Some , speaker knowledge, and subkinds Andrew Weir University of - - PowerPoint PPT Presentation

Some, speaker knowledge, and subkinds

Andrew Weir

University of Massachusetts Amherst aweir@linguist.umass.edu

ESSLLI StuS University of Opole, August 15th 2012

Introduction Epistemic conditions on some Modeling the epistemic condition Some and subkinds Further work/conclusion

Introduction

I investigate the properties of English some paired with singular NPs. This triggers an epistemic effect: ‘speaker doesn’t know which NP is the witness’ (1) Some file on this computer is infected, but I don’t know which one/#namely, this one here. In some cases, though, the epistemic effect is different: ‘speaker knows which NP is the witness, but not what kind of NP it is’ (2) Some plant is growing through the wall of my room. I propose a semantics for some, following Alonso-Ovalle & Men´ endez-Benito 2010b (AM)’s analysis of Spanish alg´ un as containing an ‘anti-singleton’ presupposition on its prejacent NP, to capture both these readings.

Andrew Weir Some, speaker knowledge, and subkinds

Introduction Epistemic conditions on some Modeling the epistemic condition Some and subkinds Further work/conclusion

(Initial) generalization

A speaker uses some NP to signal that they cannot identify which member of NP is the intended referent. (3) Some file on this computer is infected, but I don’t know which one/#namely, this one here.

Andrew Weir Some, speaker knowledge, and subkinds

Introduction Epistemic conditions on some Modeling the epistemic condition Some and subkinds Further work/conclusion

Counterexamples

Cases where you clearly can ‘point out’ the intended referent: (4) (adapted from Alonso-Ovalle & Men´ endez-Benito 2003) [We are at an after-conference social event. Suddenly, we see someone – whom we haven’t seen before, but from whose age we can guess is a professor – dancing on a table.] — Some professor is dancing the lambada on the table! You might even know the name: (5) [As above, but the professor is wearing a namebadge.] — Some professor called Thaddeus Q. Smythe is dancing the lambada on the table!

Andrew Weir Some, speaker knowledge, and subkinds

Introduction Epistemic conditions on some Modeling the epistemic condition Some and subkinds Further work/conclusion Human/thing split Differentiation condition

‘Human/thing’ split

Some, however, seems very sensitive to whether its NP complement is a human or a ‘thing’. (6) a. Some professor is dancing the lambada! b. I saw some guy hanging about outside. (7)

• a. ??Some statue is in the middle of the square. [looking at it]
• b. ??I saw some building on my way through the desert.
• c. ??There’s some letter in my mailbox. [looking at it]

Andrew Weir Some, speaker knowledge, and subkinds

Introduction Epistemic conditions on some Modeling the epistemic condition Some and subkinds Further work/conclusion Human/thing split Differentiation condition

About ‘knowing who/what’?

There are lots of means of ‘knowing who’ (Bo¨ er & Lycan 1986, Aloni 2001): knowing a name, knowing someone’s job, knowing what they’re doing here. . . Maybe there are just many more such ways for people than for things? However, even when ‘things’ do bear more means of ‘knowing’ than are normal for ‘things’, this still seems inaccessible for ‘some’ For example, ‘don’t know the name’ can license ‘some’ for people, but this is much less easy for things, even things with names

Andrew Weir Some, speaker knowledge, and subkinds

Introduction Epistemic conditions on some Modeling the epistemic condition Some and subkinds Further work/conclusion Human/thing split Differentiation condition

About ‘knowing who/what’?

Two diplomats from Peru are delegates to a conference you are at. One is a man and one a woman. You see them both several times, and know that they’re both from Peru, but never catch their names. (8) At dinner, I was sat across from a/some delegate from Peru. You are lost. You know the town you’re in has only two squares in it, but you don’t know their names. You keep coming across both. You can tell them apart because one has a fountain and the other doesn’t, but you can’t see any street signs. You end up in the fountainless one. Your friend phones you: (9) A: Where are you? B: I’m in a/?#some square in the city.

Andrew Weir Some, speaker knowledge, and subkinds

Introduction Epistemic conditions on some Modeling the epistemic condition Some and subkinds Further work/conclusion Human/thing split Differentiation condition

Epistemic conditions on some

It seems as if the condition on some with things is simpler than with humans. (10) Differentiation condition on ‘some NP’ A speaker uses some NPthing to signal that she could not, if presented with the extension of NP, reliably differentiate the witness of the existential claim from everything else in the extension of NP. No claim is made in the above with respect to the names of things,

• r other conditions – just one’s ability to distinguish the witness of

the claim from other things in the extension of NP.

Andrew Weir Some, speaker knowledge, and subkinds

Introduction Epistemic conditions on some Modeling the epistemic condition Some and subkinds Further work/conclusion AM: alg´ un and antisingleton constraints Anti-singleton constraint in English

Modeling the epistemic condition

Alonso-Ovalle & Men´ endez-Benito 2010b (AM)’s proposal for Spanish alg´ un (11) alg´ un = λfet,etλPe,tλQe,t : antisingleton(f ).∃x[f (P)(x) & Q(x)] A generalized quantifier which first combines with a subset selection function f (Schwarzschild 2002) f restricts the domain P to some subset AM’s suggestion: alg´ un imposes a presupposition that f may not restrict the domain to a singleton set

Andrew Weir Some, speaker knowledge, and subkinds

Introduction Epistemic conditions on some Modeling the epistemic condition Some and subkinds Further work/conclusion AM: alg´ un and antisingleton constraints Anti-singleton constraint in English

Anti-singleton constraint on alg´ un (AM)

(12) a. John is hiding in un room of the house (namely the kitchen). b. ∃x.f (room of the house)(x) & John is hiding in(x) f can narrow room of the house down to a singleton set; compatible with the speaker knowing which room John is in. (13) a. John is hiding in alg´ un room of the house (#namely the kitchen). b. ∃x.f (room of the house)(x) & John is hiding in(x) Presupposition: antisingleton(f ) f cannot narrow room of the house down to a singleton set: by saying alg´ un, speaker signals that she cannot specify which room of the house satisfies the existential claim.

Andrew Weir Some, speaker knowledge, and subkinds

Introduction Epistemic conditions on some Modeling the epistemic condition Some and subkinds Further work/conclusion AM: alg´ un and antisingleton constraints Anti-singleton constraint in English

Anti-singleton constraint in English

Preliminary move: we can take over the definition of alg´ un to English some, at least when it combines with non-human NPs. (14) some = λfet,etλPe,tλQe,t : antisingleton(f ). ∃x[f (P)(x) & Q(x)]

Andrew Weir Some, speaker knowledge, and subkinds

Introduction Epistemic conditions on some Modeling the epistemic condition Some and subkinds Further work/conclusion Data Extending the analysis Unitary definition for some

But what about these examples?

(15) a. I saw some contraption in the copy room this morning. b. I came home to find some plant growing through a hole in my wall. c. [What’s in that bottle?] Some adhesive that John uses to put his models together. In all of these cases, the differentiation condition is satisfied; you would be able to distinguish the contraption, plant, adhesive that is the witness

• f the claim. Why is some licensed in the above examples?

Andrew Weir Some, speaker knowledge, and subkinds

Introduction Epistemic conditions on some Modeling the epistemic condition Some and subkinds Further work/conclusion Data Extending the analysis Unitary definition for some

Subkinds

Intuitively, the lack-of-knowledge being signaled in this examples concerns the subkind in question. (16) a. I saw some kind of contraption. . . b. I found some kind of plant . . . c. It’s some kind of adhesive . . . . . . but I don’t know what kind.

Andrew Weir Some, speaker knowledge, and subkinds

Introduction Epistemic conditions on some Modeling the epistemic condition Some and subkinds Further work/conclusion Data Extending the analysis Unitary definition for some

Modeling the epistemic condition on subkinds

If the ‘differentiation’ condition on subkinds is the same as with things, then we can model it with the same antisingleton constraint. (17) A speaker says some NPsubkind to signal that she cannot restrict to a singleton the set of subkinds within the extension of NP to which the witness of the existential claim belongs.

Andrew Weir Some, speaker knowledge, and subkinds

Introduction Epistemic conditions on some Modeling the epistemic condition Some and subkinds Further work/conclusion Data Extending the analysis Unitary definition for some

Formalizing it

Krifka 1995, Kratzer 2008: nouns (as we hear them) are ambiguous between entity readings and subkind readings. (18) DP D this NP CLind/kind √plant (19) √plant = ΣPlant (the kind ‘plant’, the sum of all plants) (20) a. CLind = λxλy.kind(x) & individual(y) & yΠx b. CLkind = λxλy.kind(x) & kind(y) & yΠx

Andrew Weir Some, speaker knowledge, and subkinds

Introduction Epistemic conditions on some Modeling the epistemic condition Some and subkinds Further work/conclusion Data Extending the analysis Unitary definition for some

Formalizing it

Alter the definition of some a little: (21) some = λfet,etλPe,tλQe,t : antisingleton(f ).∃y[(f (P))(y) & ∃z[zΠy & Q(z)]]

Andrew Weir Some, speaker knowledge, and subkinds

Introduction Epistemic conditions on some Modeling the epistemic condition Some and subkinds Further work/conclusion Data Extending the analysis Unitary definition for some

Formalizing it

(22) DP D’ D some f NP CLkind √plant (23) NP = λx.kind(x) & x ≤ ΣPlant = {x|kind(x) & x ≤ ΣPlant} = {ΣPoisonIvy, ΣCreeperVine, ΣRhododendron, . . .} (24) some plant = λQ : antisingleton(f ).∃y.f ({x|kind(x) & x ≤ ΣPlant})(y) & ∃z.zΠy & Q(z)

Andrew Weir Some, speaker knowledge, and subkinds

Introduction Epistemic conditions on some Modeling the epistemic condition Some and subkinds Further work/conclusion Data Extending the analysis Unitary definition for some

Formalizing it

(25) some plant = λQ : antisingleton(f ).∃y.f ({x|kind(x) & x ≤ ΣPlant})(y) & ∃z.zΠy & Q(z) The antisingleton constraint operates on the set of subkinds, not on the set of entities. some NPsubkind signals that the speaker cannot restrict the set of subkinds in the extension of NP down to a singleton There may not be any uncertainty about how to differentiate the entity; there is uncertainty in the subkind the entity belongs to.

Andrew Weir Some, speaker knowledge, and subkinds

Introduction Epistemic conditions on some Modeling the epistemic condition Some and subkinds Further work/conclusion Data Extending the analysis Unitary definition for some

Unitary definition for some

This same definition of some can combine with individual-denoting NPs. (26) DP D’ D some f NP CLind √ file (27) a. NP = λx.ind(x) & x ≤ ΣFile = {x|ind(x) & x ≤ ΣFile} = {file1, file2, file3 . . .} b. some file = λQ : antisingleton(f ).∃y.f ({x|ind(x) & x ≤ ΣFile})(y) & ∃z.zΠy & Q(z)

Andrew Weir Some, speaker knowledge, and subkinds

Introduction Epistemic conditions on some Modeling the epistemic condition Some and subkinds Further work/conclusion Data Extending the analysis Unitary definition for some

Unitary definition for some

(28) some file = λQ : antisingleton(f ).∃y.f ({x|ind(x) & x ≤ ΣFile})(y) & ∃z.zΠy & Q(z) This is a generalized quantifier over parts z of some individual file y But the individual files are mereological atoms, without proper parts. Only one z is an individual part of y – y itself Therefore, uncertainty about identity of the referent is preserved, and the expression as a whole still quantifies over individual files

Andrew Weir Some, speaker knowledge, and subkinds

Introduction Epistemic conditions on some Modeling the epistemic condition Some and subkinds Further work/conclusion Data Extending the analysis Unitary definition for some

Those not-so-bad-sentences

(29)

• a. ??Some statue is in the middle of the square. [looking at it]
• b. ??I saw some building on my way through the desert.
• c. ??There’s some letter in my mailbox. [looking at it]

These are degraded, but they’re not dreadful. They are OK just to the extent that you can meaningfully insert kind of in there. (30) a. (??)Some kind of statue is in the middle of the square. b. (??)I saw some kind of building on my way through the desert. c. (??)There’s some kind of letter in my mailbox.

Andrew Weir Some, speaker knowledge, and subkinds

Introduction Epistemic conditions on some Modeling the epistemic condition Some and subkinds Further work/conclusion

For future work

How does some + singular NP differ from some + plural NP (if at all?)

see Alonso-Ovalle & Men´ endez-Benito 2010a on Sp. algunos (alg´ un + plural) – epistemic effect goes away It goes away in English too: Some files are infected, namely these

• nes

Will the proposed (re)definition of some work the same way as AM’s analysis of algunos?

How is the human/thing dichotomy derived – in the grammar or due to basic constraints on ‘knowing who’? Cross-linguistic variation in epistemic indefinites:

Epistemic indefinites show cross-linguistic differences when paired with humans; for example alg´ un is bad in the ‘lambada professor’ examples (see Aloni & Port 2010 for extensive discussion) Differences in plurality: some + pl, algunos + pl have no epistemic effect, Ger. irgendwelche + pl does (Alonso-Ovalle & Men´ endez-Benito 2010a, Aloni & Port 2010); why?

Andrew Weir Some, speaker knowledge, and subkinds

Introduction Epistemic conditions on some Modeling the epistemic condition Some and subkinds Further work/conclusion

Conclusions

Some + singular NP has a context-dependent epistemic effect when paired with humans, but a ‘differentiation/picking-out’ epistemic effect when paired with things. This can be modeled with an anti-singleton constraint, following Alonso-Ovalle & Men´ endez-Benito 2010b. A similar ‘differentiation’ epistemic effect obtains when some pairs with subkinds. This can be modeled by assuming polysemy of common nouns between sets of individuals and sets of subkinds, and adding an additional existential quantification over parts of subkinds; and this can extend to some + individual-denoting noun without loss of generality.

Andrew Weir Some, speaker knowledge, and subkinds

Introduction Epistemic conditions on some Modeling the epistemic condition Some and subkinds Further work/conclusion

Thank you!

Andrew Weir Some, speaker knowledge, and subkinds

Introduction Epistemic conditions on some Modeling the epistemic condition Some and subkinds Further work/conclusion

References I

Aloni, Maria. 2001. Quantification under conceptual covers: University of Amsterdam dissertation. Aloni, Maria & Angelika Port. 2010. Epistemic indefinites cross-linguistically. Presented at NELS 41. http://staff.science.uva.nl/~maloni/NELS2010-handout.pdf. Alonso-Ovalle, Luis & Paula Men´ endez-Benito. 2003. Some epistemic

• indefinites. In Makoto Kadowaki & Shigeto Kawahara (eds.),

Proceedings of NELS 34, 1–12. Amherst, MA: GLSA. Alonso-Ovalle, Luis & Paula Men´ endez-Benito. 2010a. Domain restriction, modal implicatures and plurality: Spanish algunos. Journal

• f Semantics 28(2). 211–40.

Alonso-Ovalle, Luis & Paula Men´ endez-Benito. 2010b. Modal indefinites. Natural Language Semantics 18. 1–31.

Andrew Weir Some, speaker knowledge, and subkinds

Introduction Epistemic conditions on some Modeling the epistemic condition Some and subkinds Further work/conclusion

References II

Bo¨ er, Steven E. & William G. Lycan. 1986. Knowing who. Cambridge, MA: MIT Press. Kratzer, Angelika. 2008. On the plurality of verbs. In Johannes D¨

• lling,

Tatjana Heyde-Zybatow & Martin Sch¨ afer (eds.), Event structures in linguistic form and interpretation, 269–300. Berlin: Walter de Gruyter. Krifka, Manfred. 1995. Common nouns: a contrastive analysis of Chinese and English. In Gregory N. Carlson & Francis J. Pelletier (eds.), The generic book, 398–411. Chicago: University of Chicago Press. Schwarzschild, Roger. 2002. Singleton indefinites. Journal of Semantics 19(3). 289–314.

Andrew Weir Some, speaker knowledge, and subkinds

Introduction Epistemic conditions on some Modeling the epistemic condition Some and subkinds Further work/conclusion

Appendix: subkinds pattern with ‘things’

If subkind-denoting NPs otherwise patterned like human-denoting NPs when paired with some, then this would not be an interesting

• bservation.

Some here might just signal ‘don’t know the name’, as it can with people; or some other contextual factor that enters into ‘identity questions’ (Aloni 2001) I will argue that subkind-denoting NPs don’t generally pattern with humans, but rather with ‘things’.

Andrew Weir Some, speaker knowledge, and subkinds

Introduction Epistemic conditions on some Modeling the epistemic condition Some and subkinds Further work/conclusion

Appendix: subkinds pattern with ‘things’

The game: show that being able to distinguish subkinds from one another is sufficient to de-license some, even if other epistemic uncertainties (e.g. the name) remain.

Andrew Weir Some, speaker knowledge, and subkinds

Introduction Epistemic conditions on some Modeling the epistemic condition Some and subkinds Further work/conclusion

Appendix: subkinds pattern with ‘things’

(31) Able to distinguish subkinds Katniss, having grown up on her wits, is intimately familiar with all the plants in her district, and how they can be used for medicinal purposes. She’s never had any formal schooling or parental teaching of herbal lore, though, so she doesn’t know any of their names. She applies one to heal Gale’s burns. Gale: What’s that? Katniss: A/?#some plant that’s good at soothing burns. (32) Unable to distinguish subkinds Katniss is in the Hunger Games Arena, far from home, where there are new types of plant that she’s never seen before. She discovers through experimentation that one type is good for healing burns. She applies it to heal Rue’s burns. Rue: What’s that? Katniss: A/some plant that’s good at soothing burns.

Andrew Weir Some, speaker knowledge, and subkinds

Introduction Epistemic conditions on some Modeling the epistemic condition Some and subkinds Further work/conclusion

Appendix: subkinds pattern with ‘things’

Learning the name will often de-license some, but (I argue) that’s an artifact of the fact that learning the name will often let you differentiate the subkind. If the name doesn’t help you do that, then some remains licensed. (33) A: There’s some plant growing through my wall! B: That’s poison ivy. A: Oh! [Later, A talks to C] A: There was a/?#some plant growing through my wall. (34) A: There’s some plant growing through my wall! B: That’s Toxicodendron radicans. A is unmoved [Later, A talks to C] A: There was a/some plant growing through my wall.

Andrew Weir Some, speaker knowledge, and subkinds

Introduction Epistemic conditions on some Modeling the epistemic condition Some and subkinds Further work/conclusion

Appendix II: Number-neutrality

This account predicts that some plant, on the kind reading, should be number-neutral. (35) some plant = λQ : antisingleton(f ).∃y.f ({x|kind(x) & x ≤ ΣPlant})(y) & ∃z.zΠy & Q(z) No requirement in the above that z be an atomic part of the subkind y. This prediction is borne out.

Andrew Weir Some, speaker knowledge, and subkinds

Introduction Epistemic conditions on some Modeling the epistemic condition Some and subkinds Further work/conclusion

Appendix II: Number-neutrality

(36) A: What’s this warehouse for? B: There’s some contraption in there. There are shelves upon shelves of the things, all the same. I don’t know what they do. (37) [I take a delivery of 100 plants, but they are not the type I

• rdered; they are all the same type of plant, but I don’t

recognize what type.] A: Did you get the plants you ordered? B: They did deliver some plant. I have 100 of the things clogging up the office. But I’ve no idea what they are, they’re not what I ordered.

Andrew Weir Some, speaker knowledge, and subkinds

Introduction Epistemic conditions on some Modeling the epistemic condition Some and subkinds Further work/conclusion

Appendix II: Number-neutrality

But the items in question do all have to belong to the same subkind; if not, then the some sentences become bad. (38) A: What’s this warehouse for? B: #There’s some contraption in there. Three shelves for three different things, but I don’t know what any of them do. (39) [I take a delivery of 100 plants, not the type I ordered, and not all the same type of plant; I don’t recognize any of the types of plants.] A: Did you get the plants you ordered? B: #They did deliver some plant.

Andrew Weir Some, speaker knowledge, and subkinds