By David Norris
The census of England and Wales for 1891 introduced a question on language spoken for inhabitants of Wales and Monmouthshire. This followed the successful use of such a question in the Scottish census to assess the use of the native language. Perhaps surprisingly for modern observers, the inclusion of the language question was subject to debate in the House of Commons. The question used was as follows:

Language spoken

  • If able to speak English only write “English”.
  • If able to speak Welsh only write “Welsh”.
  • If able to speak both English and Welsh write “Both”.

It appears a straightforward question, however, as soon as the census was completed questions began to be asked about the reliability of the responses recorded. The question was asked in much the same format in 1901 and 1911, and the concerns continued. This article uses some very local evidence to illuminate some of the issues.

At the time of the 1901 census the land I now farm was split between two tenants. Interestingly, the land was not divided into two self-contained parcels. Instead, the fields were divided into four blocks so that each tenant had some of the better draining land and some of the wetter land.1 The first tenant was Hannah Howells, a widow, living at Tŷ Canol. The house is long gone, the only evidence left is the remains of an orchard and fruit bushes which formed part of the kitchen garden. Her answer to the language spoken question was “Welsh”. However, the Pembrokeshire Archives contain papers lodged by the solicitors of the owner’s guardian, Mrs Annie Richards.2 On 15 January 1914, Hannah Howells wrote to Mrs Richards regarding a land dispute with a neighbour. The letter is in impeccable English!

The second tenant was William Davies who lived at Bryn with his wife and his mother. Bryn was also known as Blaentail in this period, reflecting an interesting sense of humour. For all three residents of Bryn the language question was answered “Welsh”. Once again things are not straightforward. In October 1906 William Davies wrote to Mrs Richards as he was giving up the tenancy of Bryn. The two letters in the archives deal mainly with his claim for compensation for improvements made to the land and farm buildings. For the purposes of this article two matters of interest arise. Firstly, the letters show that a very friendly relationship seemed to exist between landlord and tenant. In particular, the letters refer to a social visit by Mrs Richards’ daughter. Secondly, the letter is written in excellent English. While the census returns show that Mrs Richards was bilingual, her daughter spoke English only.

These two local examples both appear to fit with concerns raised at the time regarding the reliability of Welsh only responses to the language question. The Registrar General reported on the 1891 census that “abundant evidence was received by us that it was either misunderstood or set at naught by a large number of those Welshmen who could speak both languages, and that the word “Welsh” was returned when the proper entry would have been “Both”; on the ground, it may be presumed that Welsh was the language spoken habitually or preferentially.”3 The same report also stated that “so desirous do many householders appear to have been to add to the number of monoglot Welshmen, that they returned themselves as speaking Welsh, that is Welsh only, but made similar returns as to infants who were only a few months or even a few days old.” The latter comment drew complaints from Welsh MPs and the Registrar General apologised saying he had not intended to accuse the Welsh people of untruthfulness or of a deliberate intention to make fraudulent returns. There were also suggestions that the census enumerators were too keen on the use of “ditto” when transcribing household returns into their books (which were the basis of the published census statistics in 1891 and 1901) and too often ascribed the response of the head of household to all other occupants. When in 1911 the returns completed by the head of household were used directly to compile census statistics, there was a very significant increase in the number of non-responses to the language question (from 2,757 in 1901 to 58,517 in 1911).4

The availability to researchers of head of household returns for the 1911 census allows some further light to be shed on my two local cases. Although Hannah Howells is shown as the head of household, the return was completed by her son. While she was returned as monoglot Welsh, her son was bilingual. A simple explanation for the apparent discrepancy between the census returns and the letter in the archives may be that her son wrote the letter on his mother’s behalf. The case of William Davies is more interesting. As we have seen, he had left Bryn by 1911 but he continued to farm in the parish. He completed the 1911 census return as head of household. The response to the language question remained “Welsh”. Why? His relationship with his landlord’s non-Welsh speaking daughter suggests strongly that he spoke English (even if his letters were written on his behalf which their very personal nature leads one to doubt). Furthermore, a number of years ago I was fortunate enough to meet his granddaughter who seemed sure that he could both speak and write English. Perhaps his 1911 census return provides a clue. The return also asked about the nationality of every person born in a foreign country. The enumerator put a line through William Davies’s response to this question – “Welsh”! Perhaps he was not the only Welshman to use the census return to make a wider statement.


1.    See the maps drawn up for the Duties on Land Values, Part 1 Finance (1909/10) Act 1910
2.    Pembrokeshire Archives, D/LI/159 and D/EE/R/1/6
3.    Census of England and Wales 1891 General Report, vol. iv, p.81
4.    Pryce, W.T.R. The British Census and the Welsh Language, Cambria, vol. 13, 1979