Hi! Today’s topic comes from a friend of mine who posed a question to the world of Facebook in his status. That’s the best kind of topic.
He asks, “Can someone please explain vy many Indian people pronounce ‘v’ as ‘w’ and wice wersa? I’m wery curious what causes this.”
A similar question is asked by many people who’ve heard someone speak a second language: why are some sounds so difficult for non-native speakers? A few people followed up on his post to give some other examples: Korean speakers substitute ‘p’ for English ‘f’, Arabic speakers substitute ‘b’ for ‘p’, and so on. And believe me, English speakers are just as horrible in other languages. So what’s going on here?
To start with, every language has a limited inventory of sounds—in theory, around what’s needed to differentiate the language’s words. Everyone learns their own language, and ends up being able to make and distinguish all of its sounds. So why can’t they learn another language’s sounds?
It’s not just a matter of pronouncing it right. There’s another major obstacle: they often can’t even perceive the sound properly. A language doesn’t just have an inventory of real sounds, which we call “phones”. It also has an inventory of abstractions from the sounds; we call these “phonemes”. Once you’ve learned your native language, your phoneme inventory is more or less set.
What does this mean, “abstractions from the real sounds”? Think about all the situations in which you hear language spoken: in a quiet room, but also at the bus stop, in a bustling cafeteria, in an echo-prone lecture hall, and so on. The complex torrent of sound waves coming at you is different in each. There’s also variation in who we talk to—their pitch, their exact pronunciation, their volume, the timbre of their voice, their accent, their dialect, and so on. Take the ‘b’ sound; what a miracle we can recognize all the million variations of the sound waves that pass for ‘b’ as the same sound!
How do we do it? One fairly well-evidenced theory is that the brain learns to discard all the minor variations and assimilate all those inputs under one “heading”, one abstraction; this is the phoneme. “Aha,” says your unconscious, “those sound waves I’m hearing are really ‘b’. I know to ignore all these irrelevant physical details that would only distract me from recognizing it.”
All well and good—for speaking one language. What happens when you encounter a language where some of those details aren’t irrelevant? What if one of those details is the tiny physical difference between the sound waves for ‘p’ and ‘b’? If your language doesn’t make the distinction, your brain has been unconsciously trained to consider it the same phoneme.
Here’s an example. Briefly pronounce “s” and “z” and you’ll feel how just about everything in your mouth stays the same between them. The same is true of “sh” and “zh” (the sound in “measure”). The friend who raised this question speaks Polish, a language with three different pairs in the range between the two English pairs I just named. I’d bet anything that English speakers faced with Polish will only perceive two pairs, even when they’re hearing three.
So the question is, can we ever train ourselves to hear new sounds?
First, a bit of background on how we learn our native language. It’s a well-accepted fact that our language-learning ability as children is a superpower compared to what we have in our adult life. When a baby is born, it has no “preset” language; it’s ready for anything. Some fascinating studies found that in their first few months, babies can distinguish between any given sounds (at least, among those found in human language). After a few months, that ability starts to diminish. As they become a fluent speaker of their language, an inventory of only the phonemes they need solidifies.
Changing it after that is very, very hard. What do you do if you can’t even perceive the sound you want to learn? It is possible. It depends partly on awareness of your language and its sounds (for example, many English speakers are surprised to learn that English even has a “zh” sound, perhaps because we don’t have a spelling dedicated to it).
Which sound you’re learning is also a factor, and probably the most relevant for my friend’s question. I once asked Professor Syme of the English department how it is that he can say “th” perfectly, but sometimes chooses the wrong sound between “w” and “v”. He explained that for German speakers, “th” is actually easy to learn, because they don’t have any phonemes to confuse it with. But to him, “w” and “v” sound similar and feel similar to pronounce.
Thus, because in English “w” contrasts with ‘v”, we’ve learned to carefully distinguish them. But speakers of languages that never need to make that distinction—like the Indian languages my friend mentioned—are unconsciously conditioned to ignore the differences.
And this particular case is quite common. We know that “w” and “v” are acoustically and perceptually similar, partly because we frequently see fluctuations between them in the histories of languages. One person commented on the post that Latin had only “w”. True—and yet most of the languages descended from Latin have “v” where Latin had “w”, and it’s not that surprising a change. In the big picture, English is the odd one out for contrasting them.
Maybe one day we won’t.
(P.S. Someone once generously said I should pitch this column to some big paper. I said I’m just a student—my current handle on linguistics wouldn’t cut it in the outside world. Case in point: last week I said the fact that you change a real thing in the real world just by saying “I swear” is illocution. No doubt the illocution also has to do with swearing, but by the definitions I gave, what matters here is clearly perlocution. Thanks to the perceptive reader who pointed this out!)