There’s been a fair amount in the news recently about the number of communities in Wales where the proportion of bilinguals is >70%, >50% and so on. But I’ve been thinking about these figures. Why are 70% and 50% such important figures?
Then it struck me.
If you assume that conversations between individuals take place at random, then 70% takes on tremendous significance. Because it’s the level of community language competence at which you’d expect the number of Welsh-language conversations in the street to dip below half. How can that be?, I hear you ask.
So 70% of the population is bilingual, and 30% is monoglot English speakers. The proportion of conversations in this hypothetical community is as follows:
- 0.7 x 0.7 = 0.49 (49%) between two bilinguals – which for the most part means they will speak Welsh to one another.
- 0.7 x 0.3 = 0.21 (21%) between a bilingual and a monoglot (conversation in English)
- 0.3 x 0.7 = 0.21 (21%) between a monoglot and a bilingual (conversation in English)
- 0.3 x 0.3 = 0.09 (9%) between two monoglots – conversation in English
This shows the power of deferring to English as the common language. People have commented for donkeys’ years that the willingness of bilinguals to defer to English has been (at least partly) responsible for in-migrants not bothering to learn Welsh. Perhaps here’s a statistical demonstration of why that might be detrimental to bilingualism in a community – because all of a sudden more than half the conversations in an overwhelmingly bilingual community are in English.
As it happens, conversations don’t just happen at random. According to bilingual friends of mine, it’s common for bilinguals to tend to preferentially socialise with other bilinguals – and to do so through the medium of Welsh. The scale of that preference varies according to the level of bilingualism in a community, so I’m told. But I can’t help thinking there’s something in this simple statistical model that should ring warning bells in communities in the west and north.
And the significance of 50%? Well, this is a bit easier. Clearly if you’re one of the 50% who is bilingual then where you live in a community of majority bilinguals it makes sense for you to start conversations with unfamiliar people in Welsh. As soon as it dips below 50% then the hassle of more often than not being told that the recipient doesn’t speak Welsh means that you’re unlikely to bother starting conversations in Welsh. Which leads to Welsh not being heard on the streets and an increase in the perception that it’s not a community language. Perhaps this is one of the reasons for the apparent failure of the (former) Welsh Language Board’s “Start all conversations in Welsh” campaign.
But back to the Vale. Ward-level results have been published for Welsh language competence. So let’s delve into the data…
- Llandow/Ewenni – skills 23.6% – speak 14.8%
- Baruc – skills in Welsh 19.8% – can speak Welsh 13.4%
- Wenvoe – skills 18.6% – speak 13.3%
- Peterston-super-Ely – skills 18.8% – speak 12.6%
- Cowbridge – skills 19.1% – speak 11.9%
- Stanwell – skills 17.4% – speak 11.8%
- Buttrills – skills 17.0% – speak 11.8%
- Plymouth – skills 16.6% – speak 11.4%
- Illtyd – skills 16.4% – speak 11.1%
- Cornerswell – skills 17.0% – speak 11.0%
- Dyfan – skills 16.2% – speak 10.9%
- Cadog – skills 15.1% – speak 10.9%
- St. Augustine’s – skills 16.8% – speak 10.8%
- Court – skills 15.3% – speak 10.7%
- Rhoose – skills 15.8% – speak 10.5%
- St. Bride’s Major – skills 18.0% – speak 10.4%
- Dinas Powys – skills 15.2% – speak 10.0%
- Gibbonsdown – skills 14.2% – speak 9.7%
- Llantwit Major – skills 15.2% – speak 9.6%
- Castleland – skills 14.0% – speak 9.4%
- Sully – skills 13.6% – speak 8.7%
- Llandochau* – skills 14.1% – speak 8.4%
- St. Athan – skills 12.8% – speak 8.0%
*I will call Llandochau by its proper name henceforth (reasoning by Dic Mortimer)
So for the many people who are thinking of moving to the Vale (4,400 annually) but who want to live in as Welsh-language a community as possible, the answer appears clear. In Llandow/Ewenni ward in rural western Vale a shade under one in four people has Welsh-language skills, and more than one in seven people speaks Welsh. The chance of a random conversation in the street being bilingual? Slightly greater than 2% in Llandow/Ewenni – although of course for people who are bilingual it’ll be 14.8%.
But if someone is dead set on Penarth and wants to find fellow bilinguals, their preference should be Stanwell where 11.8% of people are bilingual. Throughout Penarth town the proportion of bilinguals is greater than 10%, although it’s disappointing to see Sully and Llandochau in single figures, scrabbling around for last place with St. Athan.
Given that the average proportion of bilinguals in the Vale is 10.8%, it’s nice to see that Penarth town is either at or above that figure. Why do I consider that a good thing? Because even if the chance of random conversations in the street being in Welsh is little above 1%, it adds to the recognition that we live in a country blessed with two languages.
Finally, it’s worth the recap that far and away the highest proportion of bilinguals is in our young people. So while just 4.2% of people aged 75-79 in the Vale are bilingual, that figure is more than eight times higher among the 10-14 cohort (35%). This figure is surprisingly high given that 13% of Vale children are in Welsh medium primary schools, and 9% in Ysgol Bro Morgannwg. It seems likely that parents in the Vale are suffering from Torfaen syndrome.
And a final note of disappointment that the Welsh Government has decided to pull all funding from Menter y Fro today.
This post has been modified to rectify my error that indicated Baruc to have the highest proportion of bilinguals. My thanks to IJ for pointing out this error.