Skip to main content

Disjunction in a Predictive Theory of Anaphora

Part of the Lecture Notes in Computer Science book series (LNCS,volume 13524)


In this paper I develop a dynamic semantics for a first-order fragment, which incorporates insights from work on anaphora in logic, and the trivalent approach to presupposition projection. The resulting system—EDS—has interesting features which set it apart, both conceptually and empirically, from earlier iterations of dynamic semantics. Conceptually, the meanings of the logical connectives are derived by systematically generalizing the Strong Kleene connectives into a dynamic setting—the system is thereby predictive, drawing a tight connection between the logic of presupposition projection and patterns of anaphoric accessibility. On the empirical side, EDS diverges sharply from earlier proposals. In this paper, I focus mainly on disjunction, arguing that EDS provides a simple and elegant account of the dynamics of disjunction, including traditionally problematic cases such as Partee disjunctions and program disjunctions.


  • Disjunction
  • Presupposition
  • Anaphora

This is a preview of subscription content, access via your institution.

Buying options

USD   29.95
Price excludes VAT (USA)
  • DOI: 10.1007/978-3-031-25894-7_4
  • Chapter length: 23 pages
  • Instant PDF download
  • Readable on all devices
  • Own it forever
  • Exclusive offer for individuals only
  • Tax calculation will be finalised during checkout
USD   39.99
Price excludes VAT (USA)
  • ISBN: 978-3-031-25894-7
  • Instant PDF download
  • Readable on all devices
  • Own it forever
  • Exclusive offer for individuals only
  • Tax calculation will be finalised during checkout
Softcover Book
USD   54.99
Price excludes VAT (USA)
Fig. 1.
Fig. 2.


  1. 1.

    This will later prove useful when embedding EDS in a concrete discourse pragmatics.

  2. 2.

    The definition of relational composition (which is totally standard) is given below:

    $$\begin{aligned} R \circ S := \{(g,i)|\exists h[(g,h) \in R \wedge (h,i) \in S]\} \end{aligned}$$

    This operation plays a central role in both DPL and EDS.

  3. 3.

    G&S importantly assume that a multi-sentence discourse is translated into DPL as a conjunctive sentence.

  4. 4.

    N.b. the universal quantifier is defined as the dual of the existential.

  5. 5.

    As emphasized by Mandelkern and Rothschild [10] the kind of situation-based e-type approach to anaphora developed in [11] and refined in [12, 13] does not (in its current state, at least) consitute a viable alternative. E-type theories have not addressed in detail how to capture notions of anaphoric accessibility in complex sentences, beyond donkey sentences, and as shown in [14], were they to do so, they would require entries for the logical connectives which manipulate minimal situations in an apparently arbitrary fashion.

  6. 6.

    If \(\phi \) and \(\lnot \lnot \phi \) are equivalent, then choosing to use a sentence of the form \(\lnot \lnot \phi \) is naturally expected to trigger a Manner implicature. I leave the interesting question of the pragmatics of doubly-negated sentences to future work.

  7. 7.

    (16) is in fact modelled after the following well-known example used to motivate existential readings of donkey sentences (attributed by [17, p. 63] to Robin Cooper).

    figure q


  8. 8.

    This observation is often attributed to the later [18]. The intuition behind the analysis is already implied by the fact that the two indefinites are annotates with the same variable.

  9. 9.

    For an excellent recent overview of DPL, which expands on many issues which I don’t have the space to discuss here, see [21].

  10. 10.

    Since I’m exclusively concerned with anaphoric presuppositions here, I make the simplifying assumption that all predicates are bivalent, i.e., if all of the values of the variables are known then an atomic sentence is always either true or false. One way of extending the logical language in order to model (non-anaphoric) presuppositions while maintaining bivalent predicates would be to incorporate Beaver’s unary presupposition operator [23].

  11. 11.

    Note that there are different ways in which to interpret the third truth value in a trivalent setting, e.g., as standing in for undefinedness. Here, it is explicitly referred to as “unknown”, since this framing is a natural fit for the Strong Kleene logic of indeterminacy, which is exploited extensively later in the paper. Undefinedness typically goes together with Weak Kleene logic. I’m grateful to an anonymous reviewer for pressing me to clarify this point.

  12. 12.

    This builds on the dynamic system developed in [25, 26], in which outputs are paired with bivalent truth-values.

  13. 13.

    The generalization of Strong Kleene trivalent semantics to a dynamic setting will out of necessity remain rather impressionistic in this paper. The procedure of lifting truth-functional operators into a dynamic setting has however been made precise in important work by Charlow [26]. Simon Charlow (p.c.) points out that the recipe for lifting the Strong Kleene semantics used here can be formalized as a lifting of the Strong Kleene connectives into the State.Set applicative, following [26]. See [20] for more details.

  14. 14.

    In order to keep the definitions relatively terse, I take advantage of the convention that is understood as .

  15. 15.

    I’ll simply note here that the validity of de Morgan’s in the presence of anaphoric dependencies seems independently desirable given our intuitions about natural language. The following sentences are all arguably truth-conditionally equivalent.

    figure aq
  16. 16.

    Assuming a Strong Kleene semantics for material implication. Something interesting to note here is that EDS predicts existential truth conditions for donkey sentences, unlike, e.g., [3]. Egli’s corrolary therefore doesn’t hold.

    This is by no means a bad prediction—it has been widely reported that such existential readings are attested for donkey sentences, as alluded to in fn. 7 (see also [17, 27]). The following example, for example, is clearly true if Gabor owns two credit cards but only pays with one of them.

    figure ar

    The empirical picture is however much more complicated, and donkey sentences do often have stronger, universal readings. Relatedly, [15] reports that Partee disjunctions have universal readings—unlike donkey sentences, to my knowledge very little work has been done examining the distribution of existential and universal readings of Partee disjunctions. A detailed discussion of universal readings will have to wait for another occasion, but see [28] for one recent approach.

  17. 17.

    This is what von Fintel calls ‘Stalnaker’s bridge’ [30], in the context of a dynamic setting.

  18. 18.

    Various pragmatic justifications can be given for the formal contingency requirement stated in (43). What is important for my purposes is that if a disjunctive sentence is asserted by a speaker \(s\) in a context which trivializes one of the disjuncts, the assertion is judged to be ‘odd’.

  19. 19.

    There seem to be information-structural constraints on program disjunctions in natural language which are beyond the remit of EDS. For example, it seems that some degree of parallelism is required to hold between the disjuncts. Singular anaphora, by my reckoning, is extremely difficult in the following example:

    figure ax

    I speculate that this is related to constraints on co-indexing. I leave this interesting issue to future work on program disjunctions.

  20. 20.

    This kind of suggestion raises the issue of how exactly natural language sentences can be mapped compositionally to EDS Logical Forms. After all, one doesn’t want to allow for too much flexibility, otherwise the resulting grammar won’t be sufficiently constrained.


  1. Karttunen, L.: Discourse referents. In: McCawley, J.D. (ed.) Syntax and Semantics, vol. 7, pp. 363–386. Academic Press (1976)

    Google Scholar 

  2. Heim, I.: The Semantics of Definite and Indefinite Noun Phrases. PhD thesis, University of Massachusetts - Amherst (1982)

    Google Scholar 

  3. Groenendijk, J., Stokhof, M.: Dynamic predicate logic. Linguist. Philos. 14(1), 39–100 (1991)

    CrossRef  MATH  Google Scholar 

  4. Simons, M.: Disjunction and Anaphora. Semantics Linguist. Theory 6, 245–260 (1996).

    CrossRef  Google Scholar 

  5. Heim, I.: On the projection problem for presuppositions. In: Proceedings of WCCFL 2, pp. 114–125, Stanford University (1983)

    Google Scholar 

  6. Soames, S.: Presupposition. In: Gabbay, D., Guenthner, F. (eds.) Handbook of Philosophical Logic: Topics in the Philosophy of Language, vol. 4. pp. 553–616. Springer, Netherlands, Dordrecht (1989).

  7. Schlenker, P.: Be articulate: A Pragmatic Theory of Presupposition Projection. Theoretical Linguist. 34(3) (2008).

  8. Schlenker, P.: Local contexts. In: Semantics and Pragmatics, vol. 2 (2009).

  9. Schlenker, P.: Local contexts and local meanings. Philoso. Stud. Int. J. Philos. Analytic Tradition 151(1), 115–142 (2010).

    CrossRef  Google Scholar 

  10. Mandelkern, M., Rothschild, D.: Definiteness projection. Nat. Lang. Seman. 28(2), 77–109 (2019).

    CrossRef  Google Scholar 

  11. Heim, I.: E-type pronouns and donkey anaphora. Linguist. Philos. 13(2), 137–177 (1990).

    CrossRef  Google Scholar 

  12. Elbourne, P.: Situations and Individuals. PhD thesis, Massachussetts Institute of Technology (2005)

    Google Scholar 

  13. Elbourne, P.: Definite Descriptions. Oxford University Press, Oxford (2013)

    CrossRef  Google Scholar 

  14. Mandelkern, M., Rothschild, D.: Independence day? J. Semant. 36(2), 193–210 (2019).

    CrossRef  Google Scholar 

  15. Krahmer, E., Muskens, R.: Negation and disjunction in discourse representation theory. J. Semant. 12(4), 357–376 (1995).

    CrossRef  Google Scholar 

  16. Gotham, M.: Double negation, excluded middle and accessibility in dynamic semantics. In: Schlöder, J.J., McHugh, D., Roelofsen, F. (eds.) Proceedings of the 22nd Amsterdam Colloquium, pp. 142–151 (2019)

    Google Scholar 

  17. Chierchia, G.: Dynamics of Meaning - Anaphora, Presupposition, and the Theory of Grammar. University of Chicago Press, Chicago (1995)

    CrossRef  Google Scholar 

  18. Stone, M.D.: ‘Or’ and Anaphora. Semantics Linguist. Theory 2, 367–386 (1992).

    CrossRef  Google Scholar 

  19. Rothschild, D.: A trivalent approach to anaphora and presupposition. In: Cremers, A., van Gessel, T., Roelofsen, F. (eds.) Proceedings of the 21st Amsterdam Colloquium, pp. 1–13 (2017)

    Google Scholar 

  20. Elliott, P.D.: Towards a Principled Logic of Anaphora. Unpublished manuscript (2020)

    Google Scholar 

  21. Gillies, A.S.: On groenendijk and stokhof’s “dynamic predicate logic”. In: McNally, L., Szabó, Z.G. (eds.) A Reader’s Guide to Classic Papers in Formal Semantics, pp. 121–153. Springer International Publishing, Cham (2022).

  22. Van den Berg, M.: Full Dynamic plural logic. In: Proceedings of the Fourth Symposium on Logic and Language (1996)

    Google Scholar 

  23. Beaver, D.I.: Presupposition and Assertion in Dynamic Semantics. CSLI Publications (2001)

    Google Scholar 

  24. Van den Berg, M.H.: Some Aspects of the Internal Structure of Discourse. The Dynamics of Nominal Anaphora (1996)

    Google Scholar 

  25. Charlow, S.: On the Semantics of Exceptional Scope. PhD thesis, Rutgers University, New Brunswick (2014)

    Google Scholar 

  26. Charlow, S.: Static and dynamic exceptional scope. To Appear J. Semant. (2020)

    Google Scholar 

  27. Kanazawa, M.: Weak vs. strong readings of donkey sentences and monotonicity inference in a dynamic setting. Linguist. Philos. 17(2), 109–158 (1994)

    Google Scholar 

  28. Champollion, L., Bumford, D., Henderson, R.: Donkeys under Discussion. Semant. Pragmatics 12, 1 (2019).

  29. Veltman, F.: Defaults in update semantics. J. Philos. Log. 25(3), 221–261 (1996)

    CrossRef  MathSciNet  MATH  Google Scholar 

  30. Von Fintel, K.: What is presupposition accommodation, again?*. Philos. Perspect. 22(1), 137–170 (2008).

    CrossRef  MATH  Google Scholar 

  31. Heim, I.: File change semantics and the familiarity theory of definiteness. In: Meaning, Use, and Interpretation of Language, pp. 164–189. De Gruyter (1983).

  32. Elliott, P.D.: Partee Conjunctions and Free Choice with Anaphora. Handout from a talk given at LFRG, MIT (2022)

    Google Scholar 

  33. Mandelkern, M.: Witnesses. Linguist. Philos. (2022)

    Google Scholar 

  34. Hofmann, L.: The anaphoric potential of indefinites under negation and disjunction. In: Schlöder, J.J., McHugh, D., and Roelofsen, F. (eds.) Proceedings of the 22nd Amsterdam Colloquium, pp. 181–190 (2019)

    Google Scholar 

  35. Hofmann, L.: Anaphora and Negation. PhD thesis, University of California Santa Cruz (2022)

    Google Scholar 

  36. Muskens, R.: Combining montague semantics and discourse representation. Linguist. Philos. 19(2), 143–186 (1996).

    CrossRef  MathSciNet  Google Scholar 

  37. Stone, M.: Reference to Possible Worlds. Technical Report 49, Rutgers University Center for Cognitive Science (1999)

    Google Scholar 

  38. Peters, S.: A truth-conditional formulation of karttunen’s account of presupposition. Synthese 40(2), 301–316 (1979).

    CrossRef  MathSciNet  Google Scholar 

  39. George, B.R.: Predicting Presupposition Projection - Some Alternatives in the Strong Kleene Tradition

    Google Scholar 

  40. George, B.R.: A New Predictive Theory of Presupposition Projection. In: Proceedings of SALT 18, pp. 358–375. Cornell University, Ithaca, NY. (2008)

    Google Scholar 

Download references


Aspects of this work have been presented in various venues, including at Rutgers, NYU, MIT, ENS, and most recently at the Third Tsinghua Interdisciplinary Workshop on Logic, Language, and Meaning. I’m grateful to participants on all such occasions for insightful and challenging feedback on this material, which has shaped the current form. The logic outlined in this paper was developed for the Spring 2022 Topics in Semantics seminar at MIT, and I’m especially grateful to Filipe Hisao Kobayashi and Enrico Flor for their input. I remain solely responsible for any mistakes.

Author information

Authors and Affiliations


Corresponding author

Correspondence to Patrick D. Elliott .

Editor information

Editors and Affiliations

Rights and permissions

Reprints and Permissions

Copyright information

© 2023 The Author(s), under exclusive license to Springer Nature Switzerland AG

About this paper

Verify currency and authenticity via CrossMark

Cite this paper

Elliott, P.D. (2023). Disjunction in a Predictive Theory of Anaphora. In: Deng, D., Liu, M., Westerståhl, D., Xie, K. (eds) Dynamics in Logic and Language. TLLM 2022. Lecture Notes in Computer Science, vol 13524. Springer, Cham.

Download citation

  • DOI:

  • Published:

  • Publisher Name: Springer, Cham

  • Print ISBN: 978-3-031-25893-0

  • Online ISBN: 978-3-031-25894-7

  • eBook Packages: Computer ScienceComputer Science (R0)