{ "paper_id": "Y01-1025", "header": { "generated_with": "S2ORC 1.0.0", "date_generated": "2023-01-19T13:32:55.546706Z" }, "title": "Negation,. VP Ellipsis, and VP Fronting in English: A Construction-HP S G Analysis", "authors": [ { "first": "Jong-Bok", "middle": [], "last": "Kim", "suffix": "", "affiliation": { "laboratory": "", "institution": "Hee University", "location": { "addrLine": "1 Hoegi-dong" } }, "email": "" } ], "year": "", "venue": null, "identifiers": {}, "abstract": "It is well-known that the English auxiliaries are sensitive to the so called NICE (Negation, Inversion, Contraction, and Ellipsis) phenomena. Based on these empirical properties of auxiliaries, this paper argues for the existence of the construction aux-headph(rase) whose subtypes include negation-ph, inversion-ph, ellipsis-ph, vp-filler-ph, and the like. Each of this subtype has its own construction-specific constraints as well as those inherited from its supertypes. The present analysis uses grammatical constructions with declarative constraints and posits a rich network of inheritance relations among them. This enables us to provide a clean analysis for some of the puzzling phenomena in English such as negation, VP ellipsis, and VP fronting, while capturing new levels of generalizations among these seemlingly unrelated phenomena. 'The theory this paper assumes is Construction-HPSG, roughly Head-driven Phrase Structure Grammar, augmented with a theory of constructions as in Ginzburg and Sag to appear.", "pdf_parse": { "paper_id": "Y01-1025", "_pdf_hash": "", "abstract": [ { "text": "It is well-known that the English auxiliaries are sensitive to the so called NICE (Negation, Inversion, Contraction, and Ellipsis) phenomena. Based on these empirical properties of auxiliaries, this paper argues for the existence of the construction aux-headph(rase) whose subtypes include negation-ph, inversion-ph, ellipsis-ph, vp-filler-ph, and the like. Each of this subtype has its own construction-specific constraints as well as those inherited from its supertypes. The present analysis uses grammatical constructions with declarative constraints and posits a rich network of inheritance relations among them. This enables us to provide a clean analysis for some of the puzzling phenomena in English such as negation, VP ellipsis, and VP fronting, while capturing new levels of generalizations among these seemlingly unrelated phenomena. 'The theory this paper assumes is Construction-HPSG, roughly Head-driven Phrase Structure Grammar, augmented with a theory of constructions as in Ginzburg and Sag to appear.", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "Abstract", "sec_num": null } ], "body_text": [ { "text": "The English auxiliaries are sensitive to the so called NICE (Negation, Inversion, Contraction, and Ellipsis) phenomena, as seen from the contrast with main verbs in (1) and (2):", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "Introduction", "sec_num": "1" }, { "text": "(1) a. John may not leave Seoul.", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "Introduction", "sec_num": "1" }, { "text": "b. Will John leave Seoul? c. John can't leave Seoul. d. Mary will leave Seoul, and John will, too.", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "Introduction", "sec_num": "1" }, { "text": "(2) a. *John not left Seoul. b. *Left John Seoul? c. *John leftn't Seoul.", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "Introduction", "sec_num": "1" }, { "text": "d. *Mary wants to leave Seoul, and John wants.", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "Introduction", "sec_num": "1" }, { "text": "Based on these empirical properties of auxiliaries, this paper argues for the existence of the construction aux-head-phrase) whose subtypes include negation-ph, inversion-ph, ellipsis-ph, and vp-filler-ph. Each of this subtype has its own construction-specific constraint as well as those inherited from its supertypes. The paper ]. shows that if we accept this view of the English auxiliary system in terms of construction types and declarative type constraints on them, we can offer a more straightforward and explicit explanation for English negation, VP ellipsis, and VP fronting phenomena. This paper starts with an analysis for English negation. Following Kim and Sag (1995) , Kim (2000) , and Warner (2000) , this paper assume that the English negative marker not leads two lives -one as an adverbial modifier (as constituent negation) and the other as the complement of a finite auxiliary verb (as sentential negation):", "cite_spans": [ { "start": 662, "end": 680, "text": "Kim and Sag (1995)", "ref_id": "BIBREF9" }, { "start": 683, "end": 693, "text": "Kim (2000)", "ref_id": "BIBREF8" }, { "start": 700, "end": 713, "text": "Warner (2000)", "ref_id": "BIBREF17" } ], "ref_spans": [], "eq_spans": [], "section": "Introduction", "sec_num": "1" }, { "text": "[adv Clear support for this position lies in VP deletion (VPE) phenomena, as we will see in due course. Such a system also provides a simple analysis for Verb Phrase Fronting (VPF) phenomenon. The analysis provided here for these two constructions is built upon grammatical constructions with declarative constraints and a rich network of inheritance relations among them. This modifier property can be represented by a partial lexical description represented in the feature structure system of HPSG:", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "Introduction", "sec_num": "1" }, { "text": "(5) not", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "Introduction", "sec_num": "1" }, { "text": "This simple lexical information in (5) specifies that not modifies only a nonfinite VP and takes the modified VP's meaning as its semantic argument. This will be enough to capture the positional possibilities of the modifier not in nonfinite clauses: Independent principles guarantee that modifiers of this kind precede the elements they modify, thus ensuring the grammaticality of (6)a and (7)a, where not is used as a VP[nonfin] modifier. But the examples (6)b, (6)c and (7)b are ungrammatical since the modifier not fails to appear in the required position-i.e. before all elements of the nonfinite VP. In addition to these distributional properties, there is further evidence that not modifies a nonfinite VP. Given the general assumption that modification is recursive, our treatment predicts the possibility of double occurrences of negation in infinitival phrases. This prediction appears to be correct: It is an important semantic fact that a VP modifier never outscopes a higher verb. Thus in examples like (9) and (10), the finite verb always outscopes the adverb. The lexical entry for not must therefore include the information that the modified element (which corresponds to the element that the negation adjoins to) be within the scope of the negation. This lexical property of not also explains the interaction of coordination and negation: Various phenomena illustrate that the negator modifying a nonfinite VP is constituent negation whereas the one right after a finite auxiliary is sentential negation. For example, (13.) could have two interpretations: When not is a VP modifier as in (14)b, it has a narrow scope but it has a wide scope reading when serving as sententional negation as in (14)a.", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "Introduction", "sec_num": "1" }, { "text": "(13) The president could not approve the bill.", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "Introduction", "sec_num": "1" }, { "text": "(14) a. It would not be possible for the president to approve the bill. b. It would be possible for the president to approve the bill. Such a scope difference could also be found in wh-cleft sentences. In (15)a, the sentential negation has only a wide scope whereas in (15)b the modifier not takes a narrow scope. 2Notice, by contrast, that if we accept the general assumption that only categorially identical constituents can be coordinated, then these same facts pose a dilemma for the NegP hypothesis: the examples in (10) would be coordinations of NegP and VP or VP and NegP. And even if these nonidentical constituents are somehow allowed, we still lack an explanation for the impossibility of other cross-categorial coordinations, e.g. CP and IP. One solution to this dilemma might be to posit an additional functional projection such as PolP (Polarity Phrase, Culicover 1991) or E Phrase (cf. Laka 1990) . This would of course entail generating a phonetically unexpressed element as the head of such a phrase in every nonnegative sentence, a consequence that lacks independent justification. 0 1:1VP[bse}) (15) a. What the president could not do is ratify the treaty.", "cite_spans": [ { "start": 900, "end": 910, "text": "Laka 1990)", "ref_id": "BIBREF10" } ], "ref_spans": [], "eq_spans": [], "section": "Introduction", "sec_num": "1" }, { "text": "b. What the president could do is not ratify the treaty.", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "Introduction", "sec_num": "1" }, { "text": "A more clear difference is observed in VP deletion (VPE) phenomena. A VP after the sentential negation not can be elided as in (16)a, but this kind of ellipsis process is not possible after a VP modifier negation as in (16)b.", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "Introduction", "sec_num": "1" }, { "text": "(16) a. Though his supporters asked him to approve the bill, the president could not . b. *Though his supporters had asked him to ratify the treaty, the president could have not", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "Introduction", "sec_num": "1" }, { "text": "The most economical way to differentiate sentential negation from constituent negation seems to assume that the sentential negation is a syntactic complement of a finite auxiliary verb (cf. Kim and Sag 1995 , Kim 2000 , Warner 2000 . I claim that this English specific property comes not from lexical properties but from construction constraints on the type negation-ph, which is a subtype of aux-head-ph, as represented in 17.", "cite_spans": [ { "start": 190, "end": 206, "text": "Kim and Sag 1995", "ref_id": "BIBREF9" }, { "start": 207, "end": 217, "text": ", Kim 2000", "ref_id": "BIBREF8" }, { "start": 218, "end": 231, "text": ", Warner 2000", "ref_id": "BIBREF17" } ], "ref_spans": [], "eq_spans": [], "section": "Introduction", "sec_num": "1" }, { "text": "(17) negation-ph H word VFORM fin] HEAD AUX + NEG [NEG COMPS (E1ADV MOD I KEY CONT I NOT (\u00a9) -4", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "Introduction", "sec_num": "1" }, { "text": "The construction negation-ph in English is thus peculiar in that it selects an adverbial element such as not and a VP [base] as its complement. 3 The present analysis will then generate the following two structures for constituent negation and sentential negation respectively:", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "Introduction", "sec_num": "1" }, { "text": "a. Constituent Negation:", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "Introduction", "sec_num": "1" }, { "text": "VP jnonfini Adv [MOD EIVP[nonfini] not EWP[nonfin} b.", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "Introduction", "sec_num": "1" }, { "text": "Sentential Negation:", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "Introduction", "sec_num": "1" }, { "text": "VP[fin} V DAdv (DVP[bse] VFORM fin AUX + ] COMPS ( [NEG +], a)", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "Introduction", "sec_num": "1" }, { "text": "not 3The constraint also specifies that in terms of semantics the negation not takes the meaning of the VP as its argument, as represented by the feature KEY. KEY is the semantic relation introduced by the lexical head of the phrase and is passed up from the head of a phrase to the mother. See Copestake et al. 1999. In the syntactic structure (18)b that the analysis generates, the finite auxiliary and the following negator do not form a constituent. This nonconstituent property of the two elements predicts the impossibility of examples like 19 This `antifocus' property of sentential negation is presumably also lexically registered, as a constraint either on the lexical type, or else a condition on relevant constructions (see Kim 2000 for some discussion).", "cite_spans": [ { "start": 295, "end": 317, "text": "Copestake et al. 1999.", "ref_id": "BIBREF2" }, { "start": 735, "end": 743, "text": "Kim 2000", "ref_id": "BIBREF8" } ], "ref_spans": [], "eq_spans": [], "section": "Introduction", "sec_num": "1" }, { "text": "More clear support for this construction-based analysis comes from the explanation of VPE. The standard generalization that VPE is possible only after an auxiliary verb as represented in the following contrast:", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "VP Deletion", "sec_num": "3" }, { "text": "(23) a. Kim can dance, and Sandy can , too. b. Kim has danced, and Sandy has , too. c. Kim was dancing, and Sandy was , too.", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "VP Deletion", "sec_num": "3" }, { "text": "4 Following Zwicky and Pullum (1983) , we take n't to be an inflectional suffix. Under this inflectional analysis, partial formal regularities can be accomodated, exceptional forms (e.g. won't, don't) can be listed, and anomalies like *amn't and *mayn't can be treated as paradigmatic gaps. In addition, dialectal variants (e.g. %usen't, %ain't) are treated as simple lexical differences. These illustrate that VPE is also sensitive to the presence of an auxiliary verb. This leads to postulate the existence of the construction ellipsis-ph as a subtype of aux-head-ph with the following constraint (cf. Sag 2000) .6 (25). When this element is instantiated in an ellipsis-ph, its VP{bse] complement need not appear in syntax, as represented in (27).", "cite_spans": [ { "start": 12, "end": 36, "text": "Zwicky and Pullum (1983)", "ref_id": "BIBREF16" }, { "start": 604, "end": 613, "text": "Sag 2000)", "ref_id": "BIBREF13" } ], "ref_spans": [], "eq_spans": [], "section": "VP Deletion", "sec_num": "3" }, { "text": "(27)", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "VP Deletion", "sec_num": "3" }, { "text": "INI -;-*------.\u2022- VP ellipsis-ph HEAD COMPS ( ) V [ HEAD E(AUX COMPS (VP[bse]) can", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "VP Deletion", "sec_num": "3" }, { "text": "In the structure of (27), the head daughter's COMPS list (VP[bse]) is elided and is not realized in syntax. ? The sentences in (23)b and (23)c are also such cases: the verbs such as has and was are all auxiliary verbs ([+AUX]) and subcategorize for a VP complement. Thus, the VP complement of all these verbs can be elided but not that of the main verbs in (24). This analysis will easily generate examples like (28) in the same manner. The infinitive marker to (following Pullum 1982) is an auxiliary verb selecting a VP[bse] complement. This fits the partial description for the ellipsis-ph constraint in (25). One important constraint on VPE is that it cannot apply immediately after an adverb, as as illustrated in (30).", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "VP Deletion", "sec_num": "3" }, { "text": "(30) a. Tom has written a novel, but Peter never has i. b. *Tom has written a novel, but Peter has never .", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "VP Deletion", "sec_num": "3" }, { "text": "One simple fact we can observe from (30) is that adverbs cannot modify an empty VP. In the framework of HPSG, VP modifying adverbs carry at least the lexical information given in (31).", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "VP Deletion", "sec_num": "3" }, { "text": "The lexical entry in (31) simply states that the adverb with this lexical information modifies a VP. The head feature MOD guarantees the fact that the adverb selects the head VP it modifies. This then entails that when the VP that an adverb modifies is not syntactically realized, as in (30)b, there is no VP for the adverb to modify. This explains the unacceptability of VPE after an adverb. Given Sag and Fodor's (1994) HPSG has a small set of schemata, analogous to X' schemata, which specify partial information about universally available types of phrases. The adjunct schema is one of the universally available options for well-formed phrases. This adjunct schema roughly says that an adjunct and the head it selects through its modifier feature (MOD) forms a well-formed phrase. Now look at the structure in (32). In the present lexical theory where a VP modifier (e.g. always and never in (30)a,b) selects its head VP through the head feature MOD(IFIED), the absence of this VP then means that there is no VP the adverb can modify. And this results in an ill-formed structure: no universal schema in HPSG renders such a structure acceptable, thus explaining the ungrammaticality of (30)a,b. This analysis then provides a clean anlaysis to the striking property of not with respect to VPE we have discussed: not can be stranded only after a finite auxiliary. If the negator not in (33)a and (33)b were identically taken to be a modifier, we would predict both of these examples to be unacceptable since in both cases there is no VP for the negative marker to modify. Given the constructional constraint on negation-ph, the negator not in (33)a (but not the one in (33)b) is the complement of a finite auxiliary verb as represented in (34). 8 Sag and Fodor (1994) reexamine empirical motivations for phonetically empty categories which have been important theoretical foundations in modern GB analyses. They show that all independent arguments for the existence of traces such as auxiliary contraction, wanna contraction, and position of floated quantifiers are neither satisfactory nor well-grounded. They also present positive arguments for terminating filler-gap dependencies by lexical heads, not by traces. See Sag and Fodor (1994) Nothing blocks this structure. One may ask whether it is acceptable not to satisfy the MOD feature of the adverb not in such a case. But note here that the structure (35) is not an adjunct structure, but a head-complement structure because the negator is now converted to a complement. The HPSG theory says nothing about what happens when a complement has a MOD value. Thus its presence in a complement does not affect the well-formedness of the given phrase. Under this analysis, the ungrammaticality of (34)b and (34)c also falls out naturally. The negators in (34)b and (34)c are just modifiers. They cannot be complements. We have seen that an adverb requires the VP it modifies to be present in order to form a well-formed structure. But the VPs that the adverbs modify are absent here. The present construction and constraint-based analysis thus gives us a simple and explicit explanation for these VPE facts. The puzzle of the VPE with negation directly follows from our treatment of negation and ellipsis.", "cite_spans": [ { "start": 399, "end": 421, "text": "Sag and Fodor's (1994)", "ref_id": "BIBREF14" }, { "start": 1749, "end": 1769, "text": "Sag and Fodor (1994)", "ref_id": "BIBREF14" }, { "start": 2222, "end": 2242, "text": "Sag and Fodor (1994)", "ref_id": "BIBREF14" } ], "ref_spans": [], "eq_spans": [], "section": "VP Deletion", "sec_num": "3" }, { "text": "Now, let us consider the phenomenon often referred to as VP fronting (henceforth, VPF).", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "VP Fronting", "sec_num": "4" }, { "text": "(36) They swore that Lee might have been using heroin, and a. using heroin he might have been b. *been using heroin he might have ! c. *have been using heroin he might ! (Akamajian et al. 1979) The first thing we can notice is that VPF cannot be identified with VPE because of the ungrammatical cases such as those given in (36)b and (36)c. If any constituent that can undergo VPE can also be VP-fronted, we would expect that VP's headed by been or have could not be elided. A simple generalization we can obtain from such data is that a fronted VP cannot be headed by an auxiliary element. This will block examples like (36)b and (36)c. However, this simple approximation is counterevidenced by examples like the following (data from Gazdar et al. 1982 We thus accept the idea that VPF is different from VPE in that the former is a subtype of a head-filler:ph construction constituting a filler and a sentence with this filler as a gap. This would allow us to generate the following structure for (37)d. The constraint in (38) also requires that the head phrase (sentence) be headed by an auxiliary verb. This restriction on the head verb's AUX value will block us from overgenerating cases such as given in (40). As assumed, the progressive be and perfective have in (41)a and b are [+ASP] . The lexical rule specifies that the VP headed by either of these aspect verb cannot undergo the fronting process. Taking VP fronting to be a special case of topicalization, the analysis also allows unbounded examples like (42) (data from Gazdar et al. 1982 . 9The filler also as the constraint that its should not be infinitive. Such a constraint will block examples like (38)a and (40)a.", "cite_spans": [ { "start": 170, "end": 193, "text": "(Akamajian et al. 1979)", "ref_id": null }, { "start": 735, "end": 753, "text": "Gazdar et al. 1982", "ref_id": "BIBREF6" }, { "start": 1285, "end": 1291, "text": "[+ASP]", "ref_id": null }, { "start": 1532, "end": 1550, "text": "Gazdar et al. 1982", "ref_id": "BIBREF6" } ], "ref_spans": [], "eq_spans": [], "section": "VP Fronting", "sec_num": "4" }, { "text": "Notice that this constructional analysis explicitly factors out the similarities and differences between VPF and VPE phenomena. VPF is relevant only to the phrase whose head is an auxiliary. This restriction similarly holds in VPE too. But the restriction on the aspectual value of the gapped VP complement makes them different: the head of the VP that undergoes fronting should be nonaspectual. This accounts for the difference between VPF and VPE (relevant data repeated here). Then, what does this VP fronting analysis predict concerning negation? Recall that our treatment allows not to be either a VP modifier or a syntactic complement of a finite auxiliary, and that we permit the negator not to be stranded only if it becomes a complement of a finite auxiliary, i.e., only when it occurs immediately after a finite auxiliary. This prediction is borne out:", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "VP Fronting", "sec_num": "4" }, { "text": "(45) a. They all said that John was not being followed, and [being followed] he was not . b. They all said that John was not being followed, and [not being followed] he was .", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "VP Fronting", "sec_num": "4" }, { "text": "The negative marker not in (45)a is the complement of the auxiliary be, and its VP complement being followed is fronted. In (45)b, the same VP is fronted and not is modifying the fronted VP. But notice a different behavior of the modifier not.", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "VP Fronting", "sec_num": "4" }, { "text": "(46) Kim said she would be not eating spinach, and a. *[eating spinach] she will be not . b. [not eating spinach] she will be .", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "VP Fronting", "sec_num": "4" }, { "text": "The negator not (46)a can be only a modifier. As noted in the previous section, the modifier not cannot be stranded, since the modifier not does not satisfy its MOD requirement. Meanwhile, nothing blocks (46)b in which not is a modifier. Within the present analysis, examples like (47) will be predicted to be unacceptable in the present analysis:", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "VP Fronting", "sec_num": "4" }, { "text": "(47) Kim said she would not be eating spinach, and a. *[be eating spinach] she will not . b. *[not be eating spinach] she will .", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "VP Fronting", "sec_num": "4" }, { "text": "The fronted VP is headed by the aspectual head, the progressive be. This violates the condition on the assumed vp-filler-ph.", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "VP Fronting", "sec_num": "4" }, { "text": "The overall organization of the phrasal types we have posited and reviewed is sketched in (48). Since each subtype inherits the constraints of its supertypes, we could minimize the constraints declared upon each subtype. This type inheritance mechansim allows us to avoid stating redundant information and to capture cross-classifying generalizations among constructions while at the same time accommodating the idiosyncratic properties of individual constructions. This theory of grammar which uses grammatical constructions and posits a rich network of inheritance relations among them gives us a clean analysis for some of the puzzling phenomena in English such as negation, VP ellipsis, and VP fronting. It has been a common practice that the English negation phenomena is idiosyncratic in terms of its distributional properties. This has lead to adopt a English particular rule like do-support or introduce functional projections such as NegP (cf. Pollock 1989) together with empty categories. The construction, constraint-based analysis presented in this paper makes it unnecessary to resort to such abstract machinery and eventually enables us to find new levels of generalizations within the English auxiliary system that has often been regarded as a storehouse of peculiarities.", "cite_spans": [ { "start": 953, "end": 966, "text": "Pollock 1989)", "ref_id": "BIBREF11" } ], "ref_spans": [], "eq_spans": [], "section": "Conclusion", "sec_num": "5" } ], "back_matter": [ { "text": "The idea developed in this paper was conceived from Kim (2000) and discussions with many people. In particular, I would like to thank Chung Chan, Byung-Soo Park, Ivan Sag, Peter Sells, among others. All errors remain mine. I gratefully acknowledge the support of Kyung Hee University Research Fund in the program year of 2000.", "cite_spans": [ { "start": 52, "end": 62, "text": "Kim (2000)", "ref_id": "BIBREF8" } ], "ref_spans": [], "eq_spans": [], "section": "Acknowledgements", "sec_num": null } ], "bib_entries": { "BIBREF0": { "ref_id": "b0", "title": "The Category AUX in Universal Grammar", "authors": [ { "first": "Adrian", "middle": [], "last": "Akmajian", "suffix": "" }, { "first": "Susan", "middle": [], "last": "Steele", "suffix": "" }, { "first": "Thomas", "middle": [], "last": "Wasow", "suffix": "" } ], "year": 1979, "venue": "Linguistic Inquiry", "volume": "10", "issue": "", "pages": "1--64", "other_ids": {}, "num": null, "urls": [], "raw_text": "Akmajian, Adrian, Susan Steele, and Thomas Wasow. 1979. The Category AUX in Universal Grammar. Linguistic Inquiry 10: 1-64.", "links": null }, "BIBREF1": { "ref_id": "b1", "title": "press. Satisfying Constraints on Extraction and Adjunction. Natural Language and Linguistic Theory", "authors": [ { "first": "Gosse", "middle": [], "last": "Bouma", "suffix": "" }, { "first": "Rob", "middle": [], "last": "Malouf", "suffix": "" }, { "first": "Ivan", "middle": [], "last": "Sag", "suffix": "" } ], "year": null, "venue": "", "volume": "", "issue": "", "pages": "", "other_ids": {}, "num": null, "urls": [], "raw_text": "Bouma, Gosse, Rob Malouf, and Ivan Sag. In press. Satisfying Constraints on Extraction and Adjunction. Natural Language and Linguistic Theory.", "links": null }, "BIBREF2": { "ref_id": "b2", "title": "Minimal Recursion Semantics: An Introduction", "authors": [ { "first": "A", "middle": [], "last": "Copestake", "suffix": "" }, { "first": "D", "middle": [], "last": "Flickinger", "suffix": "" }, { "first": "C", "middle": [], "last": "Pollard", "suffix": "" }, { "first": "I", "middle": [], "last": "Sag", "suffix": "" } ], "year": 1999, "venue": "", "volume": "", "issue": "", "pages": "", "other_ids": {}, "num": null, "urls": [], "raw_text": "Copestake, A., D. Flickinger, C. Pollard, and I. Sag. 1999. Minimal Recursion Semantics: An Introduction. Ms. Stanford University and Ohio University.", "links": null }, "BIBREF3": { "ref_id": "b3", "title": "Polarity, Inversion, and Focus in English", "authors": [ { "first": "Peter", "middle": [], "last": "Culicover", "suffix": "" } ], "year": 1991, "venue": "Proceedings of ESCOL", "volume": "8", "issue": "", "pages": "46--68", "other_ids": {}, "num": null, "urls": [], "raw_text": "Culicover, Peter. 1991. Polarity, Inversion, and Focus in English. Proceedings of ESCOL 8, 46-68.", "links": null }, "BIBREF4": { "ref_id": "b4", "title": "Ellipsis and Higher-order Unification", "authors": [ { "first": "Mary", "middle": [], "last": "Dalrymple", "suffix": "" }, { "first": "Stuart", "middle": [], "last": "Shieber", "suffix": "" }, { "first": "Fernando", "middle": [ "C" ], "last": "Pereira", "suffix": "" } ], "year": 1991, "venue": "Linguistics and Philosophy", "volume": "14", "issue": "", "pages": "399--452", "other_ids": {}, "num": null, "urls": [], "raw_text": "Dalrymple, Mary, Stuart Shieber, and Fernando C. Pereira. 1991. Ellipsis and Higher-order Unification. Linguistics and Philosophy 14: 399-452.", "links": null }, "BIBREF5": { "ref_id": "b5", "title": "Inversion and Constructional Inheritance", "authors": [ { "first": "Charles", "middle": [], "last": "Fillmore", "suffix": "" } ], "year": 1999, "venue": "Lexical and Constructional Aspects of Linguistics Explanation", "volume": "", "issue": "", "pages": "113--128", "other_ids": {}, "num": null, "urls": [], "raw_text": "Fillmore, Charles. 1999. Inversion and Constructional Inheritance. In G. Webelhuth, J.P Koenig, and A. Kathol (eds.), Lexical and Constructional Aspects of Linguistics Explana- tion, 113-128. Stanford: CSLI Publications.", "links": null }, "BIBREF6": { "ref_id": "b6", "title": "Auxiliaries and Related Phenomena in a Restrictive Theory of Grammar", "authors": [ { "first": "", "middle": [], "last": "Gazdar", "suffix": "" }, { "first": "Geoffrey", "middle": [ "K" ], "last": "Gerald", "suffix": "" }, { "first": "Ivan", "middle": [], "last": "Pullum", "suffix": "" }, { "first": "", "middle": [], "last": "Sag", "suffix": "" } ], "year": 1982, "venue": "Language", "volume": "58", "issue": "", "pages": "591--638", "other_ids": {}, "num": null, "urls": [], "raw_text": "Gazdar, Gerald, Geoffrey K. Pullum, and Ivan Sag. 1982. Auxiliaries and Related Phenomena in a Restrictive Theory of Grammar. Language 58: 591-638.", "links": null }, "BIBREF7": { "ref_id": "b7", "title": "To appear", "authors": [ { "first": "Jonathan", "middle": [], "last": "Ginzburg", "suffix": "" }, { "first": "Ivan", "middle": [ "A" ], "last": "Sag", "suffix": "" } ], "year": null, "venue": "English Interrogative Constructions. Stanford", "volume": "", "issue": "", "pages": "", "other_ids": {}, "num": null, "urls": [], "raw_text": "Ginzburg, Jonathan, and Ivan A. Sag. To appear. English Interrogative Constructions. Stan- ford: CSLI Publications.", "links": null }, "BIBREF8": { "ref_id": "b8", "title": "The Grammar of Negation: A Constraint-Based Approach", "authors": [ { "first": "Jong-Bok", "middle": [], "last": "Kim", "suffix": "" } ], "year": 2000, "venue": "", "volume": "", "issue": "", "pages": "", "other_ids": {}, "num": null, "urls": [], "raw_text": "Kim, Jong-Bok. 2000. The Grammar of Negation: A Constraint-Based Approach. Stanford: CSLI Publications.", "links": null }, "BIBREF9": { "ref_id": "b9", "title": "The Parametric Variation of English and .French Negation", "authors": [ { "first": "Jong-Bok", "middle": [], "last": "Kim", "suffix": "" }, { "first": "Ivan", "middle": [ "A" ], "last": "Sag", "suffix": "" } ], "year": 1995, "venue": "Proceedings of the Fourteenth West Coast Conference on Formal Linguistics (WCCFL)", "volume": "", "issue": "", "pages": "303--317", "other_ids": {}, "num": null, "urls": [], "raw_text": "Kim, Jong-Bok, and Ivan A. Sag. 1995. The Parametric Variation of English and .French Negation. In Jose Camacho, Lina Choueiri, and Maki Watanabe (eds), Proceedings of the Fourteenth West Coast Conference on Formal Linguistics (WCCFL), 303-317. Stanford: SLA CSLI Publications.", "links": null }, "BIBREF10": { "ref_id": "b10", "title": "Negation in Syntax: On the Nature of Functional Categories and Projections. Doctoral Dissertation, MIT", "authors": [ { "first": "Itziar", "middle": [], "last": "Laka", "suffix": "" } ], "year": 1990, "venue": "", "volume": "", "issue": "", "pages": "", "other_ids": {}, "num": null, "urls": [], "raw_text": "Laka, Itziar. 1990. Negation in Syntax: On the Nature of Functional Categories and Projections. Doctoral Dissertation, MIT.", "links": null }, "BIBREF11": { "ref_id": "b11", "title": "Verb Movement, Universal Grammar, and the Structure of IP", "authors": [ { "first": "Jean-Yves", "middle": [], "last": "Pollock", "suffix": "" } ], "year": 1989, "venue": "Linguistic Inquiry", "volume": "20", "issue": "", "pages": "365--424", "other_ids": {}, "num": null, "urls": [], "raw_text": "Pollock, Jean-Yves. 1989. Verb Movement, Universal Grammar, and the Structure of IP. Linguistic Inquiry 20: 365-424.", "links": null }, "BIBREF13": { "ref_id": "b13", "title": "Rules and Exceptions in the English Auxiliary System", "authors": [ { "first": "Ivan", "middle": [ "A" ], "last": "Sag", "suffix": "" } ], "year": 2000, "venue": "", "volume": "", "issue": "", "pages": "", "other_ids": {}, "num": null, "urls": [], "raw_text": "Sag, Ivan A. 2000. Rules and Exceptions in the English Auxiliary System. Manuscript. Stanford University.", "links": null }, "BIBREF14": { "ref_id": "b14", "title": "Extraction Without Traces", "authors": [ { "first": "Ivan", "middle": [ "A" ], "last": "Sag", "suffix": "" }, { "first": "Janet", "middle": [], "last": "Fodor", "suffix": "" } ], "year": 1994, "venue": "Proceedings of the West Coast Conference on Formal Linguistics", "volume": "13", "issue": "", "pages": "365--384", "other_ids": {}, "num": null, "urls": [], "raw_text": "Sag, Ivan A. and Janet Fodor. 1994. Extraction Without Traces. In Raul Aranovich, William Byrne, Susanne Preuss, and Martha Senturia (eds.), Proceedings of the West Coast Con- ference on Formal Linguistics 13, 365-384. Stanford: Stanford Linguistic Association.", "links": null }, "BIBREF15": { "ref_id": "b15", "title": "Syntactic Theory: A Formal Approach", "authors": [ { "first": "Ivan", "middle": [ "A" ], "last": "Sag", "suffix": "" }, { "first": "Tom", "middle": [], "last": "Wasow", "suffix": "" } ], "year": 1999, "venue": "", "volume": "", "issue": "", "pages": "", "other_ids": {}, "num": null, "urls": [], "raw_text": "Sag, Ivan A. and Tom Wasow. 1999. Syntactic Theory: A Formal Approach. Stanford: CSLI Publications.", "links": null }, "BIBREF16": { "ref_id": "b16", "title": "Cliticization vs. Inflection: English n't. Language", "authors": [ { "first": "Arnold", "middle": [ "M" ], "last": "Zwicky", "suffix": "" }, { "first": "K", "middle": [], "last": "Geoffery", "suffix": "" }, { "first": "", "middle": [], "last": "Pullum", "suffix": "" } ], "year": 1983, "venue": "", "volume": "59", "issue": "", "pages": "502--513", "other_ids": {}, "num": null, "urls": [], "raw_text": "Zwicky, Arnold M. and Geoffery K. Pullum. 1983. Cliticization vs. Inflection: English n't. Language 59: 502-513", "links": null }, "BIBREF17": { "ref_id": "b17", "title": "The Nature and Function of Syntactic Categories", "authors": [ { "first": "Anthony", "middle": [ "R" ], "last": "Warner", "suffix": "" } ], "year": 2000, "venue": "Syntax and Semantics", "volume": "32", "issue": "", "pages": "167--220", "other_ids": {}, "num": null, "urls": [], "raw_text": "Warner, Anthony R. 2000. English Auxiliaries Without Lexical Rules. In R. Borsley, ed., Syntax and Semantics, Volume 32: The Nature and Function of Syntactic Categories, 167-220. San Diego and London: Academic Press.", "links": null } }, "ref_entries": { "FIGREF0": { "num": null, "type_str": "figure", "uris": null, "text": "(Everyone's turning the offer down, but I'm wavering...) ?I can't believe you would consider [not [not [taking advantage of the offer]]]." }, "FIGREF1": { "num": null, "type_str": "figure", "uris": null, "text": "33) a. Kim said he could have heard the news, but Lee said that he could not . b.*Kim said he could have heard the news, but Lee said that he could have not ." }, "TABREF0": { "num": null, "text": "John [[will] [not] [leave Seoul]]. (Sentential negation) b. John wants [not [to leave Seouln.", "content": "
(3) a.
[HEADMOD VP[nonfin]:
CONT NOT (n)
", "html": null, "type_str": "table" }, "TABREF1": { "num": null, "text": "VP constituent NegationThe English negator not, like an adverb never, can function as a modifier of a phrasal element like a nonfinite VP as in (4).", "content": "
2 Negation
2.1 (4) a. Kim regrets [never/not [having seen the movie]).
b. We asked him [never/not [to try to call us again]].
", "html": null, "type_str": "table" }, "TABREF3": { "num": null, "text": "(9) a. Kim seems [never [to be alone]]. b. Kim seems [not [to like anchovies]]. c. Pat considered [always [doing the homework assignmentjj. d. Pat considered [not [doing the homework assignment]].", "content": "", "html": null, "type_str": "table" }, "TABREF5": { "num": null, "text": "a.4 (19) a. *[Would not] he leave the city? b. Wouldn't he leave the city?As we have seen in (8), the treatment of not as VP constituent negation predicts the recursive modification by not in nonfinite clauses. But since the system allows only one sentential negation, the second not in (20)b is correctly analyzed as constituent negation.The introduction of the feature NEG for the verbal lexemes that select not as a complement accounts for the constraint that English allows only one sentential not per clause.The two scope patterns discussed earlier are analyzed in terms of two lexical classes with differing scope constraints lexically encoded. 5 A further property of such not-selecting forms is that they cannot be focused (seeKim 2000 for some discussion.):(21) a. *They WILL not be there.b. *Leslie CAN not do that. c. *Leslie CAN so/too do that.This is supported by tag questions. As noticed in (22), the negation following the stressed auxiliary does not function as sentential negation:(22) a. He CAN not go to school tomorrow, (can't he/*can he)? b. He MAY not go to school tomorrow, (mayn't he/*may he)?", "content": "
(20) Kleptomaniacs [will] [not] [NOT [steal]].
", "html": null, "type_str": "table" }, "TABREF6": { "num": null, "text": "Kim must not drink the wine on the table. b. Kim may not drink the wine on the table. (24) a. *Kim considered joining the navy, but I never considered . b. *Kim got arrested by the CIA, and Sandy got , also. c. *Kim wanted to go and Sandy wanted , too.", "content": "
5 There are cases with no ambiguity:
(i) a.
", "html": null, "type_str": "table" }, "TABREF7": { "num": null, "text": "What this constraint allows is an instance of ellipsis-ph whose head is an auxiliary selecting a list of complements (represented by 2a and an XP. It is this phrasal XP element that is missing in syntax. For example, the auxiliary can would have the following lexical information.", "content": "
ellipsis-ph
H
[-4word AUXO
COMPSOnelist(XP)
", "html": null, "type_str": "table" }, "TABREF10": { "num": null, "text": "for details. (34) [COMPS (ADV[NEG +], VP[bse])]Once this auxiliary head is realized in an ellipsis-ph, then the rightmost element (VP[bse]) can be unrealized in syntax, yielding the following structure.", "content": "
VP
[negation-ph &
VDAdv
AUX +[NEG +
COMPS (H, VP[bse])MOD VP[nonfin]]
couldnot
Notice that the phrase [could not] in (35) forms a well-formed head-complement structure where
", "html": null, "type_str": "table" }, "TABREF11": { "num": null, "text": "Notice here that though the fronted VP in (37)a-c carries the [+AUX] value inherited from the head (to, be, have and being), they are all unacceptable. To overcome this issue, we assume that the progressive be and perfective have are aspect verbs ([+ASP(ECT)]) (cf.Gazdar et al. 1982) together with the following constructional constraint:", "content": "
(35) (38) vp-filler-ph
fillerH
ASPECT-[AUX+
VFORM LOCALinfSLASH{U
):
(37) a.*... and [to go] he is
b.*... and [be going] he will.
c.*... and [have gone] he will.
d. ... and [being evasive] he was.
", "html": null, "type_str": "table" }, "TABREF12": { "num": null, "text": "They said he would have finished it, and vp [+Asp] [have finished] he will . c. They said he would be noisy, and vp [ ._ Asp] [being noisy] he was .", "content": "
(40) a. *I never thought that he would want to go, but [to go] he wanted.
b. *I never thought Lee would help move the chair, but [move the chair] Lee helped
c. *I never thought Lee would stop feeding the dog, but [feeding the dog],
Lee stopped.
The constructional constraint in (39) also imposes constraints on its filler. The slashed VP
should be [-ASP], i.e., not headed by progressive be or perfective have. 9 By requiring such
restriction on the filler VP, we can account for the grammatical contrast in examples like the
following:
(41) a.*They said he would go, and V P[--AS P] [be going] he will.
b. *
", "html": null, "type_str": "table" }, "TABREF13": { "num": null, "text": "). (42) a. ... and [go], I think he will . b. ... and [going], I believe Kim knew he was . C. ... and [being evasive], I believe he knew he was", "content": "", "html": null, "type_str": "table" } } } }