Semantic Scope Ambiguity
ฝัง
- เผยแพร่เมื่อ 21 ต.ค. 2014
- Why do people interpret the same sentence multiple ways? What is it about semantics that leads us to more than one meaning? This week, The Ling Space takes on semantic scope and talks about how the most innocent-seeming words in your sentence are fighting it out to bestow upon you an interpretation where they come out on top, as well as how we avoid being lost in an ambiguous fog all the time.
This is Topic #8!
This week's tag language: Czech!
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-8/
We also have forums to discuss this episode, and linguistics more generally!
Looking forward to next week!
"Why don't we feel confused all the time?" That struck me on a level.
Best introduction on formal semantics on TH-cam. Thanks a lot.
Thank you for helping me get back on track before my semantics exam!! My professor's explanation of scope didn't quite *click* in my brain and I might have skipped the past many lectures during which I had time to ask him about it.
I like this example sentence:
"All Quentin's friends went to a doctor."
The ambiguity upon reflection still exists. But 'a doctor' is a more mundane, everyday thing vs a fantasyland. So this example makes it more readily clear that the peculiarities of a fantasyland (such as not existing) aren't causing the ambiguity.
Yeah... I think this may have been one of the times where trying to use fandom examples maybe didn't do the best job for making the point clear. I think we do better with this now, hopefully.
There may be a twist to this in what one might call a parable where the indirect scope is the one that has the dominant scope in that the giver of the parable does that to make one think more deeply about the meaning of the parable..
Can you tell me what a wide scope vs narrow scope reading would be? I am hoping to do some work on prosody in relation to the Epistemic Containment Principle for my undergrad and something I have found a hard time understanding is scope. I have taken an intro to syntax course where where we learned about Generative Grammar from an A. Carnie textbook, but we didn't touch on scope. I figure you probably covered it in the video without actually using the terms "wide scope" and "narrow scope" , both terms I frequently find in the academic literature but can't seem to get a handle on.
The rules we use to process meaning of sentences (scoping) are really different than the ones for order of operations in math or symbolic logic. Which makes sense, you really need more precision in the latter, but it was interesting for me to discover how we actually do have systems for language to, they just rely on context and rules of conversation covered in a prior video.
Greg Sanders Thanks for the comment! Yeah, this is really an interesting point. There are rules to be obeyed within pragmatics, that we assume people are using, even if they can be flouted if people want. The core semantics system itself, though, actually does work from a pretty formal logic kind of position - interpretations are often written out in terms of lambda calculus, and formal semantics classes often include a lot of set theory. That said, though, because of how syntax and semantics interact, you do end up with these kinds of ambiguities. And I feel like that's a worthwhile price to pay. It's just good to look at. ^_^
First video I´ve seen from you and you say "goodbye" in my native language? Destiny? Na shledanou!
Thanks a lot great job! merci bien
Hey! Really a fan of this topic :) One thing that's been bugging me of late is coming across the expression "An experience of a lifetime". I was always under the impression it should be "The experience of a lifetime" or it wouldn't mean anything (because too vague, could have so many of them, so nothing special at all). I never used to hear the former until recently, and I thought the latter was the only one valid as an expression. Is the former one a thing now? Is the language evolution boat passing me by? Please enlighten me!
emroyka I agree with you that "an experience of a lifetime" is pretty vague! In a way, everything that we do is an experience of a lifetime in one sense. The use of "the" does carry with it an idea of uniqueness, so "the experience of a lifetime" would definitely be quite memorable!I do feel, though, that I still get an interpretation of something great and wonderful with both of them being "a"/"an", and that makes me wonder whether "experience of a lifetime" might be becoming like a bit of an idiom, such that the phrase itself now has the meaning that it's a wonderful time, or at least something really memorable.I'll keep an ear out for this now, though! Thanks for the question. ^_^
Thaaaaaaank you for making my BA thesis research easier.
Cal Drogo Glad to be able to help! We hope your research goes well. ^_^
Is language primarily deductive, or would it be abductive, or inductive?
Hi, I am asking if the sentence (john bought a flower for mary )is ambigous or not?
Yep, It is syntactically ambiguous.
Thank you for making this channel, it's very helpful! However, I have a question I was hoping someone could help me with. I'm not 100% sure if it has to do with scope or not, but here it is:
Why do we intuitively understand sentences like "every free man and every free woman is a citizen", without having to spell out "every free man is a citizen and every free woman is a citizen"?
It's not really a scope issue, but a coordination issue! So the subject of the sentence here is just "every free man and every free woman". You can have coordination of any two like objects, and "every free man" and "every free woman" are the same kind of phrase (either DPs or NPs, depending on your framework). So then the whole thing just operates as a big subject chunk to go with "is a citizen". So we'd expect "is a citizen" to apply to both of the conjuncts from the structure of the sentence. Hope this helps!
Wow, thank you very much!
@@thelingspace it seems the subject is plural so why isn't it "every free man and every free woman ARE a citizen"？😭
Someone might use terms like signifier in a cultural studies class, or referent in a philosophy class but semantics is really a complex science.
What about the piece that verbal emphasis plays in it?
Diana Kennedy It's definitely true that emphasizing one word or another can push us towards one interpretation or another! I'm thinking you probably mean like stressing the "a," so like "Everyone went to A fantasy land." But this doesn't eliminate the ambiguity entirely; it just tells people that you're trying to promote a particular reading. That's particularly helpful if you don't want the most broad interpretation, but you can still perform the same trick of thinking about it and finding the ambiguity in it. This sort of ambiguity isn't resolvable from just the sentence itself, but people are definitely susceptible to being pushed one way or the other.But you're right that those sorts of extra cues, from pragmatics or from prosody, really can make a difference in interpretation, and we definitely do use those to let people know what we want. Thanks for the question! ^_^
Hello! I am studying semantics and was wondering if you'd be able to answer a few questions via email. I need to conduct an interview for my course, in addition to understanding more about my topic. Thanks
i love quantifier raising
Out of curiosity, are there terms for more complex orders of dominance when sentences have three or more quantifiers? Is that even possible?
12tone You can definitely have interactions between three or more quantifiers! Really, our sentences are full of the things, since every "a" or "the" or "some" or "all" or "none" has that power. But all the interactions between them can be tricky to separate: when you have 3 or more quantifiers interacting, you get lots of equivalencies and entailments between the various scopes. So even if you have 3 quantifiers, you don't necessarily get 6 fully distinct interpretations. And the terms don't really change, either; we still can talk about surface vs. inverse scope, or one quantifier taking scope over another. Hope this helps! ^_^
Cool, thanks! I've really been enjoying catching up with these so far. I met Adele at VidCon and the channel sounded really interesting, and I've learned a lot from it already. Keep up the good work!
12tone Thanks! Glad you're enjoying them. Let us know if there's anything else you'd like to hear about! ^_^
Well, I do videos about music theory and composition, so I'd be fascinated to hear a linguistic perspective on how we learn and process music, given that melody and harmony have some language-like communicative functions. I don't really know how much linguistics has to say on the topic, but it'd be awesome to find out!
12tone Cool! Yeah, there are interesting overlaps in music and language stuff; for example, brain reactions to violations of phrasing in music are similar to those of grammatical violations in language. There's a bunch of research going on in here these days! I'll add it to the list. ^_^
On the other hand, if they all went to fantasy lands, that eliminates one possible reading.
Not to mention if this sentence was spoken rather then written, it could mean "Fantasy Land" in capitals and might refer to a theme park...
+Frahamen Yeah, we can find even more ambiguity out there. We were focusing on semantic scope ambiguity here, but you could definitely do similar things for syntactic ambiguities, or for lexical ambiguity. That we're so able to work out what people mean is really amazing, when you think about all the possible complications.
Ahhhh Bee and Key!!!
Haha, yeah. We're all about Brakebills this week. ^_^
Logically, "a" has scope over "all" by merit of two facts: 1. "all" quantifies by type, not by token - to quantify by token, it would have to be "each", and 2. both the use of the article "a" and the singular rather than plural inflection make the object explicitly singular. For the alternate interpretation to be logically valid, it would have to be either "Each of Quentin's friends went to a fantasy land", mapping each instance to the object individually, or "All of Quentin's friends went to fantasy lands", making the object explicitly plural. If half of Quentin's friends went to Oz and half to Wonderland, then they didn't all go to a fantasy land, because Oz and Wonderland can not be described collectively as a singular fantasy land. For it to be logically ambiguous whether it's referring to one fantasy land or more than one fantasy land, it would have to be "Each of Quentin's friends went to a fantasy land."
Bottom line: "all" is not distributive, "a" and "land" are singular. Therefore, your argument is invalid.
+notoriouswhitemoth Here, I don't think I really agree. This is gonna get pretty long, though. Acknowledging that there'll be interpretive differences between and within speakers, a plural object certainly isn't required for surface scope; moreover, the availability of inverse scope can't possibly be because "all" is necessarily non-distributive (because it definitely isn't).
First, we see what are - I think - uncontroversial examples of surface scope involving singular objects. Though (1) is technically ambiguous, the inverse scope reading (where everybody wore the same tuxedo) is fairly implausible.
(1) All my friends made sure to wear a tuxedo to the wedding.
In fact, if we pluralize the object, we introduce another (potentially unwanted) ambiguity, where it now becomes possible - though not particularly plausible - that each of my friends wore more than one tuxedo.
(2) All my friends made sure to wear tuxedos to the wedding.
So it looks like plural objects aren't needed for surface readings, here, and employing them in the service of forcing other readings can even backfire - leading to strange results.
Additionally, "all" cannot be exclusively collective; it has to be somewhat distributive. Although it's compatible with certain collective predicates:
(3a) *Each of the girls gathered in the hall
(3b) All of the girls gathered in the hall
It doesn't work with others:
(4a) *Each of the girls were numerous
(4b) *All of the girls were numerous
These facts, taken together with the (hopefully salient) distributive reading that "all" receives in (1), suggest a more complex story.
Having said that, it's fair to spend a little bit of time recognizing just how complex that story must be. For one, "all" and "each" (alongside "every" and "any") have very different syntactic distributions, in spite of their intuitive interrelatedness:
(5a) Every student read 1Q84
(5b) Each student read 1Q84
(5c) *All student read 1Q84
(6a) *Every of the students read 1Q84
(6b) Each *(of the) students read 1Q84 (where "of" and "the" are obligatory)
(6c) All (of the) students read 1Q84 (where "of" and "the" are optional)
(7a) 1Q84 was assigned to each/every student
(7b) 1Q84 wasn't assigned to each/every student
(7c) *1Q84 was assigned to any student
(7d) 1Q84 wasn't assigned to any student
And, of course, there are corresponding differences in meaning. Capturing the entire picture, though, would require delving pretty deep into the semantics of distributivity/cumulativity/collectivity, as well as the semantics of plurality and the mass/count distinction, both of which make use of some fancy math.
It even means digging into event semantics, as sentences like (8) have been argued to have at least three (though I think more) distinct readings - where everybody sang the same song at the same time, where everybody sang the same song at different times, and where everybody sang different songs at different times.
(8) All the soldiers sang a victory song
And all of this is without even wading into the entailment relations between different scope readings, which can really gum things up, as well as cases where we might naively expect multiple readings, but find only one:
(9a) A girl admires every teacher (both scopes possible)
(9b) A girl who admires every teacher recently graduated (only surface scope possible)
(10a) I gave a book to each girl (both scopes possible)
(10b) I gave a girl each book (only surface scope possible)
The discussion gets very thorny, very quickly! We do plan on dealing with at least a couple of these issues (syntactic islands and negative polarity items) at some point in the not-too-distant future, and our next semantics episode (coming soon!) will definitely cover some of the more fundamental issues involved in natural language quantification, and explore why the tools of predicate logic just aren't enough to account for all the variability and creativity we see in language. ^_^
(Edited to fix some formatting issues)
The Ling Space t
I didn't think that pennies had hair. I appreciate that u are thinking about this stuff but please realize that every word has multiple meanings.
+Michael Byrd Ah, yeah, we were making a reference to the character from the Magicians - that was the theme fandom for the episode. But we do try to keep it from getting confusing. Sorry if it didn't work here!
Y
This is a good explanation, but your example doesn't really work for me. The next interpretation I think of is along the lines of "all of Quentin's friends daydreamed", where the scope of "a" isn't an issue. I guess metaphor is ruled out by your initial assertion that we convert sentences into logic, which is fine if you're talking about a model, but not so great if you're talking about peoples actual experience.
+Lucas Wilkins Yeah, that leads us into just "fantasyland" as a different lexically ambiguous morpheme. Personally, with the "a" there, I don't really have access to that reading, but if you can get it, then sure! But coming up with other sentences with scope ambiguities should work pretty well, I hope - how about "every horse didn't jump over the fence", to mean either "no horse jumped over the fence" or "some horses jumped over the fence, but not all of them"? Those'd give you the same two scope options.
+The Ling Space Yeah, that's a far more concrete example. Though, can't you just say "all the horses jumped over a fence" - I guess you're wanting it to have a very clear preferred interpretation, rather than one that is about 50/50 (or so it seems to me).
+Lucas Wilkins Sure, your sentence has the same scope ambiguity - it works fine for me! And it's closer to the one in the video. I just thought I'd use one with negation instead. But they both illustrate the point pretty well, I think. ^_^
Lucas Wilkins
So is this the reason why people have trouble solving riddles like "The man was afraid to go home because the man with the mask was there"? The most obvious meaning is not the intended one at all.
Beverly Barkon That's actually a different kind of ambiguity than the scope ambiguity that we're talking about here. That one hinges on lexical ambiguity, and in particular, the two meanings of "home." It's hard to get the meaning of the riddle because the much more common meaning of "home", like the place where you live, is what you try to interpret it around, and the whole scenario - is he a burglar? Is he an assassin?! - seems plausible.You need to actually be thinking that there's a trick to get the other meaning of home out, and realize (RIDDLE SPOILERS) that this is a baseball scenario, and that the guy's trying to score, but the catcher is there. We can work this out once we know that we're supposed to be able to interpret this, but it's based around the fact that we need the less-common reading of the word. So it's a different kind of ambiguity, but it's still a common one, and one worth accounting for. Thanks for the question!
Linguistics is very interesting, but what is its real usage in the real world?
President Trump supports this message.
just change the sentence to "All his friends went to their own fantasy land" if thats what you want to say, christ.
I'm going to sound sarcastic now, but no sarcasm is intended.
But what if several of them, but not all, went to, say, Middle Earth? And a couple of other ones went to Hogwarts?
In your example, it sounds like each and every single one of them went to their own fantasy land and no one else went there except for one person per fantasy land. The sentence did not imply an even distribution.