J. van Eijk and H. Kamp. Representing discourse in context. In J. van Benthem and A. ter Meulen, editors, Handbook of Logic and Linguistics, pages 179--237. Elsevier, 1997.  Home/Search   Document Details and Download   Summary   Related Articles   Check

....in the refinement of the notion of meaning the theory is based on. Thus, by considering how DRT can be made a compositional theory of text interpretation, it is possible to determine how the meaning of sentences should be represented in order to account for the role sentences play in texts. In [vEK97] a compositional version of DRT has been introduced which accounts for the sharing of discourse markers across DRSs which have been associated with different sentences. Let A be a set of constants, V a set of discourse markers and P a set of n ary predicates. The language of pDRS (proto DRSs) can ....

.... to account for the sharing of discourse markers is the semantics of the composition of pDRS, which is defined as follows: X ,F = XX , f F : fF f[X ]f Given a pDRS K i , in order to characterize the set of the discourse markers which can be referred to also in a different pDRS K j , in [vEK97] three different kinds of discourse markers occurrences have been distinguished which roughly correspond to different types of variables in imperative programming languages: i) discourse markers whose interpretation is fixed in a larger context (variables in read memory) ii) discourse markers ....

[Article contains additional citation context not shown here]

van Eijk, J. and Kamp, H. 1997, "Representing Discourse in Context", in Handbook of Logic and Language, (van Benthem, J. and ter Meulen, A., eds.), Amsterdam: Elsevier Science, pp. 181-237.

....the semantic content of one picture and 9x that of the other. Note that ii (but not i) forces variables shared between and to corefer. It is essentially this property of ii that is exploited in the various formalisms of dynamic semantics, such as Discourse Representation Theory (e.g. [van Eijck and Kamp, 1997]) which try to capture certain laws governing anaphora in natural language. Implementing the approach outlined in this paper in a full blown document generation system requires that a number of issues are resolved that cannot be discussed here. For example, the issue of locating sequences of ....

J.van Eijck and H.Kamp. Representing Discourse in Context. In J.van Benthem and A.ter Meulen (Eds.), Handbook of Logic and Language. Elsevier Science and The MIT Press.

....Hathout, 1998] Unfortunately, these approaches do not deal explicitly with the problem of tempo1 ral anaphora. In [Bos et al. 1994] we nd an interesting solution for the resolution of nominal anaphora, but it cannot be used directly to solve the temporal anaphora. In their description of DRT, [van Eijck and Kamp, 1997] give a compositional account of tense and aspect, in a formalism similar to the one proposed in [Muskens, 1996] but they do not give a complete and satisfactory solution. In this paper, we present a compositional implementation of DRT where the resolution of temporal anaphora is directly ....

....we will interpret that Mary cried after John hurt her and also that she cried yesterday. Thus, contrary to the implementation of the nominal anaphora resolution of [Bos et al. 1994] there are more than one element involved in the temporal anaphora. In the compositional approach proposed in [van Eijck and Kamp, 1997] a register is used to memorize the last event introduced in the contextual DRS. The new eventuality expressed by the current sentence is related with this event pointed by this registry. The registry is updated if the new eventuality is an event. This is a solution to only a part of the problem. ....

[Article contains additional citation context not shown here]

van Eijck, J. and Kamp, H. (1997). Representing discourse in context. In van Benthem, J. and ter Meulen, A., editors, Handbook of Logic and Language. North-Holland, Amsterdam. 20

....the question should be addressed which variable to use for that introduction. This of course depends on the variables already in use for the discourse referents that are part of the current context . If there is no explicit representation of the current context, the problem remains elusive. In [21] an attempt was made to deal with it by devising strategies for merging representations, but in retrospect I consider the merge problem as a typical example of a pseudo problem caused by a wrong choice of primitives. One way of avoiding the destructive assignment pseudo problem is by ....

.... Treatment of Context in NL A Montagovian treatment of the dynamics of context change should not start out from a first order logic with destructive assignment and extend this to a typed logic, like in Dynamic Montague Grammar [26] or the typed versions of discourse representation theory in [21, 39, 48], but from a process of context extension. What we need instead of a destructive Montague Grammar is a Contextual or Incremental Montague Grammar. The Proper Treatment of Context for NL developed in [17] in terms of polymorphic type theory (see, e.g. 30, 43] uses type specifications of ....

[Article contains additional citation context not shown here]

J. van Eijck and H. Kamp. Representing discourse in context. In J. van Benthem and A. ter Meulen, editors, Handbook of Logic and Language, pages 179--237. Elsevier, Amsterdam, 1997.

....Incremental Processing Dynamic versions of Montague grammar use some version of typed logic. Most typed logics assume a basic type for discourse referents or indices, and impose suitable axioms to ensure that these behave like the variables of dynamic predicate logic. Examples of this approach are [12, 5, 19, 29, 30, 28, 6, 9, 24, 25, 3]. Closest to our approach is [3] where a typed dynamic logic is proposed that handles the singular plural distinction and treats anaphoric linking. The main di#erence is that [3] follows the lead of DPL ( 13] whereas we remain much closer to the spirit of DRT. Because of the reliance on ....

J. van Eijck and H. Kamp. Representing discourse in context. In J. van Benthem and A. ter Meulen, editors, Handbook of Logic and Language, pages 179--237. Elsevier, Amsterdam, 1997.

....6= and [ i 1 ] c 6= i 2 ] c : If c 2 D n and d 2 D we use c d for the context c 2 D n 1 that is the result of appending d at the end of c. The ID interpretation of formulas can now be given as a map in D , P(D ) a partial function, because of the possibility of unde nedness) [9]] c) fc d j d 2 Dg [ P i 1 i n ] c) if 9j(1 j n and [ i j ] c = fcg if M j= c P i 1 i n ; if M =j c P i 1 i n [ i 1 = i 2 ] c) if [ i 1 ] c = or [ i 1 ] c = fcg if M j= c i 1 ; if M =j c i 1 [ c) fcg if [ c) ....

....Incremental Processing Dynamic versions of Montague grammar use some version of typed logic. Most typed logics assume a basic type for discourse referents or indices, and impose suitable axioms to ensure that these behave like the variables of dynamic predicate logic. Examples of this approach are [12, 5, 19, 29, 30, 28, 6, 9, 24, 25, 3]. Closest to our approach is [3] where a typed dynamic logic is proposed that handles the singular plural distinction and treats anaphoric linking. The main di erence is that [3] follows the lead of DPL ( 13] whereas we remain much closer to the spirit of DRT. Because of the reliance on ....

J. van Eijck and H. Kamp. Representing discourse in context. In J. van Benthem and A. ter Meulen, editors, Handbook of Logic and Language, pages 179-237. Elsevier, Amsterdam, 1997.

.... 22 Related Work Rational Reconstructions of DRT based on DPL When dynamic semantics for NL rst was proposed in [25] and [20] the approach invoked strong opposition from the followers of Montague [31] Rational reconstructions to restore compositionality were announced in [16] and carried out in [15, 7, 22, 33, 34, 32, 8, 12, 28, 29]. All of these reconstructions are based in some way or other on DPL [16] and they all inherit the main aw of this approach: the destructive assignment problem. Interestingly, DRT itself did not su er from this problem: the discourse representation construction algorithms of [25] and [26] are ....

J. van Eijck and H. Kamp. Representing discourse in context. In J. van Benthem and A. ter Meulen, editors, Handbook of Logic and Language, pages 179-237. Elsevier, Amsterdam, 1997.

....of terms under provable equality as elements of the canonical model) 7. A Variation on the Theme To justify the plural in the title of this paper, first note that the approach to axiomatising dynamic predicate logic carries over without a hitch to the analysis of Discourse Representation Theory [8, 5]. The version for DRT looks even more elegant because the shift rules are absorbed by the set theoretic notation for DRSs. This reflects the fact that the syntax of DRT is more abstract than that of DPL. To keep as close as possible to the previous calculus, we consider a version of DRT where DRS ....

J. van Eijck and H. Kamp. Representing discourse in context. In J. van Benthem and A. ter Meulen, editors, Handbook of Logic and Language, pages 179--237. Elsevier, Amsterdam, 1996.

....result of this is that the subsequent sentence (2) can now use this discourse information to pick up the references: 2 He 1 kissed her 2 . 2 Extensions to Typed Logic Attempts to incorporate DPL stype dynamic semantics in mainstream Montague style natural language semantics [10] can be found in [5, 1, 7, 12, 13, 11, 2, 3, 8, 9]. The type hierarchy employed has basic types for entities (e) truth values(t) and markers (m) The states themselves can be viewed as maps from markers to suitable referents, i.e. a state has type m e. We abbreviate this as s. This can either be built into the type system from the start (see ....

J. van Eijck and H. Kamp. Representing discourse in context. In J. van Benthem and A. ter Meulen, editors, Handbook of Logic and Language, pages 179-237. Elsevier, Amsterdam, 1997.

....dependent types. Attempts to incorporate dynamic semantics in mainstream Montague style natural language semantics were also made. These attempts preserve much more of the original model theoretic avour of Montague grammar. Examples can be found in (Groenendijk and Stokhof, 1990; Muskens, 1994; van Eijck, 1997; van Eijck and Kamp, 1997; Kohlhase et al. 1996; Kuschert, 2000) What these attempts at incorporation have in common is the representation of context as a list of variables, and the use of dynamic quanti cation over these variables as context update. A serious disadvantage of dynamic quanti ....

....Attempts to incorporate dynamic semantics in mainstream Montague style natural language semantics were also made. These attempts preserve much more of the original model theoretic avour of Montague grammar. Examples can be found in (Groenendijk and Stokhof, 1990; Muskens, 1994; van Eijck, 1997; van Eijck and Kamp, 1997; Kohlhase et al. 1996; Kuschert, 2000) What these attempts at incorporation have in common is the representation of context as a list of variables, and the use of dynamic quanti cation over these variables as context update. A serious disadvantage of dynamic quanti cation is that update of a ....

J. van Eijck and H. Kamp. 1997. Representing discourse in context. In J. van Benthem and A. ter Meulen, editors, Handbook of Logic and Language, pages 179-237. Elsevier, Amsterdam.

....Semantics [11] in terms of the theory of dependent types. Attempts to incorporate dynamic semantics in mainstream Montague style natural language semantics were also made. These attempts preserve much more of the original model theoretic flavour of Montague grammar. Examples can be found in [7, 23, 29, 32, 17, 18]. What these attempts at incorporation have in common is the representation of context as a list of variables, and the use of dynamic quantification over these variables as context update. A serious disadvantage of dynamic quantification is that update of a reference marker u i makes the previous ....

J. van Eijck and H. Kamp. Representing discourse in context. In J. van Benthem and A. ter Meulen, editors, Handbook of Logic and Language, pages 179--237. Elsevier, Amsterdam, 1997.

No context found.

J. van Eijk and H. Kamp. Representing discourse in context. In J. van Benthem and A. ter Meulen, editors, Handbook of Logic and Linguistics, pages 179--237. Elsevier, 1997.

No context found.

J. van Eijck and H. Kamp. Representing discourse in context. In J. van Benthem and A. ter Meulen, editors, Handbook of Logic and Language. North{Holland, 1997.

No context found.

J. van Eijck and H. Kamp. Representing discourse in context. In Handbook of Logic and Language. J. van Benthem and A. ter Meulen, Cambridge, 1997.

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