What is a language?: When Easy Questions Demand Tough Answers : 13.7: Cosmos And Culture Is there really a fact of the matter about where English stops and Spanish begins? Yes, but these differences are not linguistic. In the end, they reflect our values. And it's worthwhile to ask ourselves what these values are.
NPR logo What is a language?: When Easy Questions Demand Tough Answers

What is a language?: When Easy Questions Demand Tough Answers

Does the inability to comprehend someone mean they're speaking a different language? Don Emmert/AFP/Getty Images hide caption

toggle caption
Don Emmert/AFP/Getty Images

Does the inability to comprehend someone mean they're speaking a different language?

Don Emmert/AFP/Getty Images

How many languages are there? Authorities vary on the exact amount, but there's fairly widespread agreement that the answer is somewhere between four and six thousand. Languages die every day, we are told, and with these passages something is lost, something precious.

But what is a language? And is this really the sort of thing you can count? Let's compare this to a different question: How many objects are there? Simple question. So there must be a simple answer.

Right? No.

And for a simple reason: The answer depends on why you are asking.

Do you remember when Microsoft was getting sued for integrating its web browser in its operating system, back before Google was King? Courts faced the question: Is the web browser one product, distinct from the operating system, or is the browser, in fact, a feature of the operating system? The answer to this question, surely, ought to be: Well, for some purposes it is reasonable to think of the browser+OS as one, but not for others. Millions were spent thinking about this question, though today it can seem difficult to remember why.

And so with objects. Is my pen one object? Is the pen cap a distinct object from the pen itself? What about the ink cartridge, or the ink in the cartridge, or the water in the ink? And if the pen is one thing — the sum of the parts out of which it is made — then why isn't the pen plus my car also a discrete entity (what philosophers have called a mereological sum).

The point is not that these questions have no answers. The point is they have many. Before we can count objects, we need to specify the criterion of individuation — the principle of counting — that interests us. And we can't do that in advance. That is, we can't do that before we take a stand on what interests us, on why we are asking. We need to think about values.

It is a big mistake — maybe the mistake of Western Philosophy — to think that questions of this sort can be answered once and for all, independently of our interests and concerns. The mistake is not to think that there is a determinate answer to the question, how many objects are there? The mistake is to think that Nature intimates to us what question we are really asking!

Back to language. What are we asking when we ask how many languages there are? What interests us? What is our criterion? What are our values?

A famous linguist once said: a language is a dialect with an army. His thought was that the only difference between a language like German and a dialect such as that spoken by a Swabian peasant is that the one way of speaking, but not the other, is associated with dictionaries, school systems, standards, media, and, in general, the massive prestige and power of the state. The supposed difference between a language and a mere dialect is bunk, he was suggesting; or rather, it is not a linguistic difference.

Which brings us back to the question, what is a linguistic difference? Do we have any grip on what language is apart from our ideas, ideologies, and socially shared attitudes about what a language is? Why do we say that Swabian is a dialect of German (even if it is also an autonomous language)? Why do we say the Dutch farmer just north of the border with Germany is speaking Dutch at her table whereas the German farmer to the south is speaking German? Surely part of the answer to these questions is precisely to do with the fact that the Dutch farmer organizes her linguistic practices with reference to magazines and TV programs and dictionaries made in Amsterdam, not Frankfurt or Berlin. The Dutch farmer supports a different football team and serves in a different army. The Dutch farmer believes she is doing something different when she talks; she believes she is speaking Dutch.

Is there a way of marking differences between languages other than with regard to the beliefs, the ideologies, of speakers? The natural suggestion to make would be mutual intelligibility. But this cuts both ways. Remember, the farmer in Holland and her cousin south of the border can understand each other, whereas the Swabian peasant and the north German banker may not be able to.

Mutual intelligibility doesn't do the job for us because it isn't a fixed point. Part of what it is to be a speaker is to adjust one's linguistic behavior continuously in ways that modify one's intelligibility to others. I try to make myself intelligible to my children, and my students, to my elderly neighbor, and to the Serbian mechanic who works on my car. Is my language intelligible to my children? Yes of course. But not independently of my active efforts to make it so. So intelligibility isn't a language-independent feature we can use to mark off languages. It's just one of language's different moments.

So maybe we should say that none of us speak one language after all. We are all multi-lingual, all the time. But then the mark between the linguistic variations that count as variations in English, say, and those that involve switching to a different language — French, or German, or Cantonese — are no less fluctuating, and arbitrary. There isn't a principled ground for saying that English stops here and becomes Spanish, or Yiddish, or Cantonese. Or rather, there isn't a linguistic ground for drawing these boundaries. Which is not to say that there are not other grounds. What are they?

I don't want to minimize the importance of language in our lives. Think of the work an immigrant has to do to cope with new linguistic surroundings, or the efforts the German or Chinese business person has to make to learn the language that will enable him or her get ahead and succeed in the international business world.

But when it comes to language, it's ideology, and armies, all the way down.