Ancestral Hide-and-Seek in the Roaring Twenties

Where were YOUR ancestors on the evening of Sunday 19 June, 1921? It’s not long to go now before you can find out. Excitement is building in the genealogy world as one of the most hotly-anticipated record releases in a long while edges closer to public view, promising to unfold countless family history stories across the nation and beyond.

Due to arrive early next year, the 1921 Census of England & Wales was originally slated for digital release in January 2022. FindMyPast, the commercial partner handling the release, have since advised that slight delays may occur due to disruption arising from the global Covid pandemic. By contrast, the separate 1921 Census of Scotland is due to appear on ScotlandsPeople later this year. I’ll focus on the England & Wales side of things in this article.

Forewarned, as they say, is forearmed: even though we can’t view the actual returns yet, we can understand a fair bit about the 1921 Census of England & Wales from public documents which tell us how it was planned, or which report on its findings. Today I’m going to identify some of its major components and highlight potential stumbling blocks to bear in mind, ready for when we finally see those coveted returns.

I’ll explain how your ancestors’ holidays, their marital status and their concerns about privacy made their mark on the 1921 returns – along with a decades-old incident which may affect legibility in some enumeration districts. And for the mapping aficionados amongst you, we’ll also take a quick look at how the occupational information in the 1921 census can help you to understand the day-to-day address networks of your ancestors’ lives.

First, let’s familiarise ourselves with what the household returns from the 1921 Census actually look like. The image above provides an annotated example (clicking on this image will allow you to see it full-screen), but I’d recommend downloading a zoomable image of a sample form from the Vision of Britain website to look at it properly.

What’s New?

This 1921 household return involved a few key changes compared to previous years. Let’s take a look at some of the main points.

• People’s ages were stated in years AND months, rather than the age-in-years format used throughout 1851-1911. These will give you a more accurate estimate of an ancestor’s date of birth (allowing, of course, for errors or lies!), helping you to locate them in other record sets such as civil birth registrations;
• Women undertaking unpaid domestic duties were instructed to enter Home Duties in the Occupation field of column K, rather than entering No Occupation on the return;
• Those aged 16 or over had to describe their marital status as Single, Married, Widowed or Marriage Dissolved by Divorce (more on that later!), whilst entries for children aged 15 or under had to indicate whether both their parents were living, if their father or mother had died, or if they were orphaned.
• Those in education had to state whether this was Whole Time or Part Time in Column H.
• Occupations received quite a lot of attention, including a statement of workplace address in Column M;
• Married men, widows and widowers were required to state their total number of living children and step-children in Column N and indicate their ages by entering crosses into boxes in final Column O.

So given the revised format and new data fields for the 1921 Census, and its timing, what can we expect when we finally get to see the returns for ourselves? Drawing on knowledge from government planning, the official 1921 census reports and newspaper articles from the census year and surrounding years, here are my Top Tips for planning your 1921 ancestral search…

I Spy a Holidaymaking Ancestor

Coal stoppages in 1921, a precursor to the General Strike of 1926, brought much of the country to a halt and resulted in the census being delayed. The original date of 24 April was scrapped and as the stoppages rolled on, newspapers speculated as to what the alternative date might be. Administrators were keen to avoid the July to September period, when many families employed in industry would be taking their holidays, and there was even talk of it a postponement until October. Eventually a revised date was set for 19 June.

The choice of date for any census is carefully managed, because you want to capture the population on a “normal” day, living where they usually would. The England & Wales census hadn’t been held in June since 1841, the convention shifting to March/April from 1851 onwards in order to avoid agricultural workers being away from home during the harvest-time. Although harvest-related disruption was less of a concern by 1921, a thriving summer holiday tradition loomed large.

By avoiding the bulk of the main July-September holiday period, the 1921 Census should still depict many families in their usual residence, but there is potential for early summer holidays to throw a spanner in the works. The June census weekend in 1921 enjoyed warm, settled weather and some families chose to make the most of this by going away for a few days. If initial searches don’t turn up your forebears in the town you’re expecting to find them in, examine any potential matches located in typical holiday destinations, particularly coastal resorts which were a popular choice during this time.

Take, for instance, the seaside resort of Blackpool in Lancashire, which in the 1920s was at the height of its popularity. The town experienced a suspicious surge in population in the 1921 census: where the town had recorded 62,302 residents in 1911, it claimed 101,110 in 1921. The apparent growth over 1911-21 was due to many of its 1921 census inhabitants not being permanent residents. Evidently the imposition of a June census date in 1921 found some of our forebears strolling along the promenade eating sticks of rock and getting sand in their toes, rather than at home preparing to go into work as usual the following day!

Lift the Lid on Marriage, Loss & Divorce

The impact of war and pandemic influenza upon the population demanded a focus upon slightly different marriage data by 1921. The age profile of widows had changed dramatically, and gone were the “fertility questions” of 1911, replaced by a more detailed declaration of living children and their ages, and the capacity to indicate divorced status. Unfortunately this new format did not require information about children who had died, providing the modern-day genealogist with fewer hints and clues to infant mortality.

Expect to see more widows & widowers

“There are but few questions today upon which guidance can be sought of the last Census across the great gulf of War which lies between…” The Introduction to the Preliminary Census Report of 1921 could not avoid mention of the catastrophic loss of life which had occurred in the intervening years since the last census. 886,000 fatalities sustained during the First World War produced a great many widows, and the 228,000 deaths that followed in the influenza pandemic of 1918-19 left few families untouched by grief.

The UK’s population swelled from just over 42 million in 1911 to just over 44 million in 1921. Despite this, the catastrophic losses in the intervening period created gaps in many family units: the 1921 snapshot of the population identified 1,621,758 widows, compared to 642,311 widowers. If you’re looking to prepare your research ahead of the 1921 Census, now is a good time to revisit all your 1911 family groups and write down the names of each individual. This will leave you ready to identify people who are no longer with the family unit in 1921, whether because they’ve moved on elsewhere or have died.

And remember: ten years is a long time when a war comes in between, enough for someone to be married, widowed and remarried. When working with the 1921 Census, we’ll need to remain mindful of the range of possibilities for people’s lives over that period: young widows who have tied the knot again may not be living under the surname you’re expecting (recall from my previous article Summer Lovin’ that the median age of widows at remarriage saw a substantial drop after both world wars, because the war created so many young widows who subsequently remarried).

Wartime fatalities on WW1 battlefields also led to an inevitable excess of “women of marriageable age” in the population and 1921’s official report on the census remarked that “the reduced opportunity for marriage in the case of a large number of women will be felt in an extreme degree“. The situation drew attention in the newspapers, with The Times of 30 August 1921 suggesting that English and Scottish women might consider moving abroad to “the Dominions”, including Australia and New Zealand, to procure themselves a husband. So in addition to young war widows, you may also find more young, unmarried women in your family units within the 1921 census.

Cast a critical eye over declarations of marital status

The 1921 Census was the first to indicate a person’s marital status as a divorcé(e). Newspapers reported that the D on the census form was originally meant to stand for “Divorced”, but was eventually labelled “marriage dissolved by divorce” to avoid offensive implications that the person was the party at fault. Yet where there is the potential for perceived shame or stigma, people tend not to tell the truth – so despite this careful choice of words, the social stigma of divorce in the 1920s rendered the census data on divorced status highly unreliable.

“The expected numbers [of divorced persons] might well be put at a figure twice as large as the total recorded…it appears more than probable therefore that a large number of persons failed to return the desired information.”

England & Wales Census, 1921: General Report with Appendices (1927)

Of the 37,886,699 inhabitants of England and Wales enumerated in 1921, 16,682 were recorded as having had their marriage dissolved by divorce. Yet this failed to give an accurate picture of the situation. When the 1921 returns were tabulated and the aggregate totals of divorced persons were compared with totals for the divorces known to have been granted over the preceding years, it seemed that only half of the nation’s divorcé(e)s had been truthful.

So how should this inform our research when we’re working with the 1921 Census? Don’t automatically take the entries in the marital status column at face value. Divorcé(e)s only accounted for around 0.088% of the population of England & Wales, so even the ones who didn’t declare it were relatively few in number. Nonetheless, you should remain on your guard for untruths. Entries indicating “D” are likely to be correct (those willing to make a socially difficult admission are unlikely to be fabricating), but bear in mind the possibility of dishonest statements claiming another status until you can confirm otherwise.

Prepare for errors on the “dependency” question

The combined effect of World War One and the 1918-19 influenza pandemic had claimed many lives: it was hoped that clearer statistics on financial dependency would help in administering widows’ and orphans’ pensions and other state support such as invalidity allowances and workmen’s compensation [1]. This prompted the inclusion of questions on orphanhood (column E, where children under 16 were stated as having both, one or no parents living) and the number and ages of living children and step-children – even non-resident ones – associated with a widow(er) or married man (columns N and O).

It seems that the grid format of column O, where householders had to insert crosses to indicate the age of each child, caused a lot of confusion. Indeed the official report on the census, published six years later, noted that this question “gave rise to an amount of difficulty which was bound to detract from the completeness and accuracy of the returns.” Such errors included householders simply not filling this part of the form in; including grandchildren or other children who were not meant to be counted in the grid; and faulty placement of crosses within the boxes [1].

So once you’re finally working with 1921 census returns, be aware that your ancestors may have made mistakes in this final column. Be sure to cross-reference any age data with other sources, such as civil registration records and parish registers, to identify any mistakes. And since single men weren’t required to complete the final column, illegitimate children they might have fathered won’t be declared on their return!

Watch Out for Private Returns

Previous censuses allocated one household return to each domestic unit, to be filled out by (or under instruction from) the head of the household. This could cause privacy issues where servants and lodgers were concerned however, and increased the chance of inaccurate information being entered for residents who were not part of the family.

The 1921 Census was the first to allow a resident to request a separate return form if they wished to submit their personal information without giving their details to the head of the house. If this happened, the head of the house was supposed to list the person’s forename, surname and relationship to the head in columns A and B (as indicated in my earlier schematic). If you identify someone on a 1921 return who is only detailed in these first two columns, search for their private return, which should contain the other details.

No doubt the announcement of separate, private returns was met with relief by many domestic servants – including the maid in the news extract below, who feared losing her job if she admitted her true age and place of birth to her employer.

The servant who does not wish to disclose private information to the head of the household has also been provided for. Mr. Vivian [the Registrar-General] has received a letter from a maid at Clapham, S.W., who points out that her age was more than she led her mistress to believe, and that she was born in a workhouse. “If I were to disclose this,” she wrote, “I should lose my situation.” People in cases such as this should apply to their local registrar…and they will be given a [separate] form which they should hand to the enumerator when he calls.

“Census Paper Secrets”, Daily Mail, 22 January 1921

Join Your Ancestor’s Daily Commute

“At the present time,” begins the Workplaces section of the Official Report on the 1921 Census, “in many parts of the country masses of population move in tides of daily ebb and flow.” By this time, urban industrial expansion had left many workers with a longer commute. Knowledge of typical commuting patterns had a bearing not only on the planning of transport networks, but also upon occupational health. Consequently, occupations received quite a bit of attention in the 1921 Census. If you know my love of genealogical mapping, then you might appreciate just how excited I am at the prospect of the information in Column M: the workplace address.

Both workers and retirees were meant to complete this part of the form, so it’s a golden opportunity to pinpoint either a current workplace for those of working age, or the final workplace for older members of the family. Identifying the workplace by name and address may provide a research direction to pursue in the archives. Follow up the company in local business directories and archive catalogues. Even if surviving company records don’t mention your ancestor by name, they may still create a picture of your forebear’s working life.

And if you enjoy using my Outlier Method for genealogical mapping (catch this video for details), plot 1921 residential and workplace addresses on a suitable map* alongside other historical addresses you have for an individual, to see how the 1921 places fit in with what you already know about the network of their day-to-day life. And while you’re at it, use the map to examine the streets where your ancestor lived and worked: can you work out a possible route for their daily commute?

Prepare for Soggy Reading

Consider brushing up on your palaeography skills – you may have some difficult reading on your hands! In a letter reporting on the (entirely separate) destruction of the 1931 Census in a fire at the Office of Works warehouse, W A Derrick notes that the 1921 Census – mercifully stored at a separate location from the 1931 returns – sustained water damage during its time in storage and was subsequently dried out [2,3] – a transcript of which is archived on TNA’s site**. There have been no public statements yet from FindMyPast as to the extent of the damage or the registration districts affected, but we should know more in 2022. At the very least, be prepared for reduced clarity on any water-damaged returns.

An Action Plan

So where does all of this leave us? And what can you do to prepare for FindMyPast’s release of the 1921 Census of England & Wales come early 2022?

Make an ancestral hit list

Find time to revisit your genealogy research over the coming months and take stock of where each of your lines left off in 1911. Think of this like a missing persons enquiry, only instead of the details of ‘last seen wearing…‘ you’ll want to identify each person’s name, age, occupation and address ready for your onward search.

1. Return to all the families in your tree who were alive in the 1911 census.
2. Make a list or spreadsheet containing the personal details of each individual in these family groups.
3. Check whether you have any civil registration (or other) information for them over the decade 1911-1921: this is great for identifying name changes and births/deaths of family members over that interim period and hence making a ‘best guess’ for how they might appear in the 1921 census.
4. Take a moment to consider each person and family unit in your list. Write down any good research questions for 1921 which come to mind based on the information you already have.

Once you can access the 1921 returns, the spreadsheet you’ve compiled can act as your to-do list and help you get started with your searches.

1921 Census: a Strategy Roadmap

Once you’re immersed in 1921 census research next year, remember the following:

1. Extract all ages in years and months and use these to pinpoint a D.O.B. for missing birth registrations or baptismal records (as usual, allowing for possible errors or false information).
2. If you can’t find your ancestors in the town you’re expecting, they may have relocated elsewhere, but alternatively they might be on holiday – check search matches located in holiday towns, particularly coastal resorts.
3. Found a D in the marital status column? Follow up the divorce through court reports in contemporary newspapers and file series J77 (divorce cases) at the UK National Archives.
4. Divorces were comparatively rare in 1921, but since half of the nation’s divorcé(e)s lied about their status on their census return, don’t rule out a divorce in your ancestor’s past purely because they have stated otherwise on the census.
5. Extend any genealogical mapping you have by plotting data points for your ancestors’ places of work. Analyse how this fits in with the characteristic network formed by other key locations in your ancestor’s life.

Just as the 1920s represented a new chapter in the lives of our ancestors, so will the advent of the 1921 census herald an exciting new phase in our family history research: there are stories a-plenty on the horizon. But as any seasoned researcher knows, locating an ancestor who either doesn’t want to be found or who has thwarted discovery through poor handwriting (or subsequent indexing errors) can also bring frustration for a time. Bear these pointers in mind and I hope they help you unravel some of the brick walls and mysteries which arrive when we’re finally allowed to delve into our 1921 family secrets…

Have you started planning your 1921 family history sleuthing? Is there a family mystery or a brick wall you’re hoping these census returns will help solve? Share your stories in the comments section below!

Footnotes & Reading

* I’d advise starting with the Ordnance Survey collections at the National Library of Scotland.

** My thanks to Audrey Collins of TNA for pointing me to the online transcription of the Derrick letter.

If you’d like to explore the 1921 Census further, or to follow up on any of the references I’ve mentioned in this article, here are some useful resources:

[1] General Report of the 1921 Census of England and Wales (pub. 1927). Vision of Britain website.

[2] Destruction of 1931 Census Schedules by Fire, UK National Archives, RG20/109 (date 22 December 1942). Letter from W A Garrick refers to water-damaged 1921 returns, which were not stored with the destroyed 1931 records. Typewritten copy available online via archived pages at TNA.

[3] A History of Census Taking in the UK, Oliver Duke-Williams (2017). Chapter within The Routledge Handbook of Census Resources, Methods and Applications: Unlocking the UK 2011 Census.