A. Visser and C. Vermeulen. Dynamic bracketing and discourse representation. Notre Dame Journal of Formal Logic, 37:321--365, 1996. 15  Home/Search   Document Details and Download   Summary   Related Articles   Check

....theory. 10] demonstrates that reference resolution can be brought within the compass of dynamic semantics in a relatively straightforward way, and that very simple means are enough to implement something quite useful. 19 Visser Style Contexts and Referent Systems Visser s context theories [36, 38, 37, 39] also start out as rational reconstructions of DRT. Visser s view of contexts is considerably more abstract than the simple minded approach taken here. Our approach illustrates how little one has to assume about contexts for a working system. Referent systems [34, 33, 35] are a mechanism for ....

A. Visser and C. Vermeulen. Dynamic bracketing and discourse representation. Notre Dame Journal of Formal Logic, 37:321--365, 1996.

....syntactic units. For this we have the rst position in assignment tuples, but also the index keeping track of the current level during updating. We change levels with [ and ] Here is the The use of levels is closely related to Visser and Vermeulen s work on dynamic bracketing, cf. [VV96]. semantics. i [ hn 1; S hn 1; S i [ hn; S Next, I will consider centering theory as our candidate for integration with the formal system. 3 Centering Theory Centering Theory [GJW95] is a theory of coherence in local discourse. In this paper we will embrace ....

A. Visser and C. Vermeulen. Dynamic bracketing and discourse representation. Notre Dame Journal of Formal Logic, 37:321-365, 1996. 11

....theory. 10] demonstrates that reference resolution can be brought within the compass of dynamic semantics in a relatively straightforward way, and that very simple means are enough to implement something quite useful. 19 Visser Style Contexts and Referent Systems Visser s context theories [36, 38, 37, 39] also start out as rational reconstructions of DRT. Visser s view of contexts is considerably more abstract than the simple minded approach taken here. Our approach illustrates how little one has to assume about contexts for a working system. Referent systems [34, 33, 35] are a mechanism for ....

A. Visser and C. Vermeulen. Dynamic bracketing and discourse representation. Notre Dame Journal of Formal Logic, 37:321-365, 1996.

....with the formal system. 3 Centering Theory Centering Theory [GJW95] is a theory of coherence in local discourse. In this paper we will embrace the basic ideas behind centering but nothing prevents The use of levels is closely related to Visser and Vermeulen s work on dynamic bracketing, cf. [VV96]. The semantics of the closing bracket as given here is no doubt problematic, since it allows multiple utterances to be represented on the same level. In this paper however, the ] action plays no r ole. Still, to give a hint of an alternative de nition we might propose that the closing ....

A. Visser and C. Vermeulen. Dynamic bracketing and discourse representation. Notre Dame Journal of Formal Logic, 37:321-365, 1996. 13

....in the second fragment is an extension of the basic context logic. It should be compared with the treatment of pronoun resolution in DRT proposed in the second volume of [6] as well as with the earlier proposal for pronoun resolution in DRT in [43] Visser Style Contexts Visser s context theories [39, 41, 40, 42] also start out as rational reconstructions of DRT. Visser s view of contexts is considerably more abstract than the simpleminded approach taken here. The Centering Approach to Reference Resolution Central claim of the centering theory of local coherence in discourse [19, 18] is that pronouns are ....

A. Visser and C. Vermeulen. Dynamic bracketing and discourse representation. Notre Dame Journal of Formal Logic, 37:321-365, 1996.

.... formulations of DRT (e.g. Groenendijk and Stokhof [5] feature an explicit merge operation (e.g. relational composition) allowing a clear separation between the components i (so that i need not incorporate 1 Delta Delta Delta i Gamma1 ) See also van Benthem [1] and Visser and Vermeulen [14]. The notion of incrementality here is very much in the spirit of Kempson et al. [9] where an account is developed of how a sentence is interpreted as each word in the sentence is read from left to right. The present work proceeds at a somewhat different level, away from details of subsentential ....

....expressions as follows. For ( e; 2 E m Theta Phi m where m n, we backtrack to (ffl; ffl) e; ei n ( e 0 ; 0 ) iff (ffl; ffl) eei ( e 0 ; 0 ) 8 That said, there is no denying the richness of monoidal processing, as brought out in Visser and Vermeulen [14], compared to which the interpretation in Proposition 8 is incomparably crude. 21 (recalling that (ffl; ffl) is always assumed to belong to Stage) Otherwise, for all ( ee 1 Delta Delta Delta e n ; 1 Delta Delta Delta n ) and ( e 0 ; 0 ) 2 Stage, define ( ee 1 Delta ....

A. Visser and C.F.M. Vermeulen. Dynamic bracketing and discourse representation. Notre Dame Journal of Formal Logic, 37(2), 1996. 23

....the same. However, in (10b) we have lost the possibility to refer back to the dog. Hence, the dynamic meanings of the two formulae are not identical. This is rather unfortunate. What can be done A solution to this circle of problems was outlined by Kees Vermeulen and Albert Visser in [13] and [14]. Since the second paper is rather advanced and technical we will concentrate on the rst one, which introduces the so called referent systems. Referent systems will solve the problem only partly but that will be enough for our purposes. Our solution is clearly intended by [13] and [14] though the ....

....in [13] and [14] Since the second paper is rather advanced and technical we will concentrate on the rst one, which introduces the so called referent systems. Referent systems will solve the problem only partly but that will be enough for our purposes. Our solution is clearly intended by [13] and [14] though the actual details might di er. First of all, referent systems take a step back from Montague semantics in using no expressions. The variable management that was left implicit in Montague semantics is now made fully explicit. So we will actually translate tall by tall(x) and man by ....

Kees F. M. Vermeulen and Albert Visser. Dynamic bracketing and discourse representation. Notre Dame Journal of Formal Logic, 37:321 - 365, 1996. 29

....stage it is impossible to answer the question, but it is clear that several researchers feel the limitations of relational methods. For example, in their search for a precise specification of the dynamic interpretation process of natural language texts in humans and machines, Visser and Vermeulen [43] take the radical position that one must be able to interpret any chunk of text, and that the interpretation of larger chunks is a function of the interpretations of the smaller chunks. They argue that category theory is the proper format for the description of the flow of interpretation, and for ....

A. Visser and K. Vermeulen. Dynamic bracketing and discourse representation. Technical report, Department of Philosophy, Utrecht University, 1995.

No context found.

C. Vermeulen and A. Visser. 1996. Dynamic bracketing and discourse representation. Notre Dame Journal of Formal Logic, 37(2):321-365.

....with control. If this is true, then the present work is a move in the right direction. An important methodology of the present paper is the use of monoids to programme the interaction of meanings. In this paper we will not try to make the semantics fully monoidal. The reader is referred to [11] for some discussion of the aims and claims of monoidal semantics, 2 What is a Dynamic Predicate Logic Our aim of this section is to specify what we mean by a variant of Dynamic Predicate Logic. We will call such variants dynamic predicate logics or dpl s. Thus, somewhat awkwardly, DPL itself ....

....: g ffl [ 1] fpol: Gamma g The proper way to view 1 is as a dynamic bracket, signalling a switch of polarity. It is somewhat analogous to the bracket of L A T E X, compare 1 Gamma Gamma 1 with plain math plain. An important difference between 1 and the brackets developed in [11] is that nothing is thrown away when we switch polarity via 1. What has been stored is remembered and can be further extended as soon as we return. In contrast a level that is popped by one of the dynamic brackets of [11] cannot be returned to. In subsection A.3 of appendix A we show how to ....

[Article contains additional citation context not shown here]

A. Visser and C. Vermeulen. Dynamic bracketing and discourse representation. Notre Dame Journal of Formal Logic, 37:321--365, 1996.

....out to be extremely compatible. 5 Alternatively one could use (update) functions: interpreting OE in state I results in state O iff f OE (I) O. In such a functional set up of dynamic semantics the semantic universe will consist of (update) functions. cf. Veltman [16] 6 Or categories: cf. [19]. Zeevat [20] provides the algebraic formulation of the data base approach and Visser and Vermeulen [19] develop this idea further. Thus we obtain two main styles of formalisation in dynamic semantics: one describes meanings as programs, the other describes meanings as data bases. In both ....

....I results in state O iff f OE (I) O. In such a functional set up of dynamic semantics the semantic universe will consist of (update) functions. cf. Veltman [16] 6 Or categories: cf. 19] Zeevat [20] provides the algebraic formulation of the data base approach and Visser and Vermeulen [19] develop this idea further. Thus we obtain two main styles of formalisation in dynamic semantics: one describes meanings as programs, the other describes meanings as data bases. In both schools quite some thought has been given to the treatment of presuppositions. We will see the basics of the ....

A. Visser and C. Vermeulen. Dynamic bracketing and discourse representation. Notre Dame Journal of Formal Logic, 37:321--365, 1996.

....to the pieces of a jig saw puzzle. The forms of the pieces regulate the way they are put together. What we see on the pieces is the content. However, we often employ what we see on the pieces to find neighbouring pieces. The distinction of underlying file and file name is worked out in [14] and [17]. The solution of the problem of combining DPL with Veltman s Update Semantics is given in [6] The reader is referred to [11] and [7] for the study of a stacking semantics. See [17] for the idea of iterated storage. In terms of the three aspects, the present treatment can be characterized as ....

....the pieces to find neighbouring pieces. The distinction of underlying file and file name is worked out in [14] and [17] The solution of the problem of combining DPL with Veltman s Update Semantics is given in [6] The reader is referred to [11] and [7] for the study of a stacking semantics. See [17] for the idea of iterated storage. In terms of the three aspects, the present treatment can be characterized as follows. We assume that one discourse referent is associated rigidly with a fixed string, so that we may confuse the referent with the string. We identify occurrences as the same, simply ....

[Article contains additional citation context not shown here]

A. Visser and C. Vermeulen. Dynamic bracketing and discourse representation. Notre Dame Journal of Formal Logic, 37:321--365, 1996.

No context found.

A. Visser and C. Vermeulen. Dynamic bracketing and discourse representation. Notre Dame Journal of Formal Logic, 37:321--365, 1996. 15

No context found.

K. Vermeulen and A. Visser. 1996. Dynamic bracketing and discourse representation. Notre Dame Journal of Formal Logic, to appear.

No context found.

K. Vermeulen and A. Visser. 1995. Dynamic bracketing and discourse representation. To appear.

No context found.

A. Visser and C. Vermeulen. Dynamic bracketing and discourse representation. Notre Dame Journal of Formal Logic, 37:321-365, 1996.

Online articles have much greater impact   More about CiteSeer.IST   Add search form to your site   Submit documents   Feedback  

CiteSeer.IST - Copyright Penn State and NEC