So let's talk about personality.
Even if you've known someone for a long time, they might still be able to surprise
you — maybe with fears that you never knew they had, or a legion of special skills they've
been keeping under wraps.
Often, it's just about asking the right question.
And when it comes to language, that's exactly what we aim to do: Even something as familiar as
a verb phrase can have secrets tucked away inside.
I'm Moti Lieberman, and this is The Ling Space.
The framework known as X-Bar Theory essentially
says that humans come with a rough syntactic sketch inside their heads, in order to explain
how we build our sentences.
It provides a template which at first might seem too simple to do much good, but when
it's put to proper use, it actually has the ability to generate each and every expression
that makes up part of someone's language, without a misstep.
In fact, it's that very simplicity that makes X-Bar Theory so powerful.
Not only can it deftly deal with the different categories that make up the structures in
any given language, it'll happily handle whatever word order you throw at it, no matter
if it's German, Irish, or Mi'gmaq.
Like, X-Bar Theory can tackle transitive verbs just fine, like in "His father read the book."
But there's a sinister problem lurking at its core: it doesn't have any good way of
coping with ditransitive verbs — so, words like "give" and "put."
These have not one, but two objects following them — one direct, and one indirect, as
in "give the pen to Clark" or "put him in the swimming pool."
For a template that has only one slot after the verb, that's a bit of a shock, so let's
have a closer look at where things go awry, and how we can solve this conundrum.
Direct and indirect objects are both arguments of the verb, which means that they represent participants
in the action or the event being described.
And these sorts of phrases are pretty much indispensable, since you can't say things
like "give to Clark" or "put him" by themselves.
In order to represent this requirement, X-Bar Theory situates both objects next to the verb, like so.
But the branches of the trees holding our sentences together are typically divided into two,
if they branch at all, so a three-way split definitely breaks with tradition.
Of course, we could chalk it up to ditransitives just being an exception.
We could even force them into our existing template, if we really wanted to.
But the problem is more than just aesthetic: it actually runs against some really well-established
grammatical facts. Let's say that Melanie has a mirror.
Then, we could easily imagine saying a sentence like "Melanie showed David himself," with
"himself" effortlessly referring back to David.
But according to the principles of Binding Theory, which spell out when and where pronouns
like "himself" can appear, it should be just as easy for us to use "Melanie showed
himself David" to express that very same thought.
If this is the right shape for our tree, either one should work, since even if the ordering's
different, the structural relationships in each case are exactly the same.
But in the second case, "himself" just can't refer to David, no matter how hard it tries to.
This strongly suggests that, somehow, the first object is actually higher up in the
tree than the second — that it sort of 'outranks' it.
So what could this possibly look like?
To answer that, we'll have to take a short detour through the syntax of causation.
In English, we use verbs like "make" and "cause" to represent the fact that somebody
made something happen.
So, we can take a sentence like "They fight back" and add in a third party, like in
"Sydney made them fight back."
But this isn't the only way languages introduce causers into the mix.
There's at least one class of verb that has this causative meaning built right into
it — so-called lexical causatives.
These verbs, like "melt" and "roll," connect people to the events that they set in motion.
"Oliver melted the ice" means that he caused the ice to melt, while "She rolled the oven
down the alley" draws a similar kind of connection.
These sorts of verbs are different from ordinary transitive verbs like "punch" or "kick,"
both in their meaning and their syntax.
While you can say "Oliver punched the ice, but nothing happened to it," you can't
say "Oliver melted the ice, but nothing happened to it."
The causative meaning is too much a part of the word to be cast aside.
More to the point, these verbs can optionally skip over the causer and go straight to what
was caused; so, "The ice melted" or "The oven rolled down the alley".
So how can we account for the behaviour of these verbs?
Do we have to suppose there are two separate versions of the verb stored in our head, each
associated with a slightly different structure?
Well, one big clue comes from languages that have a richer morphology than English.
When we look at, say, Malagasy, we see a similar pattern, but where the mechanism is much more visible.
In shifting from the transitive verb "hide something" to the intransitive "hide
yourself," Malagasy swaps out the beginning of the word.
In Tagalog, the intransitive verb "fall" shows up as this, while actively knocking
someone down looks like this — again, with a different set of sounds at the start.
And in Japanese, the intransitive verb 凹む"hekomu," meaning "to be dented," turns into the
transitive 凹ます "hekomasu," meaning "to dent something."
The root verb is the same in either case, but its function inside a sentence changes
by way of a special affix that inserts itself between the verb and the tense marker.
Let's see how this works in tree form.
So, if we had a sentence like "kikai-ga hekomu," for "the machine will be dented,"
our tree would look like something like this.
Since "kikai" is the only participant in this sentence, it goes on to become the
subject, while the verb moves up the tree to join together with the tense marker, like
the verbs do in French.
In contrast, speakers also have the option of inserting another layer into the structure,
headed by the morpheme "-as," which then paves the way for an agent to come in to cause
the event in question.
In "David-ga kikai-o hekomasu," David's now denting the machine.
He then goes on to become the subject of the sentence as a whole, while the verb once again
climbs the tree to collect all of its suffixes and become a full-fledged word.
In this setup, there's no need to suppose that two versions of each verb exist.
Instead, whenever an agent appears, it's simply the result of a special kind of phrase
found sitting just on top of the regular verb phrase — a kind of shell wrapped around
the main VP, whose job is to invite other personalities into the mix.
Importing the idea over into English, we can make an educated guess that verbs like "melt"
and "roll" follow the same blueprint.
These verbs have to get involved with inserting all the relevant players into the action,
so now they get split across two different sections of the sentence.
And when the part of the tree tasked with introducing the subject doesn't show up,
the object is all that's left.
We might even suppose that ordinary transitive verbs like "keep" or "erase" make
use of this kind of structure, too.
When we look at how much influence subjects and objects have over the meanings of their
verbal hosts, we find that objects seem to be a lot more closely connected to them than subjects.
Phrases like "throw a toaster" and "throw a fit" denote very different kinds of throwing;
and the same with "take an aspirin" and "take a nap."
But switching one subject for another doesn't really have the same effect; "Lenny took
a nap" and "Benny took a nap" describe fundamentally the same event — just with different nap-takers.
Our 'split VP' idea matches up with this nicely.
Best of all, we finally have a satisfying explanation of what the structures of ditransitive verbs must be like.
If agents and causers really do make their ways into sentences separately from objects,
this opens up the space that we need to fit the ditransitive puzzle piece into our overall
picture of sentence structure.
Let's have a look at how this actually works.
So, we've argued in the past that the structure of an English verb phrase has to look something
like this, with the subject starting off at the top, and then later moving up even
higher to become the subject of the sentence as a whole.
But you can see that this doesn't really leave any room for the indirect object — at
least, not unless we use the three-way branching.
But now, we toss in our new agent-adding structure, which surrounds our original verb phrase.
Call it a 'little vP' , to mark the difference between it and the main one.
Since the subject now starts off a bit higher in the tree, there's extra space inside
the lower verb phrase — enough to hold both objects.
Then, just like in Japanese, the verb moves up into the top floor, giving us the word order we want!
Ditransitives aren't really exceptions after all, and with the first object in a position
that's higher up in the tree than the second one, the facts about reflexives, or words
pointing back to other parts of the sentence, make total sense; a pronoun like "himself"
can't show up first, because that would put it up too high in the tree to be able
to refer back to David, whereas the other way around is fine.
And even though we can't see this extra structure with the naked eye, we can test
to see whether it's really there.
If verb phrases really are split in two, there end up being twice as many spots for adverbs
like "again" to attach to, which means that we expect to find differences in meaning.
And sure enough, a sentence like "Dr. Poole closed the door again" has two subtly different
interpretations: one where he'd closed the door before, and then did it again, and another
where someone else had closed it first, and after it was opened, he closed it again.
As you can see, different languages can reveal different sides to the same Universal Grammar
that we all possess.
And when they work together, there's no secret that they can't uncover.
So, we've reached the end of The Ling Space for this week.
If you cracked open your verb phrase to see what's inside, you learned that certain
words, like "give" and "put," pose a problem for our current theory of syntax;
that some languages use a special kind of structure, to introduce agents and causers
into the mix; and that if English has these structures too, it helps explain ditransitive
verbs, how they interact with reflexives, and why subjects and objects seem so different.
The Ling Space is made by all of these amazing people over here.
If you want to learn more about *how* these ideas tie into the passive voice, check back
on our website!
And while you're there, why not check out our store? You can even get the Super Schwa shirt I'm wearing!
We're also on Tumblr, Twitter, and Facebook, and if you want to keep expanding your own
personal Ling Space, please subscribe.
And tap that little notification bell down below if you want to find out right when we
post a new video. See you next time! Bar lo kom!
Không có nhận xét nào:
Đăng nhận xét