As the latest releases of information from the 2011 census for the Welsh language are a little bit difficult to piece together, I've put the information correlating language ability with country of birth and national identity together in a way that I hope makes things a little clearer. I've produced a full spreadsheet which can be downloaded here, and distilled that information into this table:
I don't think any of the information will come as too much of a surprise, but it might be worth making a few comments about it.
The biggest factor affecting the figures is migration. The numbers are huge. 28.1% of the population of Wales aged three years or older was not born in Wales. In Fflint and Powys, Welsh-born people are in a minority; and in Conwy, Ceredigion, Denbighshire and Monmouthshire the percentage of non-Welsh-born people is greater than 40%.
The figures show that people born in Wales are very much more likely to speak Welsh than those not born in Wales. On average the difference is about three to one (23.3% to 8.0%) but in some of the counties with very high levels of immigration, those born in Wales are about five times more likely to be able to speak Welsh (44.5% to 7.9% in Conwy; 74.6% to 15.0% in Ceredigion) than those not born in Wales.
However I want to repeat something I've said before. I don't think that immigration in itself is the problem, it is a symptom of a bigger problem: economic disadvantage. It is lack of economic opportunity that forces an inordinately high number of people to leave their local communities, either to find work or to find somewhere affordable to live. This in turn creates a vacuum which makes it easier for those who don't work or don't need to pay for a home (those who have retired, or those on benefits) to move into those same communities. Any solutions we propose should therefore not be directed at immigration (and certainly not against individual immigrants) but at improving the economic fortunes of the areas concerned, and at providing a more appropriate mix of housing tailored to meet the local needs of those areas.
It's also worth saying that the two main reasons why the 2011 census figures are lower than those for 2001 are the reduction in the number of children, and the migration of young people away from Wales. I've no doubt that the overall number of Welsh speakers went up between 2001 and 2011, but that tens of thousands of young Welsh speakers moved away from Wales and therefore weren't counted. In time many of them will move back; perhaps when they've finished university, perhaps when they've spent a few years seeing the world, perhaps when they decide to settle down and raise a family, perhaps when they retire. And, of course, some won't come back at all. This has been the pattern for several decades now, but the difference is that in the decade before last a far greater percentage of young people learned to speak Welsh at school than ever did before (which accounted for the rise in the 2011 census) and therefore a far greater number of Welsh speakers has now left Wales.
It's no surprise that the percentage of people who think of their nationality as Welsh should go hand-in-hand with the fact that they can speak Welsh. But it's interesting to see that there isn't very much difference in the percentages between those who think of their nationality as Welsh only and those who think of their nationality as Welsh and British. However we do need to bear in mind that in numerical terms those who think of their nationality as Welsh only is eight times bigger than those who think of themselves as Welsh and British, and nearly four times bigger than those who think of themselves as British only.
As we know that about 21% of those who live in Wales were born in England, the British only group is likely to include both people who we might consider Welsh (i.e. born and raised in Wales, but thinking of their nationality as British even though they would consider themselves Welsh in a more general sense) and those from other parts of Britain who prefer to think of their nationality as British than English, Scottish or Cornish ... either for the same reason or because they prefer to adopt an identity that doesn't set them apart from their neighbours now that they live in Wales.
Choosing to Learn Welsh
Whether we look at things in terms of country of birth or national identity, one thing that I think is particularly important is that the percentage of non-Welsh/non-Welsh-born people in Wales who can speak Welsh is not zero, or anywhere near zero. It shows us that immigrants can and do learn Welsh.
Sadly the figures don't tell us when those who weren't born in Wales moved to Wales. Those that moved when they were children will have had the same opportunities to learn Welsh through the education system as everyone else in Wales; but fully half of the 66,000 or so who speak Welsh but weren't born in Wales are over 35 years old, so they were in school well before Welsh was a compulsory subject in schools. It is therefore clear that there is some pattern of adult immigrants learning Welsh in order to better integrate into their adopted communities.
The percentages for those born outside Wales but able to speak Welsh are 20.4% in Gwynedd, 17.6% in Ynys Môn, 15.0% in Ceredigion and 13.2% in Sir Gâr. Now of course I'd like these figures to be higher and I certainly wouldn't disagree with Cefin Campbell when he says, in this article, that we need to do more to raise awareness of the language among those who come to live in Wales. However I think these figures are fairly encouraging, especially if we bear in mind that for every person who learns Welsh to the extent that they can say they speak Welsh, there will be many more who are learning or trying to learn but have not yet got far enough to move from thinking of themselves as being a "Welsh learner" to thinking of themselves as being a "Welsh speaker".
The willingness appears to be there, so I'd suggest that the problem isn't so much about raising awareness or getting an initial taste of the language, but in the way we go about teaching Welsh to adults who have expressed that willingness to learn. There is a very high drop out rate from adult Welsh courses, as shown in this graph from a paper by Heini Gruffudd and Steve Morris:
Decrease in the number of learners by age and learning level
And part of the reason for this high drop out rate is that the intensity of the courses is very gentle compared with that in other countries in a similar situation. The pace of learning is so slow that people drop out because they don't feel they're getting anywhere. The most typical model in Wales is about 2 hours of teaching a week; in Euskadi 73% of those learning Euskara study for more than 6 hours a week, and 65% of them for more than 10 hours a week. So we could, and should, be doing much more to improve the way we teach Welsh to adults.
Stepping back and looking at the big picture, I don't think teaching Welsh to adult immigrants is the most important factor in the growth of Welsh. School-age education will always be more important because it's easier to teach puppies new tricks than it is to teach them to old dogs. But the scale of immigration into Wales is so large that we urgently need to put more resources into teaching Welsh to adults as well.