Working with Discourse Representation Theory Patrick Blackburn

1
Working with Discourse Representation
TheoryPatrick Blackburn Johan Bos Lecture
4Pronouns and Presupposition
2
Today

• Pronouns
• Anaphoric pronouns
• Binding constraints
• Presupposition
• Triggers
• Problems
• Van der Sandt
• Implementation

3
PRONOUNS
4
Pronouns

• We will concentrate on 3rd person singular
  personal pronouns in English
• he/him/himself
• she/her/herself
• it/itself

5
Anaphoric Pronouns

• We will focus on anaphoric pronouns
• Anaphoric pronouns find their antecedent in the
  preceding text
• Anaphora -- backwards Vincent looked at Mia.
  She dances.

6
Anaphoric Pronouns

• We will focus on anaphoric pronouns
• Anaphoric pronouns find their antecedent in the
  preceeding text
• Anaphora -- backwards Vincent looked at Mia.
  She dances.
• She is the anaphor

7
Anaphoric Pronouns

• We will focus on anaphoric pronouns
• Anaphoric pronouns find their antecedent in the
  preceeding text
• Anaphora -- backwards Vincent looked at Mia.
  She dances.
• Mia is the antecedent

8
Anaphoric Pronouns

• We will focus on anaphoric pronouns
• Anaphoric pronouns find their antecedent in the
  preceeding text
• Anaphora -- backwards Vincent looked at Mia.
  She dances.
• How far backwards?

9
Cataphoric Pronouns

• We will not deal with cataphora
• Cataphoric pronouns find their antecedent in the
  text following the pronoun
• ExampleAfter he lost the match, Butch left
  town.

10
Cataphoric Pronouns

• We will not deal with cataphora
• Cataphoric pronouns find their antecedent in the
  text following the pronoun
• ExampleAfter he lost the match, Butch left
  town.

11
Deictic Pronouns

• Pronouns referring to objects in the situation,
  rather than linguistic objects
• ExamplesI, you, we, here, there, etc.

12
Pleonastic use of pronouns

• ExampleIts about nine oclock in the morning.

13
Grammatical agreement

• In English, pronouns come with a gender and
  number feature
• Only refer to antecedents carrying the same
  feature values
• he (singular, male)
• men/boys, male animals
• she (singular, female)
• women/girls, female animals, things regarded
  as female, e.g. vehicles or ships
• it (singular, neuter) things, animals, children

14
Pronouns and Ambiguity

• Butch threw a TV at the window.It broke.
• Butch threw a vase at the wall.It broke.

15
Pronouns and Ambiguity

• Butch threw a TV at the window.It broke.
• Butch threw a vase at the wall.It broke.

16
Pronouns and Ambiguity

• Butch threw a TV at the window.It broke.
• Butch threw a vase at the wall.It broke.

17
Pronouns and Ambiguity

• Butch threw a TV at the window.It broke.
• Butch threw a vase at the wall.It broke.

18
Pronouns and Ambiguity

• Usually many candidate antecedents available
• ExampleButch walks into his modest kitchen.
  He opens the refrigerator. He takes out a milk
  and drinks it.

19
Pronouns and Ambiguity

• Usually many candidate antecedents available
• ExampleButch walks into his modest kitchen.
  He opens the refrigerator. He takes out a milk
  and drinks it.

20
Pronouns and Ambiguity

• Usually many candidate antecedents available
• ExampleButch walks into his modest kitchen.
  He opens the refrigerator. He takes out a milk
  and drinks it.

21
Pronouns and Ambiguity

• Usually many candidate antecedents available
• ExampleButch walks into his modest kitchen.
  He opens the refrigerator. He takes out a milk
  and drinks it.

22
Pronouns and Ambiguity

• Usually many candidate antecedents available
• ExampleButch walks into his modest kitchen.
  He opens the refrigerator. He takes out a milk
  and shuts it.

23
Pronouns and Ambiguity

• Usually many candidate antecedents available
• ExampleButch walks into his modest kitchen.
  He opens the refrigerator. He takes out a milk
  and shuts it.

24
Pronouns and Ambiguity

• Usually many candidate antecedents available
• ExampleButch walks into his modest kitchen.
  He opens the refrigerator. He takes out a milk
  and shuts it.

25
Reflexive Pronouns

• Examples
• Vincent goes to the toilet, and Jules enjoys
  himself.
• Vincent enters the restaurant, and Jules watches
  him.

26
Reflexive Pronouns

• Examples
• Vincent goes to the toilet, and Jules enjoys
  himself.
• Vincent enters the restaurant, and Jules watches
  him.

27
Reflexive Pronouns

• Examples
• Vincent goes to the toilet, and Jules enjoys
  himself.
• Vincent enters the restaurant, and Jules watches
  him.

28
Binding rules

• Behaviour of pronounsButch likes
  himself.Butch likes him.Butch likes his
  chopper.

29
DRT and pronouns 1/3

• Vincent did not dance with a woman.She

x u
xvincent u ???
?
y e
woman(y) dance(e) agent(e,x) patient(e,y)
30
DRT and pronouns 2/3

• Vincent did with every woman. She

x u
xvincent u ???
e
dance(e) agent(e,x) patient(e,y)
y
woman(y)
?
31
DRT and pronouns 3/3

• Vincent did with no woman. She

x u
xvincent u ???
y e
woman(y) dance(e) agent(e,x) patient(e,y)
?
32
Summing up

• We have looked at anaphoric pronouns
• It is unlikely that we can solve all the problems
  related to resolving pronouns
• However, we can deal with some important aspects
• Semantic constrains gender
• Binding constrains reflexivity
• DRT constrains pronoun resolution, but only
  partially

33
PRESUPPOSITION
34
Presupposition

• Presupposition vs. Entailment
• Look at some examples of presupposition
• Look at the typical problems associated with
  presuppositions
• Concentrate on a DRT based approach due to Rob
  van der Sandt

35
What is presupposition?

• It is hard to pin down precisely what
  presuppositions are or how they behave
• Presuppositions are a bit like entailment but not
  quite

36
Entailment

• ConsiderVincent has a car. A car is a
  vehicle.
• This entails Vincent has a vehicle.

37
Entailment

• ConsiderVincent has a red car.
• This entails Vincent has a car.

38
Entailment and negation

• Entailments are typically not preserved under
  negation.

39
Entailment

• ConsiderVincent has no car. A car is a
  vehicle.
• This does not entail Vincent has a vehicle.

40
Entailment

• ConsiderVincent does not have a red car.
• This does not entail Vincent has a car.

41
Presupposition

• ConsiderVincent cleaned his car.
• This entailsVincent has a car.

42
Presupposition

• ConsiderVincent did not clean his car.
• This entailsVincent has a car.

43
Entailment or presupposition

• We call implications preserved under negation
  presuppositions
• We call implications not preserved under negation
  entailments

44
Presupposition triggers

• In English, presuppositions are usually triggered
  by lexical items
• There are several tricks to find out whether a
  lexical item is a presupposition trigger or not
• These tests are
• The negation test
• The conditional test
• The question test

45
Presupposition trigger test

• Consider the sentenceAlex is a bachelor.
• This sentence implies that Alex is male.
• But are we dealing with a presupposition or
  entailment?

46
Presupposition test

• Alex is a bachelor.Does this presuppose Alex is
  male?

47
Presupposition test

• Alex is a bachelor.Does this presuppose Alex is
  male?
• Negation Alex is not a bachelor.Implies Alex
  is male? YES

48
Presupposition test

• Alex is a bachelor.Does this presuppose Alex is
  male?
• Negation Alex is not a bachelor.Implies Alex
  is male? YES
• Conditional If Alex is a bachelor, then
  ...Implies Alex is male? YES

49
Presupposition test

• Alex is a bachelor.Does this presuppose Alex is
  male?
• Negation Alex is not a bachelor.Implies Alex
  is male? YES
• Conditional If Alex is a bachelor, then
  ...Implies Alex is male? YES
• Question Is Alex is a bachelor?Implies Alex is
  male? YES

50
Presupposition test

• Alex is a bachelor.Does this presuppose Alex is
  male?
• Negation Alex is not a bachelor.Implies Alex
  is male? YES
• Conditional If Alex is a bachelor, then
  ...Implies Alex is male? YES
• Question Is Alex is a bachelor?Implies Alex is
  male? YES
• Conclusion being a bachelor presupposes being
  male.

51
Presupposition trigger test

• Consider the sentenceAlex is a man.
• This sentence implies that Alex is male.
• But are we dealing with a presupposition or
  entailment?

52
Presupposition test

• Alex is a man.Does this presuppose Alex is
  male?

53
Presupposition test

• Alex is a man.Does this presuppose Alex is
  male?
• Negation Alex is not a man.Implies Alex is
  male? NO

54
Presupposition test

• Alex is a man.Does this presuppose Alex is
  male?
• Negation Alex is not a man.Implies Alex is
  male? NO
• Conditional If Alex is a man, then ...Implies
  Alex is male? NO

55
Presupposition test

• Alex is a man.Does this presuppose Alex is
  male?
• Negation Alex is not a man.Implies Alex is
  male? NO
• Conditional If Alex is a man, then ...Implies
  Alex is male? NO
• Question Is Alex is a man?Implies Alex is
  male? NO

56
Presupposition test

• Alex is a man.Does this presuppose Alex is
  male?
• Negation Alex is not a man.Implies Alex is
  male? NO
• Conditional If Alex is a man, then ...Implies
  Alex is male? NO
• Question Is Alex is a man?Implies Alex is
  male? NO
• Conclusion being a man does not presuppose
  being male.

57
Presupposition trigger test

• Consider the sentenceButch knows that Zed is
  dead.
• This sentence implies Zed is dead.
• But are we dealing with a presupposition or
  entailment?

58
Presupposition test

• Butch knows that Zed is dead.Does this
  presuppose Zed is dead?

59
Presupposition test

• Butch knows that Zed is dead.Does this
  presuppose Zed is dead?
• Negation Butch does not know that Zed is
  dead.Implies Zed is dead? YES

60
Presupposition test

• Butch knows that Zed is dead.Does this
  presuppose Zed is dead?
• Negation Butch does not know that Zed is
  dead.Implies Zed is dead? YES
• Conditional If Butch knows that Zed is dead,
  then ...Implies Zed is dead? YES

61
Presupposition test

• Butch knows that Zed is dead.Does this
  presuppose Zed is dead?
• Negation Butch does not know that Zed is
  dead.Implies Zed is dead? YES
• Conditional If Butch knows that Zed is dead,
  then ...Implies Zed is dead? YES
• Question Does Butch know that Zed is
  dead?Implies Zed is dead? YES

62
Presupposition test

• Butch knows that Zed is dead.Does this
  presuppose Zed is dead?
• Negation Butch does not know that Zed is
  dead.Implies Zed is dead? YES
• Conditional If Butch knows that Zed is dead,
  then ...Implies Zed is dead? YES
• Question Does Butch know that Zed is
  dead?Implies Zed is dead? YES
• Conclusion knowing P presupposes P.

63
Presupposition triggers

• Presupposition triggers are not rare
• English comes with a large variety of
  presupposition triggers

64
Possessives

• ExampleMia likes her husband.Mia does not
  like her husband.
• PresuppositionMia has a husband.

65
To regret

• ExampleVincent regrets that he left Mia
  alone.Vincent does not regret that he left Mia
  alone.
• PresuppositionVincent left Mia alone.

66
To like

• ExampleMia likes Vincent.Mia does not like
  Vincent.
• PresuppositionMia knows Vincent.

67
To answer

• ExampleButch answered the phone.Butch did not
  answer the phone.
• PresuppositionThe phone was ringing.

68
Only

• Example Only Jules likes big kahuna
  burgers.Not only Jules likes big kahuna
  burgers.
• PresuppositionJules likes big kahuna burgers.

69
Again

• ExampleButch escaped again.Butch did not
  escape again.
• PresuppositionButch escaped once before.

70
To manage

• ExampleButch manage to start the
  chopper.Butch did not manage to start the
  chopper.
• PresuppositionButch had difficulties starting
  the chopper.

71
Third

• ExampleButch lost for the third time.Butch
  did not loose for the third time.
• PresuppositionButch lost twice before.

72
Continue

• ExampleButch continued his race.Butch did not
  continue his race.
• PresuppositionButch interrupted his race.

73
To win

• ExampleGermany won the world cup.Germany did
  not win the world cup.
• PresuppositionGermany participated in the
  world cup.

74
Another

• ExamplePeter wants another beer.Peter does
  not want another beer.
• PresuppositionPeter had at least one beer.

75
To lie

• ExampleButch lied to Marsellus.Butch did not
  lie to Marsellus.
• PresuppositionButch told something to
  Marsellus.

76
Cleft construction

• ExampleIt was Butch who killed Vincent.It was
  not Butch who killed Vincent.
• PresuppositionSomeone killed Vincent.

77
Proper names

• ExampleButch talked to Marsellus.Butch did
  not talk to Marsellus.
• PresuppositionThere is someone named
  Marsellus.

78
Definite NP

• ExampleButch talked to the boss.Butch did not
  talk to the boss.
• PresuppositionThere is a boss.

79
Dealing with Presupposition

• OK, so presuppositions are fairly common. But
  whats the big deal?
• Problems related to presupposition
• The Binding Problem
• The Denial Problem
• The Projection Problem
• Presupposition may convey new information
• Accommodation

80
The Binding Problem

• ExampleButch nearly escaped from his
  apartment.
• Trigger his apartment presupposes that Butch
  has an apartment.

81
The Binding Problem

• ExampleA boxer nearly escaped from his
  apartment.
• Trigger his apartment presupposes that a boxer
  has an apartment.
• But which boxer? A boxer? Any boxer?

82
The Denial Problem

• Vincent does not like his wife.

83
The Denial Problem

• Vincent does not like his wife.
• Vincent does not like his wife, because Vincent
  does not have a wife!

84
The Denial Problem

• Vincent does not regret killing Zed, because he
  did not kill Zed!

85
The Denial Problem

• Vincent does not regret killing Zed, because he
  did not kill Zed!
• Alex is not a bachelor, because she is a woman!

86
The Denial Problem

• Vincent does not regret killing Zed, because he
  did not kill Zed!
• Alex is not a bachelor, because she is a woman!
• Butch did not lie to Marsellus,because he did
  not tell him anything!

87
The Projection Problem

• Consider
• Mias husband is out of town.
• Presupposes that Mia is married.

88
The Projection Problem

• Consider
• If Mia has a husband, then Mias husband is out
  of town.
• Does NOT presuppose that Mia is married.

89
The Projection Problem

• Consider
• If Mia is married, then Mias husband is out of
  town.
• Does NOT presuppose that Mia is married.

90
The Projection Problem

• Consider
• If Mia dates Vincent, then Mias husband is out
  of town.
• Does presuppose that Mia is married.

91
The Projection Problem

• Complex sentences sometimes neutralise
  presuppositions
• Complex meaning here sentences with
  conditionals, negation, or disjunction, modals
• These sentences make it difficult to predict
  whether a presupposition projects or not

92
Accommodation

• ExampleVincent informed his boss.
• Presupposition Vincent has a boss.
• What if we dont have a clue whether Vincent has
  a boss or not?
• Accommodation incorporating missed information
  as long as this is not conflicting with other
  information

93
Solutions

• There is a rich literature on presupposition
• There are many different attempts to solve the
  problems related to presupposition
• Many-valued logics
• Default logics
• Pragmatic theories
• Non-monotonic reasoning

94
Van der Sandts Theory

• Presuppositions are essentially extremely rich
  anaphoric pronouns
• Presuppositions introduce new DRSs that need to
  be incorporated in the discourse context
• It is a good way of dealing with the binding,
  projection, and denial problems

95
Van der Sandts Theory

• Presuppositions introduce new DRSs that need to
  be incorporated in the discourse context
• There are two ways to resolve presuppositional
  DRSs
• By binding
• By accommodation

96
Two birds with one stone

• The presupposition as anaphora theory handles
  anaphoric pronouns and presuppositions in
  essentially the same way Presupposition
  Anaphora Anaphora Presupposition

97
One mechanism

• Essentially one mechanism to deal with pronouns,
  proper names, definite descriptions, etc.
• The differences are accounted for in the way they
  can accommodate and bind
• Pronouns do not accommodate
• Proper names always accommodate globally
• Definite descriptions can accommodate anywhere

98
Presuppositions in DRT

• We need to carry out two tasks
• Select presupposition triggers in the lexicon
• Indicate what they presuppose
• We will use a new operator, the alpha-operator,
  ?
• If B1 and B2 are DRSs, the so is B1?B2
• B1 is the presupposition of B2

99
Preliminary DRSs

• She dances
• Mia dances
• The woman dances

x
female(x)

dance(x)
?
x
mia(x)

dance(x)
?
x
woman(x)

dance(x)
?
100
Presupposition in the lexicon

• She
• Mia
• The woman

x
female(x)
? p_at_x
?p.
x
mia(x)
? p_at_x
?p.
x
woman(x)
? p_at_x
?p.
101
Indefinite vs. Definite NP

• A woman
• The woman

x
woman(x)
p_at_x
?p.
x
woman(x)
? p_at_x
?p.
102
The algorithm

• After constructing a preliminary DRS for an input
  sentences, we still have to resolve the
  presuppositions
• After resolution we will have an ordinary DRS
  that we can use for our inference tasks
• Resulting DRS needs to be consistent and
  informative

103
Binding Presuppositions

• ExampleVincent danced with a woman.

x y e
vincent(x) dance(e) agent(e,x) with(e,y) woman(y)
104
Binding Presuppositions

• ExampleVincent danced with a woman.The woman
  collapsed.

x y e
vincent(x) dance(e) agent(e,x) with(e,y) woman(y)

collapse(z)
z
woman(z)
?
(
)
105
Binding Presuppositions

• ExampleVincent danced with a woman.The woman
  collapsed.

x y e
vincent(x) dance(e) agent(e,x) with(e,y) woman(y)

collapse(z)
z
woman(z)
?
(
))
(
merge
106
Binding Presuppositions

• ExampleVincent danced with a woman.The woman
  collapsed.

x y e
vincent(x) dance(e) agent(e,x) with(e,y) woman(y)

collapse(z)
z
woman(z) zy
?
(
))
(
pick antecedent
107
Binding Presuppositions

• ExampleVincent danced with a woman.The woman
  collapsed.

x y e z
vincent(x) dance(e) agent(e,x) with(e,y) woman(y) woman(z) zy

collapse(z)
(
)

move
108
Binding Presuppositions

• ExampleVincent danced with a woman.The woman
  collapsed.

x y e z
vincent(x) dance(e) agent(e,x) with(e,y) woman(y) woman(z) zy collapse(z)
merge reduction
109
Accommodating Presuppositions

• ExampleIf Mia dates Vincent, then her husband
  is out of town

x y
mia(x) vincent(y)
z
husband(z) of(z,x)

date(x,y)

out(z)
?
?(
)
110
Global accommodation

• ExampleIf Mia dates Vincent, then her husband
  is out of town

x y
mia(x) vincent(y)
z
husband(z) of(z,x)

date(x,y)

out(z)
?
?(
)
111
Global Accommodation

• ExampleIf Mia dates Vincent, then her husband
  is out of town

x y z
mia(x) vincent(y) husband(z) of(z,x)

date(x,y)

out(z)
?
112
Non-global accommodation

• Performing global accommodation is saying that
  something is presupposed.
• But recall the projection problem.
• Presuppositions can be neutralised by binding and
  non-global accommodation.

113
Non-global Accommodation

• ExampleIf Mia is married, then her husband is
  out of town

x
mia(x)

married(x)
z
husband(z) of(z,x)

out(z)
?
?(
)
114
Non-global Accommodation

• ExampleIf Mia is married, then her husband is
  out of town

x
mia(x)

married(x)
z
husband(z) of(z,x)

out(z)
?
?(
)
115
Non-global Accommodation

• ExampleIf Mia is married, then her husband is
  out of town

x
mia(x)
z
married(x) husband(z) of(z,x)

out(z)
?
116
Preferences

• Binding is preferred to accommodation
• Global accommodation is preferred to local
  accommodation

117
Van der Sandts Algorithm

1. Generate a DRS for the input sentence, with all
   elementary presuppositions marked by ?
2. Merge this DRS with the DRS of the discourse so
   far processed
3. Traverse the DRS, and on encountering an ?-DRS
   try to
4. Bind the presupposed information to an accessible
   antecedent, or
5. Accommodate the information to a superordinated
   level of DRS
6. Remove those DRSs from the set of potential
   readings that violate the acceptability
   constraints

118
The acceptability constraints

• DRSs should obey the binding rules
• DRSs should not contain free variables
• DRSs should be consistent and informative
• DRSs should also be locally consistent and
  locally informative

119
Free Variable Check

• Consider the exampleEvery man likes his car
• DRS obtained with global accommodation

y
car(y) of(y,x)
x
man(x)

like(x,y)
?
120
Free Variable Check

• Consider the exampleEvery man likes his car
• DRS obtained with global accommodation

y
car(y) of(y,x)
x
man(x)

like(x,y)
?
121
Free Variable Check

• Consider the exampleEvery man likes his car
• DRS obtained via intermediate accommodation

x y
man(x) car(y) of(y,x)

like(x,y)
?
122
Free Variable Check

• Consider the exampleEvery man likes his car
• DRS obtained with local accommodation

y
car(y) of(y,x) like(x,y)
x
man(x)
?
123
The projection problem solved

• Recall our exampleIf Mia is married, then her
  husband is out of town
• Local constraints play a crucial role here!

x z
mia(x) husband(z) of(z,x)

out-of-town(z)

married(x)
?
124
The projection problem solved

• Recall our exampleIf Mia is married, then her
  husband is out of town
• Local constraints play a crucial role here!

x z
mia(x) husband(z) of(z,x)
Locally uninformative

out-of-town(z)

married(x)
?
125
The projection problem solved

• Recall our exampleIf Mia is married, then her
  husband is out of town
• Local constraints play a crucial role here!

x
mia(x)
Locally informative
z
married(x) husband(z) of(z,x)

out-of-town(z)
?
126
Denial

• ExampleVincent does not like his dog.He does
  not have a dog!

x
vincent(x)
y
dog(y) of(y,x) like(x,y)
?
127
The binding problem solved

• ExampleA boxer nearly escaped from his
  apartment.
• Preliminary DRS

z
apartment(z) of(z,x)

nearly-escaped-from(x,z)
x
boxer(x)
?
(
))
(
128
The binding problem solved

• ExampleA boxer nearly escaped from his
  apartment.
• Preliminary DRS

z
apartment(z) of(z,x)

nearly-escaped-from(x,z)
x
boxer(x)
?
(
))
(
x z
boxer(x) apartment(z) of (z,x) nearly-escaped-from(x,z)

• Final DRS

129
Proper Names

• Proper Names can be treated as presupposition
  triggers
• Only global accommodation is permitted for proper
  names
• This assures they will always end up in the
  global (outermost) DRS, accessible for subsequent
  pronouns

130
Proper Names

• ExampleEvery man knows Mia. She is Marsellus
  wife.

know(x,y)
y
mia(y)
x
man(x)
?
?
131
Proper Names

• ExampleEvery man knows Mia. She is Marsellus
  wife.

y
mia (y)

know(x,y)
x
man(x)
?
132
Implementation

• The Curt system
• Small fragment of English
• Pronouns, presupposition triggers
• Uses theorem prover
• Bliksem
• Uses model builder
• Mace
• Does all inference tasks