grosz and sidner 1986 attention, intention and structure of discourse.pdf

Upload: zoltan-markos

Post on 03-Apr-2018

222 views

Category:

Documents


0 download

TRANSCRIPT

  • 7/28/2019 grosz and sidner 1986 attention, intention and structure of discourse.pdf

    1/31

    Attention, Intentions, and the Structure of Discourse

    (Article begins on next page)

    Citation Grosz, Barbara J. and Candace L. Sidner. 1986. Attention,intentions, and the structure of discourse. ComputationalLinguistics 12(3): 175-204.

    Published Version http://www.aclweb.org/anthology-new/J/J86/J86-3001.pdfAccessed October 11, 2011 6:01:32 PM EDTCitable Link http://nrs.harvard.edu/urn-3:HUL.InstRepos:2579648Terms of Use This article was downloaded from Harvard University's DASH

    repository, and is made available under the terms and conditionsapplicable to Other Posted Material, as set forth athttp://nrs.harvard.edu/urn-3:HUL.InstRepos:dash.current.terms-of-use#LAA

    http://www.aclweb.org/anthology-new/J/J86/J86-3001.pdfhttp://nrs.harvard.edu/urn-3:HUL.InstRepos:2579648http://nrs.harvard.edu/urn-3:HUL.InstRepos:dash.current.terms-of-use#LAAhttp://nrs.harvard.edu/urn-3:HUL.InstRepos:dash.current.terms-of-use#LAAhttp://nrs.harvard.edu/urn-3:HUL.InstRepos:dash.current.terms-of-use#LAAhttp://nrs.harvard.edu/urn-3:HUL.InstRepos:dash.current.terms-of-use#LAAhttp://nrs.harvard.edu/urn-3:HUL.InstRepos:dash.current.terms-of-use#LAAhttp://nrs.harvard.edu/urn-3:HUL.InstRepos:2579648http://www.aclweb.org/anthology-new/J/J86/J86-3001.pdf
  • 7/28/2019 grosz and sidner 1986 attention, intention and structure of discourse.pdf

    2/31

    A T T E N T I O N , I N T E N T I O N S , A N D T H E S T R U C T U R E O F D I S C O U R S E

    B a r b a r a J . G r o s zA r t i f i c i a l I n t e l l i g e n c e C e n t e r a n d

    C e n t e r f o r t h e S t u d y o f L a n g u a g e a n d I n f o r m a t i o nS R I I n t e r n a t i o na lM e n l o P a r k, C A 9 4 0 2 5

    C a n d a c e L . S i d n e rB B N L a b o r a t or i e s I n c.C a m b r i d g e , M A 0 2 2 3 8

    I n t h i s p a p e r w e e x p l o r e a n e w t h e o r y o f d i s c o u r s e s t r u c t u r e t h a t s t r e s s e s t h e r o l e o f p u r p o s e a n dp r o c e s s i n g i n d i s c o u r se . I n t h is t h e o r y , d i s c o u r s e s t r u c t u r e i s c o m p o s e d o f t h r e e s e p a r a t e b u t i n t e r r e -l a t e d c o m p o n e n t s : t h e s t r u c t u r e o f t h e s e q u e n c e o f u t t e r a n c e s ( c a l l e d t h e l i n g u i s t i c s t r u c t u r e ) , a s t r u c -t u r e o f p u r p o s e s ( c a l l e d t h e i n t e n t i o n a l s t r u c t u r e ) , a n d t h e s t a t e o f f o c u s o f a t t e n t i o n ( c a l l e d t h eat ten t ion a l s ta te ) . T h e l i n g u i s t i c s t r u c t u r e c o n s i s t s o f s e g m e n t s o f t h e d i s c o u r s e i n t o w h i c h t h e u t t e r -a n c e s n a t u r a l ly a g g r e g a t e . T h e i n t e n ti o n a l s t r u c t u r e c a p t u r e s t h e d i s c o u r s e - r e l e v a n t p u r p o s e s ,e x p r e s s e d i n e a c h o f t h e l in g u is t ic s e g m e n t s a s w e ll a s r e l a t io n s h i p s a m o n g t h e m . T h e a t t e n t i o n a l s t a t ei s a n a b s t r a c t i o n o f t h e f o c u s o f a t t e n t i o n o f t h e p a r t i c i p a n t s a s t h e d i s c o u r s e u n f o ld s . T h e a t t e n t i o n a ls t a t e , b e i n g d y n a m i c , r e c o r d s t h e o b j e c t s , p r o p e r t i e s , a n d r e l a t i o n s t h a t a r e s a l i e n t a t e a c h p o i n t o f t h ed i s c o u r s e . T h e d i s t i n c ti o n a m o n g t h e s e c o m p o n e n t s i s e s s e n t i a l t o p r o v i d e a n a d e q u a t e e x p l a n a t i o n o fs u c h d i s c o u r s e p h e n o m e n a a s c u e p h r a s e s , r e f e r r i n g e x p r e s s i o n s , a n d i n t e r r u p t i o n s .

    T h e t h e o r y o f attent ion , in tent ion , a n d a g g r e g a t i o n o f u t t e r a n c e s i s i l l u s t r a t e d i n t h e p a p e r w i t h an u m b e r o f e x a m p l e d i s c o u r se s . V a r i o u s p r o p e r t i e s o f d is c o u r s e a r e d e s c r i b e d , a n d e x p l a n a t i o n s f o r t h eb e h a v i o r o f c u e p h r a s e s , r e f e r r i n g e x p r e s s i o n s , a n d i n t e r r u p t i o n s a r e e x p l o r e d .

    T h i s t h e o r y p r o v i d e s a f r a m e w o r k f o r d e s c r i bi n g t h e p r o c e s s i n g o f u t t e r a n c e s i n a d i s c o u rs e .D i s c o u r s e p r o c e s s i n g r e q u i r e s r e c o g n i z i n g h o w t h e u t t e r a n c e s o f t h e d i s c o u r s e a g g r e g a t e i n t o s e g m e n t s ,r e c o g n i z in g t h e i n t e n t i o n s e x p r e s s e d i n t h e d i s c o u r s e a n d t h e r e l a t i o n s h ip s a m o n g i n te n t i o n s , a n d t r a c k -i n g t h e d i s c o u rs e t h r o u g h t h e o p e r a t i o n o f t h e m e c h a n i s m s a s s o c i a t e d w i t h a t t e n t i o n a l s t a t e . T h i sp r o c e s s i n g d e s c r i p t i o n s p e c i f i e s i n t h e s e r e c o g n i t i o n t a s k s t h e r o l e o f i n f o r m a t i o n f r o m t h e d i s c o u r s ea n d f r o m t h e p a r t i c i p a n t s ' k n o w l e d g e o f t h e d o m a i n .

    1 INTRODUCTIONp a p e r p r e s e n t s t h e b a si c e l e m e n t s o f a c o m p u t a -

    t h a t s i m p f if i e s a n dw o r k . B y s p e c i fy i n g t h e b a s i c

    u t t e r a n c e i n t h e d i s c o u rs e m u s t b e r e l a te d .

    A n a c c o u n t o f d i s c o u r s e s t r u c t u r e i s c l o s e l y r e l a t e d t ot w o q u e s t i on s : W h a t i n d i v i d u a t e s a d i s c o u r s e ? W h a tm a k e s i t c o h e r e n t ? T h a t i s , f a c e d w i t h a s e q u e n c e o fu t t e ra n c e s , h o w d o e s o n e k n o w w h e t h e r t h e y c o n s t it u t e as i n gl e d i sc o u r s e , s e v e r a l ( p e r h a p s i n t e r l e a v e d ) d i s c o u r s e s ,o r n o n e ? A s w e d e v e l o p i t, t h e t h e o r y o f d i s c o u rs e s t r u c -t u r e w i l l b e s e e n t o b e i n t i m a t e l y c o n n e c t e d w i t h t w on o n l i ng u i s ti c n o ti o n s : i n t e n t io n a n d a t t e n t io n . A t t e n t i o ni s a n e s s e n t i a l f a c t o r i n e x p l i c a t i n g t h e p r o c e s s i n g o fu t t e r a n c e s i n d i s c o u r se . I n t e n t i o n s p l a y a p r i m a r y r o l e i ne x p l a i n i n g d i s c o u r s e s t r u c t u r e , d e f i n i n g d i s c o u r s e c o h e r -e n c e , a n d p r o v i d i n g a c o h e r e n t c o n c e p t u a l i z a t i o n o f t h et e r m " d i s c o u r s e " i t s e l f .

    h t 1 9 8 6 b y t h e A s s o c i a t i o n f o r C o m p u t a t i o n a l L i n g u i s t i c s. P e r m i s s i o n t o c o p y w i t h o u t f e e a l l o r p a r t o f t h i s m a t e r i a l is g r a n t e d p r o v i d e d t h a tCL r e f e r e n c e a n d t h i s c o p y r i g h t n o t i c e a r e in c l u d e d o n t h e f i rs t p a ge . T o c o p y

    e s a f e e a n d / o r s p e c if i c p e r m i s s io n .0 3 6 2 - 6 1 3 X / 8 6 / 0 3 0 1 7 5 - 2 0 4 5 0 3 . 0 0

    n a l L i n g ui s ti c s , V o l u m e 1 2 , N u m b e r 3 , ~ l u l y - S e pt e m b e r 1 9 8 6 1 7 5

  • 7/28/2019 grosz and sidner 1986 attention, intention and structure of discourse.pdf

    3/31

    B a r b a r a J . G r o s z a n d C a n d a c e L . S i d n e r A t t e n t i o n , I n t e n t i o n s , a n d t h e S t r u c t u r e o f Dis

    T h e t h e o r y i s a f u r t h e r d e v e l o p m e n t a n d i n t e g r a t i o n o ft w o l i n e s o f r e s e a r c h : w o r k o n f o c u s i n g i n d i s c o u r s e( G r o s z 1 9 7 8a , 1 9 7 8 b , 1 9 8 1 ) a n d m o r e r e c e n t w o r k o ni n t e n t i o n r e c o g n i t i o n i n d i s c o u r s e ( S i d n e r a n d I s r a e l1 9 8 1 ; S i d n e r 1 9 8 3; 1 9 8 5 ; A l l e n 1 9 8 3 , L i t m a n 1 9 8 5 ;P o l l a c k 1 9 8 6 ) . O u r g o a l h a s b e e n t o g e n e r a l i z e t h e s ec o n s t r u c t s p r o p e r l y t o a w i d e r a n g e o f d i s c o u r s e t y p e s .G r o s z ( 1 9 7 8 a ) d e m o n s t r a t e d t h a t t he n o t i o n s o f fo c u s in ga n d t a s k s t r u c t u r e a r e n e c e s s a r y f o r u n d e r s t a n d i n g a n dp r o d u c i n g t a s k - o r i e n t e d d i a l og u e . O n e o f t h e m a i ng e n e r a l i z a t i o n s o f p r e v i o u s w o r k w i l l b e t o s h o w t h a td i s c o u r s e s a r e g e n e r a l l y i n s o m e s e n s e " t a s k - o r i e n t e d , "b u t t h e k i n d s o f " t a s k s " t h a t c a n b e e n g a g e d i n a r e q u i t ev a r i e d - s o m e a r e p h y s i c a l, s o m e m e n t a l , o t h e r s l i n g u is -t ic . C o n s e q u e n t l y , t h e t e r m " t a s k " i s m i s l e a d i n g ; w et h e r e f o r e w i l l u s e t h e m o r e g e n e r a l t e r m i n o l o g y o fin tent ions ( e .g . , w h e n s p e a k i n g o f d i s c o u r s e p u r p o s e s ) f o rm o s t o f w h a t w e s a y.

    O u r m a i n t h e s i s is th a t t h e s t r u c t u r e o f a n y d i s c o u r s e i sa c o m p o s i t e o f t h r e e d i s t in c t b u t i n t e r a c t i n g c o m p o n e n t s : t h e s t r u c t u r e o f t h e a c t u a l s e q u e n c e o f u t t e r a n c e s i n t h e

    d i s c o u r s e ; a s t r u c t u r e o f i n t e n t i o n s ; a n a t t e n t i o n a l s t a t e .T h e d i s t i n c t io n a m o n g t h e s e c o m p o n e n t s i s e s s e n t i a l t oa n e x p l a n a t i o n o f i n t e r r u p t i o n s ( s e e S e c t i o n 5 ) , a s w e l l a st o e x p l a n a t i o n s o f t h e u s e o f c e r t a i n t y p e s o f r e f e r r i n ge x p r e s s i o n s ( s e e S e c t i o n 4 . 2 ) a n d v a r i o u s o t h e re x p r e s s i o n s t h a t a f f e c t d i s c o u r s e s e g m e n t a t i o n a n d s t r u c -t u r e ( s e e S e c t i o n 6 ) . M o s t r e l a t e d w o r k o n d is c o u r s es t r u c t u r e ( in c l u d in g R e i c h m a n - A d a r 1 9 8 4 , L i n d e 1 9 7 9 ,L i n d e a n d G o g u e n 1 9 7 8 , C o h e n 1 9 8 3 ) f a i ls t o d i s ti n -g u i s h a m o n g s o m e ( o r , i n s o m e c a s e s , a l l ) o f t h e s ec o m p o n e n t s . A s a r e s u lt , s i g n if i c a n t g e n e r a l i z a t i o n s a r el o s t , a n d t h e c o m p u t a t i o n a l m e c h a n i s m s p r o p o s e d a r em o r e c o m p l e x t h a n n e c e s s a r y . B y c a r e f u l l y d i s t in g u i s h -i n g t h e s e c o m p o n e n t s , w e a r e a b l e t o a c c o u n t f o r s i g n i f -i c a n t o b s e r v a t i o n s i n t h i s r e l a t e d w o r k w h i l e s i m p l i f y i n gb o t h t h e e x p l a n a t i on s g i v e n a n d c o m p u t a t i o n a l m e c h -a n i s m s u s e d .

    I n a d d i t i o n t o e x p l i c a t i n g t h e s e l i n g u i s t i c p h e n o m e n a ,t h e t h e o r y p r o v i d e s a n o v e r a l l f r a m e w o r k w i t h i n w h i c h t oa n s w e r q u e s t i o n s a b o u t t h e r e l e v a n c e o f v a r i o u ss e g m e n t s o f d i s c o u r s e t o o n e a n o t h e r a n d t o t h e o v e r a l lp u r p o s e s o f t h e d i s c o u r s e p a r t i c i p a n t s . V a r i o u s p r o p e r -t i e s o f t h e i n te n t i o n a l c o m p o n e n t h a v e i m p l i c a t i o n s fo rr e s e a r c h i n n a t u r a l - l a n g u a g e p r o c e s s i n g i n g e n e r a l . I np a r t i c u l a r , t h e i n t e n t i o n s t h a t u n d e r l i e d i s c o u r s e a r e s od i v e r s e t h a t a p p r o a c h e s t o d i s c o u r s e c o h e r e n c e b a s e d o ns e l e c t in g d i s c o u r s e r e l a t i o n s h i p s f r o m a f i x e d s e t o f a l t e r -n a t i v e r h e t o r i c a l p a t t e r n s ( e .g . , H o b b s 1 9 7 9 , M a n n a n dT h o m p s o n 1 9 8 3 , R e i c h m a n 1 9 8 1 ) a r e u n li k e l y t o s u f f ic e .T h e i n t e n t i o n a l s t r u c t u r e i n t r o d u c e d i n th i s p a p e rd e p e n d s i n s t e a d o n a s m a l l n u m b e r o f s t r u c t u r a l r e l a t i o n st h a t c a n h o l d b e t w e e n i n t e n t i o n s . T h i s s tu d y a l s o r e v e a l ss e v e r a l p r o b l e m s t h a t m u s t b e c o n f r o n t e d i n e x p a n d i n gs p e e c h - a c t - r e l a t e d t h e o r i e s ( e . g . , A l l e n a n d P e r r a u l t1 9 8 0 , C o h e n a n d L e v e s q u e 1 9 8 0 , A l l e n 1 9 8 3 ) f r o m

    c o v e r a g e o f i n d i v i d u a l u t t e r a n c e s t o c o v e r a g e o f e x t es e q u e n c e s o f u t t e r a n c e s i n d i s c o u r s e .

    A l t h o u g h a d e f i n it i o n o f d i s c o u r s e m u s t a w a i t f ud e v e l o p m e n t o f t h e t h e o r y p r e s e n t e d i n th is p a p e r , p r o p e r t ie s o f th e p h e n o m e n a w e w a n t t o e x p la i n m us p e c i f i e d n o w . I n p a r t i c u la r , w e t a k e a d is c o u rs e t o p i e c e o f l a n g u a g e b e h a v i o r t h a t t y p i c a l l y i n v o l v e s mp l e u t t e r a n c e s a n d m u l t i pl e p a r t i c i p a n t s . A d is c om a y b e p r o d u c e d b y o n e o r m o r e o f t h e s e p a r t i c ip a ns p e a k e r s o r w r i te r s ; t h e a u d i e n c e m a y c o m p r i s e o nm o r e o f t h e p a r t i c i p a n t s a s h e a r e r s o r r e a d e r s . B e ci n m u l t i - p a r t y c o n v e r s a t i o n s m o r e t h a n o n e p a r t i cm a y s p e a k ( o r w r i t e ) d i f f e re n t u t t e r a n c e s w i t hs e g m e n t , t h e t e r m s s p e a k e r a n d h e a r e r d o n o t d i f ft i a t e t h e u n i q u e r o l e s t h a t t h e p a r t i c i p a n t s m a i n t a i ns e g m e n t o f a c o n v e r s a t i o n . W e w il l t h e r e f o r e u st e r m s i n i t i a t i n g conversat ional part ic ipant ( I C P ) a n dconversat ional part ic ipant(s) (OCP) t o d i s t i n g u i s h t h e a t o r o f a d i s c o u r s e s e g m e n t f r o m i t s o t h e r p a r t i c i pT h e I C P s p e a k s ( o r w r i te s ) t h e f i r s t u t t e r a n c e s e g m e n t , b u t a n O C P m a y b e t h e s p e a k e r o f s o m e s uq u e n t u t t e r a n c e s . B y s p e a k i n g o f I C P s a n d O C P s , wh i g h li g h t t h e p u r p o s i v e a s p e c t o f d i s c o u r s e . W e w it h e t e r m s speaker a n d hearer o n l y w h e n t h e p a r t is p e a k i n g / h e a r i n g a c t i v i t y i s i m p o r t a n t f o r t h e p o i n t m a d e .

    I n m o s t o f t h i s p a p e r , w e w i l l b e c o n c e r n e d d e v e l o p i n g a n a b s t r a c t m o d e l o f d i s c o u r s e s t r u c t u rp a r t i c u la r , t h e d e f i n i t i o n s o f t h e c o m p o n e n t s w i l l a b sa w a y f r o m t h e d e t a i ls o f t h e d i s c o u r s e p a r t i c i pW h e t h e r o n e c o n s t r u c t s a c o m p u t e r s y s t e m t h a tp a r t i c i p a t e i n a d i s c o u r s e ( i . e . , o n e t h a t is a l a n gu s e r ) o r d e f i n e s a p s y c h o l o g i c a l t h e o r y o f l a n g u a g et h e t a s k w i ll r e q u ir e t h e a p p r o p r i a t e p r o j e c t i o n o fa b s t r a c t m o d e l o n t o p r o p e r t i e s o f a l a n g u a g e u s e r ,s p e c i f i c a t i o n o f a d d i t io n a l d e t a i l s ( e . g ., s p e c i f y i n g mr y f o r li n g ui s ti c s t r u c t u r e , m e a n s f o r e n c o d i n g a t t e n ts t a t e , a n d a p p r o p r i a t e r e p r e s e n t a t i o n s o f i n t e n ts t r u c t u r e ) . W e d o , h o w e v e r , a d d r e s s o u r s e l v e s d i r e cc e r t a i n p r o c e s s i n g i s s u e s t h a t a r e e s s e n t i a l t o t h e c ot a t i o n a l v a l i d i ty o f t h e [ a b s t r a c t ] m o d e l a n d t o i t s u tt i o n f o r a l a n g u a g e - p r o c e s s i n g s y s t e m o r p s y c h o l ot h e o r y .

    F i n a l ly , i t i s i m p o r t a n t t o n o t e t h a t a l t h o u g h d i s cm e a n i n g i s a s i g n i f i c a n t , u n s o l v e d p r o b l e m , w e w i la d d r e s s i t i n t h is p a p e r . A n a d e q u a t e t h e o r y o f d i s cm e a n i n g n e e d s t o r e s t a t l e a s t p a r t i a l l y o n a n a d e qt h e o r y o f d i s c o u r s e s t r u c t u re . O u r c o n c e r n i s w i t h p ri n g t h e l a t t e r .

    T h e n e x t s e c t i o n e x a m i n e s t h e b a s i c t h e o rd i s c o u r s e s tr u c t u r e a n d p r e s e n t s a n o v e r v i e w o f e at h e c o m p o n e n t s o f d i s c o u r s e s t ru c t u r e . S e c t i oa n a l y z e s t w o s a m p l e d i s c o u rs e s - a w r i t t e n t e x t af r a g m e n t o f t a s k - o r i e n t e d d i al o g ue - f r o m t h e p e rt i v e o f t h e t h e o r y b e i n g d e v e l o p e d ; t h e s e t w o e x a ma r e a l s o u s e d t o i l l u s t r a t e v a r i o u s p o i n t s i n t h e r e m ao f t h e p a p e r . S e c t i o n 4 i n v e s t i g a t e s v a r i o u s p r o c e

    17 6 C o m p u t a t i o n a l L i n g u i s t i c s , V o l u m e 1 2 , N u m b e r 3 , J u l y - S e p t e m b e r

  • 7/28/2019 grosz and sidner 1986 attention, intention and structure of discourse.pdf

    4/31

    J . G r o s z a n d C a n d a c e L . S i d n e r A t t e n t i o n , I n t e n t i o n s , a n d t h e S t r u c t u r e o f D i s c o u r s e

    e s t h a t th e t h e o r y r a i s e s. T h e f o l l o w i n g t w o s e c t i o n se o f th e d i s c o u r s e s t r u c t u re c o m p o n e n t s

    S e c t i o n 7 d e s c r i b e s th e g e n e r a l i z a t i o ne l t o d i s c o u r s e - l e v e l i n t e n t io n s , e s t a b -

    2 T H E B A S IC T H E O R Y

    a l i n g u is t i c s tr u c t u r e , a n i n t e n t i o n a l s t r u c -a n d a n a t t e n t i o n a l s ta t e . T h e s e t h r e e c o n s t i t u e n t s

    a d i s c o u rs e . U t t e r a n c e s - t h e a c t u a l s a y i n ga r e t h e l i n g u is t i c s t r u c t u r e ' s b a s i c e l e m e n t s . I n t e n t i o n s

    s t r u c tu r e . A t t e n t i o n a l s t a t e c o n t a i n s in f o r m a t i o no n s t h a t a r e m o s t s a l i e n t a t a n y g i v e n : p o i n t. I t i s

    a n t s ; i t s e r v e s t o s u m m a r i z e i n f o r m a t i o n f r o m

    T o g e t h e r t h e t h r e e c o n s t i t u e n t s o f d i sc o u r s e s t r u c t u r et h e i n f o r m a t i o n n e e d e d b y t h e C P s t o d e t e r m i n e

    i n e s s e n c e , e n a b l i n g t h e m t o f i g u r e o u t w h y i td a n d w h a t i t m e a n s . T h e c o n t e x t p r o v i d e d b y

    a n a d d i t io n a l p u r p o s e : n a m e l y , i t f u r n i s h e s

    2.1 LINGUI STIC STRUCTU REf i rs t c o m p o n e n t o f d is c o u r s e s t r u c t u r e i s t h e s t r u c -

    1 J u s t a s t h e w o r d s i n a s i n g l e s e n t e n c e f o r my a g g r e g a t e d i n t o d is c o u rs e s e g m e n ts . T h e u t t e r -a r r o l e s w i t h r e s p e c t t o t h a t s e g m e n t . I n a d d i t i o n ,

    r e s p e c t t o t h e o v e r a l l d i s c o u rs e . A l t h o u g hm a y b e i n t h e s a m e d i s c o u r s e

    T h e f a c t o r i n g o f d i s c o u r s e s i n to s e g m e n t s h a s b e e no b s e r v e d a c r o s s a w i d e ra n g e o f d i s c o u r s e ty p e s . G r o s z( 1 9 7 8 a ) s h o w e d t h is f o r t a s k - o r i e n t e d d i a l o g ue s . L i n d e( 1 9 7 9 ) f o u n d i t va l i d f o r d e s c r i p t i o n s o f a p a r t m e n t s ;L i n d e a n d G o g u e n ( 1 9 7 8 ) d e s c r i b e s u c h s t r u c t u r i n g i nt h e W a t e r g a t e t ra n s c ri p ts . R e i c h m a n - A d a r ( 1 9 8 4 )o b s e r v e d i t i n i n f o r m a l d e b a t e s , e x p l a n a t i o n s , a n d t h e r a -p e u t i c d i s c o u r se . C o h e n ( 1 9 8 3 ) f o u n d s im i l a r s t r u c t u r e si n e s s a y s i n r h e t o r i c a l t e x ts . P o l a n y i a n d S c h a ( 1 9 8 6 )d i s c u s s t h i s f e a t u r e o f n a r r a t i v e s .

    A l t h o u g h d i f f e r e n t r e s e a r c h e r s w i t h d i f f e r e n t t h e o r i e sh a v e e x a m i n e d a v a r i e t y o f d is c o u r s e t y p e s a n d f o u n dd i s c o u r s e - l e v e l s e g m e n t a t i o n , t h e r e h a s b e e n v e r y l i tt lei n v e s t i g a ti o n o f t h e e x t e n t o f a g r e e m e n t a b o u t w h e r e t h es e g m e n t b o u n d a r i e s li e. T h e r e h a v e b e e n n o p s y c h o l o g -i c a l s t u d i e s o f t h e c o n s i s t e n c y o f r e c o g n i t i o n o f s e c t i o nb o u n d a ri e s . H o w e v e r , M a n n ( M a n n e t a l. 1 9 75 ) a s k e ds e v e r a l p e o p l e t o s e g m e n t a s e t o f d ia l o g u e s . H e h a sr e p o r t e d [ p e r s o n a l c o m m u n i c a t i o n ] t h a t h i s s u b j e c t ss e g m e n t e d t h e d i s c o u r se s a p p r o x i m a t e l y t h e s a m e ; t h e i rd i s a g r e e m e n t s w e r e a b o u t u t t e r a n c e s a t t h e b o u n d a r i e s o fs e g m e n t s . 2 S e v e r a l s t u d i e s o f s p o n t a n e o u s l y p r o d u c e dd i s c o u r s e s p r o v i d e a d d i t i o n a l e v i d e n c e o f t h e e x i s t e n c e o fs e g m e n t b o u n d a r i e s , a s w e l l a s s u g g e s t in g s o m e o f t h el i ng u i st i c c u e s a v a i l a b le f o r d e t e c t i n g b o u n d a r i e s . C h a f e( 1 9 7 9 , 1 9 8 0 ) f o u n d d i f f e r e n c e s i n p a u s e l e n g t h s a ts e g m e n t b o u n d a ri e s . B u t t e r w o r t h ( 1 9 7 5 ) f o u n d s p e e c hr a t e d i f f e r e n c e s t h a t c o r r e l a t e d w i t h s e g m e n t s ; s p e e c hr a t e i s s l o w e r a t s t a r t o f a s e g m e n t t h a n t o w a r d t h e e n d .

    T h e l i n g u i s t i c s t r u c t u r e c o n s i s t s o f t h e d i s c o u r s es e g m e n t s a n d a n e m b e d d i n g r e l a t i o n s h i p t h a t c a n h o l db e t w e e n t h e m . A s w e d i s c u s s i n S e c t i o n s 2 . 2 a n d 5 , t h ee m b e d d i n g r e l a t i o n s h i p s a r e a s u r f a c e r e f l e c t i o n o fr e l a t i on s h i p s a m o n g e l e m e n t s o f t h e i n t e n t i o n a l s t r u c tu r e .I t is i m p o r t a n t t o r e c o g n i z e t h a t t h e l in g u is t ic s t r u c t u r e i sn o t st r i ct l y d e c o m p o s i t i o n a l . A n i n d i v i du a l s e g m e n t m a yi n c lu d e a c o m b i n a t i o n o f s u b s e g m e n t s a n d u t t e r a n c e so n l y i n th a t s e g m e n t ( a n d n o t m e m b e r s o f a n y o f i tse m b e d d e d s u b s e g m e n t s ) . B o t h o f t h e e x a m p l e s i n S e c ti o n3 e x h i b i t s u c h n o n s t r i c t d e c o m p o s i t i o n a l i t y . B e c a u s e t h el i n g u i s t i c s t r u c t u r e i s n o t s t r i c t l y d e c o m p o s i t i o n a l , v a r i o u sp r o p e r t i e s o f t h e d i s c o u r s e ( m o s t n o t a b l y t h e i n t e n t i o n a ls t r u c t u r e ) a r e f u n c t i o n s o f p r o p e r t i e s o f i n d i v i du a l u t t e r -a n c e s a n d p r o p e r t i e s o f s e g m e n t s .

    T h e r e i s a t w o - w a y i n t e r a c t i o n b e t w e e n t h e d i s c o u r s es e g m e n t s t r u c t u r e a n d t h e u t t e r a n c e s c o n s t i t u t i n g t h ed i s c o u r se : l in g u is t ic e x p r e s s i o n s c a n b e u s e d t o c o n v e yi n f o r m a t i o n a b o u t t h e d i s c o u r s e s t r u c t u r e ; c o n v e r s e l y ,t h e d i s c o u r s e s t r u c t u r e c o n s t r a i n s t h e i n t e r p r e t a t i o n o fe x p r e s s i o n s ( a n d h e n c e a f f e c t s w h a t a s p e a k e r s a y s a n dh o w a h e a r e r w i l l i n t e r p r e t w h a t i s s a i d ). N o t s u r p r i s i n g -l y , l i n g u i s t i c e x p r e s s i o n s a r e a m o n g t h e p r i m a r y i n d i c a -t o r s o f d i s c o u r s e s e g m e n t b o u n d a r i e s . T h e e x p l ic i t u s e o fc e r t a i n w o r d s a n d p h r a s e s ( e . g . , in the f i rs t p lace) a n dm o r e s u b t l e c u e s , s u c h a s i n t o n a t i o n o r c h a n g e s i n t e n s ea n d a s p e c t , a r e i n c l u d e d i n t h e r e p e r t o i r e o f l i n g u i s t i cd e v i c e s t h a t f u n c t i o n , w h o l l y o r i n p a r t , t o i n d i c a t e t h e s e

    1986 177

  • 7/28/2019 grosz and sidner 1986 attention, intention and structure of discourse.pdf

    5/31

    Barbara J . Grosz and Candace L . S idner At tent ion , In tent ions , and the S tructure of Disc

    boundaries (Grosz 1978a, Reichman-Adar 1984, Cohen1983, Polanyi and Scha 1983, Hirschberg and Pierre-humbert 1986). Reichm an (1981) discusses some wordsthat function in this way and coined the term c l u e w o r d s .We will use the term c u e p h r a s e s to generalize on herobservation as well as many others because each one ofthese devices cue the hearer to some change in thediscourse structure.

    As discussed in Section 6, these linguistic boundarymarkers can be divided according to whether they explic-itly indicate changes in the intentional structure or in theattentional state of the discourse. The differential use ofthese linguistic markers provides one piece of evidencefor considering these two components to be distinct.Because these linguistic devices function explicitly asindicators of discourse structure, it becomes clear thatthey are best seen as providing information at thediscourse level, and not at the sentence level; hence,certain kinds of questions (e.g., about their contributionto the truth conditions of an individual sentence) do notmake sense. For example, in the utterance I nc iden ta l l y ,Jane swims every day , the inc iden ta l l y indicates an inter-ruption of the main flow of discourse rather than affect-ing in any way the meaning of Jane swims every day .Jane's swimming every day could hardly be fortuitous.

    Just as linguistic devices affect structure, so thediscourse segmentation affects the interpretation oflinguistic expressions in a discourse. Refe rrin gexpressions provide the pr imary example of this effect. 3The segmentation of discourse constrains the use ofreferring expressions by delineating certain points atwhich there is a significant change in what entities(objects, properties, or relations) are being discussed.For example, there are different constraints on the use ofpronouns and reduced definite-noun phrases within asegment than across segment boundaries. Whilediscourse segmentation is obviously not the only factorgoverning the use of referring expressions, it is an impor-tant one.

    2.2 INTENTIONAL STRUCTUR EA rather straightforward property of discourses, namely,that they (or, more accurately, those who participate inthem) have an overall purpose, turns out to play a funda-mental role in the theory of discourse structure. Inparticular, some of the purposes that underlie discourses,and their component segments, provide the means ofindividuating discourses and of distinguishing discoursesthat are cohere nt from those that are not. Thesepurposes also make it possible to determine when asequence of utterances comprises more than onediscourse.

    Although typically the participants in a discourse mayhave more than one aim in participating in the discourse(e.g., a story may entertain its listeners as well asdescribe an event; an argument may establish a person'sbrilliance as well as convince someone that a claim or

    allegation is true), we distinguish one of these purpas foundational to the discourse. We will refer to it asd i s c o u r s e p u r p o s e (DP). From an intuitive perspectivediscourse purpose is the intention that underlies engain the particular discourse. This intention provides the reason a discourse (a linguistic act), rather than sother action, is being performed and the reason particular content of this discourse is being convrather than some other information. For each ofdiscourse segments, we can also single out one inten- t h e d i sc o u r s e s e g m e n t p u r po s e (DSP). Fr om an intustandpoint, the DSP specifies how this segment conutes to achieving the overall discourse purpose. assumption that there are single such intentions withe end prove too strong. However, this assumpallows us to describe the basic theory more clearly.must leave to future research (and a subsequent pathe exploration and discussion of the complications result from relaxing this assumption.

    Typically, an ICP will have a num ber of differ ent kof intentions that lead to initiating a discourse. One might include intentions to speak in a certain languagto utter certain words. Anot her might include intentto amuse or to impress. The kinds of intentions thatserve as discourse purposes or discourse segmpurposes are distinguished from other intentions byfact that they are intended to be recognized (cf. Aand Perrault 1980, Sidner 1985), whereas ointentions are private; that is, the recognition of theor DSP is essent ial to its achieving its intended efDiscourse purposes and discourse segment purpshare this property with certain utterance-level intentthat Grice (1969) uses in defining utterance mea(see Section 7).

    It is important to distinguish intentions that intended to be recognized from other kinds of intentthat are associated with discourse. Intentio ns thatintended to be recognized achieve their intended eonly if the intention is recognized. For examplcompliment achieves its intended effect only if the intion to compliment is recognized; in contrast, a screamboo typically achieves its intended effect (scaringhearer) without the hearer having to recognize the sper's intention.

    Some intention that is private and not intended torecognized may be the primary motivation for an ICbegin a discourse. For example, the ICP may intenimpress someone or may plan to teach someone. neither case is the ICP's intention necessarily intendebe recognized. Quite the opposite may be true incase of impressing, as the ICP may not want the OCbe aware of his intention. When teaching, the ICP not care whether the OCP knows the ICP is teaching or her. Thus, the intention that motivates the ICengage in a discourse may be private. By contrast,discourse segment purpose is always intended torecognized.

    17 8 C o m p u t a t i o n a l L i n g u i s t i c s , V o l u m e 12 , N u m b e r 3 , J u l y - S e p t e m b e r

  • 7/28/2019 grosz and sidner 1986 attention, intention and structure of discourse.pdf

    6/31

    B a r b a r a J . G r o s z a n d C a n d a c e L . S i d n e r A t t e n t i o n , I n t e n t i o n s , a n d t h e S t r u c t u r e o f D i s c o u r s e

    DPs and DSPs are basically the same sorts ofintentions. If an intention is a DP, then its satisfaction is amain purpose of the discourse, whereas if it is a DSP,"then its satisfaction contributes to the satisfaction of theDP. The following are some of the types of intentionsthat could serve as DP/DSPs, followed by one example ofeach type.1. Intend that some agent intend to perfor m some phys-

    ical task. Examp le: I n t e n d t h a t R u t h i n t e n d to f i x t h ef l a t t i r e .

    2. Intend that some agent believe some fact. Example:I n t e n d t h a t R u t h b e l ie v e t h e c a m p f i r e h a s s t a rt e d .

    3. Intend that some agent believe that one fact supportsanother. Example: l n t e n d t h a t R u t h b e l i e v e t h e s m e l lo f s m o k e p r o v i d e s e v i d e n c e t h a t t h e c a m p f i r e i s s t a rt e d .

    4. Intend that some agent intend to identify an object(existing physical object, imaginary object, plan,event, event sequence). Example: I n t e n d t h a t R u t hi n t e n d t o i d e n t i f y m y b i c yc l e .

    5. Intend that some agent know some property of anobject. Example: I n t e n d t h a t R u t h k n o w t h a t m y b i c y -c l e h a s a f l a t t ir e .

    We have identified two structural relations that playan important role in discourse structure: dominance ands a t i s f a c t i o n - p r e c e d e n c e . An action that satisfies oneintention, say DSP1, may be intended to provide part ofthe satisfact ion of anothe r, say DSP2. Whe n this is thecase, we will say that DSP1 c o n t r i b u t e s t o DSP2;conversely, we will say that DSP2 d o m i n a t e s DSP1 ( o rDSP2 D O M DSP1). The dominance relation invokes apartial ordering on DSPs that we will refer to as the domi-nance h ie r a r c h y . For some discourses, including task-or-iented ones, the order in which the DSPs are satisfiedmay be significant, as well as being intended to be recog-nized. We will say that DSP1 satisfaction- precedes DSP2(or, DSP1 S P DSP2) whenever DSP1 must be satisfiedbef ore DSP2. 4

    Any of the intentions on the preceding list could beeither a DP or a DSP. Furt her more , a given instance o fany one of them could contribute to another, or to adifferent, instance of the same type. For example, theintention that someone intend to identify some objectmight dominate several intentions that she or he knowsome property of that object; likewise, the intention toget someone to believe some fact might dominate anumber of contributing intentions that that personbelieve other facts.

    As the above list makes clear, the range of intentionsthat can serve as discourse, or discourse segment,purposes is open- ended (cf. Wittgenstein 1953: para-graph 23), much like the range of intentions that underliemore general purposeful action. There is no finite list ofdiscourse purposes, as there is, say, of syntactic catego-ries. It remains an unresolved research question whetherthere is a finite description of the open-ended set of suchintentions. However, even if there were finitedescriptions, there would still be no finite list of

    intentions from which to choose. Thus, a theor y ofdiscourse structure cannot depend on choosing theDP/DSPs from a fixed list (cf. Reichman-Adar 1984,Schank et al. 1982, Mann and Thompson 1983), nor onthe particulars of individual intentions. Althoug h theparticulars of individual intentions, like a wide range ofcommon sense knowledge, are crucial to understandingany discourse, such particulars cannot serve as the basisfo r d e t e r m i n i n g discourse structure.

    What is essential for discourse structure is that suchintentions bear certain kinds of structural relationships toone another. Since the CPs can never know the wholeset of intentions that,might serve as DP/DSPs, what theymust recognize is the relevant structural relationshipsamong intentions. Although there is an infinite numbe rof intentions, there are only a small number of relationsrelevant to discourse structure that can hold betweenthem.

    In this paper we distinguish between the d e t e r m i n a t i o nof the DSP and the rec o g n i t i o n of it. We use the termd e t e r m i n a t i o n to refer to a semantic-like notion, namely,the complete specification of what is int ende d by whom;we use the term recognition to refer to a processingnotion, namely, the processing that leads a discourseparticipant to identify what the intention is. These areobviously related concepts; the same information thatdetermines a DSP may be used by an OCP to recognize it.However, some questions are relevant to only one ofthem. For example, the question of when the informa-tion becomes available is not relevant to determinationbut is crucial to recognition. An analogous distinctionhas been drawn with respect to sentence structure; theparse tree (determination) is differentiated from the pars-ing process (recognition) that produces the tree.

    2 .3 A T T E N T I O N A L S T A T EThe third component of discourse structure, the atten-tional state, is an abstraction of the participants' focus ofattention as their discourse unfolds. The attentional stateis a property of the discourse itself, not of the discourseparticipants. It is inherently dynamic, recording theobjects, properties, and relations that are salient at eachpoint in the discourse. The attentional state is modeledby a set of f o c u s s p a c e s ; changes in attentional state aremodeled by a set of transition rules that specify theconditions for adding and deleting spaces. We call thecollection of focus spaces available at any one time thef o c u s i n g s t r u c t u r e and the process of manipulating spacesf o c u s i n g .

    The focusing process associates a focus space witheach discourse segment; this space contains those entitiesthat are salient - either because they have beenmentioned explicitly in the segment or because theybecame salient in the process of producing or compre-hending the utterances in the segmfnt (as in the originalwork on focusing: Gros z 1978a). The focus space alsoincludes the DSP; the inclusion of the purpose reflects the

    C o m p u t a t i o n a l Ling u is ti cs , V o lu me 1 2 , N u mb er 3 , J u ly - S ep tem b er 1 98 6 1 7 9

  • 7/28/2019 grosz and sidner 1986 attention, intention and structure of discourse.pdf

    7/31

    B a r b a r a J . G r o s z a n d C a n d a e e L . S i d n e r A t t e n t i o n , I n t e n t i o n s , a n d t h e S t r u c t u r e o f D i s c o u r s

    f a c t t h a t t h e C P s a r e f o c u s e d n o t o n l y o n w h a t t h e y a r et a l k in g a b o u t , b u t a l s o o n w h y t h e y a r e t a l k in g a b o u t i t.

    T o u n d e r s t a n d t h e a t t e n t i on a l s t a t e c o m p o n e n t o fd i s c o u r s e s t r u c t u r e , i t i s i m p o r t a n t n o t t o c o n f u s e i t w i t ht w o o t h e r c o n c e p t s . F i r s t, t h e a t te n t i o n a l s t a t e c o m p o -n e n t i s n o t e q u i v a l e n t t o c o g n i t i v e s t a t e , b u t i s o n l y o n eo f i t s c o m p o n e n t s . C o g n i t i v e s t a t e i s a r i c h e r s t r u c t u r e ,o n e t h a t i n c l u d e s a t l e a s t t h e k n o w l e d g e , b e l i e f s , d e s i r e s ,a n d i n t e n t i o n s o f a n a g e n t , a s w e l l a s th e c o g n i t i v e c o r r e - 'l a t e s o f t h e a t t e n t i o n a l s t a t e a s m o d e l e d i n t h i s p a p e r .S e c o n d , a l t h o u g h e a c h f o c u s s p a c e c o n t a i n s a D S P , t h ef o c u s s t r u c t u r e d o e s not i n c l u d e t h e i n t e n t i o n a l s t r u c t u r ea s a w h o l e .

    F i g u r e 1 i l l u s t r a t e s h o w t h e f o c u s i n g s t r u c t u r e , i n a d d i -t i o n t o m o d e l i n g a t t e n t i o n a l s t a t e , s e r v e s d u ri n g p r o c e s s -i n g t o c o o r d i n a t e t h e l i n g u i s t i c a n d i n t e n t i o n a l s t r u c t u r e s .T h e d i s c o u r s e s e g m e n t s ( t o t h e l e f t o f t h e f i g u r e ) a r e t i e dt o f o c u s s p a c e s ( d r a w n v e r t i c a ll y d o w n t h e m i d d l e o f t h ef i g u re ) . T h e f o c u s i n g s t r u c t u r e i s a s t a c k . I n f o r m a t i o n i nl o w e r s p a c e s i s us u a l ly a c c e s s i b l e f r o m h i g h e r o n e s ( b u tl e s s s o t h a n t h e i n f o r m a t i o n i n t h e h i g h e r s p a c e s ) ; w e u s ea l i n e w i t h i n t e r s e c t i n g h a s h m a r k s t o d e n o t e w h e n t h i s i sn o t t h e c a se . S u b s c r i p t e d t e r m s a r e u s e d t o i n d i c a t e t her e l e v a n t c o n t e n t s o f t h e f o c u s s p a c e s b e c a u s e t h e s p a c e sc o n t a i n r e p r e s e n t a t i o n s o f e n t i t i e s ( i .e . , o b j e c t s , p r o p e r -t i e s , a n d r e l a t i o n s ) a n d n o t l i n g u i s t i c e x p r e s s i o n s .

    P a r t o n e o f F i g u r e 1 s h o w s t h e s t a t e o f f o c u s i n g w h e nd i s c o u r s e s e g m e n t D S 2 i s b e i n g p r o c e s s e d . S e g m e n t DS 1g a v e r i s e t o F S 1 a n d h a d a s i t s d i s c o u r s e p u r p o s e D S P I .T h e p r o p e r t i e s , o b j e c t s , r e l a t i o n s , a n d p u r p o s e r e p r e s -e n t e d i n F S 1 a r e a c c e s s i b l e b u t l e s s s a l i e n t t h a n t h o s e i nF S 2 . D S 2 y i e l d s a f o c u s s p a c e t h a t i s s t a c k e d r e l a t i v e t oF S l b e c a u s e D S P 1 o f D S l d o m i n a t e s D S 2 ' s D SP , D S P2. Asa r e s u lt o f t h e r e l a t i o n s h ip b e t w e e n F S1 a n d F S 2 , r e d u c e dn o u n p h r a s e s w i l l b e i n t e r p r e t e d d i f f e r e n t l y i n D S 2 t h a ni n D S 1. F o r e x a m p l e , i f s o m e r e d b a l l s e x i s t i n t h e w o r l do n e o f w h i c h i s r e p r e s e n t e d i n D S 2 a n d a n o t h e r i n F S 1 ,t h e n the red ball u s e d i n D S 2 w i l l b e u n d e r s t o o d t o m e a nt h e p a r t i c u l a r r e d b a l l t h a t i s r e p r e s e n t e d i n D S 2 . I f ,h o w e v e r , t h e r e i s a l s o a g r e e n t r u c k ( i n t h e w o r l d ) a n d i ti s r e p r e s e n t e d o n l y i n F S 1, the green truck u t t e r e d i n D S 2w i ll b e u n d e r s t o o d a s r e f e r r i n g t o t h a t g r e e n t r u c k .

    P a r t t w o o f F i g u r e 1 s h o w s t h e s t a t e o f fo c u s i n g w h e ns e g m e n t D S3 is b e i n g p ro c e s s e d . F S 2 h a s b e e n p o p p e df r o m t h e s t a c k a n d F S 3 h a s b e e n p u s h e d o n t o i t b e c a u s et h e D S P o f D S 3 , D S P3 , is d o m i n a t e d s o l e l y b y D S P 1, n o tb y D S P 2. I n t h i s e x a m p l e , t h e i n t e n t i o n a l s t r u c t u r ei n c lu d e s o n l y d o m i n a n c e r e l a t io n s h i p s , a l th o u g h , i t m a y ,i n g e n e r a l , a l s o i n c l u d e s a t i s f a c t i o n - p r e c e d e n c e r e l a t i o n -sh ip s .

    T h e s t a c k i n g o f f o c u s s p a c e s r e f l e c t s t h e r e l a t i v e s a l i -e n c e o f t h e e n t i t i e s i n e a c h s p a c e d u r i n g t h e c o r r e s p o n d -i n g s e g m e n t ' s p o r t i o n o f t h e d i s c o u r se . T h e s t a c kr e l a t i o n s h i p s a r i s e f r o m t h e w a y s i n w h i c h t h e v a r i o u sD S P s r e l a t e ; i n f o r m a t i o n a b o u t s u c h r e l a t i o n s h i p s i s

    r e p r e s e n t e d i n t h e d o m i n a n c e h i e r a r c h y ( d e p i c t e d o n t h er i g h t i n t h e f i g ur e ) . T h e s p a c e s in F i g u r e 1 a r e s n a p s h o t si l lu s t r at i n g t h e r e s u l t s o f a s e q u e n c e o f o p e r a t i o n s , s u c ha s p u s h e s o n t o a n d p o p s f r o m a s t a c k . A p u s h o c c u r sw h e n t h e D S P f o r a n e w s e g m e n t c o n t r i b u t e s t o t h e D SPf o r t h e im m e d i a t e l y p r e c e d i n g se g m e n t . W h e n t h e D S Pc o n t r i b u t e s t o s o m e i n t e n t i o n h i g h e r i n t h e d o m i n a n c eh i e r a r c h y , s e v e r a l f o c u s s p a c e s a r e p o p p e d f r o m t h e s t a c kb e f o r e t h e n e w o n e is i n s e r te d .

    T w o e s s e n t i a l p r o p e r t i e s o f t h e f o c u s i n g s t r u c t u r e a r en o w c l e a r . F i r s t, t h e f o c u s i n g s t r u c t u r e is p a r a s i t i c u p o nt h e i n t e n t i o n a l s t r u c t u r e , i n t h e s e n s e t h a t t h e r e l a t i o n -s h ip s a m o n g D S P s d e t e r m i n e p u s h e s a n d po p s . N o t eh o w e v e r , t h a t t h e r e l e v a n t o p e r a t i o n m a y s o m e t i m e s b ei n d i c a t e d i n t h e l a n g u a g e it s e lf . F o r e x a m p l e , t h e c u ew o r d first o f t e n i n d i c a t e s t h e s t a r t o f a s e g m e n t w h o s eD S P c o n t r i b u t e s t o t h e D S P o f t h e p r e c e d i n g s e g m e n tS e c o n d , t h e f o c u s i n g s t r u c t u r e , l i k e t h e i n t e n t i o n a l a n dl i ng u is t ic s t r u c t u r e s , e v o l v e s a s t h e d i s c o u r s e p r o c e e d s .N o n e o f t h e m e x i s t s a pr i o ri . E v e n i n t h o s e r a r e c a s e s i nw h i c h a n I C P h a s a c o m p l e t e p l a n f o r t h e d i s c o u r s e p r i ot o u t t e r i n g a s i n g l e w o r d , t h e i n t e n t i o n a l s t r u c t u r e ic o n s t r u c t e d b y th e C P s a s t h e d i s c o u r s e p r o g r e s s e s . T h i sd i s c o u r s e - t i m e c o n s t r u c t i o n o f t h e i n t e n t i o n a l s t r u c t u r em a y b e m o r e o b v i o u s l y tr u e f o r s p e a k e r s a n d h e a r e r s os p o k e n d i s c o u r s e t h a n f o r r e a d e r s a n d w r i t e r s o f t e x tsb u t , e v e n f o r t h e w r i t e r , t h e i n t e n t i o n a l s t r u c t u r e i s d e v e lo p e d a s t h e t e x t i s b e i n g w r i t te n .

    F i g u r e 1 i l l us t r a te s s o m e f u n d a m e n t a l d i s t i nc t i o n sb e t w e e n t h e i n t e n t i o n a l a n d a t t e n t i o n a l c o m p o n e n t s od i s c o u r s e s t r u c t u r e . F i r s t , t h e d o m i n a n c e h i e r a r c h yp r o v i d e s , a m o n g o t h e r t h i n g s, a c o m p l e t e r e c o r d o f t h ed i s c o u r s e - l e v e l i n t e n t i o n s a n d t h e i r d o m i n a n c e ( a s w e la s , w h e n r e l e v a n t , s a t i s f a c t i o n - p r e c e d e n c e ) r e l a t i o n s h i p sw h e r e a s t h e f o c u s i n g s t r u c t u r e a t a n y o n e t im e c a n e s s e n -t i a l l y c o n t a i n o n l y i n f o r m a t i o n t h a t i s r e l e v a n t t op u r p o s e s i n a p o r t i o n o f t h e d o m i n a n c e h i e r a r c h y .S e c o n d , a t t h e c o n c l u s i o n o f a d i s c o u r s e , i f i t c o m p l e t e sn o r m a l l y , t h e f o c u s s t a c k w i l l b e e m p t y , w h i l e t h e i n t e n -t i o n a l s t r u c t u r e w i l l h a v e b e e n f u l ly c o n s t r u c t e d . T h i r dw h e n t h e d i s c o u r s e i s b e i n g p r o c e s s e d , o n l y t h e a t t e n -t i o n a l s ta t e c a n c o n s t r a i n t h e i n t e r p r e t a t i o n o f r e f e r r i n ge x p r e s s i o n s d i r e c tl y .

    W e c a n n o w a l so c l a ri f y s o m e m i s i n t e rp r e t a t io n s of o c u s - s p a c e d i a g r a m s a n d t a s k s t r u c t u r e i n o u r e a r l i e rw o r k ( G r o s z 1 9 7 8 a , 1 9 8 1 , 1 9 7 4 ) . T h e f o c u s - s p a c e h i e r -a r c h i e s in t h a t w o r k a r e b e s t s e e n a s r e p r e s e n t i n g a t t e n -t i o n a l s t a t e . T h e t a s k s t r u c t u r e w a s u s e d i n t w o w a y s :1. t o r e p r e s e n t c o m m o n k n o w l e d g e a b o u t t h e t a s k ;2 . a s a s p e c i a l c a s e o f t h e i n t e n t io n a l s t r u c t u r e w e p o s i

    i n t h i s p a p e r .A l t h o u g h t h e s a m e r e p r e s e n t a t i o n a l s c h e m e w a s u s e d f o re n c o d i n g t h e f o c u s - s p a c e h i e r a r c h i e s a n d t h e t a s k s t r u c -t u r e ( p a r t i t i o n e d n e t w o r k s : H e n d r i x 1 9 7 9 ) , t h e t w os t r u c t u r e s w e r e d i s ti n c t.

    1 8 0 C o mp u ta t io n a l L ing u is ti cs , V o lu me 1 2 , N u m b er 3 , J u ly - S ep tem b er 1 98 6

  • 7/28/2019 grosz and sidner 1986 attention, intention and structure of discourse.pdf

    8/31

    B a r b a r a J . G r o s z a n d C a n d a c e L . S i d n e r A t t e n t i o n , I n t e n t i o n s , a n d t h e S t r u c t u r e o f D i s c o u r s e

    D I S C O U R SE S E G M E N T S F O C U S S P A C E S T A C K D O M I N A N C E H I E R A R C H Y

    f ~DS1

    D S 2 ~,M,,tM m i. lh ,i. /~ I l l ./ ~. IW

    D S 3 a . J

    T E X T G I V E SR I S E T O F S 2

    T E X T G I V E ~ , ~% R I S E T O F S 1

    P R O P E R T I E SO B J E C T SR E L A T I O N SDSP 2 F S 2P R O P E R T I E SO B J E C T SR E L A T I O N SDSP 1 F S l

    D SP 1 D O M I N A T E S D S P 2

    ( a )

    D I S C O U R SE S E G M E N T S F O C U S SP A C E S T A C K D O M I N A N C E H I E R A R C H Y

    ~ ' l ~ T E XT G IV E SDS1 ~ . ~ R I E T F"~"-"~'* ~'~.I S 0 S~D S 2 ~ , ~ /

    - - ' - \ T : X ; TO F >

    P R O P E R T I E SO B J E C T SR E L A T I O N SDSP 3 F S 3P R O P E R T I E SO B J E C T SR E L A T I O N SDSP 1 F S l

    D S P 1 D O M I N A T E S D S P 3D S P 1 D O M I N A T E S D S P 2

    ( b )

    F ig u r e 1. D i s c o u r s e S e g m e n t s , F o c u s S p a c e s a n d D o m i n a n c e H i e r a rc h y .

    C o m p u t a t i o n a l L i n g u i s t ic s , V o l u m e 1 2 , N u m b e r 3 , J u l y - S e p t e m b e r 1 9 8 6 1 8 1

  • 7/28/2019 grosz and sidner 1986 attention, intention and structure of discourse.pdf

    9/31

    B a r b a r a J. Grosz a n d C a n d a c e L . S i d n e r A t t e n t i o n , I n t e n t i o n s , a n d t h e S t r u c t u r e o f D i s c o u r s e

    S e v e r a l r e s e a r c h e r s ( e .g . , L i n d e a n d G o g u e n 1 9 7 8 ,R e i c h m a n - A d a r 1 9 8 4 ) m i s i n t e r p r e t e d t h e o r i g i n a lr e s e a r c h i n a n u n f o r t u n a t e a n d u n i n t e n d e d w a y : t h e yt o o k t h e f o c u s - s p a c e h i e r a r c h y t o i n c l u d e ( o r b e i d e n t i c a lt o ) t h e t a s k st r u c tu r e . T h e c o n f l a t i o n o f th e s e t w o s t r u c -t u r e s f o r c e s a s in g le s t r u c t u r e t o c o n t a i n i n f o r m a t i o na b o u t a t t e n t i o n a l s t a t e , i n t e n t i o n a l r e l a t io n s h i p s , a n dg e n e r a l t a sk , k n o w l e d g e . I t p r e v e n t s a th e o r y f r o ma c c o u n t i n g a d e q u a t e l y f o r c e r t a i n a s p e c t s o f d i s c o u r s e ,i n c l u d i n g i n t e r r u p t i o n s ( s e e S e c t i o n 5 ) .

    A s e c o n d i n s t a n c e o f c o n f u s i o n w a s t o i n f e r ( i n c o r -r e c t l y ) t h a t t h e t a s k s t r u c t u r e w a s n e c e s s a r i l y a p r e b u i l tt r e e . I f t h e t a s k s t r u c t u r e i s t a k e n t o b e a s p e c i a l c a s e o fi n t e n t i o n a l s t r u c t u re , i t b e c o m e s c l e a r t h a t t h e t r e e s t r u c -t u r e i s s i m p l y a m o r e c o n s t r a i n e d s t r u c t u r e t h a n o n em i g h t r e q u i re f o r o t h e r d i s c o u r se s ; t h e n a t u r e o f t h e t a s kr e l a t e d t o t h e t a s k - o r i e n t e d d i s c o u r s e i s s u c h t h a t t h ed o m i n a n c e h ie r ~ ir c hy o f t h e i n t e n t i o n a l s t r u c t u r e o f t h ed i a lo g u e h a s b o t h d o m i n a n c e a n d s a t i s f a c t i o n - p re c e d e n c er e l a t io n s h i p s , 5 w h i l e o t h e r d i s c o u r s e s m a y n o t e x h i b i ts i g n i f i c a n t p r e c e d e n c e c o n s t r a i n t s a m o n g t h e D S P s .F u r t h e r m o r e , t h e r e h a s n e v e r b e e n a n y r e a s o n t o a s s u m et h a t t h e t a s k s t r u c t u r e s i n t a s k - o r i e n t e d d i a l o g u e s a r ep r e b u i lt , a n y m o r e t h a n t h e i n t e n t i o n a l s t r u c t u r e o f a n yo t h e r k i n d o f d i s c o u r s e s. I t i s r a t h e r t h a t o n e o b j e c t i v e o fd i s c o u r s e t h e o r y ( n o t a t o p i c c o n s i d e r e d h e r e , h o w e v e r )i s t o e x p l a i n h o w t h e O C P b u i l d s u p a m o d e l o f t h e t a s ks t r u c t u r e b y u s i n g i n f o r m a t i o n s u p p l i e d i n t h e d i s c o u r s e .

    H o w e v e r , i t i s i m p o r t a n t t o n o t e t h a t c o n f l a t i n g t h ea f o r e m e n t i o n e d t w o r o l e s o f i n f o r m a t i o n a b o u t t h e t a s ki t s e l f ( a s a p o r t i o n o f g e n e r a l c o m m o n s e n s e k n o w l e d g ea n d a s a s p e c i a l c a s e o f i n t e n t i o n a l s t r u c t u r e ) w a s r e g r e t -t a b l e , a s i t f a i l s t o m a k e a n i m p o r t a n t d i s t i n c t i o n .F u r t h e r m o r e , a s i s c l e a r w h e n i n t e n t i o n a l s t r u c t u r e s a r ec o n s i d e r e d m o r e g e n e r a ll y , s u c h a c o n f l a t i o n o f r o l e sd o e s n o t a l lo w f o r d if f e r e n ce s b e t w e e n w h a t o n e k n o w sa b o u t a t a s k a n d o n e ' s i n t e n t i o n s f o r ( o r w h a t o n e m a k e se x p l ic i t in d i s c o u r s e a b o u t ) p e r f o r m i n g a t a s k .

    I n s u m m a r y , t h e f o c u s i n g s t r u c t u r e i s t h e c e n t r a lr e p o s i t o r y f o r t h e c o n t e x t u a l i n f o r m a t i o n n e e d e d t o p r o c -e s s u t t e r a n c e s a t e a c h p o i n t i n t h e d i s c o u rs e . I t di s t in -g u i s h e s t h o s e o b j e c t s , p r o p e r t i e s , a n d r e l a t i o n s t h a t a r em o s t s a l i e n t a t t h a t p o i n t a n d , m o r e o v e r , h a s l i n k s t or e l e v a n t p a r t s o f b o t h t h e l in g u is t ic a n d i n t e n t i o n a l s t r u c -t u r e s. D u r i n g a d i s c ou r s e , a n in c r e a s i n g a m o u n t o f i n f o r -

    m a t i o n , o n l y s o m e o f w h i c h c o n t i n u e s t o b e n e e d e d f o rt h e i n t e r p r e t a t i o n o f s u b s e q u e n t u t t e r a n c e s , i s d i s c u s se d .H e n c e , i t b e c o m e s m o r e a n d m o r e n e c e s s a r y t o b e a b l et o i d e n t i f y r e l e v a n t d i s c o u r s e s e g m e n t s , t h e e n t i t i e s t h e ym a k e s a l i e nt , a n d th e i r D S P s. T h e r o l e o f a t t e n t i o n a ls t a t e i n d e l i n e a t i n g t h e i n f o r m a t i o n n e c e s s a r y f o r u n d e r -s t a n d i n g i s t h u s c e n t r a l t o d i s c o u r s e p r o c e s s i n g .

    3 T w o E X A M PL E ST o i l l u s t r a t e t h e b a s i c t h e o r y w e h a v e j u s t s k e t c h e d , w ew i l l g i v e a b r i e f a n a l y s i s o f t w o k i n d s o f d i s c o u r s e : a na r g u m e n t f r o m a r h e t o r i c t e x t a n d ~i t a s k - o r i e n t e dd i a l o g u e . F o r e a c h e x a m p l e w e d i s c u s s t h e s e g m e n t a t i o no f t h e d i s c o u r s e , t h e i n t e n t i o n s t h a t u n d e r l i e t h i s s e g m e n -t a t i o n , a n d t h e r e l a t i o n s h i p s a m o n g t h e v a r i o u s D S P s . I ne a c h c a s e , w e p o i n t o u t s o m e o f t h e l i n g u i s t i c d e v i c e su s e d t o i n d i c a t e s e g m e n t b o u n d a r i e s a s w e l l a s s o m e o ft h e e x p r e s s i o n s w h o s e i n t e r p r e t a t i o n s d e p e n d o n t h o s eb o u n d a r i e s . T h e a n a l y s i s is c o n c e r n e d w i t h s p e c i f y i n gc e r t a i n a s p e c t s o f t h e b e h a v i o r t o b e e x p l i c a t e d b y at h e o r y o f d i sc o u r s e ; t h e r e m a i n d e r o f t h e p a p e r p r o v i d e sa p a r t i a l a c c o u n t o f t h is b e h a v i o r .

    3.1 AN ARGUME NTO u r f i r s t e x a m p l e i s a n a r g u m e n t t a k e n f r o m a r h e t o r i ct d x t ( H o l m e s a n d G a l l a g h e r 1 9 1 76 ) . I t i s a n e x a m p l e u s e db y C o h e n ( 1 9 8 3 ) i n h e r w o r k o n t h e s t r u c t u r e o f a r g u -m e n t s . F i g u r e 2 s h o w s t h e d i a l o g u e a n d t h e e i g h td i s c o u r s e s e g m e n t s o f w h i c h i t i s c o m p o s e d . T h e d i v i s i o no f t h e a r g u m e n t in t o s e p a r a t e ( n u m b e r e d ) c l a u s e s isC o h e n ' s , b u t o u r a n a l y s i s o f t h e d i s c o u r s e s t r u c t u r e i sd i f f e r e n t , s in c e i n C o h e n ' s a n a l y s i s , e v e r y u t t e r a n c e i sd i r e c t l y s u b o r d i n a t e d t o a n o t h e r u t t e r a n c e , a n d t h e r e i so n l y o n e s t r u c t u r e t o e n c o d e l i n g u i s t i c s e g m e n t a t i o n a n dt h e p u r p o s e s o f u t t e r a n c e s . A l t h o u g h b o t h a n a l y s e ss e g m e n t u t t e r a n c e ( 4 ) s e p a r a t e l y f r o m u t t e r a n c e s ( 1 - 3 ) ,s o m e r e a d e r s p l a c e t h i s u t t e r a n c e i n D S1 w i t h u t t e r a n c e s( 1 ) t h r o u g h ( 3 ) ; t h i s is a n e x a m p l e o f t h e k i n d o f d i s a -g r e e m e n t a b o u t b o u n d a r y u t t e r a n c e s f o u n d i n M a n n ' sd a t a ( a s d i s c u s s e d i n S e c t i o n 2 . 1 ). T h e t w o p l a c e m e n t sl e a d t o s l i g h t l y d i f f e r e n t D S P s , b u t n o t t o r a d i c a l l y d i f f e r -e n t i n t e n ti o n a l s t r uc t u r e s . B e c a u s e t h e d i ff e r e n c e s d o n o ta f f e c t t h e m a j o r t h r u s t o f t h e a r g u m e n t , w e w i l l d i s c u s so n l y o n e s e g m e n t a t io n .

    18 2 C o m p u t a t i o n a l L i n g u i s t ic s , V o l u m e 1 2 , N u m b e r 3 , J u l y - S e p t e m b e r 1986

  • 7/28/2019 grosz and sidner 1986 attention, intention and structure of discourse.pdf

    10/31

    Barbara J. Grosz and Candace L. S i d n e r A t t e n t i o n , I n t e n t i o n s , a n d t h e S t r u c t u r e o f D i s c o u r s e

    m DS0

    DS2

    $3

    5

    D S 6

    I

    I . _ _ .

    DS7

    1. The " mov i es" a re so a t t ra c t i ve t o t he g rea t Am er i can pub l i c ,2. e spec i a l l y t o young peop le ,3. t ha t i t i s t ime to t ake ca re f u l t hough t abou t t he i r e f fec t on mindand mora l s .4 . Ough t any pa re n t t o pe rmi t h i s ch i l d ren t o a t t end a mov ing p i c tu re

    show o f t en o r w i thou t be ing qu i t e ce r t a in o f t he show he pe rmi t sthem to see?5. No one can deny , o f cou r se , t ha t g rea t educa t iona l and e th i ca l

    ga in s may be made t h rough the mov ies6. becau se o f t he i r a s ton i s h ing v iv idness .7. Bu t t he impor t an t fac t t o be de t e r mined i s t he t o t a l r e su l t o f

    con t in uous and i nd i s c r im ina t e a t t en dance on shows o f t h i s k ind.8. Can i t be o the r t han ha rm fu l?9. In t he f i r s t p l ace t h e cha r ac t e r o f t he p l ays is se ldom o f t he

    bes t .10. One has on ly t o read t he ev e r -p resen t " mov i e" b i l l b oa rd t o see how

    c h e a p , m e l o d r a m a t i c a n d v u l g a r m os t of t h e p h o t o p l a y s a r e .11 . Even the bes t p l ays , moreove r , a re bound to be exc i t i ng and

    o v e r - e m o t i o n a l .12 . Wi thout spoke n words , fac i a l exp r ess io n and ges tu re mus t ca r r y t he

    mean ing :13. bu t on ly s t ro ng emot ion , o r bu f fo one ry can be rep r esen t ed t h rou ghf a c i al e x p r e s s i o n a n d g e s t u r e .

    14. The more reas onab l e and qu i e t a sp ec t s o f l ife a re ne cessa r i l yn e g l e c t e d .15. How can ou r y oung peop le d r ink i n t h rou gh the i r eyes a con t inu ous

    s p e c t a c l e o f i n t e n s e a n d s t r a i n e d a c t i v i t y a n d f e el i n g w i t h o u th a r m f u l e f f e c t s ?16. P a ren t s and t each e rs w il l do wel l t o gua rd t he you ng aga ins t

    o v e r i n d u l g e n c e i n t h e t a s t e f o r t h e " m o v i e " .

    Figure 2. The Movies Essay.

    C o m p u t a t i o n a l Linguistics, Volume 12, Number 3, July-September 1986 183

  • 7/28/2019 grosz and sidner 1986 attention, intention and structure of discourse.pdf

    11/31

    B a r b a r a J. Grosz a n d C a n d a c e L . S i d n e r A t t e n t i o n , I n t e n t io n s , a n d t h e S t r u c t u r e o f D i s c o u r s e

    F i g u r e 3 l is t s t h e p r i m a r y c o m p o n e n t o f t h e D S P f o re a c h o f t h e s e s e g m e n t s a n d F i g u r e 4 s h o w s t h e d o m i -n a n c e r e l a t i o n s h ip s t h a t h ol d a m o n g t h e s e i nt e n t i o n s. I nS e c t i o n 7 w e d i s c u s s a d d i t i o n a l c o m p o n e n t s o f t h ed i s c o u r s e s e g m e n t p u r p o s e ; b e c a u s e t h e s e a d d i t i o n a lc o m p o n e n t s a r e m o r e i m p o r t a n t f o r c o m p l e t e n e s s o f t het h e o r y t h a n f o r d e t e r m i n i n g t h e e s se n t i a l d o m i n a n c e a n d

    s a t i s f a c t i o n - p r e c e d e n c e r e l a t i o n s h i p s b e t w e e n D S P s , w eo m i t su c h d e ta i l s h e r e . R a t h e r t h a n c o m m i t o u r s e l v e s t oa f o r m a l l a n g u a g e i n w h i c h t o e x p r e s s t h e i n t e n t i o n s o ft h e d i s c o u r s e , w e w il l u s e a s h o r t h a n d n o t a t i o n a n dE n g l i s h s e n t e n c e s t h a t a r e i n t e n d e d t o b e a g lo s s f o r af o r m a l s t a t e m e n t o f t h e a c t u a l i n t e n ti o n s .

    I O : ( I n t e n d

    I 1 : ( I n t e n d

    I 2 : ( I n t e n d

    1 3 : ( I n t e n d

    1 4 : ( I n t e n d

    1 5 : ( I n t e n d

    I 6 : ( I n t e n d

    1 7 : ( I n t e n d

    I C P ( B el ie v e O C P P O ) )w h e r e P O = t h e p r o p o s i t i o n t h a t p a r e n t s a n d t e a c h e r s s h o u l d g u a r d t h e y o u n gf r o m o v e r i n d u l g e n c e i n th e m o v i e s .I C P ( B e l i e ve O C P P 1 ) )w h e r e P 1 = t h e p r o p o s i t i o n t h a t i t i s t i m e t o c o n s i d e r t h e e f f e c t o f m o v i e s o nm i n d a n d m o r a l s .I C P ( B e li e ve O C P P 2 ) )w h e r e P 2 = t h e p r o p o s i t i o n t h a t y o u n g p e o p l e c a n n o t d r i n k i n t h r o u g h t h e i r e y e sa c o n t i n u o u s s p e c t a c l e o f i n t e n s e a n d s t r a i n e d a c t i v i t y w i t h o u t h a r m f u l e f f e c t s .I C P ( B e li e ve O C P P 3 ) )w h e r e P 3 - - t h e p r o p o s i t i o n t h a t i t i s u n d e n i a b l e t h a t g r e a t e d u c a t i o n a l a n de t h i c a l g a i n s m a y b e m a d e t h r o u g h t h e m o v i e s .I C P ( B e l ie v e O C P P 4 ) )w h e r e P 4 = t h e p r o p o s i t i o n t h a t a l t h o u g h t h e r e a r e g a i n s , t h e t o t a l r e s u l t o fc o n t i n u o u s a n d i n d i s c r im i n a t e a t t e n d a n c e a t m o v i e s i s h a r m f u l .I C P ( B e l ie v e O C P P 5 ) )w h e r e P 5 = t h e p r o p o s i t i o n t h a t t h e c o n t e n t o f m o v i e s ( i. e ., t h e c h a r a c t e r o f t h ep l a y s ) i s n o t t h e b e s t .I C P ( B e l i e ve O C P P 6 ) )w h e r e P 6 = t h e p r o p o s i t i o n t h a t t h e s t o ri e s ( i. e ., th e p l a y s ) i n m o v i e s a r e e x c i t -i n g a n d o v e r - e m o t i o n a l .I C P ( B e l ie v e O C P P 7 ) )w h e r e P 7 = t h e p r o p o s it i o n th a t m o v i e s p o r t r a y s t r o n g e m o t i o n a n d b u f f o o n e r yw h i l e n e g l e c t in g t h e q u i e t a n d r e a s o n a b l e a s p e c t s o f l if e .

    F i g u re 3. P r i m a r y i n te n t i o n s o f t h e D S P s f o r M o v i e s e s s a y .

    I 0 D O M I1I 0 D O M 1212 D O M 1312 D O M 1414 D O M 1514 D O M 1616 D O M 17

    F i g u re 4 . D o m i n a n c e r e l a t i o n s h i p s f o r t h e D S P s o f t h e M o v i e s e s s a y .

    184 Com putat ional Lingu i s t i c s , Volume 12, Number 3, July-September 1986

  • 7/28/2019 grosz and sidner 1986 attention, intention and structure of discourse.pdf

    12/31

    ~ fB a r b a r a J. Grosz a n d C a n d a e e L . S i d n e r A t t e n t i o n , I n t e n t i o n s , a n d t h e S t r u c t u r e o f D i s c o u r s e

    All the primary intentions for this essay are intentionsthat the reader (OCP) come to believe some proposition.Some of these propositions, such as P5 and P6, can beread off the surface utterances directly. Other prop-ositions and the intemions of which they are part, such asP2 and 12, are moCe indirect. Like the Gric ean utter -ance'-level intentions (the analogy with these will beexplored in Section 7), DSPs may or may not be directlyexpressed in the discourse. In particular, they may beexpressed in any of the following ways:1 . exp l ic i t l y as in I i n t e n d f o r y o u t o b e l ie v e t h a t i t 's t i m e

    t o c o n s i d e r t h e e f f e c t s o f m o v i e s o n m i n d a n d m o r a l s .[which would prod uce I1 ]

    2 . d i r e c t ly , i n o n e u t t e r a n c e , as in (3) [which doesprod uce I 1 ]

    3 . d i r e c t l y , t h r o u g h m u l t i p l e u t t e r a n c e s , as in using (7)and the utterance I t c a n o n l y b e h a r m f u l to produce14,

    4 . b y d e r i v a t i o n , i n o n e o r m o r e u t t e r a n c e s w i t h a n a s s o c i -a t e d c o n t e x t, as in (15) to produce 12.

    Not only may information about the DSP be conveyedby a number of features of the utterances in a discourse,but it also may come in any utterance in a segment. Forexample, although I0 is the DP, it is stated directly only inthe last utterance of the essay. This leads to a number ofquestions about the ways in which OCPs can recognizediscourse purposes, and about those junctures at whichthey need to do so. We turn to these matters directly inSubsection 4.1.

    This discourse also provides several examples of thedifferent kinds of interactions that can hold between thelinguistic expressions in a discourse and the discoursestructure. It includes examples of the devices that may beused to mark overtly the boundaries between discoursesegments - examples of the use of aspect, mood, andparticular cue phras es - as well as of the use of referrin gexpressions that are affected by discourse segment boun-daries.

    The use of cue phrases to indicate discourse bounda-ries is illustrated in utterances (9) and (11); in (9) thephrase i n t h e f i r s t p l a c e marks the beginning of DS5 whilein (11) m o r e o v e r ends DS5 and marks the start of DS6.These phrases also carry information about the inten-tional structure, namely, that DSP5 and DSP6 are domi-nated b y DSP4. In some cases, cue phrases have multiplefunctions; they convey propositional content as well asmarking discourse segment boundaries.. The b u t in utter-ance (7) is an example of such a multiple function use.

    The boundaries between DS1 and DS2, DS4 and DS5,and DS4 and DS2 reflect changes of aspect and mood.The switch from declarative, present tense to interroga-tive modal aspect does not in itself seem to signal theboundary (for recognition purposes) in this discourseunambiguously, but it does indicate a possible line ofdemarcation which, in fact, is valid.

    The effect of segmentation on referring expressions isshown by the use of the generic noun phrase a m o v i n g

    p i c t u r e s h o w in (4). Althoug h a reference to the movieswas made with a pronoun ( t h e i r ) in (3), a full nounphrase is used in (4). This use reflects, and perhaps inpart marks, the boundary between the segments DS1 andDS2.

    Finally, this discourse has an example of the trade-offbetween explicitly marking a discourse boundary, as wellas the relationship between the associated DSPs, andreasoning about the intentions themselves. There is noovert linguistic marker of the beginning of DS7; its sepa-ration must be inferred from DSP7 and its relationship toDSP6.

    3.2 A TASK-ORIENTED DIALOGUEThe second example is a fragment of a task-orienteddialogue taken from Grosz (1981; it is from the samecorpus that was used by Grosz 1974). Figure 5 containsthe dialogue fragment and indicates the boundaries for itsmain segments. 7 Figure 6 gives the prima ry co mpo nent ofthe DSPs for this fragment and shows the dominancerelationships between them.

    In contrast with the movies essay, the primary compo-nents of the DSPs in this dialogue are mostly intentions ofthe segment's ICP that the OCP intend to perform someaction. Also, unlike the essay, the dialogue has twoagents initiating the different discourse segments. In thisparticular segment, the expert is the ICP of DS1 and DS5,while the appren tice is the ICP of DS2-4. To furnish acomplete account of the intentional structure of thisdiscourse, one must be able to say how the satisfaction ofone agent's intentions can contribute to satisfying theintentions of another agent. Such an account is beyo ndthe scope of this paper, but in Section 7 we discffss someof the complexities involved in providing one (as well asits role in discourse theory).

    For the purposes of discussing this example, though,we need to postulate two properties of the relationshipsamong the participants' intentions. These propertiesseem to be rooted in features of cooperative behaviorand depend on the two participants' sharing some partic-ular knowledge of the task. First, it is a share d beliefthat, unless he states otherwise, the OCP will adopt theintention to perform an action that the ICP intended himto. Second, in adopting the intention to carry out thataction, the OCP also intends to perform whateversubactions are necessary. Thus, once the apprentice,intends to remove the flywheel, he also commits himselfto the collateral intentions of loosening the setscrews andpulling the wheel off. Note, however, that not all thesubactions need to be introduced explicitly into thediscourse. The apprentice may do several actions that arenever mentioned, and the expert may assume that theseare being undertaken on the basis of other informationthat the apprentice obtains. The partiality of the inten-tional structure stems to some extent from these charac-teristics of intentions and actions.

    C o m p u t a t i o n a l L i n g u i s t i c s , V o l u m e 12, Number 3, July-Sept ember 1986 185

  • 7/28/2019 grosz and sidner 1986 attention, intention and structure of discourse.pdf

    13/31

    Barbara J . Grosz and Candace L . S idner At tent ion , In tentions , and the S tructure of Discourse

    DS1 (1)(2 )(3 )(4 )DS2 (5)( S )(7 )( B )( 9 )

    ( 1 0 )(11)

    DS 3 (12)( 1 3 )( 1 4 )(15)(16)(17)

    DS4 (18)(19)(z0)( z l )(zz)( z3)(24)(z5)

    E: F i r s t you have t o remov e the f l ywhee l .A: How do I rem ove the f lywhe el?E: F i r s t , l oose n t he two a l l en head se t sc rews

    hold ing i t to the shaft , then pul l i t o ff .A: OK.

    I can on ly f i nd one sc rew . Where ' s t he o the r one?E: On the hub of the f lywheel .A: Tha t ' s t he one I found . Where ' s t he o the r one?E: Abou t n ine ty deg re es a roun d the hub f rom the f i r s t one .A: I don ' t unde rs t and . I can only f ind one. Oh

    wai t , yes I th i nk I was on the wron g wheel .E: Show me what you are doing .A: I was on th e wro ng wheel and l can f ind

    them bo th now.The t oo l I have i s awkward . I s t he re ano the rtoo l t ha t I cou ld u se i n s t e ad?

    E: Show me the tool you are using .A: OK.E: Are you su re you a re u s ing t he r i gh t s i zek e y ?A: I ' l l t ry some oth ers .

    I fou nd an angle I can get a t i t.The two sc rews a re l oose , bu t I 'm hav ing t roub l ege t t i ng t he whee l o f f .

    E: Use the whee lpu l le r" Do you know how to usei t?

    A: No .E: Do you know what i t looks l ike?A: Yes.E: Show it to me please.A: OK.E: Good , Loosen t he sc rew in t he cen t e r and

    p lace t he j aws a round the hub o f t hewheel , t hen t i gh t en t he sc rew on to t hecen t e r o f t he sha f t . The whee l shou ldslide off.

    Figure 5. A segment of a task-oriented dialogue.

    As in the movies essay, some of the DSPs for thisdialogue are expressed directly in utterances. Fo rinstance, utterances (1), (5), and (12) directly expressthe primary components of DSP1, DSP2 and DSP3,respectively. The primary compone nt of DSP4 is aderived intention. The surface intention of but I'mhaving trouble getting the wheel off is tha t the apprenticeintends the expert to believe that the apprentice is havingtrouble takin g off the flywheel. 14 is derived fr om theutterance and its surface intention, as well as fromfeatures of discourse, conventions about what intentionsare associated with the 1 am having trouble doing X type

    1 86

    of utterance, and what the 1CP and OCP know about thetask they have undertaken.

    The dominance relationship that holds between I1 and12, as well as the one that holds between I1 and 13, mayseem problematic at first glance. It is not clear howlocating any single setscrew contributes to removing theflywheel. It is even less clear how, in and of itself, identi-fying ano the r tool does. T wo facts provide the link: first,that the apprentice (the OCP of DS1) has taken on thetask of removing the flywheel; second, that the appren-tice and expert share certain knowledge about the task.Some of this shared task knowledge comes from thediscourse per se [e.g., utterance (3)], but some of it

    Computat ional Linguistics, Volume 1 2 , N u m b e r 3 , J u l y - S e p te m b e r 1986

  • 7/28/2019 grosz and sidner 1986 attention, intention and structure of discourse.pdf

    14/31

    B a r b a r a J . G r o s z a n d C a n d a c e L . S i d n e r A t t e n t i o n , I n t e n t i o n s , a n d t h e S t r u c t u r e o f D i s c o u r s e

    c o m e s f r o m g e n e r a l k n o w l e d g e , p e r c e p t u a l i n f o r m a t i o n ,a n d t h e li k e. T h u s , a c o m b i n a t i o n o f i n f o r m a t i o n i s r e l e -v a n t t o d e t e r m i n i n g 1 2 a n d 1 3 a n d t h e i r r e l a t i o n s h i p s t oI 1 , i n c l u d i n g a l l o f t h e f o l l o w i n g : t h e f a c t t h a t I 1 is p a r to f t h e i n t e n t io n a l s t r u c t u re , t h e f a c t t h a t t h e a p p r e n t i c e i sc u r r e n t l y w o r k i n g o n s a t i s f y i n g I 1 , t h e u t t e r a n c e - l e v e li n t e n t io n s o f u t t e r a n c e s ( 5 ) a n d ( 1 2 ) , a n d g e n e r a l k n o w -l e d g e a b o u t t h e t a s k .

    T h e s a t i s f a c t i o n - p r e c e d e n c e r e l a t i o n s a m o n g 1 2 , 1 3 ,a n d 1 4 a r e n o t c o m m u n i c a t e d d i r e c t l y i n t h e d i a l o g u e ,b u t , l i k e d o m i n a n c e r e l a t i o n s , d e p e n d o n d o m a i n k n o w -l e d ge . O n e p i e c e o f r e l e v a n t k n o w l e d g e i s t h a t a s a t is f a c -t i o n p r e c e d e n c e r e l a t i o n e x i s t s b e t w e e n l o o s e n i n g t h es e t s c r e w s a n d pu l l i n g o f f t h e f l y w h e e l . T h a t r e l a t i o n iss h a r e d k n o w l e d g e t h a t i s s t a t e d d i r e c t l y (F ir s t loosen . . . .then p u l l ) . T h e r e l a t i o n , a l o n g w i t h t h e f a c t t h a t b o t h 1 2a n d 1 3 c o n t r i b u t e t o l o o s e n i n g t h e s e t s c r e w s , a n d t h a t 1 4c o n t r i b u t e s t o p u l l i n g o f f t h e f l y w h e e l , m a k e s i t p o s s i b l et o c o n c l u d e 1 3 S P 1 4 a n d 1 2 S P 14 . T o c o n c l u d e t h a t 1 2S P 13 , t h e a p p r e n t i c e m u s t e m p l o y k n o w l e d g e o f h o w t og o a b o u t l o o s e n i n g s c r e w - l i k e o b j e c t s .T h e d o m i n a n c e a n d s a t i s f a c t i o n - p r e c e d e n c e r e l a t i o n sf o r t h i s t a s k - o r i e n t e d f r a g m e n t f o r m a t r e e o f i n t e n t i o n sr a t h e r t h a n j u s t a p a rt i a l o r d e r in g . I n g e n e ra l , h o w e v e r ,f o r a n y f r a g m e n t , t a s k - o r i e n t e d o r o t h e r w i s e , t h i s i s n o tn e c e s s a r y .

    I t i s e s s e n t i a l t o n o t i c e t h a t t h e i n t e n t i o n a l s t r u c t u r e i sn e i t h e r i d e n t i c a l t o n o r i s o m o r p h i c t o a g e n e r a l p l a n f o rr e m o v i n g th e f l y w h e e l . I t i s n o t i d e n t ic a l b e c a u s e a p l a ne n c o m p a s s e s m o r e t h a n a c o l l e c t i o n o f i n t e n t i o n s a n dr e l a t i o n s h i p s b e t w e e n t h e m ( c o m p a r e P o l l a c k ' s ( 1 9 8 6 )c r i ti q u e o f A I p l a n n i n g f o r m a l i s m s a s t h e b a s i s f o r i n f e r -r i n g i n t e n t i o n s i n d i s c o u r s e ) . I t i s n o t i s o m o r p h i c b e c a u s et h e i n t e n t i o n a l s t r u c t u r e h a s a d i f f e r e n t s u b s t r u c t u r e f r o mt h e g e n e r a l p l a n f o r r e m o v i n g t h e f l y w h e e l . I n a d d i t io nt o t h e i n t e n t i o n s a r i s i n g f r o m s t e p s i n t h e p l a n , t h e i n t e n -

    t i o n a l s t r u c t u r e t y p i c a l l y c o n t a i n s D S P s c o r r e s p o n d i n g t oi n t e n t i o n s g e n e r a t e d b y t h e p a r t i c u l a r e x e c u t i o n o f t h et a s k a n d t h e d i a l o gu e . F o r e x a m p l e , th e g e n e r a l p l a n f o rt h e d i s a s s e m b l y o f a f ly w h e e l i n c l u d e s s u b p l a n s f o r l o o s -e n i n g t h e s e t s c r e w s a n d p u l l in g o f f t h e w h e e l ; i t m i g h ta l s o in c l u d e s u b p l a n s ( o f t h e l o o s e n i n g s t e p ) f o r f i n di n gt h e s e t s c r e w s , f i n d i n g a t o o l w i t h w h i c h t o l o o s e n t h es c r e w s , a n d l o o s e n i n g e a c h s c r e w i n d i v i d u a l l y . H o w e v e r ,t h i s p l a n w o u l d n o t c o n t a i n c o n t i n g e n c y s u b p l a n s f o rw h a t t o d o w h e n o n e c a n n o t f i n d t h e s c r e w s o r r e a l i z e st h a t t h e a v a i l a b l e t o o l i s u n s a t i s f a c t o r y . I n t e n t i o n s I 2 a n dI 3 s t e m f r o m d i f f i c u l t i e s e n c o u n t e r e d i n l o c a t i n g a n dl o o s e n i n g t h e s e t s c re w s . T h u s , t h e i n te n t i o n a l s t r u c t u r ef o r t h is f r a g m e n t i s n o t i s o m o r p h i c t o t h e g e n e r a l p l a n f o rr e m o v i n g t h e f l y w h e e l .

    U t t e r a n c e ( 1 8 ) o f f e r s a n o t h e r e x a m p l e o f t h e d i ff e r -e n c e b e t w e e n t h e i n t e n t i o n a l s t r u c t u r e a n d a g e n e r a l p l a nf o r th e t a s k . T h i s u t t e r a n c e i s p a r t o f D S 4 - n o t j u s t p a r to f DS 1 - e v e n t h o u g h i t c o n t a i n s r e f e r e n c e s t o m o r e t h a no n e s i n g l e p a r t o f t h e o v e r a l l t a s k ( w h i c h i s w h a t I 1 i sa b o u t ) . I t f u n c t i o n s t o e s t a b l i s h a n e w D S P , 1 4 , a s m o s ts a l ie n t . R a t h e r t h a n b e i n g r e g a r d e d a s a r e p o r t o n t h eo v e r a l l s t a t u s o f t h e t a s k , t h e f i r s t c l a u s e i s b e s t s e e n a sm o d i f y i n g t h e D S P . 8 W i t h i t, t h e a p p r e n t i c e t e l ls t h ee x p e r t t h a t t h e t r o u b l e i n r e m o v i n g t h e w h e e l i s n o t w i t ht h e s c r e w s . T h u s , a l t h o u g h g e n e r a l t a s k k n o w l e d g e i su s e d i n d e t e r m i n i n g t h e i n t e n t i o n a l s t r u c t u r e , i t i s n o tid en t i ca l t o i t .

    I n t h is d i a l o g u e , t h e r e a r e f e w e r i n s t a n c e s i n w h i c h c u ep h r a s e s a r e e m p l o y e d t o i n d i c a t e s e g m e n t b o u n d a r i e st h a n o c c u r in t h e m o v i e s e s s a y . T h e p r i m a r y e x a m p l e i st h e u s e o f f i r s t i n ( 1 ) t o m a r k t h e s t a r t o f th e s e g m e n ta n d t o i n d i c a t e t h a t i t s D S P i s t h e f i r s t o f s e v e r a li n t e n t i o n s w h o s e s a t i s f a c t i o n w i ll c o n t r i b u t e t o s a t i s fy i n gt h e l a r g e r d i sc o u r s e o f w h i c h t h e y a r e a p a r t .

    P r i m a r y I n t e n t i o n s :I I " ( I n t e n d Exper t ( I n t e n d Apprent i c e ( R e m o v e A f l yw h e e l ) ) )I 2 : ( I n t e n d A ( I n t e n d E (T e l l E A ( L o c a t i o n o t h e r s e t s c r e w ) ) ) )I 3 : ( I n t e n d A ( I n t e n d E ( I d e n t if y E A a n o t h e r t o o l ) ) )I 4 : ( I n t e n d A ( I n t e n d E ( T e l l E A ( H o w ( G e t o f f A w h e e l ) ) ) ) )I 5 : ( I n t e n d E ( K n o w - H o w - t o A (U s e A w h e e l p u l l e r) ) )

    D o m i n a n c e R e l a t i o n s h i p s :I 1 D O M I 2I 1 D O M I 3I 1 D O M I 4I 4 D O M I 5

    S a t i s f a c t i o n - P r e c e d e n c e R e l a t i o n s h i p s :I 2 S P I 3I 2 S P I 4I 3 S P I 4

    Figure 6 . I n t e n t i o n a l s t r u c t u r e f o r t h e t a s k - o r i e n t e d d i a l o g u e s e g m e n t .C o m p u t a t i o n a l L i n g u i s t i c s , V o l u m e 1 2 , N u m b e r 3 , J u l y - S e p t e m b e r 1986 187

  • 7/28/2019 grosz and sidner 1986 attention, intention and structure of discourse.pdf

    15/31

    B a r b a r a J . G r o s z a n d C a n d a c e L . S i d n e r A t t e n t i o n , I n t e n t i o n s , a n d t h e S t r u c t u r e o f D i s c o u r s e

    The dialogue includes a clear example of the influenceof discourse structure on referring expressions. Thephrase the screw in the center is used in (25) to refer tothe center screw of the wheelpuller, not one of the twosetscrews mentioned in (18). This use of the phrase ispossible because of the attentional state of the discoursestructure at the time the phrase is uttered.

    4 PROCESSING ISSUESIn previous sections of the paper, we abstracted from thecognitive states of the discourse participants. The variouscomponents of discourse structure discussed so far areproperties of the discourse itself, not of the discourseparticipants. To use the theory in constructing computa-tional models requires determining how each of the indi-vidual components projects onto the model of anindividual discourse participant. In this regard, the prin-cipal issues include specifying1. how the ICP indicates and the OCP recognizes the

    beginning and end of a discourse segment,2. how the OCP recognizes the discourse segmentpurposes, and3. how the focus space stack operates.

    In essence, the OCP must judge for each utterancewhether it starts a new segment, ends the current one(and possibly some of its embedding segments), orcontributes to the current one. The informatio n availableto the OCP for recognizing that an utterance starts a newsegment includes any explicit linguistic cues contained inthe utter ance (see Section 6 9 as well as the relatio nshipbetween its utterance-level intentions and the active DSPs(i.e., those in some focus space that is still on the stack).Likewise, the fact that an utterance ends a segment maybe indicated explicitly by linguistic cues or implicitly fromits utterance-level intentions and their relationship toelements of the intentional structure. If neither of these isthe case, the utterance is part of the current segment.Thus, intention recognition and focus space managementplay key roles in processing. Moreover, they are alsorelated: the intentional structure is a primary factor indetermining focus space changes, and the focus spacestructure helps constrain the intention recognition proc-ess.

    4.1 INTENTION RECOGNITIO N

    The recognition of DP/DSPs is the central issue in thecomputation al modeling of intentional structure. If, aswe have claimed, for the discourse to be coherent andcomprehensible, the OCP must be able to recognize boththe DP/DSPs 10 and relationships (dominanc e and satis-faction-precedence) between them, then the question ofhow the OCP does so is a crucial issue.

    For the discourse as a whole, as well as for each of itssegments, the OCP must identify both the intention thatserves as the discourse segment purpose and its relation-ship to other discourse-level intentions. In particular, theOCP must be able to recognize which other DSPs that

    specific intention dominates and is dominated by, and,where relevant, with which other DSPs it has satisfac-tion-pr ecedence relationships. Two issues that arecentral to the recognition problem are what informationthe OCP can utilize in effecting the recognition and atwhat point in the discourse that information becomesavailable.

    An adequate computational model of the recognitionprocess depends critically on an adequate theory ofintention and action; this, of course, is a large researchproblem in itself and one not restricted to matters ofdiscourse. The need to use such a model for discourse,however, adds certain constraints on the adequacy of anytheory or model. Pollack (1986) describes several prop-erties such theories and models must possess if they areto be adequate for supporting recognition of intention insingle-utterance queries; she shows how current AI plan-ning models are inadequate and proposes an alternativeplanning formalism. The need to enable recognition ofdiscourse-level intentions leads to yet another set ofrequirements.

    As will become clear in what follows, the informationavailable to the OCP comes from a variety of sources.Each of these can typically provide partial informationabout the DSPs and their relationships. These s ources areeach partially constraining, but only in their ensemble dot