What Does Possession Tell Us About Syntax? Determiner Phrases
ฝัง
- เผยแพร่เมื่อ 4 ต.ค. 2016
- How do we show possession in our syntax? And what does that tell us about representing nouns in our trees? In this week's episode, we take a look at determiner phrases: why we need them on top of noun phrases, how they turn some of our ideas about syntax inside out, and what they let us display in trees that we couldn't before.
This is Topic #79!
This week's tag language: Danish!
Related episodes:
What's the Structure Beneath a Sentence? X' Theory - • Syntactic Trees and X'...
What Makes a Basic Sentence? A History of Clauses - • What Makes a Basic Sen...
Last episode:
What Do You Start with in a Third Language? L3 Acquisition - • What Do You Start with...
Other of our syntax videos:
Where Do Subjects Start Off in Sentences? The VP Internal Subject Hypothesis - • Where Do Subjects Star...
What Questions Can't We Ask? Syntactic Islands - • What Questions Can't W...
What Changes in a Sentence When We Swap Verbs? Raising vs. Control Verbs - • What Changes in a Sent...
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-79/
We also have forums to discuss this episode, and linguistics more generally.
Sources:
Much of the inspiration for the episode comes from Andrew McIntyre's mini syntax textbook: www.angl.hu-berlin.de/departm...
The DP hypothesis itself is usually attributed to Steven Abney's 1987 MIT thesis 'The English Noun Phrase and its Sentential Aspect' (www.ai.mit.edu/projects/dm/the... or www.vinartus.net/spa/87a.pdf)
The Dutch example came from this paper here: the Dutch example, there's this chapter isites.harvard.edu/fs/docs/icb....
See you all in two weeks!
I've been waiting for you to talk about the DP hypothesis! I like how it accounts for Italian possessives: il mio cane, for example. Literally "the my dog".
Yeah! I think the thing is that we just have to lay groundwork before we can talk about some of the stuff that's more exciting to us. But it's good to be covering this stuff now. Glad you liked it! ^_^
I love this series, even though it goes over my head half the time
Thanks for the comment! Going to reply in more detail to the other one, but I'm really glad you like our series, even if it's tough sometimes. I really appreciate it. ^_^
As a programmer, I had a decent understanding of syntax trees before coming to this channel. Even though you have some really good videos on x-bar theory, I never really grokked it when applied to natural language. For some reason this episode finally made it click!
Awesome! Really glad to be able to help with that. I feel like sometimes it takes a different perspective to get it there for me, too. Thanks for letting us know. ^_^
I LOOOVE your channel, it's been really helpful to my linguistics major. Any chance you could talk about spectrogram reading?
We have a few videos on the basics of it! So what formants are, how to look for various sorts of consonants, and for vowels. But we haven't done a ton of practice with specific words, if that's what you're looking for! But for what we have, check some of the later phonetics ones on this playlist: th-cam.com/play/PLfLdA1jGDSu6exdSf9yQJWKgNqPviO4b4.html
A complete tangent: around 4:20 I got reminded of Finnish (surprise surprise) where some locational and temporal nouns can be compared with the meaning of "closer to". Like "beach-er" means "closer to the beach" as in "I'd like to live closer to the beach" could be said "I'd like to live beacher" and "the beachest house" is "the house closest to the beach". Also works with the word evening; "I'll call you later in the evening" is often "I'll call you eveninger".
Also consider a video on autosegmental phonology.
Thanks for the comment! Good to hear from you, as always. ^_^
I like that usage! Finnish has a ton of interesting stuff going on. I think I've used the superlative in English to mean "most like", so like "the beachest house" would be "the house most like a beach", but that's definitely non-standard.
Out of curiosity, since I'd more been thinking of doing an Optimality Theory episode, is autosegmental phonology of more interest?
¿por que no los dos?
***** The point was not to make grammatical sentences in English. Doing that would obscure the original Finnish meaning.
Seasons too... spring and autumn usually.
I've long been thinking of the genetive case as a kind of adjective. But as a kind of determiner makes a lot more sense.
Now, how does this look in other languages?
... I actually yelled in surprise when I saw your beard. it looks awesome btw!
Thanks so much! I've ended up keeping it (as you've probably seen in the more recent ones). Hope you didn't scare anyone with the yelling!
That beard looks badass.
Thanks! Still not 100% sure how I feel about it, but the response has been great. ^_^
I agree, very nice.
that's a goatee. not sure how i feel about it. do some mountain hiking and let it fill out I'm sure you'll feel happier about it :-D
While I consider the DP hypothesis valid, I don't think the part about Angela applies. The phrase "Angela a friend" is semantically equivalent to "Angela to be a friend" in this context, and I think the whole phrase "consider X to be" is a copular expression. So, "He considers Angela (to be) a friend," is really NP Copula DP, where the DP is only "a friend". Does this mean that the whole verb is "considers Angela to be"? Yes, and the merge function in Chomsky's generative grammar minimalist program would account for that.
Sorry for the delay in replying! Here's what we have to say here. So, generally, copulas are considered to be 'linking verbs' which have little to no meaning of their own. In English, this corresponds to the various forms of the verb "be." It's a 'linking verb' in the sense that it links subjects (typically noun phrases) up to their predicates (typically verb phrases, or adjectives). And you're definitely on to something in noticing that "Angela a friend" and "Angela to be a friend" have essentially the same meaning.
Now, sometimes, the label "copula" gets applied more generously, to verbs which do contribute some semantic content; words like "seem," "appear,"
"get," "become," and even "smell" and "taste" (along with many others). But even in the extended sense, "consider" doesn't easily fit into this class. Conceptually, "consider" isn't so much linking the subject "Angela" with the predicate "a friend" as it's linking the subject "He" with the idea that Angela is a friend (think about the roles that each part has to play: there's someone doing the considering, and something -- some idea -- being considered). Intuitively, it seems that the material following the word "consider" forms a kind of unit, which we've claimed is a DP. And this actually lines up nicely with other uses of the verb, since a simple non-clausal DP can also follow "consider," as in "consider her proposal," as well as full clauses, as you point out.
"Consider" just happens to be one of many verbs which allow for a wide range of complements -- a finite clause (a), a non-finite clause (b), a clausal
DP (c), and a regular DP (d), alongside a couple other categories of small clause (e).
(a) We must consider [that we only have enough money to do this once]
(b) I consider [the matter to be out of our hands]
(c) She considers [this a necessity]
(d) You should consider [the report]
(e) They consider [him lazy/above reproach]
With all that said, we do acknowledge in the extra materials that analyzing the words following "consider" as a DP isn't quite enough, since we can say something like "I consider Angela Elliot's friend." Our theory has no way of accommodating both the subject of the small clause "Angela" and the possessor "Elliot." So, some extra structure is needed to account for these sorts of sentences. But, that'll have to wait until a future episode! ^_^
Thank you very much for making this episode of _the Ling Space_, Moti Libermann, Adèle Louise Prévost and all the others! So, English likes it’s possession clitic but does English have any other clitics? Also, which languages tend to favour clitics? Germanic languages? Slavic? Bantu languages?
Thanks for watching! And yeah, there are a bunch of other clitics that exist in English - for example, the abbreviated forms of will (so 'll), am (so 'm), not (so n't), etc., those are clitics as well. My understanding is that most languages have some clitics, but there are some that do use them more extensively - I think you do generally find them more in Slavic languages, and Japanese is definitely full of them. ^_^
Thank you! I speak Russian, so I will keep an eye out for those from now on.
Great video! Also, keep the beard!
Thanks! And I have. ^_^
I've said that this series goes over my head half the time. Let's break that down a little more. One thing I've noticed in trying to learn things online: there seems to be a tendency to assume that people know the very things they're trying to learn in the first place.
A determiner can't stand on its own, it can only exist either as an isolated word to illustrate its own existence or as part of a noun phrase - so the very idea of a "determiner phrase" is inherently redundant, since a determiner is by its very nature part of a noun phrase. Illud in Latin and its various delcensions and derivatives, das in German and its various declensions and derivatives, ha, ga to, mo, ni, no, wo and similar case markers in Japanese, they can't exist as stand-alone phrases and still carry meaning, so why mark them as phrases unto themselves?
Why do determiners end up above the noun they refer to, but other modifiers typically end up below the term they're modifying? Why is the object considered part of the verb phrase when languages exist in which it's possible for the subject to separate the verb from its object?
What does the tree for a sentence in classic Hebrew look like - a VSO language where determiners are clitics rather than standalone words?
I feel like you're handing me a 700 piece jigsaw puzzle with 900 pieces in the box.
So let me first say, we've been thinking about this a lot in general. When we started the channel, we wanted to have each episode be able to serve as a standalone topic, where maybe you'd only need to watch max one of our other videos to get it. But over time, we relaxed that idea so we could cover more complex topics, and sometimes I think maybe we don't really strike the balance well enough - we do try to give some background in the video, but maybe it's not enough.
For the specifics here: The DP Hypothesis is definitely a tricky thing to get your head around; it runs against most people's intuitions that determiners form part of the noun phrase, and not the other way around. It's good to remember, though, that our intuitions don't always lead us to the right results (think of strange theories like quantum physics or relativity). What we want is a theory that accounts for as much data as possible, while making as few assumptions as it can, all while staying consistent.
To tackle your question, we can split the DP Hypothesis up into two parts: determiners form phrases, and those phrases contain noun phrases. While the second part of the theory is more controversial, the first part is generally accepted, even by people who don't take the whole thing overall to be true (like this here: www.lel.ed.ac.uk/%7Egpullum/grammar/detv_sli.pdf). Notice, for instance, that determiners can be modified by words like "hardly" and phrases like "just about", as in "[hardly any] hackers" and "[just about every] client." In each case, the determiner forms a DP with some other words, which in turn appears to modify the following noun.
But there's no real home-run argument for the second part of the hypothesis. Instead, there's a wide array of converging evidence from multiple areas, which many find convincing. We offered some evidence in our episode, but we would encourage you to have a look at the extra materials over at www.thelingspace.com/episode-79, which go into even greater depth on why it's such a handy way to think about the structures we've been calling NPs up until now.
To give a taste, notice how determiners exhibit a kind of 'transitivity' with their nouns, in the same way verbs do with their objects; so, some are obligatorily transitive, as in "the employee," some go both ways, as in "I've seen these files" or "I've seen these," and some are intransitive, as in "He sent them" (if you accept that pronouns are a kind of determiner). Without supposing that NPs are actually the 'objects' of DPs, we have no real way of accounting for why this kind of variation (both within and across languages) depends more on the determiner than the noun phrase that it's traditionally supposed to be a part of. And as you can see, just like intransitive verbs, determiners can stand on their own and still carry enough meaning to get the job done! This even seems to be true in German, at least in the following context (a), where the determiner "den" can exist independently (b).
(a) "Ich Idiot habe den Schlüssel vergessen"
(b) "Ich habe den zuhause liegen lassen"
Some of the rest of the stuff you bring up skirts into other areas - like, for example, most of the Japanese markers you bring up aren't determiners at all, but instead postpositions (like ni) or conjunctions (like to) or case markers (like ga or wo). But just because something is a clitic or an affix doesn't mean that it can't be the head of its own phrase, regardless.
For the Hebrew: I'm not 100% sure on the question here, or how it relates to the point before it, like, if there's some connection you see between the question on VPs and then on clitic determiners in Hebrew? If so, maybe try this handout, which draws parallels between clauses and possessive noun phrases in Modern Hebrew, and touches on the nature of determiners in the language, too: www.bu.edu/linguistics/UG/course/lx523-s01/handouts/SyntaxII.3.DP.Ritter.pdf
Thanks for the input, and thanks for watching. We'll keep trying to do better. ^_^
Thank you
So apparently there's some controversy about the "DP theory". Maybe that's just one of the settings of a language - whether it is "DP-dominated" or "NP-dominated"?
Yeah, this is one of the hypotheses about how to address this, assigning it a parametric setting. So for languages that don't appear to have a DP, we can just say that they actually do lack it. There'd be a lot of evidence for whether you were in a language that has DPs or not. Personally, I view this as the best option for keeping the DP hypothesis. There are other proposed fixes, but having it just be parametric is a pretty clean fix. ^_^
I am super confused by the short clause part. How is there a predicate without a verb?
But Elliot doesn't use computers! That's Hardison's job!
*adds Leverage to the possible episode theme list for the future*
Is it crazy how saying sentences backward creates backward sentences saying how crazy it is ?
I have never seen that before, but that is really pretty awesome. ^_^
I've just realised that the phrase _The director of technology's disappearance_ can mean the disappearance of the director of technology, or the person responsible for directing the disappearance of technology.
Yeah, it's weird how many things turn out to be ambiguous when you think about them a bit more. We just let the pragmatics of the situation carry us through a lot of the time. It's something I always marveled at when I'd teach these parts of syntax - how hard it can be to construct something that isn't ambiguous at all.
I wish I could identify your accent. Western Massachusetts? Or somewhere in Connecticut? I'm usually good at this kind of guessing . . . I figure that you changed your accent in college.
I kind of bounced around a lot, which makes it more difficult. I grew up in New Jersey, went to university in Maryland, and now live in Montreal. So it's kind of an amalgam, I think.
Thanks for satisfying my curiosity. I really enjoy your videos. Keep it up!
Sure! And thanks for the kind words. We'll keep going! ^_^
Good afternoon dear users.
I would like you to tell me which is the correct order of this sentence:
Her daughter's first two names are Mary Jane.
Her first two daughter's names are Mary Jane.
Thank you.
The first of them is correct if he has one daughter whose name is Mary Jane. The second is grammatical, but only correct if he has two daughters and each of them are named Mary Jane. Hope this helps! ^_^
The Ling Space thank you so much
YOU GOTTA TELL EM!
NOUN PHRASES ARE DPs!!
SOYLENT GREEN IS PEOPLE!!
We'll get it out there! ^_^
Watching several videos where you've got no beard and suddenly you have one, I got so confused I thought it was someone else X'D
Sorry for that! Over time, there'll be more and more videos where I have a beard, so hopefully it'll even out? ^_^
i cant tell if that was an undertale reference or not
you like homestuck so probably?
We did an Undertale episode where there was determination, but this was more of a Mr. Robot thing. Good idea, though. ^_^
u have very little on acquisition
We have 12 videos on it so far! I hope some of them are good for you. Check the playlist here: th-cam.com/play/PLfLdA1jGDSu79MMbAN3LN3VSUr0OwOGOg.html