Skip to main content

Agentless presuppositions and the semantics of verbal roots


Bale (2007) proposes that agentive intransitives differ semantically from agentive transitives, in that while the agent of a transitive is introduced by a functional projection and composes with its verb via Event Identification (Kratzer 1996), intransitives lexically encode their agent arguments and compose with them via Function Application. This is based on the availability of agentless repetitive presuppositions with again, with transitives permitting a repetitive presupposition excluding the agent while intransitives do not. In this paper, we challenge Bale’s claim and show that typically intransitive verbs like dance and bark, which do not usually allow agentless presuppositions, permit agentless presuppositions when they appear with an optional internal argument. To account for this, we propose that verbal roots possess an underspecified thematic role argument, along with individual and event arguments. Combined with a conservative syntax for introducing agents via VoiceP (Kratzer 1996), the analysis captures the dependence of agentless presuppositions on the presence of an internal argument without recourse to any distinction between transitive and intransitive eventive verb roots. The analysis contributes a new theory of roots lying between two theoretical poles, one that argues that roots take internal arguments (e.g., Harley 2014) and one that severs internal arguments syntactically and semantically from the verb (e.g., Schein 1993; Borer 2003, 2005).

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


  1. 1.

    Following Bale, we provide both Davidsonian and Neo-Davidsonian translations for the verbs in (12) and (13), and use the Davidsonian translation in the trees (Davidson 1967; Castañeda 1966). In other examples, we will make use of Neo-Davidsonian representations throughout. We will also depart from Bale in treating thematic roles as functions from events to individuals, in line with recent work in event semantics (Carlson 1998; Landman 2000; Champollion 2010). By this latter point, we mean that Agent, for instance, is a function that takes an event and, if defined, returns the unique individual acting as the agent of that event. This is distinct from the type <e,<v,t>> functions that the functional heads Voice and v denote in our analysis, which serve to restrict the denotation of an event predicate to that set of events with a particular individual bearing a particular role in the events in question.

  2. 2.

    We presented the contrasts in (18)-(21) to five other native speakers of English, who confirmed their existence.

  3. 3.

    We thank an anonymous reviewer for pointing this issue out to us, as well as for providing the example in (25). The anonymous reviewer notes that these examples do sound unusual for them but are not completely unacceptable. The reviewer suggests that a possible explanation is that they are interpreting again in the same way as too, which presupposes that an identical event was performed by a discourse salient alternative individual as a result of associating with a focused DP. See Spathas and Michelioudakis (2020) for a discussion of additive presuppositions and their relevance to verbal argument structure in Greek.

  4. 4.

    As an alternative to different flavors of v, one could posit that a single functional head receives different semantic interpretations based on surrounding structural contexts, i.e., contextual allosemy (e.g., Wood and Marantz 2017). On this style of analysis, v may be interpreted as an identity function in the context of particular roots, or in the presence of a specifier of a particular sort, for example. The choice between this perspective and different flavors of v does not affect the results of the analysis, though there may be other reasons to favor one approach over another. For instance, the use of contextual allosemy rules might allow more fine-grained control over the distribution of particular thematic roles, where an account making use of flavors of v may need to appeal to additional syntactic mechanisms to ensure that certain v heads are incompatible with certain roots. This said, both approaches need to rely on additional factors constraining the range of possible meanings in a given structural and semantic context, such as conceptual knowledge.

  5. 5.

    We thank an anonymous reviewer for bringing this particular case to our attention.

  6. 6.

    One might suggest that this is in fact a restitutive, rather than a repetitive, reading of again, such that it is the state of the books being in the library’s possession, rather than the donation event, that is being repeated. This possibility can be ruled out by the following context, which shows that again is infelicitous with donate when no previous donation event occurred.

    1. (1)

      Context: The library had Moby Dick, White Teeth, and The Corrections in its collection, but a forgetful borrower failed to return them, and the library subsequently removed them from their records. Years later, the borrower gave the library the books as a donation, having forgotten that they had borrowed them from there in the first place.

      # The borrower donated the books (to the library) again.

    This can be distinguished from the behavior of verbs like give, which do permit restitutive readings with again (Beck and Johnson 2004).

    1. (2)

      Context: The library had Moby Dick, White Teeth, and The Corrections in its collection, but a forgetful borrower failed to return them, and the library subsequently removed them from their records. Years later, the borrower gave the library the books as a donation, having forgotten that they had borrowed them from there in the first place.

      The borrower gave the library the books again.

  7. 7.

    We do not claim that other ditransitive verbs, specifically those that participate in the double object-dative alternation like give and send, can be given the same analysis in their dative variants. Most likely, there will be individual syntactic and semantic differences between verbs like donate, which do not alternate, and verbs like give and send, which do. See for example Rappaport-Hovav and Levin (2008) and Beavers and Koontz-Garboden (2020) for extensive discussion of semantic differences between different classes of verbs that appear in one of these frames or alternate between both.

  8. 8.

    Wood and Marantz (2017) analyze for as a prepositional root modifying a high applicative head in Pylkkänen’s (2008) sense, introducing the Benefactor thematic role. They face issues with deriving the surface word order where for precedes its DP object; Jim Wood (personal communication) indicated there might be preposition incorporation of some sort to derive the surface order.

  9. 9.

    An additional minor difference from Pylkkänen’s analysis is the addition of an event argument to the To-the-Possession-of relation, similar to Basilico (2008) and Bruening (2010). Note that Pylkkänen’s original To-the-Possession-of relation does not contain the event argument of the verb. Larson (2010) shows that since the indirect object argument denoting the possessor is not related to the event denoted by the verb, Pylkkänen’s analysis for sentences like (51) predicts that it should be felicitous in a scenario in which Lucy bought Finnegan’s Wake and another person, say John, brought it to Tom. This is clearly not a possible reading of (51), and the addition of an event argument to the To-the-Possession-of relation precludes it.

  10. 10.

    This, of course, is not a necessary assumption. One could assume that Appl is a flavor of v since it does introduce the Theme role, specifying that the individual argument is a Theme of the event denoted by the verb. If so, then the root is categorized as verbal in the Distributed Morphology sense, and it can simply combine directly with the argument interpreted as the Possessor in the possession relation introduced by Appl. This would then bring the syntax of these constructions closer to Basilico’s (2008). Nothing crucial hinges on this choice; since we merely aim to show that our analysis proposed here can be extended to the analysis of double object constructions, we will not attempt to differentiate between these two choices.

  11. 11.

    Alternatively, the second v head may introduce a Benefactor role in English, as per Pylkkänen’s (2008) analysis of high applicatives. This v head would then combine with the root and Appl constituent via Predicate Modification, endowing the DP argument introduced in its specifier with both a Possessor and Benefactor role, capturing the dual interpretations the indirect object receives in an English benefactive double object construction as observed by Basilico (2008) and Wood and Marantz (2017). Again, as far as we can tell, the choice here does not affect the core proposal and analysis in any crucial way.

  12. 12.

    This is in contrast to approaches like Hale and Keyser (1993) and Folli and Harley (2005), who treat creation verbs as involving a DP that is the complement to an agentive flavor of v they label v\(_{\textsc{do}}\).

  13. 13.

    Folli and Harley (2005) in fact assume that agent and causer arguments are introduced directly by v\(_{\textsc{do}}\) and v\(_{\textsc{cause}}\), in which case such an analysis might work. However, later work like Pylkkänen (2008) and Harley (2013) provide extensive arguments within Distributed Morphology that Voice and v should be separate, with Voice being the exclusive locus of external argument introduction.

  14. 14.

    More generally, Rappaport-Hovav and Levin (1998) and Levin (1999) term the class of verbs that allow object drop non-core transitive verbs, where the optional object is not an argument of a lower sub-event. Because of their Argument Per Subevent Condition requiring that each sub-event in an event template expresses an argument, non-core transitive verbs, which are mono-eventive and already have an agent argument expressed, allow their object arguments to be dropped.

  15. 15.

    It is possible as well to dance without performing a particular conventionalized dance, in which case one might treat the events picked out by dance as falling into the same class as those picked out by kick. In this case, what might appear to be an existentially bound implicit argument is in fact simply the existentially closed event argument of dance.

  16. 16.

    For an overview of other empirical arguments against explicitly representing existentially interpreted implicit arguments in the syntax and semantics, see Williams (2015).

  17. 17.

    Not all of these authors adopt Distributed Morphology, so in our representations illustrating the issue at hand we aim for neutrality by writing \(\sqrt{\textsc{Root}}\)/V for the object corresponding to the idiosyncratic verbal element at the base of the tree.

  18. 18.

    An anonymous reviewer points out that non-animate causers can also escape again’s presupposition like agents, producing causerless presuppositions.

    1. (1)

      Context: In 1710, an earthquake destroyed Lisbon. Today...

      A tsunami destroyed Lisbon again.

    This suggests that it is more appropriate to make reference to thematic roles broader than Agent to include Causer. The Internal Argument Generalization can hence make reference to a broader class of arguments, which one might call Initiator in Ramchand’s (2008) terms, and that Initiator arguments may escape again’s presupposition, rather than Agent arguments specifically. Since we have focused on agent arguments throughout the paper, we will maintain our reference to agents and agentless presuppositions. This, however, does not amount to us claiming that agent and causer arguments should always be given the same thematic role label and exhibit the same behavior grammatically, since we restrict our discussion to subjectless presuppositions only. As an anonymous reviewer points out, there are clear arguments in the literature for differentiating between agents and causers in terms of their general behavior and should hence be treated separately (e.g. Alexiadou et al. 2015; Martin 2020).


  1. Alexiadou, Aretmis, Elena Anagnostopoulou, and Florian Schäfer. 2015. External arguments in transitivity alternations: A layering approach. Oxford: Oxford University Press.

    Book  Google Scholar 

  2. Alexiadou, Aretmis, Florian Schäfer, and Giorgos Spathas. 2014. Delimiting voice in Germanic: On object drop and naturally reflexive verbs. In North East Linguistic Society (NELS) 44, eds. Jyoti Iyer and Leland Kusmer, 1–14. Amherst: GLSA.

    Google Scholar 

  3. Bale, Alan. 2007. Quantifiers and verb phrases: An exploration of propositional complexity. Natural Language & Linguistic Theory 25: 447–483.

    Article  Google Scholar 

  4. Basilico, David. 2008. Particle verbs and benefactive double objects in English: High and low attachment. Natural Language & Linguistic Theory 26: 731–773.

    Article  Google Scholar 

  5. Beavers, John, and Andrew Koontz-Garboden. 2020. The roots of verbal meaning. Oxford: Oxford University Press.

    Book  Google Scholar 

  6. Beck, Sigrid, and Kyle Johnson. 2004. Double objects again. Linguistic Inquiry 35(1): 97–123.

    Article  Google Scholar 

  7. Borer, Hagit. 2003. Exo-skeletal vs. endo-skeletal explanations: Syntactic projections and the lexicon. In Explanation in linguistic theory, eds. John Moore and Maria Polinsky, 31–67. Stanford: CSLI Publications.

    Google Scholar 

  8. Borer, Hagit. 2005. Structuring sense: In name only, Vol. 1. Oxford: Oxford University Press.

    Book  Google Scholar 

  9. Bresnan, Joan. 1978. A realistic transformational grammar. In Linguistic theory and psychological reality, eds. Morris Halle, Joan Bresnan, and George Miller. Cambridge: MIT Press.

    Google Scholar 

  10. Bruening, Benjamin. 2010. Ditransitive asymmetries and a theory of idiom formation. Linguistic Inquiry 41(4): 519–562.

    Article  Google Scholar 

  11. Bruening, Benjamin. 2020. Implicit arguments in English double object constructions. Natural Language & Linguistic Theory.

    Article  Google Scholar 

  12. Carlson, Gregory. 1998. Thematic roles and the individuation of events. In Events and grammar, ed. Susan Rothstein, 35–51. Dordrecht: Kluwer.

    Chapter  Google Scholar 

  13. Carlson, Gregory N. 1984. Thematic roles and their role in semantic interpretation. Linguistics 22: 259–279.

    Article  Google Scholar 

  14. Castañeda, Hector-Neri. 1966. Comments. In The logic of decision and action, eds. Nicholas Rescher and Alan Ross Anderson, 104–112. Pittsburgh: University of Pittsburgh Press.

    Google Scholar 

  15. Champollion, Lucas. 2010. Parts of a whole: Distributivity as a bridge between aspect and measurement. PhD diss., University of Pennsylvania.

  16. Davidson, Donald. 1967. The logical form of action sentences. In The logic of decision and action, ed. N. Rescher, 81–94. Pittsburgh: University of Pittsburgh Press.

    Google Scholar 

  17. Dowty, David. 1979. Word meaning and montague grammar. Dordrecht: D. Reidel Publishing.

    Book  Google Scholar 

  18. Dowty, David. 1982. Grammatical relations and montague grammar. In The nature of syntactic representation, 79–130. Dordrecht: Springer.

    Chapter  Google Scholar 

  19. Folli, Raffaella, and Heidi Harley. 2005. Flavors of v. In Aspectual inquiries, eds. Paula Marie Kempchinsky and Roumyana Slabakova, 99–120. Dordrecht: Springer.

    Google Scholar 

  20. Hale, Ken, and Samuel Jay Keyser. 2002. Prolegomenon to a theory of argument structure. Cambridge: MIT Press.

    Book  Google Scholar 

  21. Hale, Kenneth, and Samuel Jay Keyser. 1993. On argument structure and the lexical expression of syntactic relations. In The view from building 20: Essays in linguistics in honor of Sylvain Bromberger, eds. Jay Keyser Samuel and Kenneth Hale, 53–109. Cambridge: MIT Press.

    Google Scholar 

  22. Halle, Morris, and Alec Marantz. 1993. Distributed Morphology and the pieces of inflection. In The view from building 20: Essays in honor of Sylvain Bromberger, eds. Ken Hale and Samuel Jay Keyser, 111–176. Cambridge: MIT Press.

    Google Scholar 

  23. Harley, Heidi. 2002. Possession and the double object construction. Linguistic Variation Yearbook 2: 31–70.

    Article  Google Scholar 

  24. Harley, Heidi. 2013. External arguments and the Mirror Principle: On the distinctness of voice and v. Lingua 125 (34-57).

  25. Harley, Heidi. 2014. On the identity of roots. Theoretical Linguistics 40(3/4): 225–276.

    Google Scholar 

  26. Harley, Heidi, and Megan Schildmier Stone. 2013. The ‘No Agent Idioms’ hypothesis. In Syntax and its limits, eds. Raffaella Folli, Christina Sevdali, and Robert Truswell, 251–273. Oxford: Oxford University Press.

    Chapter  Google Scholar 

  27. Horvath, Julia, and Tal Siloni. 2003. Against the little-v hypothesis. In Proceedings of the Israeli Association for Theoretical Linguistics 19, ed. Yehuda N. Falk. Israel: Ben Gurion University of the Negev.

    Google Scholar 

  28. Kratzer, Angelika. 1996. Severing the external argument from its verbs. In Phrase structure and the lexicon, eds. Laurie Ann Zaring and Johan Rooryck, 109–137. Dordrecht: Kluwer.

    Chapter  Google Scholar 

  29. Kratzer, Angelika. 2005. Building resultatives. In Event arguments: Foundations and applications, eds. Claudia Maienborn and Angelika Wöllstein, 177–212. Tübingen: Niemayer.

    Chapter  Google Scholar 

  30. Landman, Fred. 2000. Events and plurality. Dordrecht: Kluwer.

    Book  Google Scholar 

  31. Larson, Richard. 2010. On Pylkkänen’s semantics for low applicatives. Linguistic Inquiry 41: 701–704.

    Article  Google Scholar 

  32. Levin, Beth. 1993. English verb classes and alternations: A preliminary investigation. Chicago: University of Chicago Press.

    Google Scholar 

  33. Levin, Beth. 1999. Objecthood: An event structure perspective. In Papers from the 35th regional meeting of the Chicago Linguistic Society, 1: The main session, eds. Sabirna J. Billings, John P. Boyle, and Aaron M. Griffith, 223–247. Chicago: Chicago Linguistic Society.

    Google Scholar 

  34. Lohndal, Terje. 2012. Without specifiers: Phrase structure and events. PhD diss., University of Maryland, College Park.

  35. Marantz, Alec. 1984. On the nature of grammatical relations. Cambridge: MIT Press.

    Google Scholar 

  36. Marantz, Alec. 2005. Objects out of the lexicon: Objects as events. Talk given at the University of Vienna.

  37. Marantz, Alec. 2007. Restitutive re-, and the first phase syntax/semantics of the VP. Talk given at the University of Maryland.

  38. Martin, Fabienne. 2020. Aspectual differences between agentive and non-agentive uses of causative predicates. In Perspectives on causation: Jerusalem studies in philosophy and history of science, eds. Elitzur A. Bar-Asher Siegal and Nora Boneh, 257–294. Cham: Springer.

    Chapter  Google Scholar 

  39. Merchant, Jason. 2019. Roots don’t select, categorial heads do: Lexical selection of PPs may vary by category. The Linguistic Review 36(3): 325–341.

    Article  Google Scholar 

  40. Nunberg, Geoffrey, Ivan A. Sag, and Thomas Wasow. 1994. Idioms. Language 70(3): 491–538.

    Article  Google Scholar 

  41. Pylkkänen, Liina. 2008. Introducing arguments. Cambridge: MIT Press.

    Book  Google Scholar 

  42. Ramchand, Gillian. 2008. Verb meaning and the lexicon: A first phase syntax. Cambridge: Cambridge University Press.

    Book  Google Scholar 

  43. Rappaport-Hovav, Malka, and Beth Levin. 1998. Building verb meanings. In The projection of arguments: Lexical and compositional factors, eds. Miriam Butt and Wilhem Geuter, 97–134. Stanford: CSLI Publications.

    Google Scholar 

  44. Rappaport-Hovav, Malka, and Beth Levin. 2008. The English dative alternation: A case for verb sensitivity. Journal of Linguistics 44(1): 129–167.

    Article  Google Scholar 

  45. Rappaport-Hovav, Malka, and Beth Levin. 2010. Reflections on manner/result complementarity. In Syntax, lexical semantics, and event structure, eds. Malka Rappaport Hovav, Edit Doron, and Ivy Sichel, 21–38. Oxford: Oxford University Press.

    Chapter  Google Scholar 

  46. Schein, Barry. 1993. Plurals and events. Cambridge: MIT Press.

    Google Scholar 

  47. Spathas, Giorgos, and Dimitris Michelioudakis. 2020. States in the decomposition of verbal predicates: Evidence from additive operators. Natural Language & Linguistic Theory.

  48. von Stechow, Arnim. 1996. The different readings of wieder ‘again’: A structural account. Journal of Semantics 13(2): 87–138.

    Article  Google Scholar 

  49. Wechsler, Stephen. 2005. What is right and wrong about little v. In Grammar and beyond—Essays in honour of Lars Hellan, eds. Mila Dimitrova-Vulchanova and Tor A. Åfarli, 179–195. Oslo: Novus Press.

    Google Scholar 

  50. Williams, Alexander. 2015. Arguments in syntax and semantics. Cambridge: Cambridge University Press.

    Book  Google Scholar 

  51. Wood, Jim, and Alec Marantz. 2017. The interpretation of external arguments. In The verbal domain, eds. Roberta D’Alessandro, Irene Franco, and Ángel J. Gallego, 255–278. Oxford: Oxford University Press.

    Google Scholar 

Download references


We would like to thank Josep Ausensi, Kenyon Branan, Noam Chomsky, Michael Yoshitaka Erlewine, Heidi Harley, Robert Henderson, Jian Gang Ngui, and Zheng Shen for discussion and comments on the work reported in this article. Additional thanks go to audiences at WCCFL 38 in Vancouver, BC, and to three anonymous NLLT reviewers, whose feedback greatly improved and expanded the original empirical domain of this paper.

Author information



Corresponding author

Correspondence to Ryan Walter Smith.

Ethics declarations

Conflict of Interest

The authors declare that they have no conflict of interest.

Additional information

Publisher’s Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Smith, R.W., Yu, J. Agentless presuppositions and the semantics of verbal roots. Nat Lang Linguist Theory (2021).

Download citation


  • Roots
  • Argument structure
  • again
  • Agentless presuppositions