CAS LX 502 Semantics - PowerPoint PPT Presentation

1 / 21
About This Presentation
Title:

CAS LX 502 Semantics

Description:

The ps is inconsistent with one of the conversational implicatures of the matrix ... If the problem has been solved, then it is Lauri who solved it. Implicature: ... – PowerPoint PPT presentation

Number of Views:30
Avg rating:3.0/5.0
Slides: 22
Provided by: buEduling
Learn more at: http://www.bu.edu
Category:
Tags: cas | semantics

less

Transcript and Presenter's Notes

Title: CAS LX 502 Semantics


1
CAS LX 502Semantics
  • 10a. Presupposition
  • 10.2

2
Presupposition
  • Frege 1892
  • Referring expressions (names, definite
    descriptions) carry the presupposition that they
    do in fact refer.
  • For a sentence to have a truth value, its
    presupposition must hold.
  • A presupposition of a sentence is also a
    presupposition of its negation.

3
Presupposition triggers
  • Lexical triggers
  • Definite noun phrases
  • The student fell asleep.
  • The student didnt fall asleep.
  • Factive verbs
  • Pat wanted to eat a sandwich.
  • Tracy thought Pat ate a sandwich.
  • Tracy realized Pat ate a sandwich.
  • Pat regretted eating a sandwich.
  • Pat liked eating a sandwich.
  • Pat ate a sandwich.

4
Presupposition triggers
  • Lexical triggers
  • Change of state verbs
  • Pat stopped eating a sandwich (at 2pm).
  • Pat started eating a sandwich (at 2pm).
  • Verbs of judgment
  • Tracy blamed Pat for eating the sandwich.
  • Tracy faults Pat for eating the sandwich.

5
Presupposition triggers
  • Structural triggers
  • Cleft constructions and focus
  • It was Tracy that ate the sandwich.
  • Tracy ate the sandwich.
  • The sandwich was eaten.
  • It was the sandwich that Tracy ate.
  • What Tracy ate was the sandwich.
  • Tracy ate the sandwich.
  • Tracy ate something.
  • Too, even
  • Even Tracy ate a sandwich.
  • Tracy ate a sandwich too.

6
Presupposition survival
  • Families of sentences.
  • The Queen of America is bald.
  • The Queen of America isnt bald.
  • Is the Queen of America bald?
  • Perhaps the Queen of America is bald.
  • If the Queen of America is bald, I will not wear
    a wig.
  • Joan has stopped drinking wine for breakfast.
  • Joan hasnt stopped drinking wine for breakfast.
  • Has Joan stopped drinking wine for breakfast?
  • Perhaps Joan has stopped drinking wine for
    breakfast.
  • If Joan has stopped drinking wine for breakfast,
    I will celebrate.

7
Non-presupposition non-survival
  • Families of sentences.
  • Its cold in here.
  • Its not cold in here.
  • Is it cold in here?
  • Perhaps its cold in here.
  • If its cold in here, John loses the bet.
  • Rocky is a squirrel.
  • Rocky is not a squirrel.
  • Is Rocky a squirrel?
  • Perhaps Rocky is a squirrel.
  • If Rocky is a squirrel, I will celebrate.

8
Non-presupposition survival
  • Families of sentences.
  • Pat, who bought something here, ate lunch.
  • Pat, who bought something here, didnt eat lunch.
  • Did Pat, who bought something here, eat lunch?
  • Perhaps Pat, who bought something here, ate
    lunch.
  • If Pat, who bought something here, ate lunch, I
    will not cook the spinach.
  • What Pat bought here was a fork.
  • What Pat bought here wasnt a fork.
  • Was what Pat bought here a fork?
  • Perhaps what Pat bought here was a fork.
  • If what Pat bought here was a fork, I will
    celebrate.

9
So
  • The families of sentences test can tell us when
    something is not a presupposition (a
    presupposition will always survive).
  • Or will it?
  • The Queen of America isnt baldthere is no Queen
    of America.

10
Metalinguistic negation, etc.
  • Matt groaning isnt standing over there, Matt
    grayning is.
  • He isnt adequate at painting eggshes fantastic
    at it.
  • Objecting to the sentence, on various grounds
    (pronunciation, implicature, connotation,
    register, ).
  • Usually comes with a special intonation.
  • If the Queen of America just opened the window,
    America must have a Queen.

11
ps projection holes and plugs
  • Joan has stopped drinking wine for breakfast.
  • Joan previously drank wine for breakfast.
  • Pat knows that Joan has stopped drinking wine for
    breakfast.
  • Joan previously drank wine for breakfast.
  • Pat believes that Joan has stopped drinking wine
    for breakfast.
  • (No presupposition)
  • PlugsDo not let presuppositions through
    (believe)
  • HolesLet presuppositions through (know)

12
ps projection filters
  • JOHN drinks too.
  • JOHN drinks too and Mary doesnt like it.
  • Bill is not present and JOHN drinks too.
  • Somebody else drinks besides John.
  • Bill drinks and JOHN drinks too.
  • No constraint on the context?
  • If JOHN drinks too, then the bottle will not last
    long.
  • If the bottle is already empty, then JOHN drinks
    too.
  • Somebody else drinks besides John.
  • If Bill drinks, then JOHN drinks too.
  • No constraint on the context?
  • Generalization so far p?q and p?q inherit the
    presuppositions of both p and q except any pss of
    q entailed by p. Conjunctions and conditionals
    are filters (letting some pss through and others
    not).

13
ps projection filters refined
  • If there is a depression, the president of GM
    will lose his job too.
  • Someone besides the president of GM will lose
    his/her job?
  • Well, no, not really. But why? There being a
    depression doesnt logically entail that someone
    loses his/her job.
  • But the normal context is one in which it is
    assumed that when there is a depression, many
    people lose their jobs. In that context, p
    entails the presupposition of qand so it is
    filtered out.
  • If there is a thunderstorm, the president of GM
    will lose his job too.
  • Filter out any ps of q which is entailed by the
    contextp.

14
ps projection local contexts
  • If there is a depression, the president of GM
    will lose his job too.
  • Someone besides the president of GM will lose
    his/her job?
  • A way to look at this local contexts
  • Suppose c is the context, p?q is our sentence.
  • To be felicitous, c must entail ps(p).
  • Add p to c, and cp is now the context against
    which q is evaluated.
  • To be felicitous, cp must entail ps(q).
  • So in a normal context where when theres a
    depression all sorts of people lose their jobs,
    cp will be a context in which all sorts of
    people lose their jobs, and this entails ps(q).
    So, the overall c is fine so long as cp entails
    ps(q).

15
ps projection filters refined
  • Bill is absent and JOHN drinks too
  • Someone besides John drinks?
  • Actually, that one was a bit misleading.
  • A I need a non-drinker to support me in avoiding
    alcohol. It is known that Bill does not drink,
    and youve just mentioned the known fact that
    Mary drinks.
  • A kind of normal context, compatible. Because
    ps(q).
  • B I need a non-drinker to support me in avoiding
    alcohol. The only possible reason for anybody to
    not be present (in this room) is to drink vodka
    in the other room.
  • But here, ps(q) seems not to be presupposed. Why?
  • The difference is that in this context p entails
    ps(q).
  • So, really, it didnt presuppose ps(q) after all,
    but rather that cp entails ps(q). If p is deemed
    irrelevant, then the context by itself has to
    entail ps(q).

16
Some difficulties with filters
  • I havent told the truth.
  • I realize that I havent told the truth.
  • I didnt realize that I havent told the truth.
  • I regret that I havent told the truth.
  • If JOHN sings too then I will celebrate.
  • If I regret that I havent told the truth, I will
    confess it to everyone.
  • If I realize that I havent told the truth, I
    will confess it to everyone.
  • If Frank realizes that he hasnt told the truth,
    he will confess it to everyone.

17
A cancellation view
  • Triggers yield potential presuppositions. A
    potential ps may or may not become an actual ps.
    In particular, not if it is canceled. This
    happens when
  • The ps is inconsistent with assumptions in the
    context.
  • The ps is inconsistent with one of the
    conversational implicatures of the matrix
    sentence containing the trigger.
  • If the problem has been solved, then it is Lauri
    who solved it.
  • Implicature truth value of the problem has been
    solved (somebody has solved the problem) not
    assumed.

18
A cancellation view
  • Triggers yield potential presuppositions. A
    potential ps may or may not become an actual ps.
    In particular, not if it is canceled. This
    happens when
  • The ps is inconsistent with assumptions in the
    context.
  • The ps is inconsistent with one of the
    conversational implicatures of the matrix
    sentence containing the trigger.
  • I dont have a dog.
  • So at least you dont have to walk your dog.
  • Problems here too.
  • If John has twins, Mary will not like his
    children.

19
Projection into ones beliefs?
  • The unicorn is waiting in the garden.
  • Yet there are no unicorns.
  • Pat knows that the unicorn is waiting in the
    garden.
  • Yet there are no unicorns.
  • Pat thinks that the unicorn is waiting in the
    garden.
  • Yet there are no unicorns (silly Pat).
  • Yet, Pat believes there are no unicorns.
  • Pat wants the unicorn to sleep in the garden.
  • If the unicorn is in the garden, Pat will be
    happy.

20
Attitude verbs
  • Patrick wants to sell his cello.
  • Patrick is under the mistaken belief that he owns
    a cello, and he wants to sell his cello.
  • Embed q under an attitude verb (SVq) requires
    that ps(q) be entailed by Ss beliefs.

21
?
  • ? ?
  • ?
  • ? ?
  • ? ?
  • ?
  • ?
Write a Comment
User Comments (0)
About PowerShow.com