What co-varies in backward variable binding? Keir Moulton McSIRG 13 May 2010

The game plan First, I want to argue that some cases of backwards variable binding are real, in the sense that they don’t involve reconstruction or “logophoric” interpretations. Then I want to examine what is special about the verbs that allow backward binding. It seems limited to causatives, with psych-verbs (the most famous instance of backward binding) just being a sub-case (Arad 1998; Pesetsky 1995). (1)

Backward Binding (adapted from Pesetsky 1995, (124a)) a. The remark (made) about his1 hair made every boy1 angry. b. Remarks about his1 hair made every boy1 angry.

My main question today is: What co-varies (as a function of the quantifier) in backward binding? Intuitively, what co-varies is not just the pronoun, but the whole subject that contains the pronoun: for every boy, there are different remarks: (2)

≈ For every boy1 there were different remarks about his1 hair that made him1 angry.

Even so, this intuitively simple approach (letting the QP QR) involves a weak cross-over violation. If it involves QR to bind the pronoun, why is it not as bad as: 1

(3)

Weak Crossover *His1 mother loves every boy1

And why doesn’t embedding the pronoun into a subject that itself co-varies make (3) better: (4)

*A friend of his1 mother loves every boy1 . 6= For every boy1 there is a friend of his1 mother that loves him1

What I want to try out—very tentatively, because this I am amateur in these matters—is that the examples in (1) really do involve QRing the QP to get covariation for the pronoun (or give the appearance of a “bound variable”). What distinguishes (1) from (3)/(4) is the nature of the subject in which the “bound pronoun” is part (I’ll call this the Pronoun Container): a WCO violation doesn’t arise if the Pronoun Container is or is part of a subject that describes situations and those situations co-vary wrt to the quantifier. I then want to argue that this allows the pronoun to be interpreted as a D-type pronoun. So we want to try out for (1) something like: (5)

a. b.

The remark about his1 hair made every boy1 angry. ≈ For every boy, there is a (causing) situation s in which there were remarks about THE BOY in s’s hair and s made the boy angry.

The idea with D-type pronouns (or E-type) is that they involve hidden determiners with elided NP content Elbourne (2005). So THE BOY’S = his. (We’ll need to make sure the causing situation contains just one boy so it is the unique boy. This is rough and the details of situation semantics will have to wait.) In (3)/(4), the Pronoun Container isn’t part of something that describes situations that co-vary wrt to the quantifier.

2

1 Backwards variable binding Causative/Psych Backward Binding One of the first and most famous places where backwards binding came up involved psych verbs (Belletti and Rizzi 1988). Often in the literature, examples were given with reflexives and reciprocals, not variables bound by quantifiers: (6)

a. Each other’s supporters worreid Freud and Jung. b. Each other’s remarks annoyed John and Mary. c. *Each other’s parents harmed John and Mary. d. *Each other’s teachers insulted John and Mary. (Pesetsky 1995, ex. 121b, 122, 127b,c)

But we can show that the same holds of bound variable interpretations: (7)

a. His1 supporters worried every psychiatrist1 . b. False descriptions of his1 life annoyed every person1 here. c. *His1 supporters harmed every psychiatrist1 . d. *False descriptions of his1 life insulted every psychiatrist1 .

That bound variable interpretations are possible will be important in a bit. The old line, of course, is that this “binding” meant that the argument structure of psych-verbs is such that at some level the theme is lower than the experiencer (Belletti and Rizzi 1988). Lots of people point out that backward binding is possible in a wider set of cases, such as causatives: (8)

a. b. c. d.

Each other’s remarks made John and Mary angry. Pictures of himself give John the creeps. Pictures of each other caused John and Mary to start crying. Each other’s criticism forced John and Mary to confront their problems. (Pesetsky 1995, ex. 124a,d, 125a,b)

Again, variable binding is possible: (9)

a. His1 supporters made every psychiatrist1 angry. b. False descriptions of his1 life gave every person1 the creeps. c. ?His1 bad behavior caused lasting harm to every psychiatrist1 . d. Damaging descriptions of his1 techniques forced every psychiatrist1 to address his problems.

So it’s not just psych verbs that show backwards binding; most suggest that psych-verbs should be considered a special case of causatives. 3

1.1 Not reconstruction As noted, one line on backward binding is to suggest that bound variable interpretations arise because of reconstruction, which means these surface subject must move from a position below the binder. (I.e. the analysis of Belletti and Rizzi (1988) for psych-verbs; see also Pesetsky (1995)). Not a possible solution for (exempt) reflexives and reciprocals because binder doesn’t have to c-command (Pollard and Sag 1992, Canc¸ado and Franchi 1999, Pesetsky 1990, and Iwata 1995). (10)

The picture of himself in Newsweek shattered the peace of mind that John had spent the last six months trying to restore. (Pollard and Sag 1992, ex. (62c))

On any attempt to put the subject below the causee (the peace of mind that . . . still fails to get the reflexive c-commanded by John. And with variable binding: (11)

The picture of his1 ugly mug in Newsweek shattered the confidence that every boy1 had spent the last six months trying to restore.

So messing around with the argument structure—and using reconstruction to get the right c-command relations—won’t help. (See Landau (2009) on what happens once the backward anaphora facts are taken out of the psych-verb literature.)

1.2 Logophors Another tradition recognizes that there is no c-command relation in backward binding at any level of representation, but suggests that the reflexives and reciprocals are “logophoric” (Iwata 1995, Sato and Kishida 2009, Landau 2009, among others). It is often suggested that perspectival or logophoric notions capture backward binding (see e.g. that Zribi-Hertz (1989). Note for instance that we are saying something about the experiencers state of mind in psychreports and some causative cases (see below on this). But does “logophoricity” give rise to co-variation?

4

(12)

a. b.

The picture of his1 assailant in Newsweek shattered the peace of mind that every guy1 had spent the last year trying to restore. Remarks about his1 hair are apt to make every boy1 angry.

It is true of course that there is certainly something to the idea that wider discourse level constraints are at work (why not!) but it is not clear whether that has to do with reflexives in particular or binding in general. The following clearly show some contrast between happy and famous, which might relate to point-of-view (or whatever): (13)

a. Pictures of herself1 made Ruth1 happy. b. *Pictures of herself1 made Ruth1 famous. (reported in Sato and Kishida 2009)

(14)

a. That book about herself1 struck Mary1 as embarrassing. b. *That book about herself1 struck Mary1 on the head. (Bouchard 1992: 295)

But does point-of-view (or whatever is going on here) interact with variable binding? Not so clearly: (15)

a.

Nude photos of him1 are not likely to make any actor1 here that much more famous. b. ?Nude photos of himself1 are not likely to make John1 that much more famous.

(16)

a. ?Several books about his1 company struck every business man1 on the head. b. *Several books about himself1 struck every business man1 on the head.

Here are some other cases where I don’t detect point-of-view being at play: (The more deeply embedded the pronoun, the better the WCO—Sportiche[1996]) (17)

a. b.

His1 poor background shouldn’t prevent any of these boys1 from being considered. The fact that he1 was not in the room doesn’t suggest/prove that any of these boys1 is innocent.

But even if there is a component of backward binding that involves notions like “logophoricity”/point of view (and it’s not clear there is for variable binding), we still have to get co-variation for the pronouns and that seems to require binding by the quantifier. 5

1.3 Summary So co-variation of the pronoun—as far as I can tell—requires the quantifier to scope high.

But then we should get a WCO violation! Interestingly, Pica and Snyder (1995) argue that really there is no WCO violations, just cases where it is really difficult to get wide scope. They show evidence that the badness of WCO is directly proportional to the availability of wide-scope for the quantifier over the position the bound pronoun is in. Pica and Snyder cite Snyder’s own experimental work in favor of this correlation. (I’ve been leaving aside wh-movement cases.) All data from Pica and Snyder (1995), p. 3-4, (3)-(8). (18)

Double object datives a. *John gave someone everything (Wide Scope on everything) b. *John gave its1 owner every paycheck1 c. *What1 did John give its1 owner t1 d. *John gave his1 own master Fido1

(19)

Perceptual reports a. *Mary saw someone greet everyone (Wide Scope on everyone) b. *Mary saw his1 host greet everyone1 c. *Who1 did Mary see his1 host greet d. *Mary saw his1 own host greet Ted1

(20)

Monotransitives a. ??Someone likes everyone (Wide Scope on everyone) b. ??His1 mother likes everyone1 c. ??Who1 does his1 mother like t1 d. ??His1 own mother likes Ted1

(21)

Prepositional datives a. Mary gave something to everyone (Wide Scope on everyone) b. ?Mary gave his1 paycheck to everyone1 c. ?To whom1 did Mary give his1 paycheck t1 d. ?Mary gave his1 own paycheck to Ted1

(22)

Put-locatives a. Mary put something on every box (Wide Scope on every box) 6

b. ?Mary put its1 label on every box1 c. ?On what box1 did Mary put its1 label d. ?Mary put its1 own label on the box1 So it’s not a new idea idea that both (23-a) and (23-b) involves crossover but that—for some reason—this is better in one and not the other. (23)

a. *His1 mother loves everyone1 . b. His1 mother should make everyone1 leave.

Is it true that inverse scope is better in the (b) case than the (a) case? (24)

a. b.

Someone loves everyone. Someone should make everyone leave.

I am not sure, and I find it hard to accept Pica and Snyder: wide scope might be hard in some cases, but the weak-crosser over is almost impossible. We need to find a difference between causative subject (incl. psych verb themes) and other kinds of subjects such that binding is possible in one and not the other:

2 Speculation #1 The difference between (23-a) and (23-b) has to do with whether the subject argument of the verb is one that can be fulfilled by an event or situation— rather than just an ordinary individual. This is true for causative constructions (a) and psych constructions (b): (25)

a. b.

{That event/Mary’s losing/the rain/the destruction } made/caused/forced /gave-cause-for everyone leave. {That event/Mary’s losing/the rain/the destruction } worried/amused/ frightened/occupied/angered everyone.

And this is not true for many run of the mill transitive verbs—those that show WCO violations.

7

(26)

*{That event/Mary’s losing/the rain/the destruction } loved/hit/. . .

Wild speculation: suppose that what co-varies in (successful) backward binding is a subject that denotes a situation. This means that even if the subject looks like a DP, it is part of a situation that is the causer (this is not intended as a serious analysis of causatives, but gets my hunch across): (27)

Nude photos of him1 are not likely to make any actor1 here that much more famous. ≈ For no actor x is there a situation s, s.t. s is (or contains) nude pictures of x, s.t. s is likely to make x famous.

But how does this help us capturing when the variable can be bound or not?

3 Speculation #2 Backwards bound variables are really D-type pronouns. They co-vary as a function of the situations they are in. D-type pronouns (aka E-type pronouns): pronouns that get a co-varying interpretation from non-commanding quantifiers: (28)

Every boy who owns a donkey beats it.

A recent and popular account is Elbourne (2005), who argues that the pronoun is really: (29)

Every boy who owns a donkey beats the donkey.

It takes a situation semantics (where uniqueness of the donkey is ensured) to capture the right interpretation, which I can’t implement right now for my cases. But: one of Elbourne’s strategies is to show that you can get co-varying interpretations of definites in those cases where a D-type analysis applies.

8

(30)

a. b.

If John own a donkey, he beats it/the donkey EVery farmer who owns a donkey, beats the donkey.

Buring (2005) argues that Binding out of DP (BOOD) involves a D-type pro¨ noun; and here we can get a dependent definite (at least as well as Buring’s BOOD cases): (31)

a. b.

Every boy1 ’s mother loves him1 . Every boy’s mother loves the boy.

Of course this isn’t wonderful, but it’s not bad. Now we can see that all those cases that involve backward binding work with the pronoun replaced by a dependent (i.e. co-varying) definite: (32)

a. b.

Pictures of the boy amused every boy Pictures of the boy amused every boy’s mother

(33)

a. b.

Pictures of the boy made every boy famous. Pictures of the the boy made every boy’s mother famous.

(34)

a. b.

the pictures they took of him made every boy famous the pictures they took of the boy made every boy famous

(35)

Having the boy’s picture taken like that made every boy famous

(36)

a. b.

The fact that the boy was in trouble puts every boy on the suspect list. cf. What puts every boy on the suspect list is the fact that the boy was in trouble.

4 Crossover and D-type pronouns D-type pronouns show WCO effects (Buring 2005) (which is to say, you can’t ¨ backward bind—here at least—a D-type pronoun): (37)

a. Every knight who courted a lady visited herD −type mother. b. *HerD −type mother visited every knight who courted a lady. (Buring 2005: 25(45a)) ¨ 9

So if the co-varying pronoun in (1) and backward binding in general involves a D-type pronoun, then we need to know why it doesn’t give rise to crossover but Buring’s kind does. The relevant difference—at least on my speculations— ¨ is that the “container” for the D-type pronoun is not in a phrase that describes a situation that co-varies. *

References Arad, Maya. 1998. VP-Structure and the syntax-lexicon interface. Doctoral Dissertation, University College London. Belletti, Adriana, and Luigi Rizzi. 1988. Psych verbs and theta-theory. Natural Language and Linguistic Theory 6:291–352. Bouchard, Denis. 1992. Psych constructions and linking to conceptual structures. In Romance languages and modern linguistic thoery, ed. Paul Hirschbuhler and ¨ Koerner Konrad, 25–44. John Benjamins. Buring, Daniel. 2005. Bound to bind. Linguistic Inquiry 36:259–274. ¨ Canc¸ado, M´arcia, and Carlos Franchi. 1999. Exceptional binding with psych verbs? Linguistic Inquiry 30:133–143. Elbourne, Paul D. 2005. Situations and individuals. MIT Press. Iwata, S. 1995. The distinctive character of psych-verbs as causatives. Linguistic Analysis 25:95–120. Landau, Idan. 2009. The Locative Syntax of Experiencers. Cambridge: MIT Monographs in Linguistics. Pesetsky, David. 1990. Experiencer predicates. Ms. MIT. Pesetsky, David. 1995. Zero syntax. Cambridge, Massachusetts: MIT Press. Pica, Pierre, and William Snyder. 1995. Weak crossover, scope, and agreement in a minimalist framework. In 13th West Coast Conference on Formal Linguistics, ed. Raul Aranovich, William Bryne, Susanne Preuss, Martha Preuss, and Senturia, 334–349. CSLI, Cambridge University Press. Pollard, Carl, and Ivan A. Sag. 1992. Anaphors in English and the scope of binding theory. Linguistic Inquiry 23:261–304. Sato, Yosuke, and Maki Kishida. 2009. Psychological predicates and the pointof-view hyperprojection. Gengo Kenkyu 135. Zribi-Hertz, Anne. 1989. Anaphor binding and narrative point of view: English reflexive pronouns in sentence and discourse. Language 65:695–727. 10

What co-varies in backward variable binding?

May 13, 2010 - that every boy1 had spent the last six months trying to restore. So messing around with the ... All data from Pica and. Snyder (1995), p.

69KB Sizes 0 Downloads 126 Views

Recommend Documents

What Co-varies in Backward Variable Binding
rn XP ] ] b. [ rn XP ]g = λxλs.[ XP ]g[n→s](x)(s) c. DP must occupy an A-position (rules out WCO). (15) a. ..... tation, University College London. Barker, Chris. 2008.

What co-varies in backward variable binding?
May 13, 2010 - The game plan. First, I want to argue that .... *Mary saw his1 host greet everyone1 c. *Who1 did Mary see his1 host greet d. *Mary saw his1 own ...

Variable binding, reconstruction and attitudes de re
The minimalist program. Cambridge, Massachusetts: MIT Press. Cresswell ... University of Massachusetts at Amherst: Graduate Linguistic Student Association.

backward control in samoan
silent one c-commands the overt one without inducing a Condition C violation, .... The following simplified tree summarizes my proposal. (3). VP. DP. Seui. V'. V.

backward control in samoan
A key element of my proposal is that the Possessor-DP—let us call it β—in PAGO ... PAGO reading to emerge (in this sense, my account, which posits two copies one of ...... some speaker variation: although my three consultants access PAGO.

Feature binding in zebrafish
fish may not support this ability at all, particularly in view of the current notion ...... a function. Philosophical Transactions of the Royal Society B, 360, 837e862.

binding site
3: College of Computing, Georgia Institute of Technology, USA .... Systems Bioinformatics CSB2007, 13-17 August 2007, University of California, San Diego.

Shallow Binding in Lisp 1.5 - Pipeline
computer languages (with the major exception of APL) use a lexical scoping rule in ... , where the redundant variable-name simplifies programs presented ..... Knuth, D. The Art of Computer Programming, Vol.

ELA backward design.pdf
distinguish what the purpose and the theme are. Having a POV helps create clarification of a. character's view of the text whether it is mood,. setting, and goals.

ePub Looking Backward Full Online
explains the answers using various methods, such as metaphors or direct comparisons with 19th-century society.Although Bellamy's novel did not discuss.

LATENT VARIABLE REALISM IN PSYCHOMETRICS ...
Sandy Gliboff made many helpful comments on early drafts. ...... 15 Jensen writes “The disadvantage of Spearman's method is that if his tetrad ..... According to Boorsboom et al., one advantage of TA-1 over IA is that the latter makes the ...... st

LATENT VARIABLE REALISM IN PSYCHOMETRICS ...
analysis, structural equation modeling, or any other statistical method. ...... to provide useful and predictive measurements, the testing industry will retain its ...

Supporting Variable Pedagogical Models in Network ... - CiteSeerX
not technical but come from educational theory backgrounds. This combination of educationalists and technologists has meant that each group has had to learn.

Supporting Variable Pedagogical Models in ... - Semantic Scholar
eml.ou.nl/introduction/articles.htm. (13) Greeno, Collins, Resnick. “Cognition and. Learning”, In Handbook of Educational Psychology,. Berliner & Calfee (Eds) ...

Binding problem
The fact that natural perception is usually multimodal (events of the world activate multiple sensory modalities) and the consequent development of multimodal interfaces raises the problem of how complex multimodal perceptual units are formed. The po

Squat Fix: Falling Backward - CrossFit
But why does it happen? And how can it be fixed? The Causes. Pat Sherwood, a Certified CrossFit Coach (CF-L4) and former member of CrossFit's Level 1 ...

Variable selection in PCA in sensory descriptive and consumer data
Keywords: PCA; Descriptive sensory data; Consumer data; Variable selection; Validation. 1. Introduction. In multivariate analysis where data-tables with sen-.

Backward Mapping: Exploring Questions of ...
county polygons, we have a new map with five district polygons. Once we formed the new boundary files for each state, we merged them all together to create a ...

Variable selection in PCA in sensory descriptive and consumer data
used to demonstrate how this aids the data-analyst in interpreting loading plots by ... Keywords: PCA; Descriptive sensory data; Consumer data; Variable ...

Ubiquitin-binding domains
At least one of these domains, the A20 ZnF, acts as a ubiquitin ligase by recruiting a ubiquitin–ubiquitin-conjugating enzyme thiolester adduct in a process that ...... systems. Nat. Cell Biol. 2, E153–E157. 5 Miller, J. and Gordon, C. (2005) The

Variable Compleja.pdf
Page 1 of 6. INSTITUTO POLITECNICO NACIONAL. SECRETARIA ACADEMICA. DIRECCION DE ESTUDIOS PROFESIONALES. ESCUELA:SUPERIOR DE ...