CLEFTS

1
CLEFTS EQUATIVESInformation Structure and
Word Order Variation

• LSA.323
• 17 July 2007

2
Introduction

• Clefts and equatives
• Clefts consist of a cleft clause and a clefted
  constituent. (Its Chris who I like.)
• Equatives consist of two DPs and a copula. (Im
  the cheeseburger.)
• Hedberg and Faddens (2007) analysis of clefts
• Two related canonical-word-order constructions
  with marked information structure

3
Hedberg Fadden (2007)

• Three types of clefts
• It-clefts (Its peace that we want.)
• Wh-clefts (What we want is peace.)
• Reverse wh-clefts (Peace is what we want.)
• (Theres also a 4th type th-clefts tbd.)

4
Hedberg Fadden (2007)

• Findings
• Wh-clefts are more constrained than either
  it-clefts or reverse wh-clefts.
• The cleft clause in a wh-cleft always represents
  the topic of the sentence, while the cleft
  constituent always represents the focus
  (topic-comment structure).

5
Hedberg Fadden (2007)

• Findings, cont.
• The clefted constituents of it-clefts and reverse
  wh-clefts are less constrained, with the clefted
  constituent and the cleft clause free to
  represent either the topic or focus of the
  utterance (topic-comment or comment-topic
  structure).

6
Canonical Word Order

• So far, we have focused on constructions that
  employ noncanonical word order as the basis for
  identifying the markedness of those
  constructions. However, word order is not the
  only indicator of a marked construction.
• Two copular constructions in English whose marked
  information status does not derive from word
  order variation DEFERRED EQUATIVES and EPISTEMIC
  WOULD EQUATIVES.

7
Examples

• context Server at Thai restaurant to group of
  customers
• A Who ordered the Pad Thai and who ordered the
  Nam Sod?
• B Im the Pad Thai.
• OP X CORRESPONDS TO Y
• A What did Chris order?
• B That would be the Pad Thai.
• OP CHRIS ORDERED X

8
Claims

• Both types are focus-presupposition
  constructions, requiring a salient OPEN
  PROPOSITION for felicity.
• They differ in the number of OP variables being
  instantiated as foci.
• The two arguments of deferred equatives
  correspond to the two instantiations of the OP
  (Ward 04), while the post-copular constituent of
  epistemic would equatives corresponds to the
  unique variable of the OP (Birner, Kaplan, Ward
  07).
• The constructions formally mark a salient OP yet
  what marks them so is not word order.

9
Part I Epistemic Would Equatives

• Claim Functionally, use of an epistemic would
  equative
• conveys a speakers strong commitment to the
  truth of the proposition expressed
• marks the postverbal constituent as the focus of
  a contextually salient OP.

10
Some Naturally-Occurring Data

• Dad Uh Whos that boy hanging out in our
• front yard, Danae?
• Danae That would be Jeffrey, my
  not-so-secret
• admirer.
• Non Sequitur comic, Chicago Tribune, Sunday,
  3/3/02, sec. 9, p. 3
• Q Can you tell us if you recognize this
  clothing?
• A That would be our standard attire,
  correct.
• Simpson transcripts, 2/7/95
• I need to talk to Andy to see if hes interested
  in going to a departure lunch for a Zenith
  employee this Friday at Little Villa the
  employee would be myself.
• message left on BBs answering machine, 9/21/00

11
Epistemic Would Equatives and Epistemic Modality

• Epistemic modality marks the speakers estimation
  of the likelihood that a certain state of affairs
  is true (or false) in the context of the possible
  world under consideration (Nuyts 2001).
• Previous accounts of epistemic would
• Perkins (1983) and Palmer (1990) conveys
  tentativeness
• Coates (1983) indicates predictability
• Sweetser (1982) is part of an implicit
  conditional with a suppressed antecedent.

12
Epistemic Would Equatives and Epistemic Modality

• Empirical evidence Ward et al. (in prep.) show
  that subjects are significantly more likely to
  rate epistemic would utterances (e.g. That would
  be Chris) presented aurally (but not visually!)
  as conveying certainty than the corresponding
  utterances with main verb be alone (e.g. Thats
  Chris).
• Of all the epistemic modals, would conveys the
  strongest degree of speaker commitment
• That would be Jeffrey.
• That must/should be Jeffrey.

13
Epistemic Would Equatives and Epistemic Modality

• In a corpus of 246 tokens
• 79 have pronominal demonstrative that as subject
• full lexical NPs are possible, e.g., The best
  teacher at the Institute would be Chris).
• 98 contain the copula be
• other intransitive verbs are possible, e.g., go,
  belong, correspond to (Those would go in the top
  drawer).

14
Epistemic Would Equatives and Epistemic Modality

• Empirical evidence Ward et al. (in prep.) show
  that subjects are significantly more likely to
  rate epistemic would utterances (e.g. That would
  be Chris) presented aurally (but not visually!)
  as conveying certainty than the corresponding
  utterances with main verb be alone (e.g. Thats
  Chris).
• Of all the epistemic modals, would conveys the
  strongest degree of speaker commitment
• That would be Jeffrey.
• That must/should be Jeffrey.

15
Epistemic Would and OPs

• Claim Epistemic would equatives require a
  salient OP for felicity.
• Recall the initial examples
• Dad Uh Whos that boy hanging out in our front
  yard, Danae?
• Danae That would be Jeffrey, my not-so-secret
  admirer.
• OP THE BOY HANGING OUT IN THE FRONT YARD IS X.

16
Epistemic Would and OPs

• Q Can you tell us if you recognize this
  clothing?
• A That would be our standard attire,
  correct.
• OP THE CLOTHING IS X.

17
Epistemic Would and OPs

• In the absence of a salient OP, only epistemic
  would is infelicitous.
• context Speaker picks up an envelope with the
  hearer not attending to the event
• This would be my new VISA card.
• This must/should be my new VISA card.
• This might/could be my new VISA card.

18
Reference to the Instantiation of the Variable of
the OP

• Claim The combination of a salient OP and the
  equative construction produces the possibility
  of
• using a demonstrative subject to refer
  deictically to the instantiation of the variable
  of the OP, and
• identifying it, via the equative construction,
  with its postcopular focus.

19
Reference to the Instantiation of the Variable of
the OP

• A (holding cup) Whose is this?
• B That would be my son. My youngest
• son, to be exact. conversation,
  2/4/01
• OP THIS CUP BELONGS TO X.

20
Reference to the Instantiation of the Variable of
the OP

• GW What is the per minute charge to Italy?
• Operator Do you have the one-rate plan?
• GW Im not sure can I find out through
  you?
• Operator No, that would be1800-466-3728.
• conversation with ATT operator, 6/23/01
• OP YOU CAN FIND OUT THROUGH X.

21
Reference to the Instantiation of the Variable of
the OP

• A The pots light.
• B That would be me. tosses in a chip
• poker game, 1/31/03
• Whats the OP here?
• OP THE PERSON WHO FAILED TO ANTE IS X.

22
Reference to the Instantiation of the Variable of
the OP

• Reference to the instantiation of the variable of
  the OP as a form of discourse deixis (Lyons 1977,
  Levinson 1983, Webber 1988)
• I bet you havent heard this story.
• Levinson 198385, ex. 88
• A Ive never seen him.
• B Thats a lie.
• Lyons 1977668, cited in Levinson
  198387

23
Reference to the Instantiation of the Variable of
the OP

• In fact, we claim that the demonstrative of
  epistemic would is actually referentially
  ambiguous -- referring to the instantiation of
  the variable of an OP is only one possibility.
  Stay tuned

24
Part II Deferred Equatives

• DEFERRED REFERENCE (Nunberg 1995)
• the metonymic use of an expression to refer
  to an entity not denoted by the conventional
  meaning of that expression.
• server to co-worker in deli
• The ham sandwich is at table 7.
• restaurant patron handing car key to
  parking valet
• This is parked out back.
• Yeats is still widely read.
• The table is made of oak.

25
Deferred Equatives

• Related work regular polysemy (Apresjan 1974,
  semantic transfer rules (Leech 1974), metonymy
  (Lakoff Johnson 1980), sense transfer (Sag
  1981), active zones (Langacker 1984), lexical
  networks (Norvig Lakoff 1987), lexical
  implication rules (Ostler Atkins 1991),
  referential metonymy and predicative metonymy
  (Stallard 1993), connectors (Fauconnier 1994),
  meaning transfer (Nunberg 1995), constructional
  polysemy and sense extension (Copestake Briscoe
  1995), logical metonymy (Pustejovsky 1995,
  Verspoor 1997), enriched composition (Jackendoff
  2002), inter alia.

26
Some Naturally-Occurring Data

• Im the Pad Thai.
• BL, in conversation, 8/10/02
• Samir Abd al-Aziz al-Najim is the four of clubs.
• Chicago Tribune, 4/19/03, p. 5.
• physician assigning interns to patients
• You and you are shortness of breath. You and you
  take vertigo. And last but not least, knee pain.
• ER, 4/25/03

27
More Naturally-Occuring Data

• DJ Are you Cubs or White Sox?
• Caller Im Cubs.
• WGN Radio, AM 720, 6/20/03
• request to unplug out-of-reach laptop power
  cable
• A Hi, can you unplug me?
• B Sure, which one are you?
• A Im the one on the end.
• Conversation, 12/12/06 token courtesy of
  Robert Daland

28
Previous Studies

• MEANING (SENSE) TRANSFER the name of a
  property that applies to something in one domain
  can sometimes be used as the name of a property
  in another domain, provided the two properties
  correspond in a certain noteworthy way
  (Nunberg 1995111)
• Two mechanisms for meaning transfer
• Predicate (VP) transfer
• Im parked out back. ?
• I'm the owner of a car parked out
  back.
• Common noun (N) transfer
• The ham sandwich is at table 7. ?
  
• The ham sandwich orderer is at
  table 7.

29
Deferred Reference

• Claim Unlike deferred non-equatives, deferred
  equatives involve no transfer of sense or
  reference rather, the two NPs of deferred
  equatives retain their literal meanings with the
  copula interpreted as expressing a noteworthy
  correspondence between the referents of those NPs.

30
Deferred Equatives and OPs

• Review an OP is a proposition with one or more
  uninstantiated variables or underspecified
  elements, corresponding to that aspect of
  information structure that constitutes
  backgrounded or presupposed information (i.e.,
  that which is not at issue).
• Im not really into sports. Football is gross.
  Baseball I find tolerable, but more for the scene
  at Wrigley Field than the actual game.
• OP I HAVE DISPOSITION X TOWARD SPORTS.
• FOCUS tolerable

31
Deferred Equatives and OPs

• For deferred equatives, the relevant OP contains
  two variables, corresponding to the two sets from
  which the instantiations of these variables are
  drawn.
• Im the Pad Thai.
• OP X CORRESPONDS TO Y (where X is a member of
  the set customers and Y is a member of the set
  orders).
• FOCI I, the Pad Thai

32
Deferred Equatives and OPs

• Claim For deferred equatives, no sense or
  reference transfer applies.
• In the absence of a salient double-variable OP,
  deferred equatives are infelicitous
• A on cell phone Sorry to interrupt your
  lunch. How is it?
• B Delicious! Im the Pad Thai.
• A Sorry to interrupt your lunch. How is it?
• B Delicious! I ordered the Pad Thai.

33
Deferred Equatives as Non-Deferred

• Non-deferred equatives (identity clauses) are
  not subject to this constraint
• That guy that you were just talking to is my
  neighbor Sam Miller.
• That building on your left is the Sears Tower.
• This painting is the same as the one hanging in
  the Louvre.

34
Deferred Equatives as Non-Deferred

• For deferred equatives, the relevant NP (e.g. the
  Pad Thai) does not undergo meaning transfer.
• John is the Pad Thai, who drives a Rolls Royce.
• John is the Pad Thai, which looks delicious.
• John is talking to the Pad Thai, who drives a
  Rolls Royce.
• The Pad Thai, who drives a Rolls Royce, is John.
• The Pad Thai, which looks delicious, is John.
• The Pad Thai, who drives a Rolls Royce, is
  talking to John.
• Lets see Youre what, the Pad Thai or the Nam
  Sod?
• Lets see Youre who, the Pad Thai or the Nam
  Sod?

35
Deferred Equatives as Non-Deferred

• Tell me honestly, who do you like more, the Pad
  Thai or the Nam Sod?
• Tell me honestly, what do you like more, the Pad
  Thai or the Nam Sod?
• What I am is the Pad Thai, not the Nam Sod.
• Who I am is the Pad Thai, not the Nam Sod.
• Dude, who you should be checking out is the Pad
  Thai, not the Nam Sod.
• Dude, what you should be checking out is the Pad
  Thai, not the Nam Sod.

36
Summary, Part II

• Deferred equatives
• Encode a pragmatic correspondence between members
  of sets of discourse entities with the copula
  expressing a correspondence rather than literal
  identity between those entities.
• Require a contextually salient OP.

37
Conclusions

• Two copular constructions deferred equatives
  and epistemic would equatives share the
  pragmatic requirement of a salient OP in context.
• They differ with respect to the number of OP
  variables being instantiated as foci, with the
  demonstrative of epistemic would being used to
  refer to the instantiation of an OP variable.

38
Conclusions

• Unlike other OP-marking constructions that rely
  on word order to signal their markedness, these
  equatives rely on other devices
• for deferred equatives, its the interpretation
  of the copula as expressing a (non-identity)
  correspondence
• for epistemic would equatives, its the presence
  of the epistemic modal would.