Saturday, January 28, 2017

The Ling Space: How Can We Tell What Roles Nouns Play? Case Theory

(The Ling Space)

(Note: This episode came out over a week ago, but I'm just getting around to reviewing it now.  Sorry, I've been distracted by other things.)

So, since September, I've committed myself to writing mini-reviews of new Ling Space episodes as they come out.  It seemed like it would be a good intellectual exercise, and that it would force me to engage more with the content.

But as it's been turning out,  99% of the time, I don't have anything intelligent to say other than:
"That was interesting" and "I didn't know any of that before."

So it was with this week's episodes, where my only thoughts are "that was interesting" and "I didn't know any of that before."

What's great about The Ling Space is that it is essentially a graduate level linguistics course completely free on Youtube.  But it is NOT easy listening.  I always have to pay careful attention to follow what they're talking about.

Like many Ling Space episodes, this episode builds on theories they've already established in previous episodes.  So you need to watch several other episodes first to understand this one.

This episode  started with the recommendation "Maybe Watch These First".  But let me make that a bit stronger.  Definitely watch the recommended videos first, or you won't have a hope of understanding what they're talking about in this video.
In my case, I had seen those previous videos before, but I still re-watched them and I'm glad I did.  I needed the review, and I doubt I would have understood this episode without that review.

In a way, I like that The Ling Space keeps making me go back and re-watch those old-videos, because I need the review and it makes me learn more.  But anyone who is short on time should be forewarned--you'll need about 30 minutes of background before you even get to this video.


It turns out this week's episode is based around a theory Noam Chomsky developed in the 1980s.  (The Ling Space folks joke around by flashing out the graphic  "#stillchomsky" .)

Someone like me, who only knows a little bit about Chomsky's linguistic work, sometimes gets the impression that after Universal Grammar in the 1950s, Chomsky didn't do anything more of significance in linguistics.  But he apparently was still coming up with interesting new theories as late as the 1980s.


In my review of the last Ling Space episode, I expressed frustration that I was not able to understand it.
I intended this as a self-criticism, but the author of the episode in question actually reached out to me by commenting on my blog post and apologizing for not explaining better.

I thought this was very generous of him to take the time out to personally interact with a viewer.  (These guys have a high profile on Youtube--at least in certain circles--so I was impressed that they actually cared what I thought.)

I'm still not convinced that it's their fault when I don't understand something.  (It could just be that I don't have an aptitude for linguistics).  But in an effort to try to provide useful feedback to them, I said that in the future I would try to be more specific about what exactly I didn't understand.
This is meant simply as feedback to give the folks at The Ling Space insight into what a typical viewer might be thinking as they watch their videos.  They are under no obligation to actually go through the work of answering these questions if they are busy with other things.

In this video, I think I was tracking with everything right up until the very end.
This last bit threw me off, though.  To quote:

In the sentence "Gabriel claims to know what's best for the Jennings family", the subject Gabriel hasn't risen out of the lower clause, since he really is the one doing the claiming.  But since he's also doing the knowing, we've argued that there must be a silent pronoun that refers back to Gabriel sitting at the front of the embedded clause.  So the sentence comes out to mean something like "Gabriel claims he knows what's best for the Jennings family."  With this new idea at our disposal, we can finally make sense of why this should be.  Because the lower clause has no tense, and the verb "claim" isn't the sort that can assign case to something following it....

1st question: Why can't claim assign case to something that follows it?  Was this explained earlier and I missed it?
Actually, come to think of it, maybe this was explained earlier.  Verbs that assign case are ones that assign the accusative case to nouns that follow it, like "I want her to eat".  Is that right?  So "claims" can't do that then?  But why not?  Was it explained why "claims" can't assign case?

...our silent pronoun is left out in the cold.  As speakers we understand it's there, but it remains pronounced because it has no way of getting case.
 But didn't you just show an example where it got case? "Gabriel claims he knows what's best for the Jennings family."   Or do you mean that it has no way of getting case as long as the "to" is in the sentence?
Actually, if that's the way it is, then why do we need this system at all?  Why don't we just say "Gabriel claims he knows what's best for the Jennings family" all the time instead of messing around with the infinitive form?

Oh, actually one more question.  In the video you say that Japanese has 9 cases.  I actually learned a bit of Japanese  and I don't remember the noun forms having cases.  The verbs could get conjugated, but the nouns always had the same form.  Or am I remembering wrong?

Other Notes
Since making contact with me, Stephan Hurtubise (the writer of The Ling Space mentioned above) has also been generous enough to try to sort out some of the grammar questions in my "Grammar Questions I Couldn't Answer" Series.

His comments make for very interesting reading.  It really makes you think about all the intricacies of English grammar.
Even though we never completely untangled it, he wrote a lot of interesting thoughts about question tags in the comments of this post.
And he gave me a very clear explanation of when "the" goes with the superlative in the comments of this post.

Check it out if you're at all interested in grammar.  Actually, even if you're not interested, it's still fascinating to see how complicated the whole system is.


Stephan Hurtubise said...

Great questions! ^_^

Regarding why "claim" can't assign accusative case to the noun phrase the follows it: we don't really know! This question might not even have a deep answer, beyond being a historical accident. As I'm sure you know, there exists a rich variety of different kinds of verbs, and they can be divided up in many ways. With so-called 'control' verbs like "claim" and "want" and "expect," some can alternate between having a tensed or an untensed clause following them (e.g., "I claim I like borscht" & "I claim to like borscht") and some can't (e.g., "I want I like borscht" vs. "I want to like borscht"); and, as you've seen, some can mark the subject of the following clause as accusative (e.g., "I want them to like borscht"), while some can't (e.g., "I claim them to like borscht"). Some control verbs lack both this tensed/untensed alternation *and* the ability to mark something as accusative (e.g., "I tried to eat borscht" vs. "I tried I eat borscht" or "I tried them to eat borscht"). Why "claim" can alternate, but cannot engage in 'exceptional case-marking' (with "expect" being able to do both) isn't terribly obvious -- at least, not to me.

And you're right about "to"; thats what seems to be stopping the "he" from showing up in that example. But to your question of why this verb should alternate at all: ¯\_(ツ)_/¯. Conceptually, it's a little easier to see the logic behind why "try" works the way it does; the only things that people can try are things they do themselves (i.e., I can't try someone else doing something, I can only try to do something myself). So, it makes intuitive sense that "try" would only ever make use of a silent pronoun that refers back to the subject of the main clause. We can, of course, claim things about people other than ourselves, and want and expect such things, too. But, we can't ever try them. (You could, I suppose, extend this idea to why "claim" has this alternation: with a tensed clause, it's actually ambiguous who "he" refers to, whereas with an untensed one, it's clear the subject is claiming something about himself/herself. But, it should be noted that linguists don't generally gravitate towards these sorts of 'functional' explanations. And it's still an open question as to why it can't mark anything as accusative.)

Finally, about Japanese, Moti (a Japanese speaker) included this line himself. I would have to ask to confirm, but I suspect he's using the term "case" here in just about the broadest sense possible. So, not only as a way of referring to affixes, but as a way to talk about any kind of mechanisms that a language uses to distinguish between different nouns or noun phrases (with Japanese employing various particles, like "wa" and "ga" and "o"). If you consider the German example at the start, it's the determiners that are doing the job, not the nouns. And, of course, some languages go on to place this information onto adjectives as well.

Joel Swagman said...

Ah yes, of course, the particles in Japanese. Yes, I see now. And yes, I think they do mark case in the way you describe.
I was thinking of case exclusively in terms of affixes when I asked that question. I understand now.

And thank you for walking me through all the rest of it as well. I think I've got it all now.
Wow, this stuff gets complicated, huh? But interesting.

It's fascinating how much my linguistic system knows about all this, without me consciously being able to figure it out.