If I say the word chair, or car, or tree, you know exactly what I mean, even though there are many kinds of trees and cars and chairs. If I want to be more specific I can say oak tree, sports car, or dining chair. This works well with objects because we’re pointing to something concrete. But things become harder when we talk about abstract ideas. Even a word like love or grief, which we all use, has no guarantee of meaning the same thing to each person. And the difficulty grows again when we move into categories such as racism, fascism, far-right or far-left. These categories have contested boundaries, so it becomes even less likely that we’re talking about the same thing. We might think we are discussing the same idea, while in reality imagining something entirely different.
Take the phrase “far right,” very much in use here in the UK at the moment. Today, as I write this, the Home Secretary, Shabana Mahmood, has announced plans to tighten illegal immigration rules, and some commentators have described the move as far-right. But the activist Tommy Robinson is also described as far-right, as is Nick Griffin, the former leader of the BNP. It is hard to argue that these individuals share the same political outlook. If the same label can be applied to such different figures, it becomes unclear what the term is actually pointing to. And without a shared understanding of even the basic category, the conversation isn’t just confusing, it becomes impossible.
Arguments often collapse because we don’t agree on what our words mean. Even in good faith, it is almost impossible to reach a conclusion if each person is effectively talking about something different. Think back to the object example: if I asked you to fetch a chair and you kept bringing me a fork, not out of stubbornness, but because your understanding of “chair” genuinely pointed to a fork, the conversation would be hopeless. Without a shared definition, there is no way to test ideas, challenge assumptions, or make progress.
The same thing happens in politics. If we try to have a serious discussion about immigration policy, but you describe the Home Secretary as “far right” and I hear that as placing them in the same category as Nick Griffin, we are not having one conversation at all. We are having two separate ones. With definitions that loose, there is no way to make progress on the substance. And when arguments cannot be resolved through shared meaning, people often stop using words to describe reality and instead use them as signals — to show which group they belong to and, by implication, which groups they oppose.
A recent UK example makes this even clearer. Following the government’s announcement on illegal immigration, the terms “asylum seeker” and “illegal migrant” are being used by different sides to describe the same group of people, but with entirely different moral assumptions attached. For some, the default category is “asylum seeker,” implying individuals fleeing danger and persecution. For others, the default is “illegal migrant,” implying people arriving for economic reasons in ways that break the rules. It is likely that both descriptions apply to different individuals within the wider group, but if the sides debating the issue cannot agree even on the basic definition, the discussion goes nowhere. Each word carries a whole moral world with it, and without clarity on what is meant, there can be no shared argument.
Words and their meanings are also not fixed in time or place. The word “liberal” in the UK is not the same as “liberal” in the US, and “conservative” has changed meaning over the past thirty years. A “nationalist” in Scotland is generally a pro-EU social democrat, which is very different from nationalist movements in parts of Europe. A “populist” can describe a right-wing party in Italy or a left-wing president in Latin America. Even “left wing” refers to different political traditions depending on where and when it is used. The labels stay the same, but what they point to drifts beneath them. Without context, words become unmoored from any stable meaning.
We also cannot rely on universal definitions, even when a word appears to be formally defined. Take “racism,” for example. The term now carries more than one accepted meaning, sometimes even within the same dictionary. If I am discussing racism with Kehinde Andrews or Ta-Nehisi Coates, the definition will typically include structural and systemic dimensions. If I am discussing the same word with Coleman Hughes or Trevor Phillips, the focus is likely to be on individual prejudice based on skin colour, with structural outcomes treated as a separate question. Both uses are established in public debate, but they refer to different things, and without clarifying which meaning is intended, the discussion cannot stay on solid ground.
There is a straightforward way to avoid these problems, but it requires discipline and clarity. We need to establish an operational definition: “for the purposes of this conversation, when I say X, I mean…”. This approach is standard in science and in the judicial system, where precision is essential. It used to be common in academic writing too, and it remains the most reliable way to keep a discussion grounded. By setting the boundaries of a term before the argument begins, we create the conditions for reasoning together rather than talking past each other.
When these boundaries are set in good faith, they make conversations coherent and stable. They allow disagreements to become meaningful, because each side knows exactly what the other is arguing for or against. They also create the conditions for ideas to be tested rather than left circling in stalemate. And if someone refuses to define their terms, that itself is revealing: it suggests either that their position is not fully thought through, or that they are not ready to engage in a shared conversation. Clear definitions cut through posturing and tribal signals, and open the space for genuine dialogue.