The syntax of Why-Stripping


This study investigates a clausal ellipsis construction involving the adverbial wh-phrase why and a non-wh-remnant, hereafter Why-Stripping. We show that Why-Stripping exhibits movement properties such as connectivity effects in the same way as Sluicing (with argument wh-phrases) and Stripping, and claim that Why-Stripping involves movement of the focused phrase (e.g. Mary) followed by clausal ellipsis. Furthermore, based on the fact that Why-Stripping does not show strict locality restrictions, unlike Sluicing with why, we claim that why in Why-Stripping is base-generated in the Spec_CP position while Sluicing with why involves adjunct wh-movement. According to this view, there are two types of why: one that moves and the other that does not move. It is shown that the latter induces focus association and participates in Why-Stripping. Thus, the investigation of Why-Stripping contributes to revealing the nature of the syntax of why itself.

This is a preview of subscription content, log in to check access.


  1. 1.

    See Merchant (2006) and Jones (2004) for details about the syntax of not in a stripping context.

  2. 2.

    In Nevins (2008), examples like (8) are reported to be acceptable in English. Among the 10 speakers we consulted, however, only 6 speakers accepted examples like (8). In this study, we follow Nevins and assume that Sluice Stripping as in (8) is possible. All judgments on Sluice Stripping are given by the 6 speakers who accepted this construction without any problem. Ultimately, a further refinement of ellipsis structures involving a wh- and a non-wh-remnant beyond Why and Sluice Stripping might be in order. The reader may see Grebenyova (2006), Van Craenenbroeck and Lipták (2006), Ince (2007), and Ortega-Santos et al. (2014) for discussion.

  3. 3.

    Note, however, that Merchant (2004:716–732) argues that being pragmatically controlled is not necessarily a hallmark for Deep Anaphora.

  4. 4.

    Case Connectivity effects are not seen in English. As shown in (i), the remnants of Why-Stripping and Stripping can bear Accusative Case even though the correlates have Nominative Case. If these constructions involve clausal ellipsis, such a Case Mismatch should not be allowed.

    1. (i)

    Still, in many environments beyond Why-Stripping and Stripping, DPs receive so-called Default Case, e.g., when a pronoun appears in a coordination structure functioning as a subject as in (ii) (see Schütze 2001 for details):

    1. (ii)

    Thus, we conclude that the apparent lack of the Case connectivity in English is due to an independent factor, namely, the availability of Default Case in this language (see Merchant 2004: 700–704 for discussion of other alternatives compatible with the ellipsis analysis).

  5. 5.

    The following example from German suggests that the ellipsis derivation is possible in the matrix Why-Stripping without a clear linguistic antecedent.

    1. (i)

    The native speaker of German we interviewed told us that if it is clear that the secretary is praised, the accusative case should be used and the dative case is quite odd. If this judgment holds for other speakers as well, it may weaken our speculation that matrix Why-Stripping without an antecedent is derived via Deep Anaphora (see Sect. 1.2.5 for discussion) and we may eventually claim that all examples of Why-Stripping involve ellipsis.

  6. 6.

    Under Sluicing, when the verb in the matrix clause is compatible with multiple prepositions, it seems that the preposition of the remnant can differ from that of the correlate as long as the interpretation of the matrix clause and the sluiced clause remains constant. For example, a verb like talk can take multiple prepositions, and the type of PP in the stripped clause can be different from the PP in the antecedent clause. However, a semantically empty preposition like of still cannot appear in the structure.

    1. (i)

    In this type of example, the PPs are “connected” to the matrix verb in terms of selection; that is to say, these prepositions are those that are selected by the matrix verb. Therefore, this observation is compatible with an ellipsis analysis.

    Furthermore, this type of example can be interpreted as evidence that a strict parallelism must hold between the material that falls into the scope of the ellipsis and the antecedent constituent, but the element that escaped the scope of ellipsis does not need to adhere to such a strict parallelism. See Merchant’s (2007, 2008) closely related discussion.

  7. 7.

    Marcel den Dikken (p.c.) pointed out to us that the Dutch counterpart of (33) is acceptable; however, when the negation is present, it is unacceptable.

    1. (i)

    Furthermore, the preposition is obligatory in the embedded context. This difference between embedded cases and matrix cases corresponds to the deep/surface anaphora distinction in the matrix and embedded context. Turning to the German data in (33), our informant told us that the P-omission is less acceptable, but it is not clear how bad the P-omission example is. It is possible that the matrix P-omission is better than the embedded P-omission. If this is the case, the German data also support the position that the matrix case may have deep anaphora derivation, but the embedded case is surface anaphora.

  8. 8.

    There are various exceptions and counterexamples against this generalization reported in the literature, e.g., Brazilian Portuguese (Almeida and Yoshida 2007) and Serbo-Croatian (Stjepanovic 2008), among others. It has been claimed that P-drop under clausal ellipsis constructions is associated with an underlying cleft, which means that the syntactic form of the ellipsis site is different from the antecedent clause (e.g., Rodrigues et. al. 2009, a.o.). If true, the apparent counterexamples to the P-stranding generalization would be explained away. See Merchant (2001), van Craenenbroeck (2010a) and Martín González (2010) for relevant discussions. Be that as it may, we would like to note that the same exception seems to hold true for the P-omission in Why-Stripping. An example from Brazilian Portuguese in (i) shows that P-omission in Why-Stripping is acceptable. Thus, such examples further support the parallelism between Sluicing and Why-Stripping.

    1. (i)

    A native speaker of Brazilian Portuguese that we consulted pointed out that P-less Why-Stripping with a name (e.g., a Maria) is slightly less acceptable than a pronoun (e.g., ela), but the acceptability of both cases does not differ compared to an example with a preposition.

  9. 9.

    See, however, Wilder (1997) for an argument that non-constituents can be target of ellipsis.

  10. 10.

    Using every as the antecedent for the pronoun (e.g. (48)) could run into the problem of E-type reading, and it could obscure the long-distance interpretation. This potential problem does not arise with the quantifier no in (49). The judgment is somewhat delicate but the native speakers who we interviewed confirmed that the bound variable reading in these examples is possible (10 out of 10 speakers).

    Another possible test for the long-distance reading is to employ the each the other dependency, where the other is licensed only if there is each as shown in the contrast between (ia) and (ib, ic). ((ia) is from Lasnik 2005a:264, while (ib) and (ic) are modified from Lasnik 2005a, 2005b; see also Fiengo and Lasnik 1973.)

    1. (i)

    Making use of the each the other dependency, we can come up with the following example.

    1. (ii)

    For both (i) and (ii), judgments are somewhat delicate. (6 out of 10 speakers provided the judgments reported above while 4 speakers did not have any clear judgments, even for simpler cases like (i).) However, as long as examples in (ii) are as acceptable as in (i), we can conclude that long-distance Why-Stripping is possible.

  11. 11.

    Collins (1991) also claims that how come, unlike why, does not allow tenseless clauses.

    1. (i)

    Under our analysis where tenseless clauses with why are instances of Why-Stripping, this fact is unexpected because how come licenses How come-Stripping; (i) could well be derived as How come-Stripping with a VP remnant.

    However, some speakers (3 out of 10 native speakers who accept How come-Stripping in the first place; see footnote 2) accept tenseless clauses with how come in some contexts. For example, they accept both examples in (ii):

    1. (ii)

    An informal Google search gives more than 300 hits of the instance of “how come VP”. (We counted the number of “how come VP” examples out of simple search of the string “how come”.) Specifically, tenseless clauses using keep as a verb (e.g., How come keep drinking it?) seem to be used productively. Although the issue of why tenseless clauses with how come are less productive than those with why remains as an open question, the data is not incompatible with our analysis.

  12. 12.

    These tenseless-clause remnants in Why-Stripping often occur without a linguistic antecedent. For discussion of Why-Stripping without a linguistic antecedent, see Sect. 1.2.

  13. 13.

    There is no agreement in the previous literature as to whether long-distance Stripping is possible in English. Lobeck (1995:27), for example, claims that it is ungrammatical in contrast to Depiante (2000). We consulted with 10 native speakers and they all accepted examples of long-distance Stripping like (62b). Therefore, we assume that long-distance Stripping is generally possible.

  14. 14.

    We consulted 10 native speakers of English, and all of them found these examples acceptable.

  15. 15.

    However, it does not seem to be the case that Stripping is sensitive to any type of island. Left-Branch island violations (Kennedy and Merchant 2000; Merchant 2001; Ross 1967, 1969, among many others) are not observed under Stripping, as shown in (ia). In this respect, Stripping patterns with Sluicing and Why-Stripping, (ib) and (ic), respectively.

    1. (i)

    Similarly, comp-trace effects (Chomsky 1986; Lasnik and Saito 1992; Merchant 2001; Perlmutter 1971, among many others) are not seen in Stripping (iia), Sluicing (iib), or Why-Stripping (iic).

    1. (ii)

    These observations may point to the generalization that Stripping can violate the so-called PF-islands (such as Left Branch Condition (LBC) and that-trace effects), but it is constrained by propositional islands (such as Complex NPs and Adjunct islands) in Merchant’s (2001) classification. Sluicing and Why-Stripping, on the other hand, are not constrained by any type of island. We have a speculative explanation for the island-sensitivity of Stripping to (propositional) islands in Sect. 3.3. At this point, we do not have an account of why PF-islands are not observed even under Stripping. Still, as pointed out by an anonymous reviewer, the LBC facts are explained under Kennedy and Merchant’s (2000) analysis of LBC. They claim that LBC is violated when a language lacks the appropriate spell-out for the D-head with a specific feature (e.g. [+wh]), causing a PF violation. Thus, if the relevant head is elided under any types of ellipsis, including Stripping, the lack of LBC can be expected.

  16. 16.

    See Nakao and Yoshida (2007) for an account of why there is no island-repair effect with Why-Sluicing.

  17. 17.

    Closely related approaches in terms of rightward movement exist for Gapping (see Jayaseelan 1990; see also Yoshida et al. 2012) and for Multiple Sluicing (Lasnik 2014). See Ortega-Santos et al. (2014) for further discussion of the asymmetries between Why and Sluice Stripping.

  18. 18.

    Furthermore, natto in (76) is contrasted to the other ‘given’ components of the event; that is to say, it is a question about why natto is being eaten, as opposed to, say, why John is the one eating it. Thus, even if the follow-up tag is absent, we take the remnant to be focused.

  19. 19.

    See, however, Horvath (2007), who argues against the position that the formal focus feature drives movement, based on the data in Hungarian.

  20. 20.

    The fact that only why and how come are compatible with Stripping, is due to their ability to be base-generated in the highest Spec_CP and thus do not induce Relativized Minimality (RM) effects (Rizzi 1990). As why is not moved to the highest Spec_CP position, a focus element does not “intervene” between why and its trace. Other wh-phrases, on the other hand, undergo movement, and thus a focus-fronted element should induce RM effects. This prediction can be tested in non-ellipsis contexts using Neg-Inversion, which involves focus fronting (Culicover and Winkler 2008). The acceptability contrast in following examples suggests that this is indeed the case. The why example is much more acceptable compared to the non-why examples.

    1. (i)

    We are grateful to Marcel den Dikken for directing our attention to this important point. See Sect. 3.1.2 for related discussion on other languages.

  21. 21.

    Note that D-linked wh-phrase can be followed by a focused phrase in some languages. We return to this point in Sect. 3.1.4.

  22. 22.

    Rizzi (2001:294) shows that come mai ‘how come’ in Italian follows this same pattern, as expected under the current approach. In the subsequent discussion on Romance languages, we will concentrate on the patterns with why for the sake of exposition, putting aside the data of the Romance counterpart of how come.

  23. 23.

    Stepanov and Tsai (2008) assume that focus operates via existential closure, unlike our proposal that focus is licensed via movement (see (101) below). They argue that Why-questions trigger different answers, (e.g. (79), (80)), because the focus operator is inside the scope of why as shown in (i), contrary to the position of other wh-phrases such as when in (ii) (e.g. (81), (82)).

    1. (i)
    1. (ii)

    Thus, their argument also reaches the same conclusion as ours: why in this type of example is base-generated in the CP domain.

  24. 24.

    Rizzi as well as Bromberger note that when why is “construed” in an embedded clause, it moves as other wh-phrases do. Based on the island facts in (84), we argue against this position. (See footnote 34 for further discussion.)

  25. 25.

    Rizzi points out that why does not trigger the Subject-Aux inversion in Italian (his data; see also Alboiu 2002; Suñer 1994 for Romanian and Spanish, respectively; see Uriagereka 1999 for Basque and Hungarian).

    1. (i)

    He takes this observation as further evidence in favor of the base-generation of why in IntP. While we accept Rizzi’s view to be essentially correct, there is crosslinguistic variation regarding Subject-Aux inversion in Why-interrogatives. Specifically, inversion is obligatory in this context in English. As an anonymous NLLT reviewer notes, Shlonsky and Soare (2012) argue that why is externally merged in the left-periphery and it undergoes short-distance movement to IntP. If correct, we could suggest that this short-distance movement of why triggers inversion in English, though for all other purposes its height (both in its external merge and the final landing site) is compatible with the observations concerning focus fronting and intervention effects caused by negation (and the locality facts to be discussed below). We leave this issue for future research.

  26. 26.

    Rizzi notes that the incompatibility between wh-phrases other than why and focused phrases in Italian is a matrix phenomenon, whereas in an embedded clause, wh-phrases can co-occur with focused phrases; i.e., in an embedded context, there can be an ‘extra’ position for the focused phrase available as shown in (i).

    1. (i)

    In contrast, such a word order is not licensed in English as (ii) shows.

    1. (ii)

    Irrespective of the treatment that the data in (i) should receive, it is worth noting that the inverted word order in which the wh-element precedes the Focused XP as in Why-Stripping is judged as fairly deviant in Italian:

    1. (iii)
  27. 27.

    This does not exclude the possibility that why is base-generated somewhere inside the matrix IP, while the focus-associated phrase is base-generated inside the embedded clause as in (96), as one of the anonymous NLLT reviewers points out. Our point, however, is that the focus association effect cannot be analyzed as an instance of base-generation of why and the focused phrase in the same clause followed by (long-distance) movement of why. Furthermore, we put forward the base-generation analysis of why based on the arguments in Rizzi (2001) and other works we cite here.

  28. 28.

    We do not argue for the position that why in the non-focus construction is never base-generated. (Example (94) provides evidence for the base-generation of why in a context without focus association.) We merely argue that why in focus constructions (Why-Focus constructions and Why-Stripping) is base-generated in the higher Spec_CP. Unfortunately, at this point, we cannot tell when why in non-focus construction is base-generated in the higher Spec_CP.

  29. 29.

    As is expected, wh-questions other than why-questions do not show differences in the scope interpretation regardless of whether a DP is focused, as the following examples show.

    1. (i)

    Both in (ia) and (ib), the adjunct wh-phrase where or when can take either the scope higher or lower than everyone, and thus, the pair-list answer is allowed in both cases.

  30. 30.

    An anonymous NLLT reviewer points out that if a long-distance construed why undergoes movement, as Rizzi and Bromberger suggest, we expect the scope ambiguity between why and the matrix subject quantifiers in the following example.

    1. (i)

    However, 5 native speakers we interviewed (all are linguists) did not find the pair-list reading and they can only find the interpretation in which why takes wider scope than everyone. This judgment further suggests that why is base-generated in the high position even when it is long-distance construed.

  31. 31.

    Unlike how come (Collins 1991), matrix why usually induces Subject-Aux Inversion (SAI). Nevertheless, the auxiliary in Why-Stripping does not escape the elided TP. This is compatible with Lasnik’s (2001) analysis where SAI does not occur under ellipsis (such as Sluicing, as shown in (i)), because the defective T, which usually causes SAI, is deleted.

    1. (i)

    This analysis assumes that the necessity of SAI is calculated at PF, where deletion of TP occurs. On the other hand, if SAI happens in overt syntax rather than PF, as Hartmann (2011) claims, a different explanation may be needed. We thank an anonymous reviewer for pointing out this latter possibility.

  32. 32.

    See discussion on the example (124) in Sect. 3.2 for the actual mechanism of how a bare V or P appears in the focus position via movement.

  33. 33.

    In Sect. 3.2, we introduce how exactly V and P survive the ellipsis in relation to the discussion of the obligatory ellipsis requirement.

  34. 34.

    Marcel den Dikken (p.c.) notes that the effect in (108) is not particularly strong possibly because there might be a tendency for speakers to interpret the stressed direct object as information focus as opposed to contrastive focus due to its final position. According to him, when this factor is controlled for, the effect is more salient:

    1. (i)
  35. 35.

    In such a special context, the focus association effect of why is not obligatory. Given that the non-focus-associated why does not have to be base-generated in [Spec_CP1] (as the low reading of why in (97a) indicates), the prediction should be that why with a focused element and no focus association could have a low reading under such a special context, in contrast to (98). We would like to see if this prediction holds.

  36. 36.

    With regard to the interaction between syntax, focus, and phonetics within the standardly accepted T-model, we follow Irurtzun (2007)’s proposal that the F-Structure of a sentence is built up derivationally from the elements that are assigned a [+F] formal feature as they enter the numeration. Within this view, narrow syntax creates a well-specified F-Structure and the interface components can “read” it and apply some operations on it.

  37. 37.

    We thank the anonymous NLLT reviewer who mentioned this point.

  38. 38.

    Similarly, an anonymous reviewer gives the following contrast to show that a focused phrase associated with only cannot be elided.

    1. (i)

    On the other hand, Beaver and Clark (2008:180) report an example where a focused phrase associated with only is elided.

    1. (ii)

    An example like (ii) suggests that it is not always the case that the focus-associated element cannot be elided. Note, importantly, that the non-ellipsis counterpart of (iiB) as in (iii), nutrapup may not receive focal stress.

    1. (iii)

    This suggests that when the focused element is repeated like in (iii), it may not receive focal stress. If this is the case, it is plausible to assume that the focus element that is elided in the ellipsis example in (ii) is a non-stressed focus element. As the element is not stressed, it does not need to escape from the ellipsis. This leads to the possibility that the driving force behind the movement of the focused phrase in Why-Stripping is phonetic effect of focus, i.e., the focal stress, rather than semantic focus.

  39. 39.

    We thank an anonymous reviewer for pointing out that this type of example may be derived from the “deaccented” counterpart of the antecedent clause: ‘John ate NATTO, but I don’t understand why he ate natto/it.’

  40. 40.

    This view is able to explain the existence of the why not construction (Merchant 2006), where not functions as a non-wh-remnant. Specifically, as pointed out by an anonymous reviewer, why can also be combined with polarity markers under clausal ellipsis.

    1. (i)

    According to Merchant (2006), in these structures why is a phrasal adverb and not, which is phrasal as well, adjoins to why. Note that why not also licenses Why-Stripping:

    1. (ii)

    This is expected as both why and not participate productively in ellipsis processes (e.g., see (1a) for a case of Stripping with not). Though a detailed study of the structure in (ii) goes beyond the goal of this paper, we hypothesize that this is a case of Why-Stripping with one additional operation, the adjunction of not to why.

  41. 41.

    Building on this observation, Merchant suggests that Fragment Answers and Stripping have the same type of derivation. Still, Stainton (2006:138) shows that if Fragment Answers function as the answer for a sluiced wh-question, they can violate islands. In this context, Merchant (2010) points out that Stainton’s examples may have different derivation from the Fragment Answers that Merchant (2004) investigated. Despite this controversy, in this study we basically assume that Stripping and Fragment Answers have the same type of derivations and are subject to the same restrictions.

  42. 42.

    Nakao (2009) points out several drawbacks of this line of analysis. For example, she argues that the assumption that only intermediate copies (but not the pronounced copy) get a * is unnatural, and claims that the difference in island sensitivity between Sluicing and Fragment Answers/Stripping comes from the timing of movement (i.e. Stripping causes island-violation in the PF component after spell-out, while the island-repair by ellipsis is calculated at the timing of Spell-out). However, because focus movement in Why-Stripping also only occurs in an ellipsis environment, its island-insensitivity would be unexpected under this approach without a further argument to dissociate Fragment Answers/Stripping and Why-Stripping.

  43. 43.

    As an anonymous reviewer notes, there are a number of alternatives in the literature that derive the presence/absence of island sensitivity in a wide array of contexts from parallelism requirements on ellipsis, (Fox and Lasnik 2003; Griffiths and Lipták 2014, to appear; Park and Park 2011; Saab 2010). Such proposals may provide for an alternative treatment of the data at hand, where the postulation of two FPs would be unnecessary in English. We leave this line of analysis for future research.


  1. Alboiu, Gabriela. 2002. The features of movement in Romanian. Bucharest: University of Bucharest Press.

    Google Scholar 

  2. Almeida, Diogo, and Masaya Yoshida. 2007. A problem for the preposition stranding generalization. Linguistic Inquiry 38: 349–362.

    Article  Google Scholar 

  3. Baltin, Mark. 1978. Toward a theory of movement rules. PhD diss., Massachusetts Institute of Technology.

  4. Beaver, David I., and Brady Z. Clark. 2008. Sense and sensitivity: how focus determines meaning. London: Wiley-Blackwell.

    Google Scholar 

  5. Birner, Betty, and Gregory Ward. 1998. Information status and non-canonical word order in English. Amsterdam: Benjamins.

    Google Scholar 

  6. Bromberger, Sylvain. 1992. On what we know we don’t know: explanation, theory, linguistics, and how questions shape them. Chicago: The University of Chicago Press.

    Google Scholar 

  7. Brunetti, L. 2003. Information focus movement in Italian and contextual constraints on Ellipsis. In Proceedings of 22nd West Conference on Formal Linguistics, eds. M. Tsujimura and G. Garding, 95–108. Somerville: Cascadilla Press.

    Google Scholar 

  8. Buesa García, Carlos. 2011. El hueco de sujeto en español y algunas consecuencias teóricas y empíricas. In The XVI Congreso Internacional de la Asociación de Lingüística y Filología de América Latina (ALFAL). Alcalá de Henares: The University of Alcalá de Henares.

    Google Scholar 

  9. Chomsky, Noam. 1986. Barriers. Cambridge: MIT Press.

    Google Scholar 

  10. Chomsky, Noam. 1995. The minimalist program. Cambridge: MIT Press.

    Google Scholar 

  11. Chung, Sandra, William A. Ladusaw, and James McCloskey. 1995. Sluicing and logical form. Natural Language Semantics 3: 239–282.

    Article  Google Scholar 

  12. Chung, Sandra. 2005. Sluicing and the lexicon: the point of no return. In Proceedings of the 31st annual meeting of the Berkeley Linguistics Society, eds. Rebecca T. Cover and Yuni Kim, 73–91. Berkeley: Berkeley Linguistics Society.

    Google Scholar 

  13. Collins, Chris. 1991. Why and how come. In More papers on wh-movement: MIT working papers in linguistics 15, eds. Lisa L. S. Cheng and Hamida Demirdache, 31–45. Cambridge: MIT Press.

    Google Scholar 

  14. Costa, João. 2004. Subject positions and interfaces: the case of European Portuguese. New York: de Gruyter.

    Google Scholar 

  15. Costa, João, and Anna Maria Martins. 2011. On focus movement in European Portuguese. Probus 23: 217–245.

    Article  Google Scholar 

  16. Craenenbroeck, Jeroen van. 2009. Simple and complex wh-phrases in a split CP. In Vol. 43 of Proceedings from the annual meeting of the Chicago Linguistic Society, eds. Malcolm Elliott, James Kirby, Osamu Sawada, Eleni Staraki, and Suwon Yoon, 155–168. Chicago: Chicago Linguistics Society.

    Google Scholar 

  17. Craenenbroeck, Jeroen van. 2010a. Invisible last resort: a note on clefts as the underlying source for sluicing. Lingua 120: 1714–1726.

    Article  Google Scholar 

  18. Craenenbroeck, Jeroen van. 2010b. The syntax of ellipsis. Oxford: Oxford University Press.

    Google Scholar 

  19. Craenenbroeck, Jeroen van, and Anikó Lipták. 2006. The crosslinguistic syntax of sluicing: evidence from Hungarian relatives. Syntax 9: 248–274.

    Article  Google Scholar 

  20. Culicover, Peter, and Susanne Winkler. 2008. English focus inversion. Journal of Linguistics 44: 625–658.

    Article  Google Scholar 

  21. Depiante, Marcela A. 2000. The syntax of deep and surface anaphora: a study of null complement anaphora and Stripping/bare argument ellipsis. PhD diss., University of Connecticut.

  22. Drummond, Alex, Norbert Hornstein, and Howard Lasnik. 2011. A puzzle about P-stranding and a possible solution. Linguistic Inquiry 41: 689–692.

    Article  Google Scholar 

  23. Erteschik-Shir, Nomi. 2007. Information structure Oxford surveys in syntax and morphology. Oxford: Oxford University Press.

    Google Scholar 

  24. Fiengo, Robert, and Howard Lasnik. 1973. The logical structure of reciprocal sentences in English. Foundations of Language 9: 447–468.

    Google Scholar 

  25. Fiengo, Robert, and Robert May. 1994. Indices and identity. Cambridge: MIT Press.

    Google Scholar 

  26. Fox, Danny. 1999. Reconstruction, binding theory, and the interpretation of chains. Linguistic Inquiry 30: 157–196.

    Article  Google Scholar 

  27. Fox, Danny, and Howard Lasnik. 2003. Successive-cyclic movement and island repair: the difference between sluicing and VP-ellipsis. Linguistic Inquiry 34: 143–154.

    Article  Google Scholar 

  28. Gallego, Ángel J. 2011. Sobre la elipsis. Ms. Madrid: Arco Libros.

  29. Grebenyova, Lydia. 2006. Multiple interrogatives: syntax, semantics and learnability. PhD diss., University of Maryland at College Park.

  30. Griffiths, James, and Anikó Lipták. 2014, to appear. Contrast and island sensitivity in clausal ellipsis. Syntax.

  31. Grosu, Alexander. 1972. The strategic content of island constraints. PhD diss., The Ohio State University.

  32. Grosu, Alexander. 1973. On the status of the so-called right roof constraint. Language 49: 294–311.

    Article  Google Scholar 

  33. Hankamer, Jorge. 1979. Deletion in coordinate structures, outstanding dissertations in linguistics. New York: Garland Publishing.

    Google Scholar 

  34. Hankamer, Jorge, and Ivan A. Sag. 1976. Deep and surface anaphora. Linguistic Inquiry 7: 391–426.

    Google Scholar 

  35. Hartmann, Jeremy. 2011. The semantic uniformity of traces: evidence from ellipsis parallelism. Linguistic Inquiry 43: 367–388.

    Article  Google Scholar 

  36. Hempel, Carl. 1965. Aspects of scientific explanation. New York: Free Press.

    Google Scholar 

  37. Heycock, Caroline. 1995. Asymmetries in reconstruction. Linguistic Inquiry 26: 547–570.

    Google Scholar 

  38. Horvath, Julia. 2007. Separating “focus movement” from focus. In Phrasal and clausal architecture: syntactic derivation and interpretation: in honor of Joseph E. Emonds, eds. Simin Karimi, Vida Samiian, and Wendy K. Wilkins, 108–145. Amsterdam: Benjamins.

    Google Scholar 

  39. Huang, Cheng-Teh James. 1982. Logical relations in Chinese and the theory of grammar. PhD diss., Massachusetts Institute of Technology.

  40. Huang, Cheng-Teh James. 1993. Reconstruction and the structure of VP: some theoretical consequences. Linguistic Inquiry 24: 103–138.

    Google Scholar 

  41. Ince, Atakan. 2007. Non-wh phrases in Sluicing in Turkish. In University of Maryland working papers in linguistics, eds. Akira Omaki, Iván Ortega-Santos, Jon Sprouse, and Matthew Wagers, 1–22. College Park: Department of Linguistics, University of Maryland.

    Google Scholar 

  42. Irurtzun, Ariz. 2007. The grammar of focus at the interfaces. PhD diss., University of Basque Country.

  43. Jayaseelan, Karattuparambil. A. 1990. Incomplete VP deletion and gapping. Linguistic Analysis 20: 64–81.

    Google Scholar 

  44. Jones, Lindsay. 2004. Negation and movement in stripping constructions. Ms., University of California, Santa Cruz.

  45. Kandybowicz, Jason. 2011. Krachi wh-in-situ and the grammar of ‘why’. In Vol. 42 of Annual Conference on African Linguistics, June, 2011. College Park: University of Maryland.

    Google Scholar 

  46. Kawamura, Tomoko. 2007. Some interactions of focus and focus sensitive elements. PhD diss., Stony Brook University.

  47. Kennedy, Christopher, and Jason Merchant. 2000. Attributive comparative deletion. Natural Language and Lignuistic Theory 18: 89–146.

    Article  Google Scholar 

  48. Kim, Jeong-Seok. 1998. Syntactic focus movement and ellipsis: a minimalist approach. PhD diss., University of Connecticut.

  49. Ko, Heejong. 2005. Syntax of why-in-situ: merge into [Spec,CP] in the overt syntax. Natural Language and Linguistic Thoery 23: 867–916.

    Article  Google Scholar 

  50. Krifka, Manfred. 2006. Association with focus. In The architecture of focus, eds. Valéria Molnár and Susanne Winkler, 105–136. Berlin: Mouton de Gruyter.

    Google Scholar 

  51. Lasnik, Howard. 1999. Pseudogapping puzzles. In Fragments: studies in ellipsis and gapping, eds. Shalom Lappin and Elabbas Benmamoun, 141–174. New York: Oxford University Press.

    Google Scholar 

  52. Lasnik, Howard. 2001. When can you save a structure by destroying it? In Vol. 31 of Proceedings of the North East Linguistics Society, eds. M. Kim and U. Strauss, 301–320. Amherst: University of Massachusetts, GLSA.

    Google Scholar 

  53. Lasnik, Howard. 2005a. Review of The syntax of silence, by Jason Merchant. Language 81: 259–265.

    Article  Google Scholar 

  54. Lasnik, Howard. 2005b. How to evade moving violations. Ms., a lecture given at LSA Summer Institute. Cambridge: MIT Press.

    Google Scholar 

  55. Lasnik, Howard. 2014. Multiple sluicing in English? Syntax 17: 1–20.

    Article  Google Scholar 

  56. Lasnik, Howard, and Mamoru Saito. 1984. On the nature of proper government. Linguistic Inquiry 15: 235–289.

    Google Scholar 

  57. Lasnik, Howard, and Mamoru Saito. 1992. Move α: conditions on its application and output. Cambridge: MIT Press.

    Google Scholar 

  58. Lobeck, Anne C. 1995. Ellipsis: functional heads, licensing, and identification. New York: Oxford University Press.

    Google Scholar 

  59. Martín González, Javier. 2010. English sluicing under minimalist assumptions: a case for copular sources. Paper presented at TABU Dag 2010, Groningen University, Groningen.

  60. May, Robert. 1991. Syntax, semantics, and logical form. In The Chomskyan turn, ed. Asa Kasher, 334–359. Oxford: Blackwell.

    Google Scholar 

  61. McCawley, James D. 1988. The syntactic phenomena of English. Chicago: University of Chicago Press.

    Google Scholar 

  62. Merchant, Jason. 2001. The syntax of silence: sluicing, islands, and the theory of ellipsis. Vol. 1 of Oxford studies in theoretical linguistics. Oxford: Oxford University Press.

    Google Scholar 

  63. Merchant, Jason. 2004. Fragments and ellipsis. Linguistics and Philosophy 27: 661–738.

    Article  Google Scholar 

  64. Merchant, Jason. 2006. Why no(t)? Style 40: 20–23.

    Google Scholar 

  65. Merchant, Jason. 2007. Voice and ellipsis. Ms., University of Chicago.

  66. Merchant, Jason. 2008. An asymmetry in voice mismatch in VP-ellipsis and pseudogapping. Linguistic Inquiry 39: 169–179.

    Article  Google Scholar 

  67. Merchant, Jason. 2010. Three kinds of ellipsis. In Context-dependence, perspective, and relativity, eds. Francois Recanati, Isidora Stojanovic, and Neftali Villanueva, 141–192. Berlin: Mouton de Gruyter.

    Google Scholar 

  68. Nakao, Chizuru. 2009. Island repair and non-repair by PF-strategies. PhD diss., University of Maryland.

  69. Nakao, Chizuru, and Masaya Yoshida. 2007. ‘Not-so-propositional’ islands and their implications for swiping. In Proceedings of WECOL 2006, eds. Erin Bainbridge and Brain Agbayani, 322–333. Fresno: University of California.

    Google Scholar 

  70. Nevins, Andrew. 2008. Sluicing ≠ stripping: evidence from P-stranding. Paper presented at the 3rd annual Moscow Student Conference on Linguistics.

  71. Nissenbaum, Jonathan W. 2000. Investigations of covert phrase movement. PhD diss., MIT.

  72. Nunes, Jairo. 2004. Linearization of chains and sideward movement. Cambridge: MIT Press.

    Google Scholar 

  73. Ortega-Santos, Iván, Masaya Yoshida, and Chizuru Nakao. 2014. On ellipsis structures involving a wh-remnant and a non-wh-remnant simultaneously. Lingua 138: 55–85.

    Article  Google Scholar 

  74. Park, Bum-Sik, and Ja-Yeon Park. 2011. Repairing *-marking and island violations. Studies in Generative Grammar 21: 517–534.

    Article  Google Scholar 

  75. Perlmutter, David M. 1971. Deep and surface structure constraints in syntax. New York: Holt, Rinehart and Winston Inc.

    Google Scholar 

  76. Pesetsky, David. 1995. Zero syntax: experiencers and cascades. Cambridge: MIT Press.

    Google Scholar 

  77. Pesetsky, David. 1997. Some optimality principles of sentence pronounciation. In Is the best good enough, eds. Pilar Barbosa, Danny Fox, Paul Hagstrom, Martha Mcginnis, and David Pesetsky, 337–383. Cambridge: MIT Press.

    Google Scholar 

  78. Prince, Ellen F. 1981. Topicalization, focus-movement, and Yiddish-movement: a pragmatic differentiation. Proceedings of the Berkeley Linguistics Society 7: 249–264.

    Google Scholar 

  79. Reinhart, Tanya. 1991. Elliptic conjunctions—non-quantificational LF. In The Chomskyan turn, ed. Asa Kasher, 360–384. Oxford: Blackwell.

    Google Scholar 

  80. Richards, Norvin. 2001. Movement in language: interactions and architectures. Oxford: Oxford University Press.

    Google Scholar 

  81. Riemsdijk, Henk. C. van 1978. A case study in syntactic markedness: the binding nature of prepositional phrases. Lisse: The Peter de Ridder Press.

    Google Scholar 

  82. Rizzi, Luigi. 1990. Relativized minimality. Vol. 16 of Linguistic inquiry monographs. Cambridge: MIT Press.

    Google Scholar 

  83. Rizzi, Luigi. 1997. The fine structure of the left periphery. In Elements of grammar: handbook in generative syntax, ed. Liliane Haegeman, 281–337. Dordrecht: Kluwer Academic.

    Google Scholar 

  84. Rizzi, Luigi. 2001. On the position int(errogative) in the left periphery of the clause. In Current studies in Italian syntax: essays offered to Lorenzo Rizzi, eds. Guglielmo Cinque and Giampaolo Salvi, 287–296. Amsterdam: Elsevier.

    Google Scholar 

  85. Rodrigues, Cilene, Andrew Nevins, and Luis Vicente. 2009. Cleaving the interactions between sluicing and P-stranding. In Romance Languages and Linguistic Theory 2006, eds. D. Torck and L. W. Wetzels, 245–270. Amsterdam: Benjamins.

    Google Scholar 

  86. Rooth, Mats. 1985. Association with focus. PhD diss., University of Massachusetts at Amherst.

  87. Rooth, Mats. 1996. Focus. In The handbook of contemporary semantic theory, ed. Shalom Lappin, 271–297. Oxford: Blackwell Sci.

    Google Scholar 

  88. Ross, John Robert. 1967. Constraints on variables in syntax. PhD diss., Massachusetts Institute of Technology.

  89. Ross, John Robert. 1969. Guess who? In Papers from the 5th regional meeting of the Chicago Linguistic Society, eds. Robert I. Binnick, A. Davison, Georgia M. Green, and James L. Morgan, 252–286. Chicago: Chicago Linguistic Society.

    Google Scholar 

  90. Saab, Andrés. 2010. Silent interactions: Spanish TP-ellipsis and the theory of island repair. Probus 22: 73–116.

    Article  Google Scholar 

  91. Schütze, Carson. 2001. On the nature of default case. Syntax 4: 205–238.

    Article  Google Scholar 

  92. Shlonsky, Ur, and Gabriela Soare. 2012. Where’s ‘why’? Linguistic Inquiry 42: 651–669.

    Article  Google Scholar 

  93. Stainton, Robert J. 2006. Words and thoughts: subsentences, ellipsis and the philosophy of language. Oxford: Oxford University Press.

    Google Scholar 

  94. Stepanov, Arthur, and Wei-Tien Dylan Tsai. 2008. Cartography and licensing of wh-adjuncts: a cross-linguistic perspective. Natural Lanaguage and Linguistic Theory 26: 589–638.

    Article  Google Scholar 

  95. Stjepanovic, Sandra. 2008. P-stranding under sluicing in a non-P-stranding language? Linguistic Inquiry 39: 179–190.

    Article  Google Scholar 

  96. Suñer, Margarita. 1994. Verb movement and the licensing of argumental wh-phrases in Spanish. Natural Language and Linguistic Thoery 12: 335–372.

    Article  Google Scholar 

  97. Takano, Yuji. 1995. Predicate fronting and internal subjects. Linguistic Inquiry 26: 327–340.

    Google Scholar 

  98. Uriagereka, Juan. 1999. Minimal restrictions on Basque movements. Natural Language and Linguistic Thoery 17: 403–444.

    Article  Google Scholar 

  99. Wilder, Chris. 1997. Some properties of ellipsis in coordination. In Studies in universal grammar and typological variation, eds. Artemis Alexiadou, and Tracy Hall, 59–107. Amsterdam: Benjamins.

    Google Scholar 

  100. Yoshida, Masaya, Honglei Wang, and David Potter. 2012. Remarks on “gapping” in DP. In Linguistic Inquiry, Vol. 43, 475–494.

    Google Scholar 

Download references


We would like to thank three anonymous reviewers and the editor of NLLT for their detailed and helpful discussions. We are in debt to the following colleagues for their helpful discussion: Setsuko Arita, Peter Baumann, John Bowers, Sandy Chung, Brady Clerk, Michael Frazier, Tomohiro Fujii, Angel Gallego, Takuya Goro, Theresa Gregoire, John Hale, Ken Hiraiwa, Norbert Hornstein, Koji Hoshi, Kyle Johnson, Stefan Kaufmann, William Ladusaw, Howard Lasnik, Jeffrey Lidz, Jim McCloskey, Jason Merchant, Maria Polinsky, David Potter, Henk van Riemsdjik, Elizabeth Smith, Luis Vicente, Matt Wagers, Gregory Ward, Andrew Weir, and Akira Watanabe. Thanks also to the audience of GLOW 35 and WCCFL 30. This work has been supported in part by NSF grant BCS-1323245 awarded to Masaya Yoshida. All the remaining errors are, of course, our own.

Author information



Corresponding author

Correspondence to Masaya Yoshida.

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Yoshida, M., Nakao, C. & Ortega-Santos, I. The syntax of Why-Stripping. Nat Lang Linguist Theory 33, 323–370 (2015).

Download citation


  • Ellipsis
  • Stripping
  • Sluicing
  • Why
  • How come
  • Connectivity effects