Why the Census is so Important (For Historians)

RT @paulisci: Live footage of Canadians hurrying to fill in the census right away. pic.twitter.com/VuJRwEpU5M

— Andrea Eidinger (@AndreaEidinger) May 5, 2016

It’s Census time! I know I was super excited to get my card in the mail, and then bitterly disappointed that I got the short-form questionnaire. Some of you (ok most of you) might be wondering what the heck is wrong with us census-lovers. In news reports on the subject, you’ve likely come across references to the first census in Canada, conducted in New France (Quebec) in the 1660s, some discussion of the abolishment of the long-form questionnaire and its subsequent return, and some about how Canadians broke the census website in their enthusiasm. But what these news stories don’t tell you is why the census is important.So this week, I’m going to talk about the history of the census in Canada, what purpose it serves, and why Canadian historians are in love with it. And I promise to not talk about numbers. 😉

Expert tip: The word “census” comes from the Latin word censere, which means “to assess.” While we mostly understand census-taking to be a modern undertaking, this is far from the reality. There are records of censuses from Ancient Egypt, Babylonia, the Roman Empire, and the Han Dynasty. One of the most famous censuses is the “Domesday Book,” which was a census undertaken in 1085 by William I after he conquered England. It has provided historians with fascinating information about life in England during the 11^th century, and many of the families mentioned in the census survive to this day.

In its most basic definition, a census is described as an official survey of the population of a given place. A government usually performs this survey and its purpose is to collect as much information as possible about the citizens of a city or country. This information is collected and analysed for the purpose of taxation and to ensure the proper distribution of resources over a geographic area. Or, in real people language, it’s basically a quiz that the government sends where you answer questions about who you are, where you live, what your profession is, etc… You are required by law to fill out this census, since it’s considered part of your responsibilities. If you don’t, people will come to your house and make you feel guilty (seriously).

Census Beginnings

The first census in North America was conducted in New France in 1666 by then Intendant (like a governor) Jean Talon. This census recorded the individuals living in New France at the time, their age, marital status, and occupation. The goal was to collect information to plan for the development of the new colony. But while the news media implies that this was the beginning of an epic census-taking tradition in Canada, most early censuses were regional, slap-dash, incomplete, and full of errors, and with little consistency between them in terms of questions asked or categories used.

This remained the case until 1871, when the first official census of Canada was conducted. The Constitution Act (previously the British North America Act), which had created the country of Canada, required that a census be taken in that year and every ten years afterwards. After 1951, this was switched to every five years, to take into account the speed at which Canada was changing.

Expert Tip: Before 1971, everyone filled out the same census. But since the population had grown so large, it was so longer feasible to collect such detailed information on every single individual. So from the 1971 census onwards, most people would receive the short-form census, which requires only the most basic of information, and a certain percentage receive the long-form, which was essentially the same as previous censuses. This technique of only having a certain percentage of the population do the long-form is called sampling.[1]

So Why do Historians Love the Census so Much?

The answer to that is simple: it provides a fascinating window into the lives of people in the past, acting as virtual time machine. Think of a census as a snapshot of a particular moment in time and space, as if you can freeze-frame history, and then take all the time you want to analyze it. That’s because a census is not simply a collection of numbers; it’s a collection of information.

The reason why most people think of the census as simply statistics is because they are unaware that there are actually two different forms of every census.

What’s the Difference between Published and Manuscript Censuses? (click to em-biggen)

The form that most people are familiar with is called the “printed census” or the “published census,” and this version appears online. The printed census is an analysis of all of the information that was collected during the census, organized in such a way that policy makers can make sense of it. This is why most people associate the census with statistics, graphs and enough numbers to make you go cross-eyed.

It’s the second form of the census that historians are interested in, the “manuscript census,” which is essentially the raw data. Before the ability to fill out the census online came about, the census had to be conducted manually. Official census takers, called enumerators, would be hired for this task and specially trained (check out the official 1921 training booklet here!). Once trained, these enumerators (official census-takers) would go from door to door across the country with a sheet of paper that listed all of the information the enumerator would be required to get, like who the head of the family was, the number and names of inhabitants at a particular address, and their ages, genders, religions, ethnicity, marital status, and occupations. The enumerator then recorded this information by hand, and once finished, moved on to the next address. The sheets of paper that contain this information together make up the “manuscript census.” It wasn’t until 1956 that the census started becoming self-reporting.

What Can the Manuscript Census Tell Us?

The manuscript censuses are crucially important for historians. Not only do they record personal information about individuals, they often capture information about individuals who do not leave historical records otherwise (like the illiterate or the poor). You can trace the history of a particular address and see who has lived there at various points in time. You can learn about the makeup of families, how many children people had, and whether or not there were differences in the makeup of families depending of ethnicity or religion. You can learn about the individuals involved in a particular occupation and get a sense of the people that would gravitate toward jobs like logging, for example. The manuscript census is like a candy shop, and historians are children. You do the math (ohhh… so punny…)

I actually found my grandfather and his family listed in the 1921 census! This gives you a great idea of just how much detailed information historians

• His father, Mendel Eidinger, was 35 years old when this census was taken.
  • He worked as a “ouvrier colporteur” (either he worked in a factory and as a peddler, or he worked as a peddler at a factory – it’s not clear).
  • He had been born in Austria, as were both of his parents, and he had emigrated to Canada in 1906.
  • He rented either a house or an apartment (I can’t tell which, due to the enumerator’s handwriting) with 5 rooms, and paid $17 per month in rent. The building he lived in was made of brick.
  • He could speak both English and French and he could read, but he could not write.
•  His mother Ethel was 28 at the time the census was taken. S
  • She was a housewife, had been born in Russia, as were both of her parents, and had emigrated to Canada in 1911.
  • She could speak English, but not French, nor could she read or write.
•  They only had two children at the time
  • Barmer (my grandfather, whom we called Bennie) and Sam, though they would eventually have nine total. Barmer was 7 while Sam was 1, and both had been born in Quebec.
  • Like their mother, both spoke only English and could not read or write, and they did not attend school.
•  Based on the entries surrounding those of my great-grandparents, they lived in a neighbourhood with mostly Jews, but also French Canadians. Isn’t that amazing?

But Wait, Does this Mean that Historians Know All About Me?

You don’t need to worry about creepy historians recording your every move, unless you happen to live with one. 😉 While the printed census is released to the public once all of the data has been analyzed, the manuscript census is held back for a certain period of time (usually 92 years), to protect the privacy of the individuals listed in the census. The idea is that everyone listed in the census will be dead by that point, so the information is no longer sensitive. At the time of this publication, manuscript censuses recorded up to and including 1921 have all been released.

Expert tip: The release of the 1906 census was actually quite controversial! The Chief Statistician of Canada refused to release it because the enumerators had been instructed that the information collected would remain private in perpetuity. This resulted in extensive debate about privacy and census records that was only resolved in 2005 with the passage of Bill S-18 An Act to Amend the Statistics Act, which stated that all censuses prior to 2006 would be released on the 92 year schedule, but that for the 2006 census and all others, consent was required from each individual regarding the release of the manuscript census after 92 years.

The Inevitability of Human Error

While censuses are great sources of information for historians, they should not be used without caution. Though it might seem that the statistics recorded in the census would be relatively straight forward, the reality is far from it. First all, it’s important to remember that censuses are conducted by humans, and humans make mistakes. Sometimes these mistakes are intentional, and sometimes they are not. The 1861 Census, for example, was undertaken in January, for some god-forsaken reason, in the middle of severe snowstorms. For instance, John Young, the enumerator for Leeds County in the 1861 Census wrote on one of the census returns:

Elizabethtown Jay. 24 1861 traveled half mile on the 11 Con. The Snow is up to the mares back must return I wont Knock the Senses out of my self and the mare for all the census in the Township into your Stable Nancy. […] The snow was so deep I could not get to the house but got the best Information I could.[2]

In other cases, the information that enumerators recorded was influenced by their personal views. This was particularly a problem for female landowners. Even when women held property in their own rights, enumerators often listed their eldest son as head of the household and went out of their way to find men to answer their questions instead of women. As one enumerator, Samuel Young, explained, women could not give accurate information:

The amount of crop raised I think is pretty cored, but the acres that produced the crop cannot be depened [sic] upon as I had to get my information a good deal from the woman which some times was not corect[sic].[3]

And don’t forget, we have no way of knowing if the people who answered the questions put to them by enumerators were being accurate or truthful.

Categories, Schmatagories

Then there is the problem of categories. A census is fundamentally an attempt to create order out of chaos. The people who create each census try to

You can see an example of the “colour” category here.  Click to em-biggen (1901 census of Canada, British Columbia, New Westminster (district 2), Richmond (subdistrict E), division 1, p. 7; LAC microfilm T-6429, item no. 163738).

come up with a list of questions and categories that will make sense of the total population. But people don’t actually work that way: Reality and life are complicated and messy. While the people who create censuses usually try to be objective, their beliefs and values influence the form that questions and possible answers take.

A well-known example comes from the 1901 census. Among the personal information required of individual was a designated “colour.” The available options were “w” for whites (Caucasians), “r” for red (Indigenous peoples), “b” for black (of African descent), “y” for yellow (of Japanese or Chinese descent). I hope I don’t need to explain why this was and is a problem. But aside from the obvious, what about the people who don’t fit neatly into categories? The instructions for this census required that enumerators list individuals who had both Caucasian and non-Caucasian ancestry as their non-Caucasian racial classification. But what about people who were partly Black and partly Asian? And how did the enumerators know if someone was white? Did they go by skin colour and just guess? Your guess is as good as mine; there is just no way of knowing.[4]

In other cases, the census categories were deliberately manipulated to provide specific answers. For example, when historians look at the 1871 Census, they need to be very careful when they look at the questions on nationality. This census was designed by the deputy minister of the Department of Agriculture, Registration, and Statistics, Joseph-Charles Taché. Taché was very invested preserving the political power of Quebec. But he worried that because French Canadians were a minority in the new country of Canada, their voices would be drowned out. So he insisted that British Canadians be categorized as Welsh, Scots, English, or Irish in origin and French Canadians as French. When the data was analysed, it looked like French Canadians were the largest “ethnic-national” group in Canada.[5]

Expert tip: It wasn’t until the 1990s that the term “Canadian” appeared as a possible answer for nationality on the census.

These kinds of problems persist to this day. The current census provides a great example of this. The 2016 Census asks respondents to list their gender, and provides you with the option to tick one of two boxes: male or female. But what about people who are intersex, gender queer, gender fluid, or transgender?

Now this might seem like I’m ragging on the census, but these are the kinds of concerns that historians raise about all historical sources. It’s tempting to think that because something is written down, printed, or published by an authority that the information contained is reliable. But people in the past are just like people today: sometimes they make mistakes, even with the best of intentions, they are always influenced by the way they see the world, and some of them have personal or political agendas. As Bruce Curtis argues, censuses are not taken, but made. However, this doesn’t diminish the value of historical sources like the census. They are still extremely valuable sources of information, but they need to be approached with caution.

So what do you think about this? Are you excited about the census because you’re a history geek like me? How have you used the census in your research or classrooms? Let me know in the comments below!

Extra Credit

• Want to go play with the census yourself? Check out the searchable databases of Canadians censuses to 1921. Just be aware that the database is only searchable by geographic location. There are also some great additional tips on how to find what you’re looking for.
• You can try searching the same censuses on ancestry.ca for specific names. Just be careful, since not all of the names listed were transcribed accurate. For example, my great-grandfather Mendel Eidinger’s entry in the 1921 census was transcribed as Mendel Edorger.
• Two important historical projects related to the census are the The Programme de recherche en démographie historique (PRDH) at UdeM, which has created a searchable database of the 1852 and 1881 censuses and made them available here, and the Canadian Century Research Infrastructure (CCRI), which is an ongoing project to create databases for the 1911, 1921, 1931, 1941, and 1951 censuses. Two of the CCRI’s team leaders are Eric Sager and Peter Baskerville, two of my former professors from UVic.
• (For a great article on the changing nature of research and pedagogy in the digital age, I recommend checking out: Eric Sager and Peter Baskerville, “Canadian Historical Research and Pedagogy: A View from the Perspective of the Canadian Century Research Infrastructure,” Canadian Historical Review 91, no. 3 (September 2010): 533-551.
• How to cite manuscript censuses from Canada that are microfilmed (Chicago Manual Style):
  • Before 1871
    • [Name Family Name] household, [year] census of [province], [city], [township], stamped page [number], line [number]; microfilm [number], Library and Archives Canada, Ottawa.
  • 1871 and after
    • [Name Family Name]  household, [year] census of Canada, [province], [district name] (district [number]), [subdistrict name) (subdistrict [letter]), division [number], p. [number], family [number]; LAC microfilm [number].

[1] To learn more about the history of the census in Canada, check out https://www12.statcan.gc.ca/census-recensement/2011/ref/about-apropos/history-histoire-eng.cfm

[2] Bruce Curtis,The Politics of Population State Formation, Statistics, and the Census of Canada, 1840-1875 (Toronto: University of Toronto Press, 2002), 212-213.

[3] Curtis, 215.

[4] http://www.bac-lac.gc.ca/eng/census/1871/Pages/about-census.aspx

[5] Curtis, 13, 250-251.

Liked this post? Please take a second to support Unwritten Histories on Patreon!