districts. There are currently in England and Wales:

• 34 English Counties continue to operate on the former model, divided into 238 districts.
• England also contains 47 'Shire Unitary Authorities'.
• The six Metropolitan Counties (West Midlands, Merseyside, Greater Manchester, South Yorkshire, West Yorkshire and Tyne & Wear) are governed by 36 'Metropolitan Authorities', with unitary powers.
• London is governed by 33 London Boroughs, plus the Greater London Authority created this year.
• Wales is entirely governed by 22 unitary authorities, plus the Welsh Assembly in Cardiff.

More information can be found at http://www.lga.gov.uk/lga/the_lga/Structure2.pdf and at http://census.ac.uk/cdu/Datasets/1991_Census_datasets/Area_Stats/General_Topics/Geography

The above system of counties and districts dates back to 1974.

Before 1974, we can usefully identify three earlier administrative geographies:

• Ancient Counties and Hundreds: The origins of this system is lost in the mists of time; it certainly dates from a time when government was far smaller than today. The country was divided, traditionally, into 55 counties (including the three Ridings of Yorkshire) and these were in turn divided into approaching a thousand 'Hundreds', so called because a very long time ago they contained a hundred households; in some parts of the country, these units were called 'lathes', 'wappentakes' or 'wards'. Towns with legal status existed as 'Boroughs' at the same level as Hundreds. These units were employed by the first four censuses, from 1801 to 1831. The following geography applied in 1801; note that the majority (9,639) of the individual 'places' listed were described as being parishes, but 4,128 were townships, sub-units of the large parishes often found in the north of England and other units like chapelries and even just 'places' were listed:

• Poor Law Unions/Registration Districts (PLUs/RDs) (and sub-Districts): Poor Law Unions were created by the Poor Law Amendment Act of 1834, and England and Wales was divided into about 630 unions by commissioners on the basis of market towns and their spheres of interest. When civil registration of births, marriages and deaths was introduced in 1837, the Registrar General used almost the same set of areas as Registration Districts (our work records the differences) but also sub-divided them into about 2,000 Registration sub-Districts. Unions and Registration Districts declined in importance in the late 19^th century, but disappeared only in 1930. The following figures come from the 1881 census:

• Local Government Districts (LGDs): Starting in the 1870s, but especially as a result of the Local Government Acts of 1888 and 1894, approximately 1,500 LGDs were formed. These were also aggregations of parishes and were usually formed by designating urban parishes as County or Municipal Boroughs, or Urban Districts, and designating the remainder of each Poor Law Union as a Rural District. The LGD system was extensively revised in the mid-1930s and abolished by the 1972 Local Government Act. In 1951, the census geography was:

This is something to be discussed in Florence, but there are clearly possible misunderstandings. A very important point about England is that the parish is not a purely religious unit: since the late 19^th century, the 'Civil Parish' has been part of the structure of local government; and before the 19^th century, the parish was a unit of 'the Church of England', a state church. For example, the only way to be legally married was in one of their churches, and the early censuses were mainly conducted by their priests. How far can English parishes be identified with 'communes'.?

(C) POPULATION CENSUSES

The first census was carried out in 1801. However, the early censuses were generally completed by local priests from their personal knowledge of the area. The first census to conduct a detailed house-to-house enumeration was in 1841, generally employing the newly established Poor Law administrators.

Since 1801, regular censuses have been carried out every ten years, with the exception of 1941 due to World War II. In general, successive censuses have gathered steadily more information. For example, information about housing conditions was first gathered in 1891, and information about unemployment in 1931. There are some interesting one-off reports; for example, in 1851 there was a separate census of religious attendance while the 1911 census paid great attention to fertility, including information on how many children women had had, and whether they had died between birth and 1911.

The published reports use broadly the units already described: the early censuses, up to 1841, provide head counts for counties, hundreds and parishes; between 1851 and 1911, the most important reporting units were c. 630 Registration Districts, which were aggregated up to 55 Registration Counties, which differed significantly from both Ancient and Administrative Counties; and from the late 19^th century, and especially from 1911, Registration units were replaced by the new geography of Local Government Districts. All census reports tabulated populations but little else for the individual parishes.

NB the above is of course a simplification. The 1871 census included tabulations for the following geographical areas: Counties; Parliamentary Divisions and Parliamentary Boroughs; Hides, Tythings, Hundreds, Wapentakes, Wards, etc.; Lieutenancy Sub-Divisions; Petty Sessional Divisions; Police Divisions; Highway Districts; Local Board Districts; Boroughs and Towns with Improvement Commissioners under Local Acts; Civil Parishes and Townships, and Extra-Parochial Places; Military Districts and Sub-Districts; Post Office Districts; Inland Revenue Districts; Poor Law Unions; Registration Districts and Sub-districts; Census Enumeration Districts. However, most of these tables were simple population head counts.

The individual replies to the censuses from 1841 onwards have been preserved in 'Enumerators' Books', held by the Public Record Office. However, they are only made available after 100 years, so the most recent we have access to are those for 1891; copies of the Enumerators' Books are widely used by family historians, and most public libraries have microfilm copies. A machine-readable version of the whole 1881 census was created by the Genealogical Society of Utah, and a copy is held by the Essex Data Archive; this includes locations down to parish-level (and indeed down to individual buildings) and one of the main justifications of the Great Britain Historical GIS Project was to create a parish-level base map for 1881 data. The Public Record Office are computerising the 1901 census in preparation for its release in 2002, and the Data Archive have negotiated access to this. They also hold partial transcriptions for other years.

The first census to be computerised when it was taken was 1961. The magnetic tapes have survived, but have not been used for a long time; a number of researchers are interested in bringing this material back to life but confidentiality is a major problem.

There is a great deal of data from the last three censuses, for 1971, 1981 and 1991, available in unpublished but machine-readable form, with well-established access routes for academic research. Information is available in three principal forms:

• Area Statistics: The Small Area and Local Base Statistics (collectively called the area statistics) consist of approximately 20,000 crosstabulated counts and constitute the richest source of 100% sample socio-economic data collected in the UK. They contain information about households and individuals covering topics as diverse as age, gender, occupation, qualifications, ethnicity, social class, employment, family structure, amenities and tenure. Small Area Statistics (SAS) are available for the 113,465 enumeration districts.
• Samples of Anonymised Records (SARS): Microdata from the 1991 Census is available as two separate files: The 2% Individual SAR covers 1.1 million individuals, located down to 278 SAR areas. They cover a full range of census topics on individuals, plus summary information about households. The 1% Household SAR contains 216 thousand households, a total of c. 1/2m. people. It permits linkage between household and family members, but to preserve confidentiality they are only located in terms of large 'Standard Regions.
• Longitudinal Study: This dataset contains linked information for about 1/2m. people, selected by date of birth, from the 1971, 1981 and 1991 censuses, plus vital registration information. It includes detailed information on locations and can be used to study migration. Because these individuals are inevitably identifiable, access is tightly controlled: researchers request specific cross-tabulations, which are then approved by a committee.

More information on the modern data is available from http://census.ac.uk/datasets/descriptions.htm.

Wrigley, E.A. (ed.), Nineteenth Century Society: Essays in the use of quantitative methods for the study of social data (Cambridge: CUP, 1972). Despite its title, this is almost entirely concerned with the census.

Office of Population Censuses and Surveys, Guide to Census Reports: Great Britain 1901-1966 (London: Her Majesty's Stationery Office, 1977). A definitive official guide.

(D) VITAL REGISTRATION

Parish priests were required to register births, marriages and deaths from the 1540s onwards, and the resulting parish registers have been the basis for extensive research by the Cambridge Group for the History of Population and Social Structure. However, information from parish registers was not systematically abstracted at the time, and many registers have disappeared. Aggregate national demographic time series are therefore problematic.

Compulsory civil registration (i.e. non-ecclesiastical) was introduced in 1937, via a new network of Superintendent Registrars and sub-Registrars around the country, controlled by the Registrar General in London.

See above. The network of Superintendent Registrars, and the post of Registrar General, still exist although the system is of course computerised.

Civil Registration was based on c. 630 Registration Districts, which in general were identical to Poor Law Unions. Some information was reported for the c. 2,000 Registration sub-Districts. The actual Certificates of Birth, Marriage and Death in fact record street addresses, but these are not systematically available for research (copies of individual certificates can be bought, sight unseen, for £5.50 each, but this is rarely a practical basis for research!). Data on individual deaths, located by postcode (i.e. almost to the individual house), is available in computerised form for the last 20 years.

(E) TAXATION RECORDS

The following are the main sets of taxation records providing national, generally village-by-village, coverage. There are of course many other records of taxes only gathered in particular places, such as ports.

• 1291 -- the Pope Nicholas Taxation (the Taxatio): This lists the tax paid by each parish in the country and provides the first comprehensive listing of parishes, including the dedication of the parish church, the deanery, archdeaconry and diocese. Published by the Record Commissioners in the early 19th century and now being computerised at the University of Manchester. We have already made one fairly successful attempt to map Taxatio data for Cambridgeshire.
• 1327, 1332, 1334 -- the Lay Subsidies: The data have been collected by Dr. Robin Glasscock of St. John's College Cambridge and converted into a database at Queen's Belfast. The database contains summary tax information (numbers of taxpayers, total tax paid etc) for over 13,000 tax paying vills. These data could be linked with the data on ecclesiastical wealth contained in the earlier Taxatio to reveal the economic geography of England on the eve of the disaster of the Black Death.
• 1341-2 -- The Nonarum Inquisitiones: Provides partial coverage of the whole country. A tax on output of corn, wool and lambs, permitting estimates of national agricultural output on the very eve of the Black Death. Currently being converted into a database at Queen's Belfast.
• 1377 -- The Poll Tax: Survival of the detailed listings of adult taxpayers constitute a 'census' of adults and are the single most important national source for the estimation of total population and population distribution, primarily by county, of medieval England. Being researched in Cambridge.
• 1440s-1490s -- Reliefs from Lay Subsidies: A record, by parishes and hundreds, of tax reductions due to poverty and distress.
• 1524-5, 1544 -- Tudor subsidies (taxes): There is little material of national significance between the 1370s and the Tudor subsidies of the early 16th century, most notably those of 1524 and 1525. Researched by Dr.John Sheail and Professor Richard Hoyle.
• 1662-6, 1669-74 -- The Hearth Taxes: A comprehensive survey of population, wealth and socio-economic status, by parish. Being researched and published, county by county, by a project directed by Prof. Margaret Spufford of the Roehampton Institute, London.

See above. NB many of the above are more geographically detailed than C19 census data, providing parish-level information.

(F) OTHER MAJOR HISTORICAL SOURCES SUITABLE FOR MAPPING

Apart from the above taxation data, major pre-census statistical data sets include:

• 1086 -- The Domesday Book: England’s most famous public record and one of the most remarkable sources of statistical information ever created. A full computerised transcript has been created by John Palmer and George Slater at Hull University. The remaining substantial task is to relate the places listed to points and areas on the ground to create a body of information which relates to modern communities. This is a very large but also a very difficult source, and this task may be impossible.
• 1279 -- The Hundred Rolls: a more detailed attempt to repeat Domesday but never completed. What survives covers a broad stretch of country from Warwickshire in the W. to Suffolk in the E. Now being worked on at the University of Sheffield.
• 1563 -- The Diocesan Returns (Numbers of households listed by parishes) and 1603 -- Diocesan Returns of Communicants (Numbers of persons receiving communion, by parish), both being edited by Prof. David Palliser at Leeds University.
• 1676 -- The Compton Census: A record of religious observance, by parish, identifying catholics ('Papists'), conformists and non-conformists. Studied by Marie Rowlands (Wolverhampton) and Keith Snell (Leicester).

Post-1800 sources are innumerable, but one in particular should perhaps be mentioned:

• 1832-1997 -- Parliamentary Election Statistics: The official record of elections contained only the names of candidates, not their parties, until relatively recently. Party labels have been researched back to 1885; before that, party affiliations were weaker than today and many elections were not contested. Obviously, mapping this source requires research into the boundaries of Parliamentary Constituencies, which were built up from parishes but were often quite different from local government districts.

In addition to statistical datasets, other material which may be worked with in a GIS framework includes:

• Bibliographic databases as a source for book history: Vast bibliographic databases have been assembled and computerised, and of course they always include a date and a place of publication. While the modern publishing industry tends to be dominated by national capitals, the early history of publishing was more local. Book historians wish to apply GIS techniques.
• Records of Early English Drama (REED): This project, based in Toronto and with large scale funding (c. $C 15m. over 20 years) has been assembling every reference to dramatic performances in England from the pre-Shakespeare period; this material has been traditionally published in a series of county volumes. The sources used are manorial account books and similar, not newspaper reviews of plays, so the data are mostly limited to places and dates where performances occurred: inerpretation needs mapping, and perhaps linkage to other data.
• Thesaurus and Atlas of Saint's Cults (TASC): This international project is concerned with the Saint's dedications of churches; with the spread of the cults of particular saints (across Europe as a whole, not just England), and the relationship between saints and types of location, such as that between John the Baptist and woodland areas.

(G) MODERN GEOGRAPHICAL FRAMEWORKS

The national mapping agency, the Ordnance Survey, has been converting its mapping into digital form since the 1970s, and this includes boundary lines. However, their digital data is generally expensive. Academic research is usually based on the Digital Boundary Data available at no charge from UKBORDERS (see below). They have very detailed digital maps of the census geographies for 1981 and 1991, but not for earlier dates.

The 1981 and 1991 Digital Boundary Data (DBDs) show Enumeration Districts (not themselves administrative units), Wards/Parishes, Districts and Counties.

The 1981 and 1991 DBDs are available, on-line and at no charge, from UKBORDERS in Edinburgh, to registered members of UK Higher Education Institutions. There may be informal reciprocal arrangements with academics in other countries, but NB the UK Ordnance Survey guards its copyrights zealously.

Until the mid-19^th century, parish boundaries were technically under the control of the church -- but the real answer seems to be that there was no clear mechanism for changing boundaries, or for that matter formally recording them. In the early 19^th century, the Ordnance Survey began systematically mapping boundaries, in the same period that population growth and the consequent expansion of towns was leading to pressure for change. From the late 19^th century, the Local Government Board had formal responsibility for boundaries, although changes involved statutory orders 'laid before Parliament' (but seldom debated). In the 20^th century, responsibility has been devolved to a standing Local Government Boundary Commission.

The Ordnance Survey maintained a detailed record of changes. This information was also supplied to the Registrar General, and abstracted in the census reports. For a definitive account, see J.R.S. Booth, Public Boundaries and Ordnance Survey 1840-1980 (Ordnance Survey, Southampton, 1980).

The statutory orders must be the definitive but very bulky source. The Ordnance Survey boundary books have also been preserved, in the Public Record Office, but are hand-written and very confused. Our own research has relied on the published abstract in the census reports, the Registrar General's Annual Reports and the Annual Reports of the Local Government Board. These date changes to the precise day, but provide limited information about the area transferred, unless it was a whole parish being moved between districts/counties. However, we have supplemented this information with many sets of maps for different dates.

The UK Ordnance Survey was created in 1791, with initial responsibility for mapping the whole of Britain and Ireland. It remains in existence.

Sporadic attempts to record administrative boundaries were made by map makers in the late eighteenth and early nineteenth centuries, but even the earliest Ordnance Survey (OS) maps, published from 1801 onwards, did not include boundaries. It was only with the Survey Act of 1841 (4&5Vict. c.30) that the OS began to systematically record administrative boundaries. This process took over half a century and was finally completed in 1893 when Yorkshire and Lancashire were re-surveyed at 25 inches to the mile. The definitive boundaries were published on maps at six inch and 25 inch to the mile, and appeared in simplified form on smaller scale maps

We can and have written whole papers on this, but some major series of maps are:

• The earliest systematic series of maps we have located are a pair of matching volumes, one discovered in the St. Catharine’s House library of the Office of National Statistics, the other in the Royal Geographical Society map room, showing firstly Wales, south west England and the midlands, and secondly southern England and East Anglia. They take the first Series OS One Inch-to-the-Mile maps as a base, and add parish boundaries as hand-drawn coloured lines. The watermarks and topography in the base maps indicate that they were probably published in the late 1830s, and the boundaries of Poor Law Unions and a rough approximation of parish boundaries have been added in coloured ink. Titled ‘Ordnance Survey Divided into Unions’, these volumes were probably compiled by the Poor Law Commissioners as reference material for the Registrar General, either when establishing the RDs or as a framework for the 1841 Census.
• Our main source for parish boundaries pre-1888, and especially for the detached portions eliminated by the Divided Parishes Acts of 1876 and 1882, were two series of unpublished maps in the Public Record Office, class RG.18. Firstly, pieces RG.18/198-829 comprise "Reference Maps of Registrar’s Districts", covering all districts in England and Wales for 1891. These maps were based on whatever OS One Inch maps were available, so they are a mixture of different editions The Registrar General’s staff created 631 separate sheets by cutting out the area of each Registration District and attaching it to a new linen backing sheet, often in segments, and then hand drawing additional information: the outline of each Registration District was highlighted in red with the boundaries of sub-districts added in pink and civil parishes in blue.
• A series of ‘Diagrams of Sanitary Districts’ was published in the late 1880s at four miles to the inch scale (1:253,440), showing Boroughs, Sanitary Districts, Unions, and Civil Parishes (with larger scale inserts of complicated areas such as small urban parishes) but their small scale is problematic. The base maps employ a different prime meridian for each county, greatly complicating assembly of a national GIS. These maps were H.C. Darby's main source for the maps in his Domesday Geography.
• In 1900, the Ordnance Survey published the first edition of the County Administrative Diagrams, at two miles-to-the-inch scale. These show the boundaries of parishes, Poor Law Unions and local government districts. The second edition of these maps, published 1906-10, were used by us as the main base map for our GIS, and subsequent editions have appeared at approximately ten year intervals up to the present.

There is a very limited amount of material in the Ordnance Survey files at the Public Record Office, but they are very hard to interpret. Of course, there must be a vast amount of material in various land charters, deeds and other legal documents, but systematic national research from such sources is clearly impractical.

(I) KNOWN PROJECTS

Our own Great Britain Historical GIS Project is by far the largest such project.

A number of other academic projects have constructed digital base maps, mainly of C19 Registration Districts, but without significant archival research. Similarly, a number of county atlases have been produced over the last ten years, including many maps using parish boundaries as a base. However, our enquiries suggest that these are not based on detailed research and the boundaries used are often much later than the data plotted.

The Victoria County History has been in preparation for over a hundred years. Based at the Institute of Historical Research in London, they have teams in many counties whose work includes researching boundaries.

Individual local researchers have made a significant contribution. For example, Paul Laxton of Liverpool University has researcher and printed (but not published) a map of Cheshire parish boundaries in staggering detail.

(J) ASSOCIATED METADATA (Gazetters, etc)

The English Place Name Society is researching and publishing, rather slowly, county volumes tracing the derivation of place names.

Many gazetteers have been published for England and Wales over the last two centuries, and can be consulted for names no longer in general use. However, as far as I am aware there is no systematic guide to placenames that have fallen out of use.

The established source for English administrative units, including Ancient Parishes and Hundreds, ecclesiastical hierarchies (parishes, deaneries, arch-deaconries, etc) and Poor Law Unions is F. Youngs' Local Administrative Units of England (Royal Historical Society, London, 1979 & 1991).

An equivalent for Wales is: Melville Richards, Welsh Administrative and Territorial Units (University of Wales Press, 1969).

© Humphrey Southall (Portsmouth, May 2000)

Go Back