Popular in Course
Popular in Linguistics
This 10 page Class Notes was uploaded by Erich Mueller on Sunday September 6, 2015. The Class Notes belongs to LIN 393 at University of Texas at Austin taught by Staff in Fall. Since its upload, it has received 66 views. For similar materials see /class/181572/lin-393-university-of-texas-at-austin in Linguistics at University of Texas at Austin.
Reviews for 4
Report this Material
What is Karma?
Karma is the currency of StudySoup.
You can buy or earn more Karma at anytime and redeem it for class notes, study guides, flashcards, and more!
Date Created: 09/06/15
Polarity and whconstructions Fall 2003 Bhatt amp Schwarz When negation gets in the way 1 Some phenomena Degree guestions Negation creates an island for whphrases in degree interrogatives Examples form Rullmann 1995 1 a I wonder how tall Marcus is b I wonder how tall Marcus isn t 2 a I wonder how heavy this piece of equipment is b I wonder how heavy this piece of equipment isn t 3 a I wonder how fast Lou can run b I wonder how fast Lou can t run Not all types of whphrases are subject to this restriction 4 a I wonder which book Theodore read b I wonder which book Theodore didn t read Multiple guestions Pesetsky 2000 observes that negation also destroys the acceptability of certain multiple whquestions 5a Who read wha b What did who read 6a Which person read which book b Which book did which person read 7 a Which person didn t read which book b Which book didn t which person read How many questions In how many interrogatives adding negation usually does not destroy acceptability But in some cases the negative sentence lacks a kind ofambiguity that is observed in its positive counterpart 8 How many books are we required to read Polarity and whconstructions Fall 2003 Bhatt amp Schwarz 9 a n X X is a book such that it is necessary that we read X n b n it is necessary that X X is a book such that we read X n If every student has to read the three books listed in the left column below as well as two books from the right column then both three and ve can be true answers to 8 Unexpectedly 11 below only allows for a true answer five not for six seven or eight 10 1 The Venetian Affair 1Anna Karenina 2 The Big Sleep 2 The Magic Mountain 3 Journey into Fear 3 The Return ofthe King 4 The Long Goodbye 5 To the Hilt 11 How many books are we not required to read 12a n X X is a book such that it is not necessary that we read X n b n it is not necessary that X X is a book such that we read X n Questioned clefts The following pattern does not seem to have been discussed in the literature 13a What did you not read b What is it that you didn t read c What isn t it that you read German separation constructions Beck 1996 discusses a family of constructions in which negation cannot intervene between two eXpressions that presumably have to be adjacent at logical form 14 Was glaubst du mit wem Hans gesprochen hat what think you with whom Hans talked has Who do you think that Hans talked to 15 Was glaubstdu nicht mitwem Hans gesprochenhat what think younot with whom Hans talked has Comparatives Von Stechow 1985 notes contrasts like the following 16a John weighs more than b John weighs more than ill weighs B Bill doesn t weigh Polarity and whconstructions Fall 2003 Bhatt amp Schwarz 2 Negative comparatives are unusable Following von Stechow 1985 Rullmann 1995 proposes that more than comparatives are interpreted as sketched below 17 18 19 20 21 John weighs more than Bill weighs more thanz Bill weighs t2 1 John weighs t1 thanz Bill weighs t2 d Bi weighs dmuch 1 John weighs t1 d John weighs dmuch more on 3 1 iffEld de amp dgtmaxoc more thanz Bill weighs t2 1 John weighs t1 1 iff Eld John weighs dmuch amp dgtmaxd Bi weighs dmuch Why does Rullmann take more to pick out the maximal element ofthe set denoted by the thanclause 22 23 24 25 John weighs more than Bi can weigh more thanz Bill can weigh t2 1 John weighs t1 thanz Bill can weigh t2 d Bi can weigh dmuch more thanz Bill weighs t2 1 John weighs t1 1 iff Eld John weighs dmuch amp dgtmaxd Bi can weigh dmuch The thanclause in 22 denotes the set of all weights that Bill can have so 22 is correctly predicted to say that John s weight is greater than the maximal element of that set 26 John weighs more than Bi doesn t weigh 27 28 29 more thanz Bill can weigh t2 1 John doesn t weigh t1 thanz Bill doesn t weigh t2 d Bi doesn t weigh dmuch more thanz Bill weighs t2 1 John weighs t1 1 iff Eld John weighs dmuch amp dgtmaxd Bi doesn t weigh dmuch Since d Bi doesn t weigh dmuch does not have a maximal element the negative comparative sentence is not assigned any truth conditions In Rullmann s account the sentence is ruled out as unusable Polarity and whconstructions Fall 2003 Bhatt amp Schwarz 3 Are negative degree questions unusable too 30a How long did it take b How long didn t it take Question denotations as sets In a view going back to Karttunen 1977 and Hamblin 1973 a question denotes a set of propositions the set of propositions that are possible answers to the question 31 Did Rajesh teach the class that Rajesh taught the class that Rajesh did not teach the class 32 Did Rajesh or Bernhard teach the class that Rajesh taught the class that Bernhard taught the class 33 Which instructor taught the class that Rajesh taught the class that Bernhard taught the class Karttunen stvle composition of whduestions 34 Wh 0c ll Xi laX is true This rule assumes an intensional semantics that is it assumes that declarative sentences denote propositions rather than truth values Accordingly oneplace predicates are taken to denote properties that is functions from the set of individuals into the set of propositions In particular we have instructor Aye D that y is an instructor and 1 t1 taught the class Aye D that y taught the class 35 which instructor 1 t1 taught the class that Bernhard taught the class that Rajesh taught the class Karttunen style semantics for know 36 Theodore knows which instructor taught the class 37 Hot know 3 that for every pe p is true gt oc knows p Exhaustivity in individual guestions 38 Which books has John read 39 1 The Venetian Affair TVA YES 2 The Big Sleep TBS YES 3 Journey into Fear JIF NO Polarity and whconstructions Fall 2003 Bhatt amp Schwarz Following Karttunen and Rullmann let us ignore the semantic effect of plural morphology and assume that book books Aye D that y is a book 40 which books 1 John has read t1 that John read TVA that John read TBS that John has read JIF 41 Theodore knows which books John has read Karttunen s semantics for questions and know predicts that this sentence can be true in our scenario without Theodore knowing that John hasn t read JIF Like Groenendijk and Stokhof1982 Rullmann takes this prediction to be incorrect 42 Theodore knows which books John has read Theodore knows that John has read TVA Theodore knows that John has read TBS l Theodore knows that John has not read JIF Exhaustivity through maximality Rullmann proposes a variant of the Karttunenstyle whquestion rule shown above He does not offer a general rule but the following seems close to what he intends 43 wh 0c 3 that maxy y is true X ocX is true This rule presupposes that the domain of individuals includes sums of individuals like TVA TBS in addition to atomic individuals like TVA or TBS The maximum ofa set of individuals is the sum ofall the individuals in this set For example the maximum of TVA TBS JIF is the sum individual TVA TBS JIF The variable x in the rule ranges not just over atomic individuals but also over sums The condition ocX is true abbreviates x is a sum of one or more individuals y for which ocy is true 44 which books 1 John has read t1 that maxy John has read y TVA that maxy John has read y TBS that maxy John has read y JIF that maxy John has read y TVA TBS that maxy John has read y TVA JIF that maxy John has read y TBS JIF that maxy John has read y TVA TBS JIF Rullmann proposes that such a variant of the Karttunenstyle rule derives the desired inference pattern for cases of know plus interrogative Polarity and whoonstructions Fall 2003 Bhatt amp Schwarz Maximality in degree guestions 45 How long did it take In the logical forms below A is taken to denote a property ofappropriate degrees 46 how A1 it did take t1 long that maxd it took dlong 1 hour that maxd it took dlong 2 hours that maxd it took dlong 3 hours Rullmann takes the maximum ofa set of degrees to be the greatest degree in that set rather than the group or sum ofall the degrees in that set 47 How long didn t it take 48 how A1 it did take t1 long that maxd it did not take dlong 1 hour that maxd it did not take dlong 2 hours that maxd it did not take dlong 3 hours Maximality in how many guestions 49 How many books has John read 50 how A1 t1 many books 2 John has read t2 that maxd John has read dmany books 1 that maxd John has read dmany books 2 that maxd John has read dmany books 3 Rullmann suggests that the maximality analysis also derives the lack of ambiguity in negative how many questions The following example illustrates 51 How many books has John not read 52 how A1 t1 many books 2 not John has read t2 that maxd there are dmany books that John has not read 1 that maxd there are dmany books that John has not read 2 that maxd there are dmany books that John has not read 3 53 how A1 not t1 many books 2 John has read t2 that maxd it is not the case that John has read dmany books 1 that maxd it is not the case that John has read dmany books 2 that maxd it is not the case that John has read dmany books 3 Polarity and whconstructions Fall 2003 Bhatt amp Schwarz Problems for the maximality account Beck and Rullmann 1999 observe that Rullmann s 1995 maximality analysis does not always make the correct predictions on interpretation and acceptability Suppose our marble cake recipe calls for two eggs 54 How many eggs are needed 55 how A1 t1 many eggs are needed that maxd dmany eggs are needed 1 that maxd dmany eggs are needed 2 that maxd dmany eggs are needed 3 We need two eggs to make the cake and we do not need any more than that Therefore the maximum ofd dmany eggs are needed is 2 56 How many eggs are suf cient 57 how A1 t1 many eggs are sufficient that maxd dmany eggs are sufficient 1 that maxd dmany eggs are sufficient 2 that maxd dmany eggs are sufficient 3 Two eggs are suf cient to make the cake lftwo eggs are suf cient then of course any greater number of eggs is suf cient as well Accordingly the set d dmany eggs are sufficient does not have a maximum LIN 393 Topics in Polarity and Whrconstructions November 12 20 even across languages 1 An impOItant argument for the ambiguity of even 11 Background semantics for even 1 WWW P a de ned ifand onlyif i 342 e C 39Q 3 4201 lj likelihoodQ gtlike1ihood1 j b truth conditions 131 l This semantics gives the right results for even in positive environments but makes incorr rect predictions in DE environments 2 a John even greeted Mary John s greeting Mary is unlikelyunexpected b John didn39t even greet Mary John s greeting Mary is likelyexpected Everyone agrees that the ordinary even whose semantics are describd in 1 cannot appear in the immediate scope of negation One way of stating this is that the even in 1 is a PPI As for how to derive the reading in 2b two ways have been proposed 3 a The Scope Theory Karttunen and Peters 1979 Wilkinson 1996 Lahiri 1998 Guerzoni 20 even takes scope over negation LF even P The Ambiguity Theory Rooth 1985 there are two homop onous even s even and evenNm The semantics of evean are such that we get the expected implicatures 7 12 Nonehomophony for evenpm and evenma The ambiguity approach receives support from the fact that in certain languages the pur tative counterparts of evenpp and evean are in fact not homophonous 4 evenpm a John even greeted Mary b Der lans hat sogar dieMaria begruesst the John has even the Mary greeted John even greeted Mary r see also Italian addin nura and Dutch zelfs 5 evenw a Nobody even greeted Mary b Niemand hat auch nur die Maria begruesst Norone has also only the Mary greeted Nobody even greeted Mary see also Italian anche 5010 Dutch 00k maar 2 Guerzoni s Response auch nur also only the putative evenNm does not actually parallel sogar the putative evenpp in the biases itintroduces in questions 6 a base sentences no bias lat der Georg die Maria begruesst7 Has the George the Mary greeted Did George greet Mary b with sogar eVerpi no bias lat der Georg sogar die Maria begruesst7 Has the George even the Mary greeted Did George even greet Mary c with auch nur putative evenNm negative bias lat der Georg auch nur die Maria begruesst7 Has the George also only the Mary greeted Did George even greet Mary 21 A quick review of Guerzoni s proposal The facts 7 a Did John even answer the easiest problem Negative Bias 7 Did John even answer the hardest problem No Bias c Did John even answer Problem 433 Backrground assumption Problem 433 is trivial A Negative Bias Backrground assumption Problem 433 is the toughest A No Bias Backrground assumption Problem 433 is in between A question is incoherent The analysis even can move to a position above the trace of Whether as in LFZ 8 a LFl whetherl t1 even John knows Italian hardPJohn knows Italian hardPJohn doesn t know Italian b LFZ whetherl even t1 ohn knows Italian hardPJohn knows Italian easyPJohn doesn t know Italian If we already have an easyP presupposition the only answer compatible with it is the negative answer in LFZ This is the source of the bias Rooth s ambiguity account can handle the presuppositional behaviour of questions with even but as it stands it cannot explain why we nd negative biases and why the negative biases correlate with easyP Note that Guerzoni s explanation does not depend upon the question negation licensing some NPI In the above example there just was no NPI But even when we have an NPI we can see that the mere presence of an NPI does not force a negative bias 9 a Did George greet anybody b Have you ever been to the LSA 22 Back t0 and nut and sugar Guerzoni attempts to provide a compositional analysis of auch 1111 10 The positive aspects of her proposal a the meaning of auch Illlf follows from the meaning of auch and the meaning of our It is not an idiom ET the NPIrnature of auch 111 does not need to be stipulated hence Guerzoni s approach is Lahirian n the bias facts in questions follow 1 1 Some new assumptions a Postulation ofa novel kind of ambiguity for only b The movement properties of Sugar 221 Onlyandjust 12 Onl zw a Presupposition is de ned if and only if 131 b Exclusivity Assertion E C A QUuklj Guerzoni proposes that under certain Circumstances 111 in German loses its exclusivity assertion In particular in these cases the presupposition and the assertion of 111 only are swi ched 13 HUYZQMP a Exclusivity Presupposition E b Assertion Pu r l magma She then notes that Englishjusl which otherwise has essentially the same semantics as only can in certain circumstances lose its exclusivity l4 equivalent to only met onlyjust Mary b I didn39t meet onlyjust Mary 15 not equivalent to only a Can you sparejust 5 minutes for me tomorrow Presupposition You cannot spare more than 5 minutes for me tomorrow b Can you spare only 5 minutes for me tomorrow References Guerzoni E 2003 WhyEVen Ask Doctoral dissertation Massachusetts Institute of Tech nology Cambridge Massachusetts Karttunen L and S Peters 1979 Conventional lmplicaturequot in CrK Oh and D A Din Presupposilion Syntax and Semantics 11 Academic Press Harcourt Brace ovanovich New York 1755 Lahiri U 1998 Questions and Answers in Embedded Contextsquot book manuscript University of Californiarlrvine Rooth M E 1985 Association with Focus Doctoral dissertation University of MassachusettsrAmherst Amherst Massachusetts Distributed by GLSA Wilkinson K 1996 The scope of evenquot Natural Language Semantics4 1937215 7 m m 7 m D m
Are you sure you want to buy this material for
You're already Subscribed!
Looks like you've already subscribed to StudySoup, you won't need to purchase another subscription to get this material. To access this material simply click 'View Full Document'