字幕表 動画を再生する 英語字幕をプリント So let’s talk about babies. Very cute, right? It’s hard to look at a baby smiling at you and not feel good. But even the most devoted parents don’t generally think of their baby as a cognitive powerhouse. But babies figure out how language works like little geniuses. They’re just born that way. I’m Moti Lieberman, and this is the Ling Space. Language is special. People can do a lot of amazing things - ride a unicycle, learn long division, walk on the moon. But maybe the most amazing thing that human beings do is language. Don’t believe me? Just wait. So the most important idea for today’s episode is the theory that the ability to learn and use language is biologically hardwired into the human brain. Babies are made to pick up the language that they hear around them, just sponging up those glorious sounds and structures and turning them into their mother tongues. The theory that language is something innate, something you’re born with, is known in linguistics as nativism or generativism, and it’s got a lot of really good evidence behind it. Before we start getting into what this means, though, let’s be perfectly clear about what it doesn’t. It doesn’t mean that we’re born with any particular language – no baby springs into the world with the ability to speak perfect English or German or Japanese. We don’t come equipped with the rules or sounds or vocabulary of any language - that’s all stuff we have to learn. What babies don’t need to figure out is how language can work – what kinds of words we can build, what types of sentences we can make, what sorts of interpretations we’re allowed. These fundamental principles of language define what’s possible and what’s not, and they’re the same for everybody. Anything that doesn’t stick to the rules will never come up in any human language, ever. Given how many different languages there are, it might seem unlikely that there's one set of principles that can rule them all. But the idea comes to us from none less than the father of modern linguistics, Noam Chomsky. He called it: Universal Grammar, or UG. There are a lot of arguments to back up the nativist position, but for today, we’re just going to focus on two of them. So the first is that babies go through the same stages in development no matter what language they’re learning. The second is that infants master language way faster than they should if they’re just little blank slate know-nothing babies. To be that fast, there’s got to be something there to help them along. Let’s start off with babbling, those adorable random syllables that little babies make. Except they’re not really random. No matter what language they’re learning, this early babbling uses the same set of sounds. One study examined the early babbling of babies from 15 different languages, including English, Thai, Japanese, Arabic, Hindi, and Mayan – languages where all they have in common is that they’re spoken by people. The study found that these babies all prefer labial consonants, or sounds made with the lips, more than other consonants; stop consonants like [p] and [b], where the air flow through the mouth is totally blocked, over others; and vowels made low in the mouth like [æ] and [ɑ] over those that are made higher up, like [i] and [u]. And all of that is independent of how often - or even whether at all - these languages make use of these sounds! Babbling Hindi-learning babies make the same amounts of the same sounds that Arabic-learning babies do. In fact, your average 8-month-old baby can differentiate between any pair of sounds used in any language in the world. Depending on your language, you might not be able to tell the difference between [t̪ɑk] and [ʈɑk], or between [lɑk] and [ɹɑk], but your baby can. It makes sense: a baby needs to be prepared to pick up any language, so they better come equipped to hear anything that could be relevant. It’s not just the way they deal with sounds that’s the same for all infants. They all pick up words at the same approximate rate and stages, too, and that's regardless of how the language they’re learning works. It doesn’t matter whether or not the babies hear motherese, that way of speaking slow and using easy words and intonation, like “What a cute baby! Where’s the kitty, baby?” It doesn’t matter if the language has tone, like Mandarin, or doesn’t, like English, or whether the verb comes at the beginning or end of the sentence. In fact, all babies, in whatever language, will start getting their first words around 10-12 months old. By 18 months, they’ve got about 50 words. And then they undergo a crazy vocabulary spurt, picking up hundreds of words over the next few months, so that by around 2, they’ll have about 500. And then they start going even faster! Your average 2 and a half year old is glomming up new words at the rate of about 10 a day. That’s faster than your average 3-credit undergraduate language course. So the ways that kids make sounds, the way they pick up words, it’s all the same worldwide, in Chicago or Tokyo or Cairo or Bangkok. And since the languages they’re learning are all so different, this tells us something fundamental about the human brain. How babies learn language is biological – our brains are configured for language. If you’re still not convinced, how about this: there’s been a lot of research done on other language acquisition theories, and the results there are just as clear. Maybe you think that we can pick up language quicker than other behavioral skills, that there’s nothing special about it except how fast we do it. Okay. First, of course kids make mistakes - calling a deer “horsie” the first time they see one, saying “I eated” instead of “I ate” - but there’s all sorts of mistakes that kids don’t make that it seems like they should. For example, when asking a question, “Teddy is happy” can turn into “Is Teddy happy?”, but “Teddy dressed up as Alice” can never turn into “Dressed Teddy up as Alice?”. Kids never make mistakes like that. Second, if language was just something you picked up without having a blueprint in your brain, it should be possible to approximate some part of it with computer modeling. That’s exactly what one linguist tried to do in 2011. She designed twenty different computer models of how the English stress system could be acquired – so what syllables should be pronounced more strongly than others, and what factors matter for deciding that. And the researcher didn’t only run the experiment only once – she did it a thousand times for each model, with different versions paying attention to different factors. So how many of these models nailed English? Three. Three out of twenty thousand total trials. And yet, basically every English speaking child gets this right. That strongly suggests there’s something very special going on with language. But it’s not just that all babies do the same things that makes us think that language is innate, that it’s something we’re born with. It’s that kids get so good at language so quickly. Let’s consider what your average 2 and a half year old knows about language. They know what sound combinations are possible for their language, so they know what a possible word sounds like. They know the word order for their language, so a Turkish kid will know that the verb comes after the object, but a French kid will know it’s the other way around. They know how to make questions, and what sorts of questions it's grammatically okay to ask. They know how to use modifiers like adjectives or adverbs. Now, think about what your average toddler knows about, like, everything else. They know a whole lot about how language works, at an age where they can’t add 2 + 2. Or tie their shoes. Or reliably use a bathroom. So why are kids so good? Why can they learn so fast, make so few mistakes, and succeed where sophisticated computer models fail? How do they know all of this despite not having it explicitly taught to them? It’s because they already know how language can work. Deep in their brains, in their genes, they have the abstract rule sets that tell them what’s possible and what’s not. All babies start the same way, with the same linguistic abilities. Then, they apply the data they hear to the Universal Grammar in their heads, and they make little linguistic miracles happen. Every single day. So we’ve reached the end of the Ling Space for this week. We’ll be coming back to talk more about child language in the future, but if you were paying attention this time, you learned that nativist or generativist ideas mean that we think language is innate, but not that any particular language is; that babies go through the same stages of development regardless of what language they’re learning; and that kids know a whole lot about how language works at a really early age. The Ling Space is written and produced by me, Moti Lieberman. It’s directed by Adèle-Élise Prévost, our production assistant is Georges Coulombe, and our music and sound design is by Shane Turner. Our educational consultants are Level-Up Learning Solutions, and our graphics team is atelierMuse. We’re down in the comments below, or you can bring the discussion over to our website, where we have some extra material on this topic. Check us out on Twitter and Facebook, and if you want to keep expanding your own personal Ling Space, please subscribe. And we’ll see you next Wednesday. Mata raishuu!