This is not the document you are looking for? Use the search form below to find more!

Report home > Education

Argument-Adjunct Asymmetries in Rhetorical Questions

0.00 (0 votes)
Document Description
Argument-Adjunct Asymmetries in Rhetorical Questions
File Details
  • Added: April, 15th 2011
  • Reads: 239
  • Downloads: 0
  • File size: 71.79kb
  • Pages: 7
  • Tags: argument, adjunct, rhetorical questions
  • content preview
Submitter
  • Name: valentina
Embed Code:

Add New Comment




Related Documents

Typology of Rhetorical Questions as a Stylistic Device in Writing

by: franciszka, 8 pages

This paper is concerned with the use of rhetorical questions (RQs), a stylistic device often not recognized as such. The problem of reading and writing in a second language in Nigeria and also using ...

Rhetorical Questions, Relevance and Scales

by: harumi, 17 pages

Rhetorical questions, and other varieties of pragmatically conditioned questions, present a challenge for a purely truth-conditional theory of the interpretation of interrogatives. In this paper, it ...

Asymmetries in the Effects of Monetary Policy: The Case of South Africa

by: shinta, 18 pages

PPP is unlikely to hold instantaneously for all commodities across the different regions of a monetary area. It is therefore possible that monetary expansions or contractions will have ...

Argument constructions and language processing: Evidence from a ...

by: fazila, 18 pages

The notion of argument construction is widely accepted in Cognitive Linguistics circles as a highly explanatory theoretical construct. It has recently been incorporated into the ...

Information asymmetry in decision from description versus decision from experience

by: shinta, 9 pages

In this paper we investigate the claim that decisions from experience (in which the features of lotteries are learned through a sampling process) differ from decisions from description (in ...

TRANSACTION COSTS AND SMALLHOLDER PARTICIPATION IN THE MAIZE MARKET IN THE NORTHERN PROVINCE OF SOUTH AFRICA

by: samanta, 5 pages

Smallholder maize farming is characterised by low levels of market participation. The objective of the paper is to determine the role of transaction costs in participation of smallholder farmers in ...

ISLAM AND DEMOCRACY IN THE MIDDLE EAST: THE IMPACT OF RELIGIOUS ...

by: erin, 31 pages

Discussions about democracy in the Arab world often include attention to the political orientations of ordinary men and women. In particular, questions are raised about whether popular attitudes and ...

Acupuncture in Drug and Alcohol Withdrawal at the Community Residential Withdrawal Unit, Footscray Hospital, Melbourne

by: bela, 8 pages

Acupuncture has been offered as an adjunct therapy in drug and alcohol withdrawal at the Community Residential Withdrawal Unit (CRWU), Western Hospital, Footscray, since 1996. Anecdotal reports from ...

Arabic Parsing Using Grammar Transforms Lamia Tounsiand Josefvan ...

by: wasil, 4 pages

We investigate Arabic Context Free Grammar parsing with dependency annotation comparing lexicalised and unlexicalised parsers. We study how morphosyntactic as well as function tag information ...

Andragogy and the Socratic Method : The Adult Learner Perspective

by: shinta, 11 pages

One of the teaching methods available to teachers in the ‘andragogic’ model of teaching is the method of ‘Socratic Seminars’. This is a teacher-directed form of ...

Content Preview
NELS 29 at the University of Delaware, October 18, 1998
(2)
a. information seeking question:
A: Who has been to Moose Jaw?
Argument-Adjunct Asymmetries in Rhetorical Questions
B: Rossy and Kika have been to Moose Jaw.
Rajesh Bhatt, Penn/MIT
b. rhetorical question:
bhatt@linc.cis.upenn.edu
A: Have you (ever) been to Moose Jaw?
B: Of course not. After all, who has been to to Moose Jaw?
Two contrasting views about where certain semantic well-formedness condi-
(= Of course not. No one has ever been to Moose Jaw)
¯
tions such as NPI-licensing and semantic islands like negative islands apply:
Other ways of disambiguation are provided by
A: the level of LF
(i) NPI licensing (test due to Sadock 1971, 1974)
B: a level of post-LF, which retains structure but permits certain kinds of logical
Ordinary questions cannot license ‘strong’ NPIs e.g. give a damn, at all, in weeks, a
inferences.
bit, budge an inch, a red cent, lift a finger, until etc. (cf. Zwarts 1996).
Han (1997) on NPI-licensing in rhetorical questions:
¯
(3)
a. * Who has given Miguel a red cent?
(1)
a. Positive information seeking -questions do not license strong NPIs:
b. * Have you given Miguel a red cent?
# Who lifted a finger to help Sam?
Rhetorical questions can license ‘strong’ NPIs.
b. Positive Rhetorical questions license strong NPIs:
After all, who lifted a finger to help Sam?
(4)
a. Who has (ever) given Miguel a red cent?
c. Negative questions license NPIs:
b. Have you (ever) given Miguel a red cent?
Who didn’t lift a finger to help Sam?
(ii) No long distance extraction of adjuncts in rhetorical questions.
d. Negative Rhetorical questions do not license NPIs:
# After all, who didn’t lift a finger to help Sam?
1.1 Actual Derivation
I provide support for View B by using data from certain asymmetries in extrac-
¯
Setting the question denotation to
tion in rhetorical questions.
¯
This approach is suggested but not quite worked out in Ladusaw (1980) (also see
Guti´errez-Rexach (1997)).
1
What are Rhetorical Questions?
(5)
a. What did John eat?
b. Question denotation =
ˇ
Ô
Ü
Ô
Ø´
ܵ
Ô
Rhetorical questions vs. Information seeking questions:
c. Question denotation
d.
Information seeking questions: A person who asks an information seeking
Ü
Ø´
ܵ
¯
question expects an answer.
Han (1997)’s approach: Since a rhetorical question asserts that the set of in-
¯
dividuals that satisfies the question is empty, she proposes that the wh-phrase
Rhetorical questions do not solicit an answer.
is mapped onto a negative quantifier which is interpreted with the scope of the
¯
Rhetorical questions assert that the extension of the question denotation is empty.
wh-phrase.
¯
Ambiguity: In principle, a given question can be interpreted as either an infor-
(6)
a. What did John eat t ?
mation seeking question or as a rhetorical question. However, intonation can be
b. Nothing did John eat t
used to disambiguate whether a given question is to be interpreted as an infor-
¼
¼
mation seeking question or as a rhetorical question.
c. (
ˇ
Ü´Ø
Ò
´Üµ
´Üµµµ´
Ü
Ø
´
Ü
µµ
d.
¼
¼
Ü
´Ø
Ò
´Üµ
Ø
´
ܵµ
2

The two approaches yield essentially the same LFs - both LFs have a wide scope
i. Information seeking question:
negation.
Why should John say that Fritz was fired? (ambiguous)
= For what reason R, should(say(j, fired(f), R))? (higher construal)
= For what reason R, should(said(j, fired(f, R))) (lower construal)?
2
Argument-Adjunct asymmetries
ii. Rhetorical question:
Why should John (ever) say that Fritz was fired? (only higher)
Long distance extraction of arguments is possible (contra Han & Siegel 1996 4).
¯
Ü
=
reason R, said(j, fired(f), R) (higher construal)
(7)
a. Who has Sven (ever) said [ t was intelligent]?
=
reason R, said(j, fired(f, R)) (lower construal)
=
said(sven, intelligent(x))
c. where
Ü
= said(sven,
intelligent(x))
Ü
i. Information seeking question:
b. What has Max (ever) believed [that Matt does well t ]?
Where did John say that Fritz saw Karl? (ambiguous)
=
believed(max, does-well(matt, x))
= At what location L, said(j, saw(f, k), L)? (higher construal)
Ü
= believed(max,
does-well(matt, x))
= At what location L, said(j, saw(f, k, L)) (lower construal)?
Ü
c. [With what job] has Mina (ever) thought that [you should be satisfied
ii. Rhetorical question:
t ]?
After all, where did John (ever) say that Fritz (ever) saw Karl.
=
thought(mina, should(be-satisfied-with(you, x)))
(only higher)
Ü
= thought(mina,
should(be-satisfied-with(you, x)))
=
location L s.t. said(j, saw(f, k), L) (higher construal)
Ü
=
location L s.t. said(j, saw(f, k, L)) (lower construal)
However, it does not seem to be possible to form rhetorical questions by extract-
¯
d. when
ing adjuncts from embedded clauses. Matrix extraction of adjuncts is, of course,
possible.
i. Information seeking question:
½
When did John say that Fritz met Karl? (ambiguous)
(8)
a. why, non-modal
= At what time T, said(j, saw(f, k), T)? (higher construal)
i. Information seeking question:
= At what time T, said(j, saw(f, k, T)) (lower construal)?
Why did John say that Fritz was fired? (ambiguous)
ii. Rhetorical question:
= For what reason R, said(j, fired(f), R)? (higher construal)
After all, when did John (ever) say that Fritz (ever) met Karl. (only
= For what reason R, said(j, fired(f, R)) (lower construal)?
higher)
ii. Rhetorical question:
=
time T s.t. said(j, saw(f, k), T) (higher construal)
Why did John (ever) say that Fritz was fired? (only higher)
=
location T s.t. said(j, saw(f, k, T)) (lower construal)
=
reason R, said(j, fired(f), R) (higher construal)
If we put an adverb like first/for the first time in the embedded clause of a when
=
reason R, said(j, fired(f, R)) (lower construal)
question such as (9a, b), only the lower clause construal of when is possible.
b. why, modal
(9) (information seeking questions)
½
The existence of rhetorical why-questions has been denied Cf. Lee (1995) and Guti´errez-
Rexach (1997). While it is true that such questions do not pattern with other rhetorical with
a. When
did John say t [that Moreau first hit Ray t ]?
£
respect to several diagnostics, it can be shown that this follows from the presuppositions asso-
b. When
did John say t [that Moreau hit Ray for the first time t ]?
ciated with the predicate cause which is implicit in the semantics of why-questions and that ‘no
£
reason’ usually means ‘no good reason’.
When this question is given a rhetorical interpretation, the lower construal van-
ishes. However, for the adverb in the embedded clause to be interpreted properly,
the when has to be construed with it. Therefore, we get ungrammaticality.
3
4

(10) (rhetorical questions)
The basic problem is that the negative force contributed by the rhetorical question
a. * After all when did John say [that Moreau (ever) first hit Ray]?
is too high to function as an intervener as it does in a negative island.
b. * After all when did John say [that Moreau (ever) hit Ray for the first
time ]?
3.1 Tests for the scope of rhetorical negation
Differential behavior of adverbs:
¯
3.1.1 Licensing of subject NPIs
Not all adjuncts behave alike. For some speakers, a lower construal is possible
with when and where but a lower construal is not possible with why (or how) for
Matrix negation does not license any NPIs in subject position. Adjunct informa-
any of the speakers consulted.
tion seeking questions do not license strong NPIs in subject position and for some
I will refer to the absence of long distance construal of certain adjuncts in rhetor-
speakers do not even license weak NPIs in subject position.
ical questions as the rhetorical island effect.
(13)
a. * A red cent wasn’t given to Marisa.
b. * Anyone didn’t give a book to Ermo.
3 Assimilation to Negative Islands
c. * When was a red cent given to Marisa?
d. ??/* When did anyone give a book to Ermo?
Sadock (1971, 1974)’s observations:
¯
Positive rhetorical questions have the force of negative assertions and negative
However, in rhetorical questions subject NPIs are licensed.
rhetorical questions have the force of positive assertions.
(14)
a. After all when was a red cent ever given to Marisa?
We can think of the semantics of rhetorical questions as involving a negation.
b. After all, when did anyone ever give a book to Ermo?
‘Argument-Adjunct’ asymmetry
¯
The ‘rhetorical island’ involves something like a negation and ‘argument-adjunct’
3.1.2 Scope of Modality w.r.t. Negation
asymmetries. Can it be assimilated to Negative Islands?
At first, it seems that the answer has to be NO: the negative force is in the wrong
In declarative sentences and questions with a deontic modal and matrix negation,
¯
place.
the deontic modal unambiguously takes scope over negation or a negative QP
Negative islands involve a negation/downward-entailing function that intervenes
that it c-commands. (test and examples from Han 1997, 7)
Ü
between a wh-phrase and its trace. (cf. Ross (1984), Rizzi (1990), De Swart (1992),
Szabolcsi & Zwarts (1991, 1993), Rullmann (1995), Beck (1996) inter alia)
(15)
a. Carlo must not eat the cake. (= It is obligatory for Carlo to not eat the
cake)
(11) Neg. Island Configuration 1:
(12) Neg. Island Configuration 2:
b. Carlo must never eat the cake. (= It is obligatory for Carlo to never eat
CP
the cake)
CP
¨À
¨
À
c. When must Carlo not eat the cake? (= When is it obligatory for Carlo
¨À
¨
À
¨
À
¨
À
¨
À
WH
C
¨
À
¼
WH
C
to eat the cake?)
¼
¨À
¨
À
¨À
d. Where must Carlo never eat the cake? (=Where is it obligatory for
¨
À
¨
À
¨
À
¨
À
?
IP
¨
À
?
IP
Carlo to never eat the cake?)
¨À
¨
À
¨À
¨
À
¨
À
¨
À
¨
À
However, in positive rhetorical questions that involve a deontic modal, the nega-
¨
À
¨
À
Subj
VP
DE-subject
VP
tion contributed by the rhetorical force takes scope over the deontic modal.
¨À
¨
À
¨À
¨
À
¨
À
¨
À
DE-function
t
t
5
6

(16)
a. What must Joan say? (= There is nothing s.t. it is obligatory for Joan
(20) The only locations under discussion are the houses in a small village. The
to say it)
village is plagued by a supernatural being named Homer who has the
b. What should Strube do? (= There is nothing s.t. it is obligatory for
habit of appearing in people’s houses and then promptly dying there.
Strube to do it)
Most of the houses in this village have already been ‘died in’ by Homer. In
order to exorcise Homer from the village, a house is needed where Homer
hasn’t died so far.
3.1.3 Scope of subject quantifiers
In the context sketched in (20), the oddness of (19b) disappears.
Certain quantifiers such as some cannot take scope under negation (cf. Kroch
D-linking (cf. Pesetsky 1987) alleviates Negative Islands. A D-linked wh-phrase
1974).
¯
e.g. which books requires that both the speaker and hearer have a set of books in
(17) Some student didn’t come
mind. The presence of the D-linked set makes the negative answer be pragmati-
=
cally informative so that there is no pragmatic oddness.
Ü
×ØÙ
Òشܵ
ÓÑ
´Üµ
=
However, not all wh-words/phrases are equally amenable to D-linking: when and
Ü
×ØÙ
Òشܵ
ÓÑ
´Üµ
where are easily interpreted as D-linked, while why and how resist D-linking.
However, it seems to be possible for some to take scope under rhetorical negation.
(21) (exs. 5-8b from Ch. 5 of Rullmann 1995)
(18) After all, when has some student ever volunteered himself for some extra
work?
a. * I wonder how Judy didn’t play with her dog.
=
b. # I wonder why Judy didn’t play with her dog. (under lower con-
Ø
Ü
×ØÙ
Òشܵ
Ú
ÓÐ
ÙÒØ
Ö
´Ü
ص
=
strual)
Ü
Ø
×ØÙ
Òشܵ
Ú
ÓÐ
ÙÒØ
Ö
´Ü
ص
c. I wonder where Judy didn’t play with her dog.
3.2 Behavioral parallels with Negative Islands
d. I wonder when Judy didn’t play with her dog.
Despite the fact that the rhetorical negation is not in the right location for it to be
¯
3.2.1 Assimilation to Negative Islands?
an ‘intervener’, we find parallels between the behavior of Negative Islands and
the rhetorical island.
Negative Islands improve if the context is fixed so as to make the ‘answer’ more
Negation has a general effect on the acceptability of questions.
reasonable. One way to do this is by fixing the context and explicitly D-linking
the relevant wh-phrase.
(19)
a. Where did Homer die?
b. ??Where didn’t Homer die?
(22)
a. why
i. For which reason did John say that Fritz had been fired?
An affirmative question like (19a) typically has a restricted set of true answers.
(ambiguous between higher and lower construal)
The corresponding negative question, cf. (19b) has a very large number of an-
swers, which are unlikely to be informative. This makes (19b) pragmatically
ii. * Why did/should John not say that [that Fritz had been fired t ]?
odd.
(non D-linked wh-phrase, context not fixed)
¾
If, however, the context is such that the number of potential answers is restricted
iii. # [For which reason] did John not say that [that Fritz had been
enough to become informative, the oddness disappears. Consider for example
fired t ]?
(19b) in the following context:
(D-linked wh-phrase, context fixed, still marginal)
iv. For which reason did/should John (ever) say that Fritz had been
¾
cf. Rullmann (1995) for a more precise explication and implementation of the idea sketched
here. Beck & Rullmann in Beck (1996) offer some arguments against the exact formulation in
fired?
Rullmann (1995) but the basic explanation is retained.
(higher construal only)
7
8

b. where
(24) After all, where did John (ever) say that Fritz (ever) saw Karl. (only higher)
i. In which cities did John say that Fritz saw Karl?
=
location L s.t. said(j, saw(f, k), L) (higher matrix clause construal)
(ambiguous between higher and lower construals)
=
location L s.t. said(j, saw(f, k, L)) (lower embedded clause construal)
ii. # Where did John not say [that Fritz saw Karl t ]?
(non D-linked wh-phrase, context not fixed)
4
Matrix infinitival why questions
iii. [In which cities] did John not say [that Fritz saw Karl t ]?
(D-linked wh-phrase, context fixed)
In many languages, it is possible to construct matrix questions with why and a
iv. After all, in which cities did John (ever) say that Fritz saw Karl.
bare infinitival clause (cf. Green 1973, Sadock 1974).
(higher construal preferred, but lower construal also possible)
(25)
a. Why leave?/ Why worry?
c. when
b. Why not sign our guestbook? (found on a web-site)
i. In what years/On what days did John say that Fritz met Karl?
(ambiguous between higher and lower construals)
Matrix infinitival why questions are rhetorical questions.
¯
Like rhetorical questions in general, they make an assertion of the opposite po-
ii. # When did John not say [that Fritz saw Karl t ]?
larity from what is ostensibly asked. This is reflected in their NPI-licensing prop-
(non D-linked wh-phrase, context not fixed)
erties.
iii. [In which years/On which days] did John not say [that Fritz saw
Karl t ]?
Positive matrix infinitival why questions license strong NPIs:
(D-linked wh-phrase, context fixed)
(26)
a. Why (even) lift a finger?
iv. After all, in which years/on which days did John (ever) say that
b. # Why did John (even) lift a finger?
Fritz hit Karl.
(higher construal preferred, but lower construal also possible)
Negative matrix infinitival why questions do not license strong NPIs:
Fixing the context and overtly D-linking the wh-phrases causes lower construals
(27)
a. # Why not (even) lift a finger?
¯
of when and where to become more accessible to many speakers.
b. Why didn’t John (even) lift a finger?
However, the lower construals do not become available in the case of why.
They can only be used to make suggestions, not to solicit information. As a
¯
result, they are incompatible with non-agentive predicates.
The rhetorical island effect displays a sensitivity to D-linking, which patterns
(28)
a. Why eat so much for breakfast (when you....)?
¯
with Negative Islands: when and where pattern together in showing an improve-
b. # Why find your lost sock under the dryer?
ment, while why does not display a corresponding improvement.
c. Why look for your lost sock under the dryer?
d. # Why seem to be so happy? (ok under agentive reading of ‘seem’)
3.3 A non-parallelism with Negative Islands
e. Why be working at home? (when you could be hanging out in Can-
cun)
Negative islands do not display a matrix-embedded distinction:
f. # Why have finished it by tomorrow?
(23) # Where didn’t Bill say that Homer died?
g. Why be a butcher when you can be a doctor?
(# under both the matrix extraction and embedded extraction reading)
h. Why be given to Mary? (okay under agentive interpretation)?
The rhetorical island effect allows extraction from the matrix clause but not from
i. # Why be being obnoxious?
the embedded clause.
j. Why be obnoxious, when you can be charming?
9
10

It turns out that exactly the set of predicates that can be used as imperatives of
Conjecture 1: Every should has a reason i.e.
the opposite polarity occur in matrix why-infinitival questions.
(34)
P(should (P)
R(because (should (P), R)))
(29)
a. Don’t eat so much for breakfast.
Now applying the contrapositive, we get:
b. # Don’t find your lost sock under the dryer.
c. Don’t look for your lost sock under the dryer.
(35)
should (leave(you))
d. # Don’t seem to be so happy. (ok under agentive reading of ‘seem’)
It is not morally necessary for you to leave.
e. Don’t be working at home (when you could be hanging out in Cancun)
The speaker is asserting that there is no reason which makes it morally necessary
f. # Don’t have finished it by tomorrow.
for the hearer to leave. Therefore the hearer is ‘free’ to not leave.
g. Don’t be a butcher.
The next step involves an implicature.
h. Don’t be given to Mary (okay under agentive interpretation)
What we want to derive is the following:
i. # Don’t be being obnoxious?
(36) Do not leave
j. Don’t be obnoxious.
(with my wishes and desires in mind) you should not leave
I suggest that you not leave.
should( (leave(you)))
4.1 Deriving the semantics of why-imperatives
The implicature is that the speaker by pointing out this freedom of the hearer is
I assume that (30a) involves a covert universal deontic modal. In this sense, it is
suggesting that the hearer stay.
substantially like (30b).
(30)
a. Why leave?
4.2 Rhetorical island effects
b. Why should you leave?
= Why [[should you leave] t ]?
Matrix infinitival why questions obey the rhetorical island effect.
¿
We get the following LF for (30):
(37)
a. Why say that Bill was fired?
=
R should.say(PRO, fired(bill), R)
(31)
R[
because (R, should(leave(you)))]
Ô
Ô
=
R should.say(PRO, fired(bill, R))
These questions are interpreted as rhetorical questions (obligatorily for (30a), op-
b. Why should Ermo say that Bill was fired?
tionally for (30b)) giving us the logical representations in (32)
= for what R should.say(ermo, fired(bill), R)
= for what R should.say(ermo, fired(bill, R))
(32)
(33)
R(because (R, should (leave(you))))
5
Summing Up
¿
There is one other plausible structure for (30a):
i Why [should [you leave t ]]
LF:
R[
because (R, should(leave(you)))
ˇ ]
Rhetorical island effects support the view that more than just LF is needed to
¯
Ô
Ô
Ô
Rhetorical interpretation:
R(because (R, should (leave(you))))
adequately handle certain semantic well-formedness effects. This is so because
In order to derive the right meaning, we have to assume something like (ii).
what makes a question a rhetorical question is its denotation and this information
is not available at LF.
ii
P(should (P)
R(should (because (R, P))))
Even though rhetorical island effects seem related to Negative Islands, it does
This gives us the right meaning cf. (35). However, (ii) does not seem very plausible to me and it
¯
is indeed possible that the representation in (i) is ruled out by the rhetorical island effect.
not seem to be the case that they can be reduced to Negative Islands.
11
12

Nature of the negative force in rhetorical question?
Kroch, A. (1974) The Semantics of Scope in English, Doctoral dissertation, Mas-
¯
Putative evidence for negation:
sachusetts Institute of Technology, Cambridge, Massachusetts. Distributed
by MIT Working Papers in Linguistics.
1. Paraphraseability by negation: in general, just because something can be a
Kroch, A. (1989)
“Amount Quantification, Referentiality, and Long wh-
paraphrased by a negation does not mean that it necessarily involves negation cf.
movement,” unpublished manuscript, University of Pennsylvania.
the case of double negatives.
Lawler, J. (1971) “Any Questions,” in D. Adams, M. A. Cambell, V. Cohen, J.
2. Licensing of strong NPIs - but then so do counterfactual conditionals.
Levins, E. Maxwell, C. Nygren, and J. Reighard, eds., Papers from the 7th Re-
gional Meeting of the Chicago Linguistics Society
, Chicago, Chicago Linguistics
(38) If Martin had lifted a finger to help Martina, she would have forgiven him.
Society, 163–173.
3. Similarity with Negative Island effects - but negation is not necessary for neg-
Lee, F. (1995) “Negative Polarity Licensing in Wh-Questions: The Case for Two
ative islands anyway. Since at least Rizzi (1990), we know that not just negation,
Licensers,” paper presented at the 69th Annual Meeting of the Linguistic
but a larger class of Downward Entailing elements trigger Negative Island effects.
Society of America.
Pesetsky, D. (1987) “Wh-in-Situ: Movement and Unselective Binding,” in E. Reu-
Evidence that rhetorical negation differs from syntactic negation: licensing of until
land and A. ter Meulen, eds., The Linguistic Representation of (In)definiteness,
(39)
a. John didn’t leave until midnight.
MIT Press, Cambridge, MA.
b. *After all who ever left until midnight?
Rizzi, L. (1990) Relativized Minimality, Linguistic Inquiry Monographs 16, MIT
c. *If John had left until midnight, he wouldn’t have missed his plane.
Press, Cambridge, MA.
Ross, J. R. (1984) “Inner Islands,” in C. Brugmann and M. Macaulay, eds., Pro-
So what we have is a Downward-Entailing and Anti-Additive operator, and not
ceedings of the
Ø
Annual Meeting of the Berkeley Linguistics Society, Berkeley,
½¼
a full structural negation.
California, Berkeley Linguistics Society.
Rullmann, H. (1995) Maximality in the semantics of Wh-Constructions, Doctoral dis-
References
sertation, University of Massachusetts-Amherst, Amherst, Massachusetts.
Distributed by GLSA.
Beck, S. (1996) Wh-constructions and transparent Logical Form, Doctoral disser-
Sadock, J. M. (1971) “Queclaratives,” in D. Adams, M. A. Cambell, V. Cohen, J.
tation, Universit¨at T ¨ubingen, T ¨ubingen, Germany. Distributed SfS Report
Levins, E. Maxwell, C. Nygren, and J. Reighard, eds., Papers from the 7th Re-
04-96.
gional Meeting of the Chicago Linguistics Society, Chicago, Chicago Linguistics
Society, 223–232.
Green, G. (1973) “How to get people to do things with words,” in R. W. Shuy,
ed., New Directions in Linguistics, Georgetown University Press, Washington,
Sadock, J. M. (1974) Towards a Linguistic Theory of Speech Acts, Academic Press,
DC, 51–81.
New York, NY.
Guti´errez-Rexach, J. (1997) “The Semantic Basis of NPI-licensing in Questions,”
Swart, H. de (1992) “Intervention effects, monotonicity and scope,” in C. Barker
in B. Bruening, ed., Proceedings of SCIL 8, volume 31, Cambridge, MA,
and D. Dowty, eds., Proceedings of the Second Conference on Semantics and Lin-
MITWPL, 359–376.
guistics Theory, OSU Working Papers in Linguistics 40, Dept. of Linguistics,
Ohio State University, Columbus, Ohio, 387–406.
Han, C. (1997) “Deriving the Interpretation of Rhetorical Questions,” in E. Curtis,
J. Lyle, and G. Webster, eds., Proceedings of the 16th West Coast Conference on
Szabolcsi, A., and F. Zwarts (1993) “Weak Islands and an Algebraic Semantics for
Formal Linguistics, Palo Alto, CA, CSLI.
Scope Taking,” Natural Language Semantics 1:3, 235–284.
Zwarts, F. (1996) “Three Types of Polarity Items,” in F. Hamm and E. Hinrichs,
Han, C., and L. Siegel (1996) “Syntactic and Semantic Conditions on NPI licensing
in Questions,” in B. Agbayani and S.-W. Tang, eds., Proceedings of the 15th
eds., Plural Quantification, Kluwer, Dordrecht. originally circulated as a 1993
West Coast Conference on Formal Linguistics, Palo Alto, CA, CSLI, 177–191.
ms.
13
14

Download
Argument-Adjunct Asymmetries in Rhetorical Questions

 

 

Your download will begin in a moment.
If it doesn't, click here to try again.

Share Argument-Adjunct Asymmetries in Rhetorical Questions to:

Insert your wordpress URL:

example:

http://myblog.wordpress.com/
or
http://myblog.com/

Share Argument-Adjunct Asymmetries in Rhetorical Questions as:

From:

To:

Share Argument-Adjunct Asymmetries in Rhetorical Questions.

Enter two words as shown below. If you cannot read the words, click the refresh icon.

loading

Share Argument-Adjunct Asymmetries in Rhetorical Questions as:

Copy html code above and paste to your web page.

loading