Ellipsis, linguistic interfaces, and the architecture ofgrammar

Jason MerchantUniversity of Chicago

2012

1 Abstract1 (PS-abstract), abstract2 (Phono-abstract)

(1) Two kinds of traditional abstractness in syntax

a. abstract1 (PS-abstract) structure: models the grouping and relations betweengroups of words that is not physically present in the speech signal (known asconstituent, geometrical, hierarchical, or simply phrasestructure (PS))

b. abstract2 (phono-abstract) structure: consists of nodes in the geometry whichmay not correspond to any pronounced elements in the speech stream (thesenodes may be equivalent to ‘words’, single lexical nodes, orto phrases).

(2) Architectural alternatives

a. All higher-order (phrasal) structures are projected from and contain only ele-ments that are pronouncedCorollary: There are no phrases or heads that consist solelyof the emptystring.‘WYHIWYG’ theory (‘What you hear is what you get’): Ginzburgand Sag 2000,Culicover and Jackendoff 2005, much work in categorial grammars, some workin Autolexical Grammar (Sadock 1991)—part of theSimpler SyntaxHypothesis

b. Some phrases and heads have no pronunciation.Corollary: Their presence can only be detected indirectly.

1

Ellipsis intro 2

2 Simplicity vs. abstractness

2.1 Two kinds of traditional abstractness in syntax

(3) a. abstract1 (phrase structurally abstract) structure: models the grouping and rela-tions between groups of words that is not physically presentin the speech signal(known as constituent, geometrical, hierarchical, or phrase structure)

b. abstract2 (phonologically abstract) structure: consists of nodes inthe geometrywhich may not correspond to any pronounced elements in the speech stream(equivalent to ‘words’, single lexical nodes, or to phrases).

Why do we need phrase structural abstractness? (Why can’t wejust model sentences aswords in some order—like beads on a string?) One answer: Structural ambiguities:

(4) Susan saw the man with the telescope.

(5) a. S

NP

Susan

VP

V

saw

NP

Art

the

N′

N

man

PP

P

with

NP

Art

the

N

telescope

3 Jason Merchant

b. S

NP

Susan

VP

VP

V

saw

NP

Art

the

N′

N

man

PP

P

with

NP

Art

the

N

telescope

Movement (word order alternations) is sensitive to phrase structure

(6) a. With the telescope, Susan saw the man. Unambiguousb. The man with the telescope, Susan saw. Unambiguousc. The man, Susan saw with the telescope. Unambiguous

(7) “this crime covers anyone whointentionallyaccesses a federal computer withoutauthorization,and by means of one or more instances of such conduct alters,damages, or destroys information”18 U.S.C. 1030(a)(5)(A) debated inUnited States v. Morrison(1991).

a. Adverb [VP and VP]:defendantb. [Adverb VP] and [VP]:plaintiff

(8) VP ellipsis in English

a. Bill should collect butterflies. Jill should collect butterflies, too.

b. Bill should collect butterflies. Jill should, too.

(9) TP

Bill T ′

should VP

collect NP

butterflies

Ellipsis intro 4

(10) Two possible analyses for the missing VP:

a. The VP is syntactically present, but unpronounced (‘elided’):TP

Jill T ′

shouldT <VPE>

collect NP

butterflies

b. The VP isn’t there at all (there is no VP node in the syntax):S

Jill should/V P

3 ‘Deep’ and ‘surface’ anaphora

Hankamer and Sag 1976,Sag and Hankamer 1984: (‘model-interpretive’ vs. ‘ellipsis’)

(11) Diagnostics:

a. extraction (A′, A, head)

b. agreement

c. inverse scope

d. Missing Antecedent Anaphora

e. pragmatic control (linguistic antecedent)

f. sloppy identity

g. split antecedents

3.1 Potent diagnostics

3.1.1 Extraction

(12) I asked him to write a report.

a. Did he agree to? (‘surface’)

b. Did he agree? (‘deep’)

(13) a. Which report did he refuse to write, and which report did he agree to?

b. *Which report did he refuse to write, and which report did he agree?

(Caveat in Aelbrecht 2010,van Craenenbroeck 2010: Beware the fallacy of denial of theantecedent.)

5 Jason Merchant

3.1.2 Agreement

(14) a. First, there were bananas available, and then there weren’t.b. First, there were going to be bananas available, and then there weren’t.

3.1.3 Inverse quantifier scope (IQS)

(15) a. A doctor examined every patient, and then a nurse did.(∃>∀, ∀>∃)b. A doctor examined every patient, and then a nurse did it. (∃>∀, *∀>∃)

3.1.4 Missing Antecedent Anaphora (MAA)

(16) Grinder and Postal 1971:a. My uncle didn’t buy anything for Christmas, but my aunt did, and it was bright

red.b. *My uncle didn’t buy anything for Christmas, so my aunt didit for him, and it

was bright red.

3.2 Problematic diagnostics

3.2.1 Pragmatic control (Use of anaphor without a linguistic antecedent)

(17) Yes, we can do it! Yes, we did it! Don’t do it! .¯Yes, we can! Yes, we did! Don’t!

(Pullum 2000, Merchant 2004)

3.2.2 Sloppy identity

(18) Abby1 cleaned her1 gun, and Beth2 did, too.a. = Beth2 cleaned her1 (=Abby’s) gun (‘strict’ reading), orb. = Beth2 cleaned her2 (=her own, i.e., Beth’s) gun (‘sloppy’ reading)

(19) a. Ralph ate his ice-cream with a spoon, and Seymour did the same thing.b. Harvey stubbed his toe on the doorstop, and it happened to Max, too.c. Undergraduates can be covered under their parents’ health plans if desired; {like-

wise for graduate students. / that goes for grad students, too.}d. A professor who pays down her mortgage with her paycheck iswiser than one

who gambles it away in online poker.

3.2.3 Split antecedents

(20) a. Our son has a BMW1 and our daughter drives a Kawasaki2 . They1+2 take up thewhole garage.

b. Wendy is eager to sail around the world and Bruce is eager toclimb Mt. Kili-manjaro, but neither of them can, because money is too tight.(Webber 1978)

Ellipsis intro 6

3.3 Architectural implications

(21) Alternatives

a. ‘Surface’ lexicalismAll higher-order (phrasal) structures are projected from and contain only ele-ments that are pronouncedCorollary: There are no phrases or heads that consist solelyof the emptystring.WYSIWYG theory (What you see is what you get): Ginzburg and Sag 2000,Culicover and Jackendoff 2005, much work in categorial grammars, some workin Autolexical Grammar (Sadock 1991)—part of theSimpler SyntaxHypothesis

b. Traditional lexicalismSome phrases and heads have no pronunciation.Corollary: Their presence can only be detected indirectly.(Winkler 2005 et multi alii; see Winkler and Schwabe 2003 foran overview)

3.4 Syntactic and semantic representations and the mappingbetweenthem

(22) VP ellipsis in English

a. Bill should collect butterflies. Jill should, too.

b. Bill should collect butterflies. Jill should collect butterflies, too.

3.4.1 ‘Deletion’/Ellipsis approachAbstract2 + function application [Fregean hypothesis]

(23) a. TP

Jill T′

shouldT <VPE>

collect NP

butterflies

7 Jason Merchant

b. should(collect(jill, butterf lies))

jill λz.should(collect(z, butterf lies))

λPλz.should(P (z)) λx.collect(x, butterf lies)

λyλx.collect(x, y) butterf lies

(24) Rulesa. Syntactic:should[ _ VP ], etc.

b. Semantics: Iff is a expression of typeτ containing one or more instances of afree variableh of typeσ andg is an expression of typeσ, thenλhσ[f τ ](gσ) f τ

h/g .

c. Phonology:J shouldKp /SUd/, JXEKp ∅, etc.

3.4.2 Indirect LicensingWYSIWYG syntax + additional mapping rules [Simpler Syntax hypothesis]

(25) a. S

Jill should/V P

b. should(collect(jill, butterflies))

(26) Rulesa. Syntactic: S→ NP I0 (VP), etc.

should[ _ (VP) ], etc.b. Semantics:

i. Argument/Modifier RuleCS: [F ... Xi ... ] ⇔default Syntax: {..., YPi, ...}

ii. R1′: If X is the meaning of the NP-daughter-of-S whose predicate meaningis PRED, then let PRED(AGENT:_, ...) = PRED(AGENT:X, ...)

iii. Bare Argument Ellipsis (C&J 2005:265)Syntax: [U XPi

ORPH ]IL Semantics: [f (Xi)]

iv. If f is an expression in CSa andf cannot be determined from SYNTAXaby application of Rules R1...Rn, then “f amounts to the presupposition of theantecedent, constructed by substituting variables for the[necessary elements]in the CS of the antecedent” (Culicover and Jackendoff 2005:276)

c. Phonology:J shouldKp /shud/, etc.

Ellipsis intro 8

4 Ellipsis: The phenomena

(27) a. “Eclipsis est defectus dictionis, in quo necessaria verba desunt” [‘ellipsis is aincompletion of speech in which necessary words are missing’] (St. Isidore ofSevilla, Etymologiarum, Liber I ‘De grammatica’, ch. XXXIV‘De Vitiis’, sec. 10)

b. “ellipsis, or speech by half-words, [is the peculiar talent] of ministers and politi-cians” (Alexander Pope, 1727,Peri Bathous, p. 115)

c. “zweimal tausendjährige Ellipsenplage” (Bühler 1934; 1978:168)

(28) ‘Headed’ (H+) ellipses (in Chao’s 1987 terminology)

a. sluicingJohn can play something, but I don’t know what.

b. VP-ellipsisJohn can play the guitar and Mary can, too.

c. pseudogappingJohn can play the guitar and Mary can the violin.

d. NP-ellipsis/‘N′’-ellipsisJohn can play five instruments, and Mary can play six.

(29) ‘Headless’ (H-) ellipses1

a. strippingJohn can play the guitar, {and Mary, too/and Mary as well/butnot Mary}.John can play the guitar better than Mary.

b. gappingJohn can play the guitar, and Mary the violin.John can play the guitar better than Mary the violin.

c. fragment answersQ: Who can play the guitar?A: (Not) John.

Compare nonelliptical counterparts:

(30) a. John can play something, but I don’t know what John canplay.

b. John can play the guitar and Mary can play the guitar, too.

c. John can play five instruments, and Mary can play six instruments.

(31) a. John can play the guitar, but Mary can’t play the guitar.

1All of these structures have been the focus of intense theoretical interest over the past four decades, andvast bibliographies can be compiled for each of the above phenomena. I can make no pretense of bibliographiccompleteness here, and refer the reader to excellent recentsurveys for a more detailed treatment of the literature,especially Hartmann 2000, Johnson 2001, Winkler and Schwabe 2003,?, Winkler 2005, Goldberg 2005, Reich2008, and the introduction to Johnson 2008. In what follows,I will examine some representative examples ofapproaches to the above and discuss their relative merits.

9 Jason Merchant

b. John can play the guitar better than Mary can play the violin.

c. John can(not) play the guitar.

Two questions:2

1. Thestructure question:Is there syntax internal to the ellipsis site? (E.g., is there an actual VP in the secondclause of (28)b?)

2. Theidentity question:The understood material is identical to some antecedent. Isthe relevant kind of identitysyntactic (defined over phrase markers of some sort) or semantic (defined over semanticrepresentations of some sort)?

Table 1 organizes a selection of the literature by the answers it proposes to these twoquestions.

Is there syntax in the ellipsis site?Yes No

Isid

entit

ysy

ntac

ticor

sem

antic

?

Syntactic

Sag 1976, Williams 1977Fiengo & May 1994 N/A (incoherent)

Chung et al. 1995, etc.Kehler 2002

Semantic

Merchant 2001 Keenan 1971, Hardt 1993,van Craenenbroeck 2010 Dalrymple et al. 1991

Aelbrecht 2010 Ginzburg & Sag 2000,Sag and Hankamer 1984Culicover & Jackendoff 2005 etc.

Table 1: Some previous research on the two ellipsis questions

2A third question, which so far has not attracted quite the attention the above two questions have, is thelicensingquestion: what heads or positions or structures allow for ‘ellipsis’, and what are the locality conditionson the relation between these structures and ellipsis? The licensing question has been addressed by Zagona1982, Lobeck 1995, Johnson 2001, and Merchant 2001 and formsa substrand of van Craenenbroeck’s work(?): these latter owe a great debt to Lobeck 1995, whose approach is based on a kind of ECP applied to a nullpro-like element.

Ellipsis intro 10

4.1 What ellipsis is not

Expressions with no antecedents (implicit or overt). [Shopen 1971, Sadock 1974, Yanofsky1978, Klein 1985, Barton 1990, 1998, Schlangen 2003, Culicover and Jackendoff 2005,Stainton 2006; see additional classes in Klein 1985, Schwabe 1993, Schlangen 2003]

(32) Special registers: telegrams, titles, headlines, weather reports, recipes, instructions(‘If no paper, turn wheel’)

(33) Short directives: Left! Higher! Scalpel!

(34) Labels: (cf. Bühler’s ‘dingfest angeheftete Namen’, Bühler 1934: sec. 10)

a. Campbell Soup.

b. Starbucks.

c. Thief! Thief! (cf. Paul 1919:1 “Wenn jemand den Angst- undHilferuf ‘Diebe’ausstößt, so will er, daßder Allgemeinbegriff ‘Diebe’ mit einer von ihm in demAugenblick gemachten Wahrnehmung in Beziehung gesetzt werde.”)

d. Fire!

e. And now: the first act of the night: The Rolling Stones!

f. To kill a Mockingbird

g. Der Zauberberg

h. Next exit: Chicago.

(35) Expressive exclamations: Wonderful! Nonsense! Fate!For Pete’s sake!

(36) Utterance idioms (Kleins “elliptische Formeln”):

a. Up yours.

b. ‘Gewitter im Mai—April vorbei’ (lit. ‘storms in May—April over’)

c. ‘Wenn schon, denn schon’ (lit. ‘if already, then already’; roughly, ‘in for a penny,in for a pound’)

d. Dutch ‘Met Jason’ (‘with Jason’) as a telephone greeting

(37) Other nonsentential partially fixed material expressions

a. So much for the light of reason.

b. Off with his head!

c. A good talker, your friend Bill.

d. Books open to page 15!

e. How about a cookie?

f. What, me worry?

g. Hey, Phil!

h. Vikings 27, Bears 3

(38) Some kinds of fragments (e.g. Schlangen 2003’s ‘explanation’ subtype)

a. Mary: Try it. It’s good for you.

b. Peter: Why?

c. Mary: Lots of vitamins.

11 Jason Merchant

5 Approaches to the syntax of ellipsis

(39)

Is there unpronounced syntacticstructure in ellipsis sites?

no yesa. Nonstructural b. Structural

approaches approachesIs there unpronounced syntactic

structure in ellipsis sitesthroughout the entire derivation?

no yesi. LF-copy, null anaphora ii. PF-‘deletion’

Nonstructural:

(40) John can play something, but I don’t know[S what ].

(41) Syntax: [S whatORPH]IL Semantics: Q[F(what)]

Structural with null elements:

(42) a. I don’t know [CP what [IP e ]] (Spell-Out)

b. I don’t know [CP what4 [IP e1 e2 e3 t4 ]]

(43) I don’t know [CP what4 [IP John can playt4 ]] (LF/interpreted structure)

Structural with nonpronounciation (‘deletion’):

(44) CP

what1C <TP>

John can playt1

5.1 Structural and nonstructural approaches compared

6 Evidence for structure in ellipsis

6.1 Locality effects

6.1.1 VP-ellipsis

Sag 1976, Haïk 1987, Postal 2001, Lasnik 2001, Fox and Lasnik2003, Kennedy and Mer-chant 2000, Merchant 2001, Merchant 2008, andKennedy 2003.

Ellipsis intro 12

(45) a. *I read every book you introduced me to a guy who did.

b. *Abby wants to hire someone who speaks a Balkan language, but I don’t re-member which (Balkan language) Ben does. <want to hire someone who speakst >

c. *Abby knows five people who have dogs, but cats, she doesn’t<know five peoplewho have>.

d. *Which film did you refuse to see because Roger was so revolted when he didafter renting?

e. *They met a five inches taller man than you did.

6.1.2 Fragment answers

Morgan 1973 and Merchant 2004, though see Culicover and Jackendoff 2005, Stainton 2006,Valmala 2007 for additional, conflicting data (see section 7.1.2 below).

(46) a. Does Abby speakGreekfluently?

b. No,Albanian.

c. No, she speaksAlbanianfluently.

(47) a. Did Abby claim she speaksGreekfluently?

b. No,Albanian.

c. No, she claimed she speaksAlbanianfluently.

(48) a. Will each candidate talkabout taxes?

b. No,about foreign policy.

c. No, each candidate will talkabout foreign policy.

(49) a. Did each candidate2 agree on who will ask him2 about taxes(at tonight’s debate)?

b. *No, about foreign policy.

c. No, each candidate2 agreed on who will ask him2 about foreign policy(at tonight’sdebate).

6.1.3 Stripping/Bare Argument Ellipsis (BAE)

(BAE is Reinhart 1991’s term; see Lechner 2001 for discussion.)

(50) a. The man stolethe carafter midnight, but notthe diamonds.

b. *They caught the man who’d stolenthe carafter searching for him, but notthediamonds.

13 Jason Merchant

6.1.4 Gapping

(Johnson 1996,?, Coppock 2001, Winkler 2005):

(51) *Some wanted to hire the woman who worked on Greek, and others Albanian.

(52) *SHE discussed my question which LETTERS we wrote and HEwhich BOOKS.(Winkler 2005:61 (22b))

6.1.5 Sluicing from inside DPs

Lasnik and Park 2003

(53) *Books were sold to John, but I don’t know on which shelf.

6.1.6 Sluicing over implicit correlates

Chung et al. 1995, and discussed in Merchant 2001 and Hardt and Romero 2004.

(54) Tony sent Mo a picture that he painted, but it’s not clearwith what.

a. = <Tony sent him the picturetwithwhat>

b. 6= <Tony sent him a picture that he [paintedtwithwhat ]>

6.1.7 Contrast sluicing

(Merchant 2001, Vicente 2008).

(55) She knows a guy who hasfive dogs, but I don’t know how manycats.

a. = <he [=the guy who has the five dogs] hast>

b. 6= <she knows a guy who hast ]>

6.2 The P-stranding generalization

Merchant 2001; (56)-(57) represent P-stranding languages(as seen in the (b) controls), while(58)-(60) illustrate non-P-stranding languages.

(56) Englisha. Peter was talking with someone, but I don’t know (with) who(m).

b. Who was he talking with?

(57) Swedisha. Peter

Peterharhas

talattalked

medwith

någon;someone

jagI

vetknow

intenot

(med)with

vem.who

‘Peter talked with someone, but I don’t know who.’

Ellipsis intro 14

b. Vemwho

harhas

PeterPeter

talattalked

med?with

‘Who has Peter talked with?’

(58) Greeka. I

theAnnaAnna

milisetalked

mewith

kapjon,someone

allabut

dhenot

kseroI.know

*( me)with

pjon.who

b. * Pjonwho

milisetalked.3s

me?with

(59) Russiana. Anja

Anjagovorilaspoke

swith

kem-to,someone,

nobut

nenot

znajuI.know

*( s)with

kem.who

b. * Kemwho

onashe

govorilatalked

s?with

(60) Germana. Er

hewolltewanted

mitwith

jemandemsomeone

tanzen,to.dance

aberbut

ichI

weissknow

nicht,not

*(mit)with

wem.who

b. * Wemwho

wolltewanted

erhe

mitwith

tanzen?to.dance

6.3 Case matching

Ross 1969, case matching effects found in sluicing (and fragment answers, Merchant 2004)

(61) Germana. Er

hewillwants

jemandemsomeone.DAT

schmeicheln,flatter

aberbut

siethey

wissenknow

nicht,not

{ *werwho.NOM

/

*wenwho.ACC

/ wemwho.DAT

}.

‘He wants to flatter someone, but they don’t know who.’

b. Erhe

willwants

jemandensomeone.ACC

loben,praise

aberbut

siethey

wissenknow

nicht,not

{ *werwho.NOM

/ wenwho.ACC

/

*wem}.who.DAT

‘He wants to praise someone, but they don’t know who.’

6.4 Complementizer deletion

Morgan 1973, Merchant 2004

(62) What does no-one believe?#(That) I’m taller than I really am.

15 Jason Merchant

a. No-one believes (that) I’m taller than I really am.

b. *(That) I’m taller than I really am, no-one believes.

(63) What are you ashamed of?(That) I ignored you.

a. *I’m ashamed of that I ignored you.

b. That I ignored you, I’m ashamed of.

6.5 Infinitivals: Raising vs. control

(64) a. *It’s [to procrastinate] that people tend.

b. Q: How do people tend to behave?A: *To procrastinate.

(65) a. It’s [to get a job in Europe] that she really wants.

b. Q: What does she really want?A: To get a job in Europe.

6.6 Predicate answers

Hankamer 1979, Merchant 2004

(66) a. A: What did he do for his sister?B: Funded *(her).

b. He did [fund(ed) her] for his sister.

As Culicover and Jackendoff (2005:11 fn 8) put it, the presence of these kinds of connectivityeffects would represent “impressive evidence of the reality of the invisible structure” (whilereporting that they don’t find consistent island effects in cases like (49b), they don’t considerthe remaining facts).

7 Evidence against structure in ellipsis

7.1 Absence of locality effects

7.1.1 Sluicing

As Ross 1969 famously first observed, the putative wh-extraction out of ellipsis sites in sluic-ing appears insensitive to islands:

(67) They want hire someone who speaks a Balkan language, butI don’t remember which.

(68) Every linguist1 argued with a philosopher who took issue with one of his1 claims, butI can’t remember which one of his1 claims. (adapted from Lasnik 2001)

(69) Bob found a plumber who fixed the sink, but I’m not sure with what.

Ellipsis intro 16

7.1.2 Fragment answers

Similar observations have been made for certain fragment answers (in Culicover and Jack-endoff 2005:244ff., Stainton 2006, Valmala 2007).

(70) Is Sviatoslav pro-communist or anti-communist these days?—Pro. [*Pro, Sviatoslav is [t-communist these days.]

(71) A: John met a woman who speaks French.B: And Bengali? [*And Bengali, did John meet a woman who speaks French t?]

(72) Sviatslav is pro-communist and Derzhinsky is anti-.

(73) a. = Did John meet a woman who speaks French and Bengali?

b. = Does she speak French and Bengali?

c. = And does she speak Bengali (too)?

d. = And what about Bengali?

e. = And how about Bengali?

f. 6= And did John also meet a different woman who speaks Bengali (in addition tomeeting the woman who speaks French)?

Casielles 2006 and Stainton 2006 also adduce fragment answer examples out of islandsthat seem quite acceptable.

7.1.3 Gapping

Culicover and Jackendoff 2005:273 also adduce one example,in (74), for which they claimacceptability; to their example I add the attested examplesin (75).

(74) Robin knows a lot reasons whydogsare good pets, andLeslie, cats.

(75) a. He spoke in the kind of tone a lawyer might use to address a jury, or a seriousprofessor of history his students. (Tom McCarthy,Remainder, Vintage: NewYork, 2005, p.236.)

b. If this narrative were a quotidian account of the history of Russia, this chapterwould be a proletarian’s account of the Great October SovietSocialist Revolutionof 1917, if a history of France, the beheading of Marie Antoinette, if a chroni-cle of America, the assassination of Abraham Lincoln by JohnWilkes Booth.(Marisha Pessl,Special topics in calamity physics, Vintage: New York, 2006,p. 311.)

c. No, this was the torturous, clammy kind, when one’s pillowslowly takes on theproperties of a block of wood and one’s sheets, the air of the Everglades. (op.cit.,p. 347.)

17 Jason Merchant

7.1.4 Ellipsis in comparatives

Kennedy and Merchant 2000

(76) a. Brio wrote a more interesting novel than Pico did.

b. *How interesting did Pico write a _ novel?

7.2 Exceptions to the P-stranding generalization

“ [i]n some cases and in some languages, it seems that speakers arewilling toaccept a bare wh-phrase in place of the PP, though I have not yet determinedwith sufficient clarity under what conditions this is possible, or whether or notthis is a systemtic property of a class of prepositions or languages” (Merchant2001:100).

3

(77) a. PietroPietro

hahas

parlatospoken

conwith

qualcuno,someone

mabut

nonnot

soI.know

?( con)with

chi.who

‘Pietro has spoken with someone, but I don’t know (with) who.’

b. * Chiwho

hahas

parlatospoken

PietroPietro

con?with

‘Who has Pietro spoken with?’

But thereare ‘elliptical’ environments where the P-stranding ban is enforced: the rem-nants in gapping and pseudogapping, the counterweight to pseudocleft clauses, fronted CPs,and in sluices withelse-modification (only the latter illustrated here):4

(78) JuanJuan

hahas

habladospoken

conwith

unaa

chicagirl

rubia,blonde

perobut

nonot

séI.know

*( con)with

quéwhat

chicagirl

más.other

‘Juan talked to a blonde girl, but I don’t know (with) what other (kind of) girl.’

pseudosluicing:

(79) a. * JuanJuan

hahas

habladospoken

conwith

unaa

chicagirl

rubia,blonde

perobut

nonot

séI.know

quéwhat

chicagirl

másother

esis

pro.it(lit. ‘Juan talked to a blonde girl, but I don’t know what other (kind of) girl itwas.)’

b. JuanJuan

hahas

habladospoken

conwith

unaa

chicagirl

rubia,blonde

perobut

nonot

séI.know

cualwhich

< esis

proit

>.

‘Juan talked to a blonde girl, but I don’t know which one.’

3Documented in Serbo-Croatian by Stjepanovic To appear, Brazilian Portuguese by Almeida and Yoshida2007, a variety of Romance languages by? and Vicente 2008, Indonesian by Fortin 2007, Polish and others bySzczegelniak 2005 and Nykiel and Sag 2008, and in several languages by van Craenenbroeck 2008.

4Variously, Stjepanovic To appear,?, Vicente 2008, and van Craenenbroeck 2008.

Ellipsis intro 18

(80) ‘Crazy-English’A possible language if the claim that P-stranding wh-movement is independent fromP-less wh-phrases in sluicing were true:

a. Who did she talk to? *To whom did she talk?

b. She talked to someone, but I don’t know { *who | to whom}.

8 Null anaphora and ‘deletion’

Deletion transformation (as in Ross 1969 and Hankamer 1979 among many others) to morerecent proposals (such as the ‘E-feature’ in Merchant 2001,?, Aelbrecht 2009, van Craenen-broeck and Lipták 2006, Toosarvandani 2008, Toosarvandani2009, Vicente 2006,? Corverand van Koppen 2007, and?; see also Johnson 2008).

(81) Amy seemed angry, but we didn’t know[PP at who]1 <she seemed angryt1 >.

(as in van Craenenbroeck and Lipták 2006 and Aelbrecht 2009).

(82) a. Someone murdered Joe, but we don’t know who.

b. CP

who1

C[E] <TP>

t1murdered Joe

(83) a. Abby didn’t see Joe, but Ben did.

b. TP

BenT[E]

did

<VP>

see Joe

(LF-copy approaches like Chung et al. 1995, Valmala 2007)

9 The identity condition on ellipsis

(84) a. Jake ate the sandwich even though his friend told him not to.

b. Jake ate the sandwich even though his friend told him not toeat the sandwich.

(85) “In the meantime, enjoy the ride.”“I am.” (John Updike,Terrorist, Ballantine: New York, 2006, p. 186.)

19 Jason Merchant

(86) A: Pirestook.2s

tinthe

tsandabag.ACC

maziwith

su?you

[Greek]

‘Did you take the bag with you?’B: Yes, I did.

9.1 Semantic identity and information structure

Although the vast majority of the generative research on ellipsis in the years from 1965to the mid 1990s (e.g. Chomsky 1965, Ross 1969, Sag 1976, Hankamer and Sag 1976,Williams 1977, Hankamer 1979, Chao 1987, Rooth 1992, Lappin1992, Fiengo and May1994, Lappin 1996, Chung et al. 1995, and many others) workedwith the assumption that theidentity relation was to be stated over phrase markers (whether D-structure, deep structure,LF, or something else—often, it should be noted,faute de mieux), since the early 1990s evermore proposals have been made that state the identity relation over semantic representations(Dalrymple et al. 1991, Hardt 1993, Hardt 1999, Kempson et al. 1999, Asher et al. 1997,Asher et al. 2001, Ginzburg and Sag 2000, Merchant 2001, Hendriks 2004, Hendriks andSpenader 2005, and many others; perhaps the earliest analysis in this vein is Keenan 1971).

9.2 Syntactic identity

Much more later to come...

9.3 Hybrid theories

include Kehler 2002 (though see Frazier and Clifton 2006 forcritical discussion), Chung2006, Merchant 2007, and van Craenenbroeck 2008.

10 Conclusions

Structural Uniformity An apparently defective or misordered structure is reg-ular in underlying structure and becomes distorted in the course of derivation.(Culicover and Jackendoff 2005:7)

Analytical Uniformity If a certain kind of meaning or use can be made in the ab-sence of syntactic guides to that meaning or use, then syntactic guides are neverneeded for computing that meaning or use. If some device D canrelate a formF and meaning M, then whenever we have M, D is implicated. (If somethingdoesn’t work for everything, it can’t work for anything. If something works forone thing, it can work for everything.)

Ellipsis intro 20

References

Aelbrecht, Lobke. 2009. You have the right to remain silent:The syntactic licensing ofellipsis. Doctoral Dissertation, Catholic University of Brussels, Brussels.

Aelbrecht, Lobke. 2010.The syntactic licensing of ellipsis. Amsterdam: John Benjamins.Almeida, Diogo, and Masaya Yoshida. 2007. A problem for the preposition stranding gener-

alization.Linguistic Inquiry38:349–362.Asher, Nicholas, Daniel Hardt, and Joan Busquets. 1997. Discourse parallelsm, scope, and

ellipsis. InProceedings of the Seventh Conference on Semantics and Linguistic Theory.Asher, Nicholas, Daniel Hardt, and Joan Busquets. 2001. Discourse parallelism, ellipsis, and

ambiguity.Journal of Semantics18.Casielles, Eugenia. 2006. Big questions, small answers. InThe syntax of nonsententials, ed.

Ljiljana Progovac, Kate Paesani, Eugenia Casielles, and Ellen Barton, 117–145. Amster-dam: John Benjamins.

Chao, Wynn. 1987. On ellipsis. Doctoral Dissertation, University of Massachusetts, Amherst.Chomsky, Noam. 1965.Aspects of the theory of syntax. Cambridge, Massachusetts: M.I.T.

Press.Chung, Sandra. 2006. Sluicing and the lexicon: The point of no return. InProceedings of

the annual meeting of the Berkeley Linguistics Society 31, ed. Rebecca T. Cover and YuniKim, 73–91. Berkeley, Calif.: Berkeley Linguistics Society.

Chung, Sandra, William A. Ladusaw, and James McCloskey. 1995. Sluicing and LogicalForm. Natural Language Semantics3:239–282.

Coppock, Elizabeth. 2001. Gapping: in defense of deletion.In Chicago Linguistics Society,ed. Mary Andronis, Christoper Ball, Heidi Elston, and Sylvain Neuvel, volume 37, 133–148. University of Chicago.

Corver, Norbert, and Marjo van Koppen. 2007. Let’s focus on verb phrase ellipsis. Ms.,University of Utrecht, UiL-OTS.

van Craenenbroeck, Jeroen. 2008. What does silence look like? On the unpronounced syntaxof sluicing. Handout, talk presented at the University of Chicago.

van Craenenbroeck, Jeroen. 2010.The syntax of ellipsis: Evidence from Dutch dialects.Oxford: Oxford University Press.

van Craenenbroeck, Jeroen, and Anikó Lipták. 2006. The crosslinguistic syntax of sluicing:Evidence from Hungarian relatives.Syntax9:248–274.

Culicover, Peter W., and Ray Jackendoff. 2005.Simpler Syntax. Oxford: Oxford UniversityPress.

Dalrymple, Mary, Stuart M. Sheiber, and Fernando C. N. Pereira. 1991. Ellipsis and higher-order unification.Linguistics and Philosophy14:399–452.

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

Fortin, Catherine. 2007. Indonesian sluicing and verb phrase ellipsis: Description and ex-planation in a Minimalist framework. Doctoral Dissertation, University of Michigan, AnnArbor.

Fox, Danny, and Howard Lasnik. 2003. Successive-cyclic movement and island repair: thedifference between Sluicing and VP-ellipsis.Linguistic Inquiry34:143–154.

Frazier, Lyn, and Jr. Clifton, Charles. 2006. Ellipsis and discourse coherence.Linguistics

21 Jason Merchant

and Philosophy29:315–346.Ginzburg, Jonathan, and Ivan Sag. 2000.Interrogative investigations. Stanford, Calif.: CSLI

Publications.Goldberg, Lotus Madelyn. 2005. Verb-stranding VP ellipsis: A cross-linguistic study. Doc-

toral Dissertation, McGill University, Montreal.Grinder, John, and Paul M. Postal. 1971. Missing antecedents. Linguistic Inquiry2:269–312.Haïk, Isabelle. 1987. Bound VPs that need to be.Linguistics and Philosophy10:503–530.Hankamer, Jorge. 1979.Deletion in coordinate structures. New York: Garland Publishing,

Inc.Hankamer, Jorge, and Ivan Sag. 1976. Deep and surface anaphora. Linguistic Inquiry7:391–

428.Hardt, Dan. 1993. Verb phrase ellipsis: Form, meaning and processing. Doctoral Disserta-

tion, University of Pennsylvania.Hardt, Daniel. 1999. Dynamic interpretation of verb phraseellipsis. Linguistics and Philos-

ophy22:185–219.Hardt, Daniel, and Maribel Romero. 2004. Ellipsis and the structure of discourse.Journal of

Semantics21:375–414.Hartmann, Katharina. 2000.Right Node Raising and gapping: Interface conditions on

prosodic deletion. Amsterdam: John Benjamins.Hendriks, Petra. 2004. Coherence relations, ellipsis and contrastive topics.Journal of Se-

mantics21:375–414.Hendriks, Petra, and Jennifer Spenader. 2005. Why be silent? Some functions of ellipsis

in natural language. InProceedings of the ESSLLI ’05 workshop on cross-modular ap-proaches to ellipsis, 29–35.

Johnson, Kyle. 1996. In search of the English middle field. Ms., University of Massachusetts,Amherst.

Johnson, Kyle. 2001. What VP ellipsis can do, and what it can’t, but not why. InThehandbook of contemporary syntactic theory, ed. Mark Baltin and Chris Collins, 439–479.Oxford: Blackwell.

Johnson, Kyle, ed. 2008.Topics in ellipsis. Cambridge: Cambridge University Press.Keenan, Edward. 1971. Names, quantifiers, and the sloppy identity problem. Papers in

Linguistics4:211–232.Kehler, Andrew. 2002.Coherence in discourse. Stanford, Calif.: CSLI Publications.Kempson, Ruth, Wilfried Meyer-Viol, and Dov Gabbay. 1999. VP ellipsis: Toward a dy-

namic, structural account. InFragments: Studies in ellipsis and gapping, ed. ShalomLappin and Elabbas Benmamoun, 227–289. Oxford University Press.

Kennedy, Chris, and Jason Merchant. 2000. Attributive comparative deletion.Natural Lan-guage and Linguistic Theory18:89–146.

Kennedy, Christopher. 2003. Ellipsis and syntactic representation. InThe syntax-semanticsinterface: Interpreting (omitted) structure, ed. Susanne Winkler and Kerstin Schwabe.John Benjamins.

Lappin, Shalom. 1992. The syntactic basis of ellipsis resolution. In Proceedings of theStuttgart Ellipsis Workshop, ed. Steve Berman and Arild Hestvik. Stuttgart.

Lappin, Shalom. 1996. The interpretation of ellipsis. InThe handbook of contemporarysemantic theory, ed. Shalom Lappin, 145–175. Oxford: Blackwell.

Ellipsis intro 22

Lasnik, Howard. 2001. When can you save a structure by destroying it? InProceedings of theNorth East Linguistic Society 31, ed. Minjoo Kim and Uri Strauss, 301–320. GeorgetownUniversity: GLSA.

Lasnik, Howard, and Myung-Kwan Park. 2003. The EPp and the Subject Condition underSluicing. Linguistic Inquiry34:649–660.

Lechner, Winfried. 2001. Reduced and phrasal comparatives. Natural Language and Lin-guistic Theory19:683–735.

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

Merchant, Jason. 2001.The syntax of silence: Sluicing, islands, and the theory of ellipsis.Oxford: Oxford University Press.

Merchant, Jason. 2004. Fragments and ellipsis.Linguistics and Philosophy27:661–738.Merchant, Jason. 2007. Voice and ellipsis. Ms., Universityof Chicago.Merchant, Jason. 2008. An asymmetry in voice mismatches in VP-ellipsis and pseudogap-

ping. Linguistic Inquiry39:169–179.Morgan, Jerry L. 1973. Sentence fragments and the notion ‘sentence’. InIssues in linguistics,

ed. Braj Kachru, Robert Lees, Yakov Malkiel, Angelina Pietrangeli, and Sol Saporta, 719–751. Urbana: University of Illinois Press.

Nykiel, Joanna, and Ivan Sag. 2008. Sluicing and stranding.Ms., University of Silesia andStanford University.

Postal, Paul. 2001. Parasitic and pseudoparasitic gaps. InParasitic gaps, ed. Peter Culicoverand Paul Postal, 253–313. Cambridge, Massachusetts: MIT Press.

Pullum, Geoffrey K. 2000. Hankamer was! InJorge Hankamer WebFest, ed. Sandra Chung,James McCloskey, and Nathan Sanders. http://ling.ucsc.edu/Jorge/: UCSC Linguistics De-partment.

Reich, Ingo. 2008. Ellipsis. InHandbook of semantics, ed. Klaus von Heusinger, ClaudiaMaienborn, and Paul Portner. Berlin: Mouton de Gruyter.

Reinhart, Tanya. 1991. Elliptic conjunctions – non-quantificational LF. InThe chomskyanturn, ed. Asa Kasher, 360–384. Cambridge, Massachusetts: Blackwell Publishers.

Rooth, Mats. 1992. A theory of focus interpretation.Natural Language Semantics1:117–121.

Ross, John Robert. 1969. Guess who? InProceedings of the Fifth annual meeting of theChicago Linguistics Society, ed. Robert I. Binnick, Alice Davison, Georgia M. Green, andJerry L. Morgan, 252–286. Chicago, Illinois.

Sag, Ivan. 1976. Deletion and logical form. Doctoral Dissertation, Massachusetts Instituteof Technology, Cambridge, Massachusetts.

Sag, Ivan, and Jorge Hankamer. 1984. Toward a theory of anaphoric processing.Linguisticsand Philosophy7:325–345.

Stainton, Robert. 2006.Words and thoughts. Oxford: Oxford University Press.Stjepanovic, Sandra. To appear. “P-stranding” under sluicing in a non-P-stranding language.

In Sluicing: Cross-linguistic perspectives, ed. Jason Merchant and Andrew Simpson. Ox-ford: Oxford University Press.

Szczegelniak, Adam. 2005. All sluiced up, but no alleviation in sight... Ms., Boston College.Toosarvandani, Maziar. 2008. Wh-movement and the syntax ofsluicing.Journal of Linguis-

tics44:677–722.

23 Jason Merchant

Toosarvandani, Maziar. 2009. Ellipsis in Farsi complex predicates.Syntax12:60–92.Valmala, Vidal. 2007. The syntax of little things. Presented at the 17th Colloquium on

Generative Grammar, Girona.Vicente, Luis. 2006. Negative short replies in Spanish. InLinguistics in the Netherlands

2006, ed. Jeroen van de Weijer and Kees Los, 199–211. Amsterdam: John Benjamins.Vicente, Luis. 2008. Syntactic isomorphism and non-isomorphism under ellipsis. Ms., Uni-

versity of California, Santa Cruz.Webber, Bonnie. 1978. A formal approach to discourse anaphora. Doctoral Dissertation,

Harvard University.Williams, Edwin. 1977. Discourse and logical form.Linguistic Inquiry8:101–139.Winkler, Susanne. 2005.Ellipsis and focus in generative grammar. Berlin: Mouton de

Gruyter.Winkler, Susanne, and Kerstin Schwabe. 2003. Exploring theinterfaces from the perspective

of omitted structures. InThe interfaces: Deriving and interpreting omitted structures, ed.Susanne Winkler and Kerstin Schwabe, 1–26. Amsterdam: JohnBenjamins.

Zagona, Karen. 1982. Government and proper government of verbal projections. DoctoralDissertation, University of Washington, Seattle.