The Statistical Gaze

About 15 years ago​ , when I was renting my first flat in London, a man from the Office for National Statistics paid me a call. A letter had arrived a week earlier informing me that my postcode had been randomly selected for the annual General Household Survey and that I could expect a visit. My interrogator was in his mid-fifties, tall, well-spoken, and wearing a long dark overcoat which he didn’t take off. He sat down with his laptop and I made him a cup of tea. I don’t remember much else about our encounter, except that his questions seemed to go on for ever and that I was anxious to please.

The General Household Survey, or GHS, which was discontinued in 2012 as inefficient compared to other ways of assessing social shifts, was a huge dataset from which policymakers, social scientists and experts of various kinds drew conclusions about British society and how it was changing (or wasn’t). Such surveys made possible general statements about who the British were, what sorts of house they lived in, how many cigarettes they smoked and how much exercise they did. The nation in aggregate was the nation imagined. As the questions kept coming from the stranger sitting on the sofa I didn’t own, it felt important not to get anything wrong.

It’s unlikely that the subjects of the first British census, the decennial population count established in 1801, would have shared this sensation. As Roger Hutchinson tells us in his very lively account, the early 19th-century equivalents of my visitor were an amateurish assortment of parish overseers, churchwardens, constables and schoolmasters who trudged through fields, knocked on doors and filled in paper schedules by hand in exchange for a shilling and sixpence. In 1801, these census enumerators were required only to record the number of men, women and children in each dwelling, and to classify each inhabitant as occupied in ‘agriculture’,‘trade, manufactures or handicraft’ or ‘not comprised in any of the preceding Classes’. Even this relatively simple task appeared to be beyond many local officials. John Rickman, who, following the lead of France and the United States, had dreamed up the idea of a nationwide head count and persuaded Parliament to back it, described the enumerators as men ‘who answer plain questions with much sincerity, but to whom difficult questions cannot be propounded without incurring the risk of retrograding instead of advancing in knowledge’.

Consequently, there were plenty of local census returns which never reached the Cockpit, the gloriously named offices in Westminster from which Rickman directed the exercise with a team of clerks. The final report, an intimidating document six hundred pages long, published in December 1801, contained large blanks: returns were incomplete in Somerset, Monmouthshire and Buckinghamshire, to name just a few, while Ireland was omitted completely. The inaccuracies continued in later counts, fuelling scepticism. When Rickman announced in 1821 that the population had surged from ten to fourteen million over the previous two decades, William Cobbett responded that ‘a man that can suck that in will believe, literally believe, that the moon is made of green cheese.’ As late as 1931, it was discovered that one enumerator, committed to a mental hospital, had prepared fictitious returns in order to avoid, in the registrar-general’s phrase, ‘the difficulty of obtaining them from the public’.

Despite their many imperfections, the early census counts provided a far more robust statistical picture than was previously possible and, by revealing the scale and pace of population growth, Rickman’s tables confounded contemporary assumptions about the relationship between demography and prosperity. Population was a controversial subject. The radical philosopher Richard Price had been convinced in the 1770s that Britain’s population was shrinking fast, with dire consequences for national wellbeing. Thomas Malthus was equally convinced that the opposite was true, making that case in his Essay on the Principle of Population in 1798. Malthus was disturbed by the possibility of unchecked growth because he believed it was a natural tendency for population to outstrip food supply, leading to famine or plague. When the harvest failed in the summer of 1800, some wondered whether Britain was indeed falling into the Malthusian trap. Rickman’s proposal for a national census, which he floated while editing the Commercial, Agricultural and Manufacturers Magazine, could not have come at a better time.

The results of the 1801 count, which put the population of England, Scotland and Wales at a healthy 10.9 million, discredited the claims of declinists like Price, who had gloomily estimated a figure less than half that size three decades earlier. The evidence of sustained population growth, as confirmed by census counts in 1821, 1831 and 1841, also appeared to lay the Malthusian ghost to rest. By mid-century, the population had almost doubled but British ingenuity had found a way not only to feed those extra mouths but to build a still expanding global empire and an advanced industrial economy which was the envy of the world.

Overseen from 1841 by the General Register Office (one of the antecedents of the Office for National Statistics), the census quickly became vital to the expanding machinery of the modern nation-state. In certain respects, it exemplified the new forms of bureaucracy that sprang up in the 19th century to govern what the historian James Vernon described as a ‘society of strangers’ in his study Distant Strangers: How Britain Became Modern (2014). Tracking population growth, urbanisation and economic change through regular counts was a means of rendering those phenomena knowable and susceptible to intervention. New kinds of data allowed men like William Farr, who was responsible for the General Register Office’s collection of medical statistics, to develop modern epidemiology, tracking death and disease and devising public health measures in response. Census figures took their place alongside the statistics recording births, marriages and deaths, which were also collected by the GRO, and the information produced by royal commissions and public inquiries on everything from literacy and crime rates to child labour and church attendance. ‘The command of data is the one circumstance which separates our legislation from the legislation on crude or mistaken principles which even great men were compelled to accept in former times,’ the Times declared in 1850, anticipating the publication of the 1851 count. For the Victorians, social improvement could scarcely be imagined without the assistance of what the Statistical Society of London called ‘facts which can be stated numerically and arranged in tables’.

Through their regularity and scale, the decennial counts, like the new arterial roads and railway lines, the penny post and the ordnance survey, shrank space and made the nation legible in new ways. In the pages of the census, according to the Times, lay the ‘foundation of a more complete self-knowledge’. This belief – that Britons learned who they were by subjecting themselves to the statistical gaze – was widely accepted in the Victorian age. And the feeling that numbers cannot – and therefore must not – lie accounted for some of my anxiety as I sat in the presence of the man from the ONS in the early 21st century.

Naturally,​ not everyone embraced the idea of the survey with enthusiasm. An early census bill in 1753 encountered vigorous opposition from MPs who denounced it as an encroachment on the liberties of the freeborn Englishman. William Thornton, the MP for York, announced that any busybody attempting to enumerate him would receive ‘the discipline of the horse-pond’. Objections that the census violated privacy periodically resurfaced, while ‘racy encounters between enumerator and occupant’, as one local official put it in 1891, remained a source of anxiety well into the 20th century.

In some places, these confrontations were deeply political. In Ireland, which was counted separately from 1821, the census was not regarded as a merely bureaucratic exercise. The predominantly Catholic lower classes were mistrustful of the largely Protestant magistrates tasked with making local returns, and the latter frequently met with resistance and non-cooperation as they made their house calls. The recruitment of Catholic priests as enumerators eased some of the tensions, but many nationalists still saw the census as the meddling of an occupying state.

In 1911, several hundred women entered their occupation as ‘domestic slave’ in a co-ordinated protest against the laws which denied them full civil and political rights. That same year Emmeline Pankhurst, leader of the militant Women’s Social and Political Union, launched an even more audacious attempt at disruption. Thousands of suffragettes, following her lead, refused to complete or spoiled their forms, while others vacated their homes on census night to evade enumeration. As Pankhurst explained to readers of the Times, ‘the census is a numbering of the people. Until women count as people for the purpose of representation in the councils of the nation as well as for purposes of taxation, we shall refuse to be numbered.’ Subversion of the census became fashionable again at the end of the century. In 1991, opponents of the poll tax took part in a mass boycott, resulting in that year’s census undercounting by roughly a million people. Ten years later, the addition of a question on religious belief prompted nearly 400,000 citizens to declare themselves ‘Jedi’.

In Britain’s imperial territories census-minded officials found themselves thwarted in different ways and no complete statistical abstract of the British Empire was ever achieved. Some populations, it turned out, were less legible than others. Attempts to classify accurately the indigenous peoples of India or Africa were defeated by the complexity of community structures, tribal norms and caste identities. These categories proved hopelessly incommensurable when Whitehall officials embarked on an epic work of synthesis, the never to be repeated Census of the British Empire of 1901. Costly, cumbersome, and with no agreed rules in operation across multiple imperial bureaucracies, the exercise exposed the limits of colonial knowledge.

The difficulty of classifying populations overseas throws into relief an important feature of the project at home. The census sought to abstract the nation, but relied on social relations that were local and particular. From the outset, Rickman envisaged that the people knocking at the door would be familiar, men of standing in their communities with personal knowledge of the people they were counting. As the Irish experience confirmed, trust was essential to co-operation. Even when increasingly professionalised and under the control of the GRO, the census found ways, to quote James Vernon again, of ‘reanimating the local’. As the information collected on individuals became more detailed, census reports could tell readers more and more about conditions in a town, city or parish. By this means, the census provided information on both the nation in aggregate and the disaggregated locality.

The report for 1861, for example, lists 23,238 residents of Witney in Oxfordshire. Of that total, 3069 had not yet reached their fifth birthday, and two had achieved a century. There were 15 inhabitants classified as blind and 21 as ‘deaf and dumb’. The workhouse was home for 121 people, while the number of prisoners returned was nil. The vast majority (19,543) had been born in the county of Oxfordshire; one had been born at sea. More than a third of all adult females in the district were ‘wives’, more than five hundred were employed in domestic service and roughly the same number in agricultural labour.

Hutchinson pitches his book as ‘a history of the nation’; readers will find familiar metanarratives of industrialisation, urbanisation and migration in the 19th century, and of falling birthrates, the ruptures of war and the rise of multiculturalism in the 20th. But his wide-angle lens is often replaced by a zoom lens in close-ups reconstructing the lives that shaped and were shaped by these larger historical forces. In this concentration on particular histories Hutchinson is adopting a method embraced by millions since 2002, the year ancestry.co.uk went live and it became possible to search online for practically every surviving census return dated between 1841 and 1911. For Hutchinson, this digital revolution democratised a major historical source, transforming the census into ‘a popular delight’ available to all.

It’s not difficult to share the author’s obvious pleasure at hunting down the last snuffer-maker in England (67-year-old William Garner of Birmingham, recorded in 1851), Charlotte Brontë’s census return for the same year (occupation: ‘none’), or the family of miners in Glamorgan who as late as 1891 spoke no language other than Welsh (39-year-old Evan James and his sons John and David). But what moves us about these census records is not curious job titles or strange placenames. The vast majority of users of ancestry.co.uk are family historians, who, as Deborah Cohen observes in Family Secrets: Shame and Privacy in Modern Britain (2013), are keen to make an ‘intimate discovery’ that will offer a connection with the dead and at the same time affirm their own identity.