Doughnut language policy: how to promote Welsh without the unintended consequences

This paper begins with a linguistic analysis of the Welsh Government’s flagship language policies, demonstrating statistically the current primary focus on increasing the number of Welsh speakers. This priority is the beating heart of all Welsh language policy, its guiding light.

The textual analysis also gauges the extent to which language policy explicitly aims to improve people’s material wellbeing in Wales. Such benefits might feel obvious, but they cannot be evidenced if they are not described and measured. A structured comparison shows a notable inattention to achieving such benefits.

Meanwhile, there is growing research evidence from other studies, of unintended negative consequences – from linguistics, sociology, education, and economics. For example, English-speaking children entering Welsh-medium schools (L2 students) achieve lower grades than their L1 Welsh-speaking peers; and Welsh-medium schools overall achieve lower average grades despite equal funding. Graduates of Welsh-medium schools are less geographically mobile. And within communities, valorisation of Welsh – in contrast to English – fuels tensions between language groups. These outcomes are absolutely not inevitable, but they are missed by policies centred on the language itself.

Pointing out these negative consequences is not an end in itself, nor part of an argument against promoting Welsh at all. Rather, it is a call for wholesale reorientation of Welsh language policy to encompass a dramatically wider set of factors. Intersectionality has been used elsewhere in the social sciences to understand how various factors come together to influence discriminations and human freedoms. Indeed, intersectionality informed the Welsh Government’s 2022 report on racism, Running Against the Wind. It can help language policy, too.

In turn, this may very well help the current focus on growing the number of Welsh speakers. Census data show even this goal is not being achieved. Quite the opposite. And the unintended consequences noted above – social hostilities, educational inequalities etc. – cannot be helping. There is a real need to rebalance all these factors into mutually supportive, equally weighted priorities: to develop intersectional policy attending to a diversity of needs.

To remedy the policy imbalance and the negative consequences, I propose a ‘doughnut’ model of language policy. This is a borrowing of a recently introduced economic theory, “doughnut economics”, designed for intersectional balance of environmental and social outcomes alongside the financial. The same intersectional approach has great potential for language policy, emphasising that wellbeing and social relations can and should be equally prioritised, and – returning to my textual analysis – explicitly stated within policy.