The 1841 Census Commissioners v. the 1821 and 1831 censuses

The 1841 census broke new ground in Irish census-taking. It was the first census to be taken on a specific day (6 June), with the forms distributed in advance, to be filled out by the householders. The previous three censuses had been viva voce inquiries, where the enumerator put the questions to the householder and took the details.

The Commissioners for the 1841 census were justifiably proud of their work, which succeeded in counting 8.175 million people. But they were troubled by an apparent glitch in the statistics. The 1821 census had reported a national population of 6.8 million, and the 1831 census 7.767 million, an increase of nearly 1 million people in a decade, but their census had upped the population by only 400,000 people.

They explained the difference in the decennial growth rates by arguing that the 1821 census had underestimated the population, since it was the first census undertaken, while the 1831 census had exaggerated the population, because its enumerators had been paid in proportion to the number of people counted. These comments gained widespread traction and have been commonly cited by historians for the past 180 years. However, I argue that both are invalid.

In the first instance, the 1821 census was not the first Irish census, but the second one. The first census had commenced in 1813 and although it failed to report a national population, and despite the survey not even commencing in some areas, it succeeded, nonetheless, in enumerating almost 5 million people during its tortured existence. Therefore, although the 1821 may have been the first experience of census taking for some, the majority of Irish people alive in 1821 would have been recorded in a broadly similar survey a few years earlier. This is not to argue that the 1821 was accurate – in fact, it is near certain that it was deficient – but simply to point out that the 1841 Commissioners’ comment regarding this being the first experience of a census for the population was not based on fact.

However, their argument that the 1831 census over-enumerated the population is almost certainly incorrect, but it is also unfair. In essence the 1841 Commissioners were disparaging the 1831 enumerators by intimating that many of the enumerators falsified their returns in order to boost their salaries. What is not widely appreciated, however, is that the 1831 census was subjected to a unique and unexpected verification process a few years after it was taken, which it passed with flying colours.

In 1834 the House of Commons established a Public Instruction Commission to inquire into the availability of religious instruction in Ireland. The commission was required to determine the religious breakdown in Ireland at the time, and they opted to reuse the 1831 census for this purpose. The census was copied in its entirety, and the copies were returned to the enumerators, with the request that they determine the religion of everybody recorded by the census. When this was done their work was made available for public inspection and public meetings were organised in each parish, to scrutinise their work. Any widespread fabrication of names in their census books would immediately have been exposed through this process. Yet, few returns failed this test.

There is no question that the 1841 census was ground breaking in its scope, scale and efficiency. But its criticisms of the two censuses that immediately predated it are invalid and speculative. The surviving census returns from all Irish censuses, including the 1821 and 1831 enumerations, are freely available online here: http://www.census.nationalarchives.ie/.

Image 2: An extract from the 1821 census for Castle & Spiddle Gate, Athenry parish, County Galway, showing John Taylor, aged 22. Taylor, who took the census of the parish, lists his occupation as ‘enumerator’.

Brian Gurrin: a biographical note

I’m a numbers man. When I’m examining a source my first thought usually revolves around how I might be able to summarise it with numbers. You show me a census return: I think – what’s the mean household size in this townland, or parish, or barony? Or, maybe – how many Catholics lived in Protestant households there, and how might that have differed from comparable figures elsewhere? You show me a sheet from Griffith’s Valuation: I think – what’s the mean rateable value of the land? Or what is the typical value of a house there? For me – it all comes down to numbers. Numbers can efficiently, accurately and effectively put order on almost any dataset, no matter how large it may be.

I’ve always liked numbers. My first third-level qualification was in Physics, and when I moved to historical research it was inevitable that I would be drawn towards studying census data. I argue that, of all the inquiries organised by governments, the census is the greatest, because it is the only official survey which aimed to obtain information on every single person alive, and sometimes on those recently dead too. Any other surveys, inquiries or commissions, no matter how wide their remits, operated with mere subsets of populations.

Doubtless, you’ll be aware that Ireland’s demographic experience is unique in Western Europe, characterised by very rapid growth in the two decades on either side of the Union, and later a cataclysmic, prolonged collapse, commencing in the latter half of the 1840s. The first successfully completed Irish statutory census reported a national population of 6,801,827, in 1821, and the 1841 enumeration counted 8,175,124 people, but the 1926 censuses (one was held in each of the two jurisdictions) could only report a paltry 4,228,553. For me, however, those numbers are more than raw statistics. Each individual person counted by those censuses, and by all other censuses, was a human being, whose life story was briefly abstracted in the census forms that they or their enumerators filled out. It’s a tragedy that so many of the early Irish census returns were destroyed, either by government orders before 1922, or by military action in June of that year. Our understanding of the past would be so much more nuanced if all the census data from and after 1821 was still available to us. But even employing mathematical techniques to abstract what information has survived can open up new vistas on past times, and on past lives.

In my free time I enjoy reading, running marathons and playing around with numbers.  I’m currently reading a book called Alex’s adventures in Numberland which is a great read about the simplicity and elegance of mathematics; and this year I was aiming to run the London and New York marathons, before our world was turned upside down. I won’t tell you my best marathon time, but if you want to look it up it’s the 5 digits that occur from location 35273 in the beautiful mathematical number π. Well; didn’t I say I’m a numbers man?