What Questions Can't We Ask? Syntactic Islands
ฝัง
- เผยแพร่เมื่อ 29 มี.ค. 2016
- How do we know what questions we can ask? What keeps us from moving words around into whatever order we want? In this week's episode, we talk about syntactic islands: what they are, what rules can allow us to move some words across long distances in a sentence and not others, and what evidence we have from different languages to back these rules up.
This is Topic #66!
This week's tag language: Sinhala!
Related videos:
Trace Evidence: Syntactic Movement - • Syntactic Movement and...
A Principled Approach: Principles and Parameters in Universal Grammar - • Principles and Parameters
Last episode:
The Magic of Words: Performative Language - • How Can Words Change t...
Other of our morphology and syntax videos:
Raising the Bar: What Changes in a Sentence When We Swap Verbs? - • What Changes in a Sent...
Organizing Meanings: Morphological Typology - • Systems of Morphemes
Referential Treatment: Pronouns and Binding Theory - • Binding Theory and Int...
If you're taking part in our little April 1st quiz, try to tell us where Kanji has gone - the first person who tells us in the comments will get a free mug of their choice!
Find us on all the social media worlds:
Tumblr: / thelingspace
Twitter: / thelingspace
Facebook: / thelingspace
And at our website, www.thelingspace.com/ !
You can also find our store at the website, thelingspace.storenvy.com/
Our website also has extra content about this week's topic at www.thelingspace.com/episode-66/
We also have forums to discuss this episode, and linguistics more generally.
Sources:
Most of the information in this episode comes from Andrew Radford's Minimalist Syntax -- Exploring the Structure of English (2004) & Andrew Carnie's Syntax -- A Generative Introduction (2007). We also refer to John R. Ross's dissertation from 1967, Constraints on Variables in Syntax, as well as Noam Chomsky's Conditions on Transformations (1973) and The Minimalist Program (1995).
Looking forward to next week!
![บุกร้านไอศกรีม VJ วุ้นเส้น !!!! [ดูคลิปเต็มหน้าช่อง] #Short | PANG ORNHIRA](http://i.ytimg.com/vi/g549nEhKZdc/mqdefault.jpg)
6:54 cat suddenly appears out of nowhere
As a native French speaker, I can say the sentence "Connais-tu pas Rousseau?" sounds weird, because in general, subject-verb inversion implies formal speech, whereas suppression of the "ne" implies a more colloquial context, so it seems kind of ungrammatical to me.
I'd either say "Ne connais-tu pas Rousseau?" (although that definitely sounds overly formal), or "Tu connais pas Rousseau?", or even, in Quebec French, "Tu connais-tu pas Rousseau?" (with the "tu", not as 2nd person singular but as a plain old interrogative particle). Keep up the great videos!
+Baptiste Loreau I'm glad you like the videos! And I agree, it's not perhaps the most natural way. But we consulted with a number of Quebec French speakers about whether this was a legit sentence to use, and they checked it out (including a friend in an applied linguistics master's program here who'd just studied French negation). We went into some good detail on why we did this in response to +Tiwinee's comment on this thread, so if you want to know more of our thoughts on it, look there! But the basic is that since we can use it (as it's all right in the dialect here), and it left the grammar of the sentence clearer for the point we wanted to make, we went ahead and left out the "ne". Thanks for the comment! ^_^
+The Ling Space
Thanks, I'm definitely no Quebec French expert, so I understand Quebec French speakers can have a different opinion about this than I have.
I knew the "ne" was already way rare in Quebec (it is a little more common in France, but only in formal context, whereas in Quebec, I guess it'd be near-frozen register), and it must be the other way around for inversion in French French, and the interrogative particle "tu", which I know is super common un Quebec, maybe helps inversion to stay well-established there. I have to admit I have a lot of trouble gauging the commonness of those things in French, and that's not helped by the fact I rarely, if ever, hear people speaking in casual Quebec French, but it stays interesting to talk about!
Also, I'd be really interested to read your friend's research! Is there anywhere I can find it?
+Baptiste Loreau Yeah, this is one of the challenging things about languages with a broad geographical distribution - the variation can make it tough to discuss! My thesis supervisor and I had a lot of arguments over prepositions that were okay for me and bad for her.
I don't know if his research is out anywhere to look at yet, but I'll check! ^_^
+The Ling Space
Oh, I see the kind of thing that can happen with prepositions, like "different to/from/than", isn't it? In French, there's something similar, like you can comfortably say "commencer quelque chose", "finir quelque chose" or "continuer quelque chose", but when you introduce a verb phrase, you have to say "commencer à" and "finir de". For "continuer", however, you can use either "à" or "de" (people are so inconsistent about this one that I don't even know which preposition is supposed to be the standard or traditional one!).
As for the broad distribution thing, that's also what happens with English here. For instance, the English dialect traditionally taught in France is RP (although I've never heard any teacher who spoke perfect RP English), so it happens quite a lot that students don't understand something when someone speaks in a totally different accent than RP (oh, here's the "different than" again) and teachers get really awkward when they try to explain because they haven't really been taught to teach other kinds of English. They're aware of the most basic differences between the most common varieties, but they couldn't tell you anything about Cockney, Indian or African American Vernacular English, for example.
CAT! :D
.. And it teleported away.
It often seems like that is one of her skills, yeah. I have no idea how she gets some of the places that she does.
Perhaps she was illustrating how some of those question words manage to quantum leap across whole phrases. By the way, I really like your videos because they go into more detail and interesting depth than many of the others out there that talk about languages. I'm a language buff, with some background in linguistics, but I always learn something from your videos that I didn't know before watching.
Yeah, she manages to bounce around a lot! And glad to hear you like the videos - we try to talk about the ways in which the ways language is structured and acquired and all that are interesting too, and we're really happy when people really get it. Thanks for the message. ^_^
bahaha best intro ever, love the LOST reference
+Visranda Yeah, the LOST part is really fun. But what else were we going to do with islands? ^_^
In Korean, it's very similar to Japanese.
누구를 언제 어디서 만났는지 기억하는가?
Whom when where met do you remember?
...... Do you remember meeting whom when where?
It's interesting that in English, Wh- words can't be put side by side.
“what should we ask how the fence will stop?” sounds fine to me; ‘how’ isn’t being applied as a request for information here, so I’d interpret it as “for what entity is the phrase ‘we should ask how the fence will stop (it)’ true?”
Now, if the sentence were “how will the fence stop the monster?” and we tried to contort it into “what will how the fence stop?” we’d have an issue.
I'm French, I love your videos, but I have to say that example in French was weird.
We use double negation in French : it's not only [Verb+pas], it's [ne+Verb+pas].
It's true that in colloquial speech the "ne" tends to disappear : "Tu ne connais pas Rousseau." can be "Tu connais pas Rousseau."
But it doesn't work with question where you invert the Verb and Subject, you have to keep the "ne" : "Ne connais-tu pas Rousseau ?" is the only possible way.
In colloquial speech though, we don't invert anything, we just keep the sentence as it is and change the intonation to make it a question : "Tu (ne) connais pas Rousseau ?" --> so much easier !
+Tiwinee He's Canadian. Maybe the Quebecois drop the ne in questions where you swap the subject and verb?
+Robin Kü I may be wrong but I have the feeling Quebecois may say "Tu connais-tu pas Rousseau?" or any of the two sentences that I previously said.
Another form of asking question in French/Quebecois is by adding "Est-ce que" at the beginning of the sentence : "Est-ce que tu connais Rousseau ?"
+Tiwinee That's true, that's another way to do it! But I'm glad you brought this up. I talked with a few native Quebec French speakers about this one, and they say at this point, for Quebec French, using both "ne" and "pas" basically never happens. It's basically just "pas" here now, to the degree that the Parisian French-speaking prof in my friend's applied linguistics department was used as an example for how weird it is to hear "ne" used everywhere.
As you note, there are a bunch of ways that we can make questions in French, and the intonation one'd probably be the most common in spoken French here, too. And the use of the question marker one (i.e. "Tu connais-tu pas Rousseau?") is also fine. But I specifically checked to make sure that we could do the inversion one here without "ne", and it checks out! It's true that inversion feels more formal, and if we used "vous" instead of "tu", the "ne" is then required. But with "tu", it's okay - again, we checked it with a few people, include a friend doing an applied linguistics degree in French, and we feel pretty safe using it!
So if we can use it, we feel like it makes the example more clear to leave the "ne" out - we haven't talked about the syntactic operations yet for how it'd get moved around. It lets us avoid complexity, so we went with it. On a more general note, I think it'll be interesting to see if "ne" ends up going the way of the simple past in French - just in written documents or the like - at least for here in Quebec. Particularly since historically, "ne" was the only negation, before "pas" (and other negatives like "point") evolved into the system! Linguistic evolution's pretty cool. ^_^
+Tiwinee Let's not forget Linguistics is a descriptive science and not a prescriptive one. What matters is what people actually say, not what they should be saying.
Sure, but that doesn't change the fact that nobody says "Connais-tu pas Rousseau ?".
Did he go to Sri Lanka? I love your videos by the way!
Hello ! Actually, in French the question is really easy to do :
Sentence (formal) : _Tu ne connais pas Rousseau._
Sentence (informal) : _Tu connais pas Rousseau._
Question (formal) : _Ne connais-tu pas Rousseau ?_
Question (informal) : _Tu ne connais pas Rousseau ?_ or _Tu connais pas Rousseau ?_
To do a question you just have to accented your sentence end (for speaking) and put a "?" (for writing).
At least for informal purpose.
Easy, didn't ^^ ?
how did that cat appear on and disappear from the bookshelf?
How did the cat on the bookshelf appear and disappear?
So Chomsky thought of both subjacency and minimal link. Has there ever been another linguist like him?
I'd like to see some examples of German sentences with multiple copies of wh words.
+impCaesarAvg Chomsky is definitely a really remarkable linguist, and it's hard to really think of other people that it'd be fair to compare him to. The only one that actually jumps to mind, though, is William Labov, who's really responsible for a lot of modern sociolinguistics and research into linguistic variation. We wrote a Tumblr post about this question recently, actually. ^_^
The German sentences are things like "Wen glaubst du, wen Peter meint, wen Susi heiratet?", which is "Who do you think Peter thinks that Susi is marrying?" It's worth noting that these sentences aren't really good in most dialects of German, but they do work for some people. If you'd like to read a paper analyzing this construction, check out Claudia Felser's 2001 paper here: www.uni-potsdam.de/fileadmin/projects/prim/papers/felser2001JGL.pdf
+The Ling Space I don't know about Labov. I'll Google him.
The Felser paper is too advanced for me, but I glanced at it. She seems to be looking at was ... w constructions. That's close to as wen ... wen ... wen, but I don't know if it's exactly the same. She notes that grammaticality differs across dialects. I don't recall seeing anything like this in textbooks when I studied German many years ago, so I don't have a feel for their acceptability in standard German.
I don't know about Tumblr, but I do enjoy your TH-cam uploads. Steven Pinker links to them in his Twitter feed.
+impCaesarAvg Yeah, we're really happy that he's been so supportive of the project. ^_^
And I agree regarding the German cases - they're not acceptable in all dialects, for sure. My Austrian syntactician friend I contacted about this bristled when I brought them up! But that it appears in some of them reliably is enough to help provide evidence that the wh-words are cycling up through the different spots as the theory is predicting.
So that Japanese sentence you translated as "Who do you remember where we met?", is that a grammatically correct way of asking two questions at once in Japanese; who we met and where we met them? So the person answering could say "We met an old man in the subway." in answer to the question, thereby answering both parts?
+Conway79 From what I have learned, that it is correct. You can see the answers on Stack Exchange: japanese.stackexchange.com/questions/3135/is-it-grammatical-to-have-more-than-one-wh-question-words-in-a-sentence
+Conway79 Yeah, for Japanese, you can more easily ask two different questions at once. And as +pepkin88 notes, you can fit more of them into the sentence - it'll work fine. However, for our sentence, there's also a question scope issue - the second verb, "remember", being at the end with the question marker particle (here, it's "no"), which allows for the translation to be "Do you remember who we met where?" Which means that you could also just answer that question "yes", as well as answering the question parts.
It's worth noting that if you do have multiple wh-words, you start encountering what are known as superiority effects, in which it's better to move out one of the wh-words than another. Once you go past two wh-words, additional ones don't seem to be worse (there's a cool paper about this, if you're curious), but with two, we find these effects. This may be something we should talk about more in the future, too! ^_^
In Korean, it's very similar to Japanese.
누구를 언제 어디서 만났는지 기억하는가?
Whom when where met do you remember?
...... Do you remember meeting whom when where?
It's interesting that in English, Wh- words can't be put side by side.
I tend to think that in English, non-auxilary verbs can't occure before a subject no matter what. But I'm likely wrong.
This problem is better resolved with POS differentiation between interrogative pro-forms and a class of something like "nominalizers," whose function is to transform sentences into nominal clauses, essentially an CP'/S → NP transformation rule.
Then, the jump can't occur because you can't perform the transformation more than once in English, and also because the proposition is wholly isolated
You might wonder P. Jack will see someone at some point.
(|→ Jack will see Kate when
|→ when Jack will see Kate)
XOR
(|→ Jack will see whom at some point
→ whom Jack will see at some point)
You might wonder whom Jack will see at some point.
The jumping problem, by the way, doesn't exist in Mandarin, because interrogatives in Mandarin involve no wh- movement, but still have the same limitation stated above.
你也許自問約翰什麼時候會見凱特？
Even though this is not ill-formed, albeit a bit strange, to write:
你也學自問約翰什麼時候會見誰？
That's because one of the pro-forms gets to be ambiguously treated as a pro-form that's not interrogative, either:
"You may be asking yourself, whom will you see at whatever time?"
OR
"You may be asking yourself, when will John see whomever?"
Mandarin speakers can resolve this by answering both types of questions, though it's optional.
Or, to put it another way, generative syntax is fundamentally deficient because it does not tag POS's appropriately.
Did the kanji go to China? I see what I believe to be a red panda. I also see a jar of Cinnamon, which was also found in China.
It could also be Egypt, as cinnamon was found in Egypt "as early as 2000 BC" (Wikipedia) and 2001: A Space Odyssey wikipedia page discusses a monolith appearing in ancient Africa, 3 million years BC. Cinnamon was used in Egyptian tea. I strongly believe that Kanji went to Egypt.
The only other possibility I can think of is India, as it has the dhole and Red panda, has a cinnamon export, and has its own species of elephant, which is on the cinnamon jar.
+Spencer T (branefreez) Egypt is a good guess! Actually, they all are. And you're right, the book, the cinnamon, and the animal are clues - although it's actually a mongoose, not a red panda. The tag language was also a clue. India's pretty close. What do you think? ^_^
CAT! :D
+Inez Allen (Valkiro) Yeah, so, we wanted to do our April Fool's fun thing again, but a lot of people found all the stuff happening on the shelves last year too distracting. So this year, we figured if we took the cat into our filming space and let her climb onto the shelf for a take, that'd be a little something, and less distracting from the rest. ^_^
+The Ling Space is that a bag of rice behind you? I can't get past that lol. I also heard that Cedric Boeckx's Syntatic Islands is pretty good resource. I'm Jay by the way.
+axisaudio Hey Jay! Yeah, it is a bag of rice, since we were trying to have a different set of props for something different than usual. And yeah, Boeckx's book on the topic is definitely worth a look - it goes a lot more in depth into a lot of these issues. ^_^
+The Ling Space Thanks I picked up a copy. As a lowly philosopher my interest in this subject came about in a strange way -- from philosophy of biology believe it or not (which I skipped out of graduate classes to teach myself). Enough about me 😊. Just one more question: have you read Chomsky and Berwick's new book Why Only Us? Keep up the good work.
+axisaudio If that's where you're coming from, maybe you'd be interested in Boeckx's more recent work on biolinguistics, as well? Although I'll admit I'm not as up on philosophy of biology, but I do think it'd be interesting for you.
I haven't read Why Only Us yet; I bought a copy of it a few weeks ago, but I haven't had time to tackle it yet. At the least, I'll write up something about it for our Tumblr once I do - it should be thought-provoking! ^_^
Why do English speakers put prepositions in the end of the sentences? I have always wondered why and think that that'd be a good subject for a video.
For example:
In English you usually say "where are you from?" rather than "from where até you?".
In Portuguese we say "DE onde você é?" and not "onde você é DE?"
I like this, though haha
Great video as always!!
(Sorry for my bad English)
Sometimes it can sound weird, cant it
Like
"Whats going on on the second floor?"
Or
"What did you bring that book that I don't want to be read to from out out about "Down Under" up for?"
I love English xd
+Pedro BR You're refering to preposition stranding. The point is that English speakers do NOT simply put prepositions at the end of sentences, but rather the objects of prepositions become fronted , which results in prepositions being left behind. en.wikipedia.org/wiki/Preposition_stranding
+Wilandovsky thanks!
+Pedro BR Glad you liked the video! And your English is fine, I have no trouble with getting your point. But it's as +Wilandovsky said - it's just that English allows you to move away the complement to the preposition, and leave the preposition behind. So it looks like it was just put there on its own, but it was left behind by the movement. A lot of languages disallow this (Japanese, for example), but English is totally fine with splitting the phrase up. ^_^
Did the cat go to Sri Lanka?
first?
What an utter nonsense all this Generative Syntax stuff is.