In,A.K. Joshi, B. L. Webber, and I.A. Sag (Eds. ) , Elements of Discourse Understanding. Cambridge: Defii~itereference aid mutud krlowledge 11 Cambridge University Press, 1981. about, it was that noun phrases like ~heseare ordil~arilyselected in ;I fittile amount of titile - in a few seconds or less. We were at an impasse. Tlie argument for an infinite amoutlt of processing time seemed im- Definite reference and mutual peccable, but so did the evidence against it. What we had was a pro- cessing paradox, which for reasons that will become clear later wc knowledge called the i~~ulunlhiwwle(1ge pnmdox. like all paradoxes, of' course, this one rests on several critical as- sumptions, and when these assun~ptionsare weakened in one way or HERBERT .H. CLARK and CATHERINE K. another, the paradox can be resolved in several ways. These differer~t MARSHALL resolutions, however, each have their own consequences, and depend- ing on which one we accept, we are led to rather different models for ~ack'thinks the production and understanding of speech. It is important to de- he does ~iot'know what lie thinks cide, then, which way the mutual knowledge paradox is most plausibly 111 thinks resolvetl. d e does not know But we ;Ire interested in this paradox only as a way of getting at the 13ut Jill tllinks Jack does know it. two cetl~ralquestions of this chapter: (a) What type of shared knowl- So Jill does not know edge is needed for language use? and (b) how is that shared kr~owl- she does not know edge in practice assessed and secured? The area of' language in which we will take up these questions is definite referel~ce,I~ut eve11 our interest in definite reference is secondary to our concern with the two questions of mutual knowledge. The way we will proceed, then, is to set out the mutual knowledge paradox, describe two ways of resolving it, and argue that one of them is the more usual resolution. We will then suggest that the answers to these two questions bear directly on Jack doesn't know he knows current theories of language structure and language use, in particular atid he doesn't know Jill does not know. on the characterization and processing of definite referewe. Jill doesn't know she doesn't know, and doesn't know that Jack doesn't know Jill does not kllow. l'hey have no problems. The mutual knowledge paradox k't~ots,by R. D. Laitig Imagine that there is a Marx brothers film festival on at the Roxy, with one filln showing each night for a week. Against this I)ackgroulid In speaking and listening people make essential use of a great deal of consider the following scenario: world knowledge that they "share" with each other. The question is, Vm.,iott 1. 011 Wed~icstl;~ytilort~ing Ann re;ltls 111ce;~rly etlitioti of the what kind of "shared" knowledge do they use, and how? Recently, in newslxlpcr which s;~ysthat Alotrkey lluritl~.r.<is 111;1yitigthat nigh^. L.;~tel. lookillg at how people plan definite reference, we came on one answer sl~esees I%obatld asks, "Have you ever seen the inovie sllowi~~g;it the Itoxy t011igh~?" to this question that made us distitlctly uneasy. It seemed to suggest that expressions like he cold usparagus, the mess I de,and lhat animal Our interest is in Ann's use of the definite referring expressio~~tip require speakers to check a list of facts or beliefs that is infinitely long. nrovie shuwi~rgat fk Roxy to~ligh~,term I, by which Ann inteiirls to refer Under the most plausible assumptions about how they would actually to Monkey Huriness, referent R. What does Ann have to assure herself' check that list, they should take an infinitely long time to decide on of in order to make this reference felicitously? That is, uncier wha~ each noun phrase. However, if there was anything we were cer~ain conditions does Arin have good reason to believe that Bob won't get 12 11. H. CLARK and c. R. MARSHALL Defi~~itereference and rnutual knowledge 13 the wrong referent or have to ask for clarification, as with "Which l)ny nl th~Rnces is showing that night at the Roxy. When the latc movie do you n~can?"The answer we will develop is that she must be etlitio~iarrives, Bob reads lhe movie section, notes tliat the film has certain that once she has made her reference he and she can establish been corrected to Moakey t?r~tine.cc,and circles it with his red pen. Later, Ann picks up the late edition, notes the co~.rection ant1 certain shared knowledge about the identity of that referent. Al- rccognites Bob's circle around it. She also realizes that Ilol) has no though not all aslxcts of this scenario are applicable to all other in- way of knowing t11at she has seen the late editior~.L.ater that dny Ann stances of definite reference, we will take up the more general case sces Bob and asks, "Have you ever seen tlie niovie showing at the 1:1ter. Roxy tonight?" An obvious first condition is that Ann herself know that the expres- 'The sce~iariosatisfies conditions (1) and (2). Ann knows that the movie sion tlw mouk showing at the Roxy tonight uniquely describes the movie , is Mm~.keyHuriness and that Bob knows that it is too. But she believes Mmlkey Hun'ne.rr - for example, there aren't two movies showing to- [.hat he believes tliat she still thinks it is A Day at thr! Races. He is very night insteild. We will describe this knowledge as "t is R," that is, "the likely to take her reference as one to A Dny n.t thp Rnces instead of niovie at the Roxy tonight is MmkPy fiurinc.~~."So, Ann must be certain Monkey ilzcsi7le.u. Her reference is infelicitous because she hasn't that after her refererice the following condition will be true:l satisfied this condition: ' (I) Ann knows that 1 is,R. (3) Ann knows that Bob knows that Ann knows tllat t is R. But is this enough? Obviously not, for what is missing is even the The third condition, however, is still not enough, as we can illustrate simplcst notion of shared knowledge. Specifically, (1) gives no assur- with yet another version of the original scenario: ance that on the basis of her reference Bob himself will realize that the movie at the Roxy tonigltt uniquely describes Monhey Burine.~,a realiza- Version 4. On Wednesday morning An11 and Bob rrad the early tion that is surely a sine qua non of a felicitous reference. The way edition of tlie newspaper and discuss the fact that it says tllat A Dny nt LIIR Rncc.c is playing that night at the Roxy. Later, Ann sees the late Ann's reference may fail can be illustrated by a variation on our origi- edition, notes that the movie lias been correctcd to Monkey Bucine.xc, nal scenario: and marks it with 11er blue pencil. Still later, as Ann watches without Vm.cior~2. On Wcdnesday morning Ann arid Rob read the early Bol) knowing it, he picks up the late edition and sees Ann's pencil edition of the newspaper and discuss the fact that it says thatA Dny nf mark. That afternoon. Ann sees Bob and asks, "Have you ever seen lh Rnrcc is showing that night at the Roxy. Later, after Bob has left, the rnovie showing at the Roxy tonight?" Ann gets the late edition, which prints a correction, which is that it is This version satisfies conditions (l), (2), and (5). Ann knows that tile Motlkcy Butinm that is actually showing that night. 1,ater. Ann sees movie is Mon.key Business; she knows that Dob knows it too she saw Ihl, a~ltlasks. "Have you ever seen the movie showirig ;I( the Roxy - totlight?" him look at the late edition; and she knows that he knows that she knows it too - she saw him notice her peticil mark on the correct Altl~oughthis version satisfies condition ( I), Ann has clearly made her movie in the late edition. Yet Ann is still not completely justified in definite reference without the proper assurances. She has no reason to thinking Bob will know she is referring to Makey Rucitle.rr. If she looks think tliat Bob will realize that the film she is referring to is Makey at the world from his point of view, she should reason like this: "She Hurine.~.He is most likely to tliink it is A Day at the Rnces. The reason knows that the niovie is Makey Btuinesr. But she thinks that I, Bob, why her reference isn't felicitous is clear. She has not assured herself think it is A Day nl the Rntes, and so by her reference, she must think I tli;~tafter she had rnade her reference Bob will know that ~htnovie will pick otlt A llay at the Races." But if her reference may get Bob to slmruin~at the Roxy tonight uniquely describes Monkey i3nurines.r. So Ann pick out A Day nt [he Races, it is infelicitous. So we must add another must satisfy this condition: condition for Ann to be sure of: ('I) Ann knows tliat Bob knows that t is R. (4) An11 knows tliat Boh knows that Arin knows tliat Bob knows that 1 is At first, conditions (1) and (2) may appear to be enot~gh,but it is I<.