Fox/Menéndez-Benito 12/7/06 Florian Schwarz: “Processing Presupposed Content” (and beyond) ♣ Agenda: 1. Discussion of “Processing Presupposed Content”. 2. Presentation of a follow-up experiment on also. 1. General questions • When are presuppositions computed? How quickly are they accessible to the parser? (late process vs. on-line computation) [note similarity with debate on implicatures]. From an update semantics perspective: when/where do contextupdates take place? • Do presuppositions interact with other factors relevant in parsing (i.e., syntactic preferences.) Cf. Crain and Steedman (1985) [see Schwarz paper for more recent references] 1) The horse raced past the barn fell (Beaver 1970) (a) [[the horse] [raced past the barn]] (b) [the horse raced past the barn] Analysis in (a) is preferred to the analysis in (b); strong garden path effect. • Garden-path theory: (i) analysis (a) is preferred to analysis (b) on the grounds of structural strategies (e.g., Fodor & Frazier’s Minimal Attachment: use as few new nodes as possible) Serial model: only preferred analysis is initially computed. • C & S’s explanation: (b) is presuppositionally more complex than (a): (b) “includes several horses rather than one, and a number of further facts about the basis on which they can be distinguished” [measure of complexity?]. Parallel model: several analyses are computed at the same time. 2) The Principle of Parsimony: If there is a reading that carries fewer unsatisfied but consistent presuppositions or entailments than any other, then, other criteria of plausibility being equal, that reading will be adopted as most plausible by the hearer, and the presuppositions in question will be incorporated in his or her model. • C & S 1985: Sentences like (3m) judged grammatical more often than sentences like (1) [off-line study; results actually compatible with both theories] 3) Horses raced past the barn fell. ♣ Note: parts of this handout are taken from Florian Schwarz’s slides for his SuB 10 presentation. You can find them at: http://people.umass.edu/florian/ 1 Cite as: Daniel Fox and Paula Menendez-Benito, course materials for 24.954 Pragmatics in LinguisticTheory, Fall 2006. MIT OpenCourseWare (http://ocw.mit.edu/), Massachusetts Institute of Technology.Downloaded on [DD Month YYYY].Fox/Menéndez-Benito 12/7/06 2. Why do experiments on presuppositions? [see Schwarz’s slides at http://people.umass.edu/florian/] (i) Provide new data for semantic theory • Questions of timing may be crucial: local vs. global issue. (ii) Inform theory of semantic processing • We know little about when/how semantic processing takes place. • Processing of presuppositions provides insights about an important part of semantic process: integration with the context. 2. Experiments • How can we test the effects of presuppositions in processing? (a) Present ambiguous sentences in which only one reading satisfies the presupposition. (b) Present sentence with a presupposition trigger in a context where the presupposition is not satisfied. • Potential problem: accommodation. • Overcoming the problem: choosing a trigger that (at the very least) stronglyresists accommodation: auch (‘also’) [Kripke 1990, Heim 1992, van der Sandt & Geurts 2001] 3) [JOHN] is having dinner in New York tonight too. 2 Cite as: Daniel Fox and Paula Menendez-Benito, course materials for 24.954 Pragmatics in LinguisticTheory, Fall 2006. MIT OpenCourseWare (http://ocw.mit.edu/), Massachusetts Institute of Technology.Downloaded on [DD Month YYYY].Fox/Menéndez-Benito 12/7/06 2.1. Questionnaire study Method and materials • Ambiguous bi-clause structures (RC and MC) • One reading syntactically preferred; presupposition of auch not satisfied. • Other reading syntactically dispreferred; presupposition of auch satisfied • Task: choose between paraphrases. 4) C1: Die Frau, die das Madchen sah, hatte auch der Mann gesehen The womanN/A whoN/A the girlN/A saw had also the manN seen C2: Die Frau, die das Madchen sah, hatte vorher der Mann gesehenThe womanN/A whoN/A the girlN/A saw had earlier the manN seen C3: Die Frau, sah das Madchen, auch das auch den Mann gesehen hatte The womanN/A saw the girl N/A who N/A also the manA seen had. C4: Die Frau, sah das Madchen, auch das vorher den Mann gesehen hatteThe womanN/A saw the girl N/A who N/A also the manA seen had. C5: Die Frau, sah das Madchen, auch die LehrerinN/A gesehen hatteThe womanN/A saw the girl N/A who N/A also the teacherN/A seen had. Condition 1 C1: Die Frau, die das Madchen sah, hatte auch der Mann gesehenThe womanN/A whoN/A the girlN/A saw had also the manN seen Main Clause is unambiguous: [the man] also had seen the woman. Presupposes: Someone else had seen the woman [if auch is unstressed – a plausible assumption- it must associate with der Mann] Relative Clause is ambiguous: (i) the woman saw the girl. [syntactically preferred (SO); presupposition not satisfied] (‘the man had also seen the woman who saw the girl’)(ii) the girl saw the woman. [syntactically dispreferred (OS); presupposition satisfied] (‘the man had also seen the woman who the girl saw’) 3 Cite as: Daniel Fox and Paula Menendez-Benito, course materials for 24.954 Pragmatics in LinguisticTheory, Fall 2006. MIT OpenCourseWare (http://ocw.mit.edu/), Massachusetts Institute of Technology.Downloaded on [DD Month YYYY].Fox/Menéndez-Benito 12/7/06 Task: choose between paraphrases: (a) The woman saw the girl and the man saw the woman. [corresponds to (i)] (b) The man and the girl saw the woman [corresponds to (ii)] Condition 2: Control. C2: Die Frau die das Madchen sah, hatte vorher der Mann gesehenThe womanN/A whoN/A the girlN/A saw had earlier the manN gesehen Condition 3 C3: Die Frau sah das Madchen, das auch den Mann gesehen hatteThe womanN/A saw the girl N/A who N/A also the manA seen had. Order of clauses switched around: MC first, ‘auch’ in RC. Main clause ambiguous: The woman saw the girl (preferred order) The girl saw the woman. (dispreferred order) RC unambiguous: the girl also had seen the man. (i) The woman saw the girl who also had seen the man. (preferred syntactic order; presupposition not satisfied) (ii) The girl who also had seen the man