This is not the first time I’ve written about rightness and wrongness in language, and that shouldn’t be very surprising. My experience before and after becoming a linguist suggests it’s an issue of probably more interest to non-linguists than almost any other question about language.1
This is really an essay, so I’m breaking it up into three, relatively short, parts. Its purpose is to try to discourage people, linguists and non-linguists, from holding silly ideas about rightness and wrongness in language.
And there are lots of silly ideas. On the one hand, some people think—to take a familiar example—that because two negatives make a positive in logic, the same should apply in language and that people who use two negative words to express negation are wrong on this basis. On the other hand, there’s the idea that it’s impossible to err in one’s use of language. Both ideas are false. Unfortunately, however, when one attempts to point out that the first idea is false, one is often accused by non-linguists of holding the second view, which does not follow. Similarly, linguists are often guilty (and I believe I have been myself, on occasion) of assuming that those who counter the second claim must hold the first view to be true. This also does not follow.
The point is that it is perfectly possible to use language and be wrong in doing so, but people are often rather muddled about what makes a given usage wrong. It may help to be a bit clearer about what language actually is and where it exists, because getting that wrong seems to be at the source of a lot of this muddled thinking.
What is language and where can we find it?
Language really encompasses two things: information contained in the brains of speakers2 that allows them to produce linguistic utterances (whether these be spoken, written, or signed) and the set of utterances that they produce. Some of these utterances are extremely fleeting; others are recorded in some way. Human language, in other words, encompasses information in the brain of all the language users in the world and everything they say or write. I’ve talked about this before, here.
A language, like English, Welsh, or Swahili, is a subset of this, and its boundaries are necessarily fuzzy (which is why people get into debates about whether Scots is a separate language or not. There’s no way of answering this question satisfactorily). A language is really a statistical cluster of speakers (or, rather, their grammars) and utterances. I am more like my brother than my aunt in the way I use language, and I am linguistically closer to my aunt than to Nicolas Sarkozy. This is no coincidence. Children learn language by listening to other speakers and working out (unconsciously) what grammar must have produced those utterances. Since geography and social connections influence who we listen to most often, we tend to end up sounding more like those people who live closer to us or who are more closely related.3
That’s what language is, and I’ll elaborate on how getting this wrong muddles our thinking in part 2…
1^Except perhaps the questions of how language originated and why it changes. Considering I have two degrees in precisely this area, one of them a PhD, I should really write about that more often…
2^Traditionally this information has been envisaged as consisting of rules—referred to as a grammar—and a set of words—referred to as a lexicon. Theories of linguistics differ on whether the grammar should be thought of in terms of rules, patterns, or constraints, but this doesn’t really matter for present purposes.
3^Although, as it happens, my own research suggests we may be biased towards imitating some speakers more than others, even if we interact equally often with them.