the trees that underlie our sentences, the ones that build up the structure that our

words slot into and let us build bigger meanings. Every time you build a sentence, every time

someone talks to you, you’re growing one of these trees. All I’ve just been saying

has planted a little grove of language in your mind. So let’s do some climbing! I’m Moti

Lieberman, and this is The Ling Space.

There’s a whole branch of linguistics that’s devoted to looking at the structure of sentences,

known as syntax. But why do we even bother?

Can’t we just stack our words one on top of another like pancakes

to build some delicious meanings?

Well, there has to be some structure, or else we’d be able to work back from a smashed-up wreck

of a sentence like “a threatening hand is who Angel missing lawyer the is,” and have

it get the same meaning as the original, “The lawyer who is missing a hand is threatening

Angel.” Clearly, one of those is a good sentence of English, and the other one is just gibberish

that happens to be made up of English words. So structure must matter.

But what kind of structure do we need? Well, whatever hypothesis we come up with, it’s

got to be really flexible. That's because it has to capture all the variation in how all the

different languages in the world put together all their different sentences. We don’t want to say

Icelandic speakers have one basic way of making sentences, but Telugu speakers have a second

one, and Cree speakers a third. Building sentences with their own internal structures is something

common to every language of the world, and so an Icelandic baby dropped off in southeastern

India will learn Telugu syntax just fine. That’s because the basic framework of syntax

is universal. In fact, it’s part of Universal Grammar, the linguistic knowledge all people share.

But with all the surface differences, finding something that can branch its way through

every human language isn’t obvious. Not only does it have to be flexible, it also

has to be abstract.

So, here are a lot of hypotheses out there, but one of the most commonly talked about

ones, is called X’ theory, first proposed in the early 70s. The X in X’ doesn’t

stand for anything; it’s a variable, like in algebra. We can use that

variable to make a basic structure, a template, like this: X can stand for any noun or verb

or adjective or any category you want to build a phrase around. You end up with chunks of

syntax that can be stacked and connected together, and you do it in a way that’s flexible enough to communicate

anything that you want, in any language that you want.

This gets a lot clearer when you start looking at some examples. Let’s start with something

really simple: a name, like “Cordelia.” Okay, so in your mental lexicon, where you

store all your words, each term belongs to a syntactic category - which is like a part of speech,

so a noun, an adverb, etc. “Cordelia” is a noun, so when we want to put “Cordelia”

in our X’ tree, we replace the Xs with Ns for nouns. In this phrase,

Cordelia is the “head”, which is the part of the phrase with the most content and meaning.

Because Cordelia’s the head of the phrase, and because it’s a noun, the whole thing

will become a noun phrase, or NP. Great! Done. Except, not really. This might work if we

never said anything more than bare nouns and verbs and things, but natural language is

a lot more involved than that. So sometimes all you want to say is Cordelia, but sometimes

you might want to say nice things about Cordelia. Maybe you want to say, “The amazing Cordelia.”

Where did those other words fit in?

Well, that’s where the “bar” part of X’ theory comes to the rescue. So between the

head and phrase level, we introduce one more layer of complexity: that's the bar level, which is written

with an apostrophe next to the letter that represents the head. The bar level is an intermediate, repeatable

stage in the template that allows us all the flexibility we need to

build bigger phrases and sentences. Let’s see how this works.

Since they’re still all still associated with the noun - they’re all to do with Cordelia

- you need to have extra room for those extra words in your noun phrase. So they need to get nestled into

the NP, and that’s where the N’ comes in. Now your sentence is shaping up.

But wait. Why bother having these intermediate stages at all? Even if we know all these words

come together to make a noun phrase, why put in all these extra levels of structure? Why

not just put in an NP at the top, and then different labels for all the words below - so an N for the noun,

an A for the adjective, etc. That’d be easier, right?

Well… here’s the thing. The reason we needed syntax in the first place was to give

structure to how come sentences mean what they do, and have the word order that they

do. All the information about what a sentence means, that is the syntax, and it has to be

visible in our diagrams, why bother drawing trees in the first place, right?

So we end up needing to branch things off two by two with bar levels,

otherwise we wouldn’t know what parts go with other parts. We can even put as many

bar levels into the structure as we want, so it'll work for any kind of sentence. For

example, if we said “the quirky, supremely intelligent Fred,” and there was no internal

structure, so everything was just flat, we wouldn’t know that supremely was supposed

to go with intelligent, and not with quirky. We wouldn’t be able to come up with any

rules to stop these things. All the rules, everything that’s okay and not okay,

has to be seen in the structure.The bar levels give us a hierarchy that allows us to make

sense of things like this. Now, we know “supremely” goes with “intelligent”, and that you

can’t just pull words out willy-nilly to make nonsense sentences.

What X’ theory shows us is the way that we can build structure

in order to capture all the facts of language, along with the flexibility to add whatever

we like. They let us add potentially infinite parts before the head, like “The bespectacled

bookish Brit Wesley,” or after it, as in “the vampire with a soul and a big black

coat.” And this sort of syntax also lets us capture facts about how we form larger

sentences, as questions, find ambiguity, and all sorts of other things, which we’ll

talk about in the future.

Linguists today have a lot of other hypotheses about syntax, too, but X’ is a great place

to start because it shows all of the hallmarks of why syntax is real and useful. It can be

applied to any type of word, in any type of sentence, in any type of language. It’s

just a template: a head with a phrase and as many intermediate stages as you’d like.

But by using that one little template, and putting it in every time you make a phrase,

you can shape a whole world of language. Shaping those little trees can tell you what

language is. And that’s worth the climb.

So we’ve reached the end of the Ling Space for this week, but if you were making your

own happy little trees, you learned that sentences must have an internal structure to them if we’re

going to capture the facts we know about them; that the basic template of that structure

needs to be flexible and universal; that the template in X’ theory consists of a head,

a phrase, and as many bar levels as you need to fit all the words you have; and that the

structure should branch off two-by-two to fit the facts about hierarchy that we feel are true.

The Ling Space is written and produced by me, Moti Lieberman. It’s directed by Adèle-Élise

Prévost, our production assistant is Georges Coulombe, our music and sound design is by

Shane Turner, and our graphics team is atelierMUSE. We’re down in the comments below, or you

can take the discussion back over to our website, where we have more information on this

topic. Check us out on Tumblr, Twitter, and Facebook, and if you want to keep expanding

your own personal Ling Space, please subscribe. And we’ll see you next Wednesday. /seləvu/!