We don’t hear accents. We hear class

Accent is often treated as a neutral feature of how we speak. But in Britain, the way someone sounds can shape assumptions about their class, intelligence and belonging, especially at our most prestigious universities.

There has been a lot written about classism at St Andrews over the past few weeks. Much of it has focused on money: the wealthy American students, the expensive dinners and balls, the private-school backgrounds, the sense that some students arrive already knowing the rules of a social world that others have never encountered.

That conversation matters. But when I read the recent accounts of Scottish students feeling out of place at St Andrews, I kept coming back to something slightly different.

I kept thinking about accents. Because we don't just hear accents. We hear class. We hear education, wealth, confidence, intelligence, competence and belonging. We hear where someone comes from, and then, often without realising it, we start filling in the rest.

This is not uniquely a St Andrews problem. It is a British one.

I spent a year at St Andrews doing my Master's. I was the only Scottish person on my particular course, surrounded largely by American students alongside English and international students. I found the whole place intimidating. It was expensive enough that I lived in Dundee rather than St Andrews itself, and there were plenty of social things I simply could not afford to participate in.

Sometimes I was conscious of being Scottish in a way I had never really been conscious of before. There were occasions when people said they couldn't understand my accent. It sounds like a small thing. Sometimes it probably was. People genuinely can struggle with unfamiliar accents.

But after hearing it often enough, you become more aware of yourself. Of the words you use. Of how quickly you speak. Of how your voice sounds when you answer a question in a room where everyone else sounds different.

The strange thing about being Scottish is that your accent can tell different stories depending on who is listening.

My own accent is relatively soft. I grew up in the Scottish Borders, with family connections to Edinburgh, and was always conscious of pronunciation. My mother's family are from Coatbridge and has much stronger West Coast accents. I have spent years moving between different Scottish places and, inevitably, picking things up along the way.

In one environment, my accent can be heard as posh. In another, it is unmistakably Scottish. In Glasgow, I can be teased for sounding too Edinburgh; back among family, I can sound too Glasgow. Put me somewhere else in Britain, and Scottishness becomes the first thing people hear.

None of these accents tells you very much about who I actually am. Yet we routinely behave as though they do.

Research and work being done at the University of Edinburgh make this point particularly well. Its accent-bias programme describes accent as something that can tell us about a person's social background or where they grew up, but not their intellect or competence. The problem arises when we turn those social clues into judgements about a person's worth.

That contrast seems obvious when you write it down. In practice, we are remarkably bad at observing it. Accent is genuinely a marker of social background. The people around us shape how we speak, and those people are often from similar socioeconomic backgrounds. In that sense, accents carry social information.

But we take that information and treat it as a verdict. Someone sounds polished, and we assume they are educated. Someone has a strong regional accent, and we assume they are less articulate. Someone uses vocabulary associated with a particular social group, and we make assumptions about whether they are intelligent, sophisticated or competent.

Sometimes the prejudice is explicit. More often, it disguises itself as practicality.

“I just can't understand them.”

“They should speak more clearly.”

“They'd sound more professional if they softened their accent.”

“People won't take them seriously if they speak like that.”

It is remarkably easy for class prejudice to present itself as a neutral observation about communication. And universities are an especially revealing place to see it happen. The University of Edinburgh's work on accent bias notes that students often arrive from places where everyone around them sounds similar, only to encounter a much wider range of accents at university. That difference can either be valued or treated as a problem.

This is significant because widening participation is often understood as a question of getting people through the door. How many students from working-class backgrounds have been admitted? How many contextual offers have been made? How many scholarships have been awarded? How many students from disadvantaged areas have enrolled?

These are important measures. But they do not tell us what happens afterwards. A student can be admitted to an elite university and still spend three or four years feeling that they are not really supposed to be there.

The Edinburgh project makes precisely this point: widening participation cannot simply mean monitoring who gets admitted. Students need support throughout university and into the future, and they need to know that they deserve to be there. That last part is harder to measure.

You can count admissions. You can count graduation rates. You cannot easily count how many students have stopped raising their hands because they are worried about sounding stupid. You cannot easily measure the number of times someone has changed the way they speak before answering a question. You cannot put a figure on the distance between being technically included and actually feeling that you belong.

“We confuse social history with social worth.”

This is what I think gets lost in the St Andrews debate. It is tempting to treat classism as something obvious: the expensive restaurant, the designer clothes, the private-school network, the student who has never had to think about the cost of a night out.

But class reproduces itself in much quieter ways. It exists in knowing what people are talking about. In knowing which societies to join. In knowing how to behave at a formal dinner. In knowing how to speak to a lecturer. In understanding which opinions will make you sound clever rather than embarrassing.

And sometimes it exists in the assumption that the way you speak tells other people whether you belong in the room. That is why accents are more interesting than simply another example of university snobbery. It exposes how social hierarchy works when nobody is explicitly trying to enforce it.

There is a particular cruelty in this because your accent is not something you selected from a menu. It is the accumulated sound of the people around you. It is your parents, your grandparents, your school, your friends, your town. It is the places you have lived and the people you have loved. It can change as you move through the world, sometimes deliberately and sometimes without noticing.

And yet people can be made to feel that the sound of their own voice is evidence that they have got something wrong. That does not mean accents tell us nothing. Of course, they tell us things. They can reveal geography, community and social background. The problem is what we do with that information. We confuse social history with social worth.

Perhaps this is also why code-switching is so common.

Most of us adjust the way we speak depending on where we are. That is not necessarily a bad thing. Language is flexible, and adapting to different people and circumstances is part of being human. I know people who speak with a very broad accent at home but soften it at work because they know they will be understood more easily. There is nothing inherently wrong with that.

The question is why they are doing it. When does adapting your speech become code-switching, and when does code-switching become the expectation that you erase the evidence of where you came from? There is a difference between choosing how to communicate and learning that one version of yourself will be taken more seriously than another.

If you change the way you speak so that somebody can understand you, that is communication. If you change the way you speak because you believe your natural voice will make people think you are less intelligent, less professional or lower class, that is something else. To code-switch to be understood is ordinary. To code-switch to be accepted is not.

And the pressure does not always run in one direction. A Scottish person with a broad accent can be judged as less educated or less professional. But a Scottish person who softens their accent can also be accused of sounding posh, putting on airs or thinking they are better than the people they came from.

There is a strange trap here: sound too Scottish, and you can be judged as unsophisticated; sound too English or too polished, and you can be judged as inauthentic. The point is not that every Scottish accent is oppressed, or that every softened accent represents some great act of social betrayal. It is that people are remarkably good at turning speech into a judgement about character. And this is not confined to universities.

One in five people with a Scottish accent reported in recent polling that they had modified or softened the way they spoke to sound more professional, compared with 13% across Britain. Around one in five people in Scotland also believed there was bias against Scottish accents.

It is not difficult to see why.

James McAvoy has spoken about the way his Scottish accent has affected how he is perceived in Hollywood, describing how he could be reduced in people's minds from someone with “infinite possibilities and potential” to simply “that Scottish person”.

That is the strange thing about accents. They are among the least meaningful things about a person, and yet they can become one of the quickest ways we categorise them.

And Scotland has its own complicated linguistic history. Scottish linguistic identity is not simply an English accent with a different geography. Gaelic was historically spoken across Scotland, while Scots became dominant in Lowland Scotland before English increasingly became the language of public and professional life. Scotland contains multiple linguistic traditions, dialects and histories.

That matters because there is something uncomfortable about celebrating Scottish linguistic culture while continuing to treat certain Scottish voices as evidence of provincialism, low status or incompetence. Our accents are part of our culture. They are part of the history of the people who speak them. They should not signify class, intelligence, bank balance, worthiness, competence or belonging.

There is no policy fix for this. There never has been. There are things institutions can do. Universities can train staff to recognise accent bias. Employers can challenge assumptions about what “professional” sounds like. People can become more conscious of the stereotypes they carry. Those interventions matter.

But no policy can make someone hear an accent without making assumptions about it.

The bias is found throughout the UK. It is not uniquely Scottish, and it is not simply a story about English people looking down on Scottish people. We reproduce these hierarchies ourselves. We judge Glasgow against Edinburgh, the west against the east, the broad accent against the softened one. We can be just as capable of deciding that someone's voice tells us something about their worth.

That is why the answer cannot simply be to teach everyone to speak differently. The answer has to be to stop treating one way of speaking as evidence that someone is more deserving of being listened to. There is a difference between linguistic flexibility and linguistic conformity. One says: you can speak differently here. The other says: you need to sound different to belong here.

Belonging matters. It is one of the most basic human needs, and it is particularly important when we talk about education and social mobility. We talk about widening participation as though the important thing is getting someone through the door. But what happens if they get through the door and immediately learn that their voice is telling everyone else that they are different? What happens if the price of belonging is learning to disguise where you came from?

That is not a problem that can be solved by another admissions target or another diversity statement. It requires something much more uncomfortable: changing what we hear when somebody speaks. Not whether they sound educated. Not whether they sound professional. Not whether they sound like us.

Just what they are actually saying. The problem with accent bias is not that people notice a difference. It is that we have learned to turn difference into hierarchy.

And perhaps that is the real lesson of the St Andrews debate. Class does not always announce itself through money. Sometimes it arrives quietly, in a voice, a word, an inflection — and in the assumption that the person making that sound belongs somewhere else.

We don't just hear accents. We hear class.