The population census

The 2001 population census was conducted on 29^th April and was presented in terms of all persons usually resident at the given addresses during that day and overnight. The usually resident population enables results to be compared with those of the previous two censuses, which have used this definition, and with reformatted data for 1971. Population data by age and council area were published on 30^th September 2002. These data enable demographic .profiles of local populations to be drawn, and provide a useful basis for the analysis of social and economic factors in areas of relevance to Gaelic language and its users. Age profiles can be drawn for the total population of Scotland as a whole, for the wider Gaidhealtachd area (Western Isles, Highland, and Argyll & Bute), and for the Western Isles separately.(See: Figures 1 - 6.) Compared with the rest of Scotland, it is quite obvious that in traditional Gaelic areas the 'young parent' age-group between 20 - 34 is quite dramatically reduced. Social and economic factors involve migration of this age-group for further and higher education and for work opportunities elsewhere. Data on Gaelic language and migration and other social and economic factors to be released later in March will provide further perspectives on this problem.

The census questions, and results to date

The General Register Office released Key Statistics on major census questions including Gaelic language (Table KS06) on 13^th February 2003. There was a simultaneous publication of The Registrar-General's Report to the Scottish Parliament. This provided more detailed and numerical data on key census questions, including Gaelic Language (RG Report Table 10). The data thus available comprise totals for population aged 3 years and over, and totals of all persons aged 3 years and over who are able to understand, speak, read and write Gaelic in Scotland as a whole and the 32 council areas.

For the first time a question was asked concerning ability to understand Gaelic. This question was posed first, followed by questions on speaking, reading and writing.

Intercensal change

Of particular interest are changes which have taken place since the previous census in 1991. Table 1: Gaelic speakers 1881 - 2001 presents information on total Gaelic speakers since a Gaelic question was first asked in the 1881 census. The picture is one of fairly consistent and rapid decline. The two instances of intercensal increases (in 1891 and 1971) were the result of changes to the census question rather than actual overall increases of speakers. The 2003 total is less than a quarter (23.1%) of the 1891 total.

However, the net loss of Gaelic population had become particularly severe by 1981-91 and was running at an average annual rate of 1,333. The decline from 65,978 speakers in 1991 to 58,652 in 2001 represents an annual rate of net speaker loss of 733. The rate of decline has considerably slowed down - and has almost halved. (55.5%) Reasons for this may lie in the improvements in provision of Gaelic-language media, and in particular in Gaelic-language education. Further census results in terms of age and other social and economic factors to be released later in March may be able to shed further light.

Geographical distribution

The data on Gaelic speakers by council area enable some broad conclusions to be drawn on the national pattern of distribution of Gaelic speakers, and some indications of trends and processes which are currently occurring. Table 4 presents a comparison of Gaelic speakers by present-day council area in 1991 and 2001. Three broad patterns can be distinguished:-

1. Traditional Gaelic areas.

   These were the areas of most substantial decline of the language. The Western Isles (Eileanan Siar) evidenced a loss of 3,823 Gaelic speakers - or 19.6% of the 1991 number. Total population dropped by 10.4% in this period. Between 1981 - 91 the most substantial losses of Gaelic speakers had been amongst young people. In 1971 and 1981 67% of children aged 3-15 spoke Gaelic. This fell to 49% in 1991. If this trend continues the proportion of Gaelic-speakers within this age-group can be expected to be substantially lower - moving to match the proportion of children in Gaelic medium units at around 25%.

   The Highland Council area also experienced substantial loss of Gaelic speakers of 2,044. This represented 13.9% of its Gaelic speakers in 1991. Over the period 1971-91 numbers of young persons aged 3 - 15 had increased at every census. It will be important to ascertain from forthcoming age-related results whether this trend has been maintained.

   Argyll & Bute experienced a loss of 709 of its Gaelic speakers. This represented a proportional loss of 14.5% over the decade 1991 - 2001. The situation is similar to that of Highland. It will be important to see from forthcoming age-related figures whether the development of Gaelic-medium education since 1991 has produced any growth of Gaelic speakers amongst its young people.

2. City and Suburban areas.

   The following areas all evidenced small increases in their Gaelic-speaking populations: Aberdeen City, Angus, Clackmannanshire, Dundee City, East Lothian, East Renfrewshire, Falkirk, Midlothian, Perth & Kinross, Stirling, and West Lothian.

   These areas are for the most part city and suburban. They represent destinations of migrants from the traditional Gaelic areas, and of city Gaels who are geographically and socially mobile in moving into suburban relocations. These areas have also developed Gaelic-medium education, and there may even have been some genuine growth of new Gaelic speakers amongst young people. Again, the release of age-related data later in March may help to ascertain whether or not this has been the case.

3. Older industrial areas and other rural areas.

   These areas can be said to include Glasgow City and other old industrial areas on Clydeside and in the West of Scotland. These were once economic attractors which drew many Gaelic-speakers in search of their economic future. Today with industrial decline, many of these areas suffer both overall population loss, and with it what Gaelic speakers they had.

Migration of Gaelic-speakers across Scotland

This has been historically a constant process over time. Economic adversity in Gaelic 'homeland' areas and the 'pull' factor of economic opportunity in urban, industrial areas have produced a Gaelic presence in very part of Scotland. The results of this process are illustrated in Table 1 and Figure 5.

In 1881 87.8% of all Scotland's Gaelic speakers were located in the 'Highland Counties' (covering the mainland Highlands, Hebrides and Clyde Islands), and some 12.2% of all Gaelic speakers lived in the remainder of Scotland (the 'Lowland Counties', and Northern Isles). Today that Highland / Lowland split is approaching the 50/50 mark: with 55.5% of all Gaelic speakers usually resident in the wider Gaidhealtachd of the Western Isles, Highland and Argyll & Bute council areas, and 44.5% in the rest of Scotland. This has important implications for Gaelic policies. If these are conceived as mainly or only directed towards Gaelic speakers and communities in the Highlands and Islands, they will fail to reach or benefit almost half of all Gaelic speakers.

From percentages in Key Statistics Table KS06, it can be estimated that some 10,000 of a total of over 92,000 persons understanding, speaking, reading, and writing Gaelic were born outwith Scotland. This may give some indication of movement between the home population of Gaelic speakers and return from the Gaelic diaspora. The total may also include incomers who have effectively learned the language.

Other Gaelic language abilities

For the first time the census asked a question regarding ability to understand spoken Gaelic. Together with abilities to speak, read and write Gaelic this produced a total of 92,396 - or over half as many again, compared with the total for speakers, as such. This is important information for providers of Gaelic-language broadcast and recorded media to know what their total potential audience comprises. This additional 33,744 persons may include some advanced learners but probably represents a large number of 'semi-speakers' who are used to hearing Gaelic in their immediate social environment but who have not been brought up as actively using Gaelic in their home situation. If they had had the opportunity of an active Gaelic education, their passive abilities might have been transformed into active abilities - and this clearly represents a challenge for our education system today.

Persons able to speak, read and write totalled 65,674 - almost 12 % more than speakers alone. These 7,022 persons may include - or even largely comprise - Gaelic learners who are confident in reading and writing the language but not speaking ability. These are the closest to the point of becoming fluent speakers, since they have had the largely self-motivation to master literate abilities. The additional 26,722 who understand but do not speak, read or write the language may present a tougher challenge in any strategy of reversing language shift. These results, and comparable data from 1971 - 1991 are shown in Table 3 and Figure 9.

Overall implications

The census results very clearly indicate the need for both local initiatives in the traditional Gaelic areas as well as a truly national strategy which reaches Gaelic speakers throughout Scotland.

Gaelic speakers are resident in every area of Scotland and the language is present. everywhere. Language planning needs to take account of this. Policies will need to differ from area to area - and even within council areas, which are far from being linguistically homogenous. In terms of developing policies with local authorities throughout Scotland, a starting point will certainly need to be in exchanging information: in making manifest what are the local linguistic situations, and seeking goodwill in what authorities see as their individual role in a national language strategy.

Language acquisition needs to be seriously addressed. Age-related data and other more detailed results may be able to indicate that Gaelic-medium education has had important effects over the past 12 years. As presently constituted it cannot effectively reverse language shift even in the case of the Western Isles. Preschool levels in the education authority and voluntary sectors taken together are approaching levels which match rates of language loss. However, these numbers do not translate into the primary sector. Even less do these numbers translate into the secondary sector. The whole situation needs a more systematic approach and the underlying philosophies call for a rethink.

Policies which have produced successful reversal of language-shift in countries like Wales, Euskadi, Catalunya and Galicia are based on high quality language censuses and detailed analysis. This has been co-ordinated with and linked to other kinds of research. We have high quality demographic statistics in the U.K. and Scotland. They can be much further and more effectively exploited .in showing what is happening within Gaelic communities and with Gaelic speakers. A start needs seriously to be made immediately on the release of further census data later in March.

   Kenneth MacKinnon

Professor Kenneth MacKinnon
SGRÙD Research 20.02.03