Syntactic Movement and Traces
ฝัง
- เผยแพร่เมื่อ 17 ก.พ. 2015
- What elements move around when we make our sentences? Is it the same in all languages? In this week's episode, we look at syntactic movement: how we know that words move around, what remnants get left behind, and how we can use these phenomena to explain surface differences between languages.
This is Topic #23!
This week's tag language: Indonesian!
Find us on all the social media worlds:
Tumblr: thelingspace.tumblr.com
Twitter: @TheLingSpace
Facebook: thelingspace/
And at our website, www.thelingspace.com!
Our website also has extra content about this week's topic at www.thelingspace.com/episode-23/
We also have forums to discuss this episode, and linguistics more generally.
Looking forward to next week!
![[TH] 2024 PMSL SEA GF D1 | Summer | แข่งที่ไทย ต้องใส่ให้สุด](http://i.ytimg.com/vi/Jkuf7nCS7jI/mqdefault.jpg)
This channel is pretty great. So much stuff to keep busy with. I always wondered if there was anyone out there actually trying to build the rules of language. This channel seems as close as it gets
You are just amazing. I finally understood why traces are necessary. Hello from Brazil!
Great series! Can I ask why you have a full negation phrase just above the VP? I always felt It was more intuitive (at least in English) to see the negation particle as going with the auxiliary (if I understand correctly, that would be as part of the IP I guess?).
Thanks in advance.
I was pleasantly surprised by Moti's fluent Indonesian phrase at the end of the video--"Sampai jumpa!" Didn't see it coming. 😁
I was checking a couple of linguistics lectures recently on TH-cam and you have such a unique and nice way of presenting compared to ALL of them :D. I wanted to ask you why did you feel like teaching people about "syntactic movement" from the transformation grammars versus than something like feature passing in unification-style grammars.Is this transformational approach still relevant to this day? I feel like when it comes to teaching people basic intuitions about the structure of language "movement" and "movement paradoxes" are equally intuitive.
loud and clear, finally! thank you very much :D
Interesting to learn that sentences leave traces of themselves when reformulated. So essentially if you hear part of a sentence and reformulate it as a question, then the structure of the original statement puts some limits on what questions you can grammatically ask.
While I'm not the biggest Jasper Fforde fan, I could totally see him writing a great forensic linguist detective short story with that bit of information as a jumping off point.
Greg Sanders That's a really cool idea! I'd definitely like reading that story. And yeah, that is basically it - sentences can't be manipulated any which way once you've made them. They still have to obey particular rules. And there's actually a decent number of rules that are to be obeyed. But these limits on movement are definitely big ones. ^_^
I am both a dragon hunter and a fox spirit.
Great work man! Exactly what I've been looking for.
your videos bring me so much joy
love your videos! Can you make one about A-movent? It would be really helpfull!!
I followed a link from Gunnerkrigg's twitter and I thought this video was really interesting :) I will say, the cuts between each sentence were disorienting to me, but the information was fascinating. I can tell you're genuinely excited about the topic, which always makes for a much more enjoyable learning experience. Keep up the good work!
dandette Thanks for the kind words! We do love our linguistics around here, as well as our Gunnerkrigg Court. We hope you stick around with us! ^_^
i didn't even realize there were any cuts before i read your comment. now all i see are the damned cuts!!! :D
Really informative n helpful...
Amazing Teacher
Hi Moti, You and Chomsky have both given examples of sentence trees where the tense marking is on the verb phrase. But as I understand there are also languages which use the noun phrase to mark the tense. How do these languages affect the tree when words get moved around? If the verb can move without leaving a trace for the tense, can verbs move more freely?
+Jim Newton Hey! Thanks for this question. We had to do some digging for this one, and here's what we (mostly our staff writer Stephan, to give credit where it's due) came up with.
Like many areas in linguistics, inflecting nouns with temporal information (see Nominal TAM over on Wikipedia) turns out to be a fascinating and complex topic, and there doesn't seem to be a real consensus about the best way to think about what's happening there yet. Which may be in part because, while descriptive grammars have known about it for a long time, there hasn't been a matching effort to analyze (or even acknowledge) it too much.
To start off with, it's worth noticing that even English, if you look at its collection of temporal and aspectual suffixes, can still talk about time in ways that go way beyond the verb. For example, the sentence in (1) below is in present tense, but the prepositional phrase at the end manages to work with the verb so as to put the whole event in the future (and adverbs can act similarly).
(1) We enter the forest at sundown.
And noun phrases can use adjectives to specify whether the thing that's being referred to is located in the past, or the future:
(2) The former / future Robot King
A more complicated question is whether we ever see nouns acting like verbs when it comes to tense, and whether this happens on top of - or even instead of - marking the temporal info on the verb. While we can definitely see what look like tense markers operating locally inside noun phrases in some languages (e.g. Tariana), some of the researchers looking at this phenomenon argue instead that in languages like Kayardild, tense marking on the noun actually contributes to the overall interpretation of the clause alongside the verb; in other languages, like Sirionó, the temporal info associated with the noun may be the only indication of its kind throughout the whole sentence. Other researchers have argued that, if you really look at it in detail, these kinds of markers on the noun don't really function the same way that verbal tense does, and never can. Instead, they may be closer to the adjectives we saw in (2).
As for whether the verbs in language that display this kind of marking might be more free to move around because the noun's shouldering some of the temporal load... that's a very interesting idea, and it could make a really great research topic! It'd definitely mean taking a very deep and lengthy dive into the semantics of tense, aspect, and mood, as well as the relevant syntactic structures, which come from a wide swath of otherwise unrelated languages. Like, that's a doctoral thesis worth of research there, if not more. If you decide to do some more digging of your own, or if you know of someone who already has, please let us know what you find! ^_^
I was surprised when you said "sampai jumpa" lmao. Nice series, thanks.
very good and helpful!! thank you
Do you have episodes about historical linguistics/language change and etymology coming up?
Pakanahymni Yep, we've got one coming up in a few weeks. Probably late March to early April on our current schedule. Looking forward to it! ^_^
you are great , thank you for the explanation.
+Siham Kouram Glad to be able to help! ^_^
1:28 WOAH NEW CAMERA ANGLE
Pashto is my mother tongue, happy for it and respect for all langauges
Nice video! I have a question: the sentence 'Who will Annie should show the forest to Parley' seems to be bad for independent reasons: 'will' and 'should' both select an subject and a VP, and since there's only one element in subject position and one VP, it's not surprising that the sentence is ungrammatical. Are there other reasons to believe that movement is real (besides the 'wanna' argument, which I think is very good)?
+Dominiekske Thanks for the question! It took a bit to get through answering this, but here we are (and again, most of the credit for this response should go to our staff writer Stephan):
To get a bit more specific than we did in the episode, there are a couple of things going wrong with this sentence. For one, it looks like it violates the Theta Criterion, which we talked about back in Episode 31 (www.thelingspace.com/episode-31). If we assume that "who" gets a θ-role from the verb "show", and then moves to the front, the trace it leaves behind is kind of like a note, saying "I was here". Without it, the verb has to go looking for another noun phrase to fill in that role, or its verbal needs won't be satisfied. And then we end up with one too many people hanging around in the forest.
You're right to notice that there are also one too many modal verbs (i.e. both "will" and "should"). But while it's true ins ome sense that modals always appear alongside exactly one subject and one verb phrase, we don't really think of words like "will" and "should" as doing much selecting here. For example, which kind of subject is allowed to appear at the start of a sentence seems to have more to do with the main verb than the modal verb, even if the main verb is farther away. Like, notice how in (1) vs. (2) below here that interpreting "it" as not referring to anything is only okay in the first case, and not in the second, even though both sentences involve "might".
(1a) It might seem that fairies live in the forest.
(1b) * Annie might seem that fairies live in the forest.
(2a) * It might expect that fairies live in the forest.
(2b) Annie might expect that fairies live in the forest.
So, it looks like the main verb has more of a hand in selecting the subject than the modal does. The real problem is that most English sentences just don't have the room for more than one modal at a time. That trace that's left behind acts like a kind of warning to other modals, that entering the forest is now forbidden. Interestingly, though, some dialects, like Southern US English, do allow more than one modal at a time, as below (and the Wikipedia section on this is pretty good: en.wikipedia.org/wiki/English_modal_verbs#Double_modals)
(3) Annie might could show the forest to Andrew.
In this case, turning the sentence into a question ends up looking a bit like our original, ungrammatical sentence (with more than one modal at the front, and one in the middle).
(4) How could Annie might do that?
So the fact that other dialects of English only allow one modal at a time is sort of a quirk - if one dialect can do it, there can't be some universal prohibition on it.
As for other kinds of evidence we have for movement, consider checking out our latest syntax episode (that'd be www.thelingspace.com/episode-66), where we talk about examples that look like they involve movement being blocked, which is a kind of indirect test for it. We'll also be talking even more about movement soon, and about the semantics behind it, so be on the lookout for that, too. ^_^
+The Ling Space Thank you for the great explanation!
The stuff you're doing, including saving my ass, is saint-tier. Thanks a bunch
Good job! From Morocco ✌
Of course, traces can be reconstrued as direct association to verb valence (as in construction grammar). This has the distinct advantage of not having to posit invisible doohickeys. It’s more parsimonious and, I would argue, more in line with the psycholinguistic evidence.
Can you explain 'tough' movement? It's quite tough to understand it.
Another question. The examples you give are from languages which are highly word-order dependent. However, highly inflected languages like Russian, have much more liberty, about which order the words go. In some Russian sentences I can put the words in any order and maintain the same meaning. I'm not a native speaker so I cannot say whether a shade of meaning changes.
How does the existence of highly inflected languages affect the theory of syntax trees and the operations we like to perform on them in english-like languages?
Well, Finnish has somewhat flexible word order. First word gets emphasis if nothing else if the usual order is changed. (Usually a verb; since the conjugation form already tells who, the pronouns tend to get dropped.) Sometimes the meaning can change even more.
+Elderscrollsswimmer wrt Finnish, does the inflected verb tell you the person and number of both the subject and the object?
Jim Newton Verbs are conjugated according to person and number of the subject. There's even a passive form that specifically does not tell who, or even how many.
Object, if any, is marked by the case of the object. (and that can take two forms depending on things like how complete the action is).
I need some help... I don’t understand what triggers some words to have movement (mostly in question sentences). I know is a thing called “tagging” that appears in the feature notation but I don’t know when and where to put tagging ... I AM SO LOSTTT
Please help me😭😭😭
"Coyote laughs joyously." is a perfectly good English sentence meaning the exact same thing as "Coyote joyously laughs.". At least by brain seems to think it is, and I swear I've heard it both ways. Can anyone corroborate?
I'd be interested too!
As a native speaker of American English, I can confirm that both sentences are valid.
They are not however, equivalent.
"Coyote joyously laughs."
"Noun adverb verb."
This sentence is using "joyously" as a modifier for "laughs".
In other words, Coyote is presently in action "laughing" and that action is characterized by its "joy". This rather than some other emotion, such as panic.
"Coyote laughs joyously."
"Noun adverb verb."
This sentence is using "laughs" as a modifier for "joyously".
In other words, Coyote is presently feeling an emotion "joy" which is being expressed through "laughing". This rather than some other action, such as smiling.
A subtle difference, but one that might be important with context.
The confusion stem from how both the adverb and verb in each sentence is modified in the same way: so that their auxiliaries are in a present tense and a continuous aspect; forms which also are used when the words act as nouns.
This problem is solved by the fact English has a (mostly) fixed word order.
English doesn't have a case system, or any way to consistently mark on the word itself to which part of speech it belongs. Therefor we must rely on the sequence in which the words occur to parse them.
If you wish to juxtapose (put beside with no extras) an adform and a form (in this case an adverb and verb), you place the adform before the form.
Visually:
(1)Adform + (2) Form
(1) Adverb + (2) Verb
How do you account for the movement of prepositional phrases with x-bar theory?
Examples:
"To really feel the joy in life, you must suffer through the pain."
"In the beginning, God made the heavens and the earth."
Thank you so much .... I now I understand this .
+Immortal Songs Glad to be able to help! ^_^
Nice reference to Gunnerkrigg's Court on your t-shirt :)
But this might be the first and only video I couldn't understand a single bit :\
Many Thanks From Morocco !
+Meryem Raoui You're welcome! Glad to be able to help. ^_^
you did a really great video, even if you got the part about French negatives wrong. The French for "Jones did not smile at the students." is "Jones n'a pas souri aux élèves." The " n' " is the contraction of "ne". Despite the negation being carried mainly by the "pas", the "ne" is mandatory in writting, and can only be dropped in colloquial speech.
The reason the "ne" is here is historic. It comes from the Latin, "non", which became "ne" in French. There was a time when it was the only carrier of negation. Then phrases like "Je ne marche pas" or "Je ne vois point", meaning "I walk no step" and "I do not see a point" appeared, and gradually negation moved over to the words "pas" and "point". The "point" fell out of use some time ago, but the "pas" is still a required part of negation.
+7000000th Yeah... as we discussed in some of the other comments here, we decided to go with the more colloquial presentation of the French example so as to make clearer where the negation was located. If we just have "pas", then it's quite apparent. And as you note, it's fine to drop the "ne" in speech. But since we were displaying the tree, it struck some people as wrong, I think. But that was our thinking.
Your second point is one I really like, actually - French started with having only the "ne" part, evolved over time to require both "ne" and some post-IP negation, and it seems like we may now be moving towards the only required negation being "pas", if the colloquial formation continues to spread. It's like a long-term crossover. It's really cool. ^_^
I understand your thinking, but the fact is that dropping the "ne" is so colloquial that most French speakers would not consider it the rule, and that's mainly due to the fact that we put it in speech more often than we omit it. In fact, when talking to strangers or anybody important, the "ne" remains. And when in casual conversation, other parts of speech disapear or are abbrieveated, such as the sentence "Vous n'avez pas fait ça comme il faut." which becomes "Z'avez pas fait ça commi faut." Note that this happens only orally, and that any French would know what part of speech is missing.
7000000th I thought this was a syntax video, not one on etymology.
Oh, poop! I caught up and can't watch any more tonight. Sigh! Really enjoying this series, as you know!
Diana Kennedy Glad you're liking them! We have one more up now, on aphasia. But after that, you'll have to wait until next week. ^_^
The Ling Space Wait, what does aphasia mean again? Just joking!!!
what is Government theory sir??
👍👍👍👍
How come when you talk about French negation you don't include the ne? I'm by NO means fluent, but my high school French teacher would have taken off points if we had left it off.
Diana Kennedy Good question! Mine would have to. The answer is that many dialects of French, particularly in more casual conversation, leave off the ne these days. The interesting thing is that historically, it was the opposite: it was just the ne, and not the pas, that marked negation. And then we got the pair of them together showing negation, and now it's evolving towards being just pas. We wrote in some more detail about this on our Tumblr here! thelingspace.tumblr.com/post/111597235969/jespersens-cycle But the other reason we chose to represent French with just the pas negation is that introducing the "ne" part of it obscures the point we wanted to make about word order. Since many speakers do without the "ne," we figured it was kosher for us to use one of those dialects, and have the example be clearer. Thanks for the question! ^_^
The Ling Space cool. Thanks. I tried to like your piece on Tumblr, but it was going to make me sign up for an account and that scares me! I really loved your list of tags, though!
you should replace my professor :)
+Abdelilah Jennane Haha, well, I'm glad we could be of service, either way. ^_^
When you use example write at the top
That's why I think English is a Norman-Anglo-Saxon creole.
What a lot of "head movement" hahaha
I can't follow these explanation - it's too fast
GUNNERKRIGG COURT
ur headings are confusing
YOU ARE good but you speak very fast I hope you speak alittle bit slower
please try to speak slow