Syntax II - Class IV

Class 4

29a may not be too bad. In this example, degredation very slight. By stipulation. CP is a barrier for t’’_I by stipulation. Though not a blocking category.

Need say nothing about going from t’’I to t’’’i, since already have barrier. N’.

Minimality only prevents government. For any governee, have only one governor.

But need not worry about intermediate traces having gamma marking since gone by LF. But cannot appeal to N’ ‘s minimality barrier because only work for government, not for movement.

In the assignment, how did Bill get the impression that Bill did it. In that case, the N’ will be a barrier for t’ (some number of primes), if we don’t say anything more. The ECP being straightforwardly violated on account of a minimality violation. We said about the CP in the complement of a noun that it is a barrier by stipulation. We needed to say that to create a subjacency effect in another instance (to derive 29a). CP in N complement is a barrier, by stipulation.

Is N’ still a minimality barrier. No. Minute say CP is a barrier, then N cannot govern t’’. Then nothing governs t’’. Notice it is an interesting give/take. Cannot simultaneously say that N is a minimality barrier and that CP is a barrier.

What is left, then, for minimality? To take care of that-trace violations, just as it did in Syntax I. Pg 10 in handout, who do you think that did it?

Only think that changes is now we can put finger on the node that is a barrier – the C’.

1) Why does it improve when drop complementizer = when not overt? Why isn’t C’ null a minimality barrier. Page 10, Chomsky …

[CP t’I [C’ that [IP t’ ] ]

[CP t’I NULL [IP t’ ]

then have C but C’ is no longer present.

But what would the tree going to be? If C head and CP, followed by IP, must be ternary branching. Could also say that just drop C, but then CP comes out of nowhere – the CP has no head. Perhaps Could instead use VP adjunction rather than CP, anf have verb select IP (now) with a finite Infl. But if have CP, then MUST have C’ and a C-head. Can just say that cannot have ampty headed governor.

Let us consider possibility of completely immunizing derivation to devastating influence of the complementizer. Barss: it was derivation #3 in our assignment. In which first lower from ti1 to ti2 and from there out. This would seem to succeed based on rules set out in Barriers. Can work assuming m-command.

Elsewhere, m-command gets Chomsky into trouble.

AP -> PRO A’

A’ -> A(nude)

John ate the meat nude.

On m-command, PRO seems to be governed!

Another context

C’ -> C(null) IP

IP -> PRO I’

I’ -> I somethingelse

As in John tried to win the race. Have to state “Infinitival is not a governor.”

With strict- c-command, cannot get out of I or out of A.

And here as well, the strictest c-command derivation will straightforwardly rule out Barss.

In this instance, t2 would be closer, because t2 needs pass through 0 max projections, which that must pass through 1.

What about Infl. Shouldn’t that be a closer governor? So problem shifts from C’ to I’ as minimality barrier.

Finally, the propriety between t2 and t3 in A’ positions across a trace in an A position. Looks like improper movement structurally.

So far so good, we had just said that the I’ could be a minimality barrier. But then if think further, if we would allow I’ to be a minimality barrier, we would be forbidding ALL adjunct movement, by limiting the ability of traces to govern.

Let us look at case of Adjunct Extraction.

[CP t’’i [C’ C [IP subject [I’ I [VP t’i VP

and there is a straightforward, unblockable, government relationship of Infl governing ti’

So that makes I’ a Minimality Barrier. So ECP crash! So I’ mustn’t be a minimality barrier. Do we know why it is not?

Thus, stipulation 37 of Chomsky: I’ is never a minimality barrier.

And also said IP never an inherent barrier. And IP must not be intermediately adjoined to.

Pre-barriers, IP/S had V as head, and was ternary branching. So extended X’ branching to Cp and IP. So emancipation of IP has failed.

One other context in which Barriers encounters major obstacles - with VP and object extraction.

t'’’ in spec VP, VP, V’, which has V, CP. Spec-CP has t’’. So V should be closer governor, and V’ as a minimality barrier to government by t’’’

so still have a problem, but this time around Infl is not the culprit.

Now page 11 in the handout. The X’ level is optional – only required when a specifier in XP projected.

Then, VP will be the minimality barrier. But irrelevant, because antecedent t’’’ is within this barrier, because it is within one segment of the VP.

What consequences are there of forgoing of the V’ to the base generation of the subject? We cannot base-generate it in VP, or else we would have a specifier and then we could not forgo the V’.

Is there a way we can base generate in an additional adjoined VP.

This is strange move on Chomsky’s part because based generates everything else there. I saw John sneeze, John in VP. Always subject in VP except where it is the complement of Infl.

Pg 12, between the diamonds. NB. How to distinguish between A and A’ positions.

If Eliminate C’ CP (and in general XP X’) distinction. But then no C’, so no minimality, since have Cp-CP as barrier. Then that-trace effects must follow from something else.

Rigid minimality – useless.

Ricci: relativized minimality. In rigid, only heads that govern. For Ricci, minimality relativized in terms of nature of alpha. If alpha is a head, and b is a head, only a head delta can intervene

a [ d [ b

if a and b are phrases, only delta phrase can intervene

within phrases:

if alpha and beta, with alpha in A position, blocked by delta in A position.

if alpha and beta, with alpha in A’ position, blocked by delta in A’ position.

This is summed up at bottom of pg 11.

Pg 12. relativized minimality formalizes position that intervener should have something in common with what is blocking.

Then all problems had disappear.

We don’t get that-trace effect from Ricci.

[CP t’ [C’ C(that) [IP t

“that” is a head, while t’ in A’ position. Ricci gets this is a different way.

If heads could never antecede traces of phrases, OF COURSE you would not think they could intercede. But is one context in which heads DO actually antecede traces of phrases.

Must ascertain that independently. That is what we will discover next time.

Please read NP movement and *** minimality, trace-licensing govt and minimized chain links, and if have time, Interaction of NP movement an Head Movement I. (that is, for

Some helpful suggestions for drawing trees with the program

I've starting using that online program to generate the trees. I have what I think are some helpful suggestions.

Firstly, to generate triangles under an NP where you do not want to draw the internal structure - this will happen automatically if you do not use brackets and have more than one element. What if you have a single word, like How_i? I would suggest is use the period beforehand. Thus:
. How_i

[NP . How_i]

Secondly, since these labelled brackets get out of control, since there are so many of them, I strongly suggest you use indentation to make things clearer. This is the standard strategy used by computer programmers. Thus, without giving away too much of the homework, the top of my labelled brackets looks something like this:

[CP [NP . How_i]
    [C' [C  [did_x] ]
        [IP [NP . John]
            [I' [I t_x]
                VP
            ]
        ]  
    ]
]

which looks much better than
[CP [NP . How_i] [C' [C [did_x] ] [IP [NP . John] [I' [I t_x] VP ] ] ]]

in the indented example, the end bracket is aligned with the opening bracket, the first daughter of a node is placed on the same line as the mother, and the second child is placed on the next line, indented over so that it is aligned with its sibling.

Using this methodology should make the act of writing and understanding labelled brackets much simpler and error-prone.

Syntax II - Class III

10c - now, even crossing one barrier gives you violation of subjacency.

last week, we abstracted entirely away from the VP, ujntil the end, at which p[oint realized that VP will cause us major trouble.

Whati [IP Bill [VP read ti]

here, Whati must cross VP and IP, so must not be any barriers on movement path, but looks as if there is in fact *two*. VP is both a blocking category and a Barrier. IP becomes both a blocking category and it inherits barrierhood fropm VP. So crosses *two* barriers.
Yet the sentence is perfectly grammatical. Must first LAND on edge of VP so that no longer dominated by VP and then move from there. That is:

Whati [Bill I [VP ti' [VP read ti

ti to ti'. ti' to Whati.

segments (these partial VPs) are never categories, and only categories may be barriers.

The sum total of VP (together) are a blocking category and barrier for ti. But antecedent of ti, that is ti', is not *excluded* by the VP, since it is partially *inside* the VP.

Similarly, ti' is not dominated by the upper VP, since does not entirely dominate it. Thus not a barrier to the antecedent Whati. The IP is a blocking category but presence of blocking category not sufficient to pose a problem for subjacency. The IP is not a barrier.

If had tried to get ti out in one fell swoop, would have had problems.

(Could have otherwise simply claimed that the VP is L-marked. "VC is not a BC or barrier for anything it dominates because VP is L-marked." This is not what Chomsky said. We would need to investigate 2 important ingredients:
1) I theta-marks VP.
2) I has to be or become lexical.
The idea that I theta-marks it VP complement is not a bad idea. It is in fact important point in the theory. Chomsky
John said that he would fix the car. and
"Fix the car, he did"
[VP fix the car]i he did ti

the trace must be properly governed.
in this particular governed, could make it antecedent governed. but there are other exmaples of "VP topicalization" that give you the feeling of subjacency effect and not an ECP effect which suggest theta-governed within its base position.

?? [VP fix the car]i, I wonder whether he will ti

not an ECP violation. therefore ti must be properly governed. can do antecedent or theta-governed. Can it be properly antecedent governed? No, because *whther* will definiately be in between. No point in adjoining to the VP because we are *moving* the entire VP. So cannot perform any intermediate landing, must go in one fell swoop across the CP. So *whether* should certainly give us a minimality effect. Subjacency violation like in 10c. But not so bad, because actually theta-governed. Not by "he." Will have to be "will," which is an I. Apparently I assigned theta-role and allows trace of fronted VP to satisfy the ECP such that it need not find a local antecedent.

So (1) does in fact seem to be true.
But (2), I does NOT seem to be lexical. Because not able to be generated by feature +-V and +-N. So I cannot BE lexical. But can *become* lexical by something lexical moving into it.
If want to add adverbial "probably" there, get "John read probably it."
Aux be (=copular be) and have. but modal verbs not. so would not be able to leave the VP for I. Perhaps as some point "read" moves from VP to I, and then retroactively marks the VP.

Whati [IP Bill I [VP ti' probably [VP read ti]

in languages where this is allowed, probably most likely raises to I. (ignoring comma intonation)
Perhaps in Lf syntactic component, read moves to I? without getting any phonological reflex of it.
Then at LF, would have lexical element, and at LF, would not be a barrier. But extracted the VP earlier than LF, at S-structure. If subjacency condition thought to be condition of LF representations, then OK, so long as L-marked by the end of LF. Generally, we say this is a condition on S-structure, overt representation. And if so, cannot undo the subjacency violation. Could imagine ECP violations like this, but not subjacency.
But if subjacency to be rethinkable as condition on LF, then we could in fact raise the "read" and lift the barrierhood.

In revisiting of theory in Lasnik and Chomsky in black book, "it is sufficient for a node to be selected for it to be a barrier." But it is essential for VP to be a barrier.

[Digression: Why does Chomsky need VP to be a barrier? Section 10 of the greenish-bluish book towards the end.

based on particular parasitic gap construction

Whoi ti [VP warned the men [CP Opi that they were about to arrest ti]
This is out. parasitic gap construction. two chains. Null Opi -ti chain which must be composed with "real" chain Whoi-ti, and the connection between the two chains must be established under zero subjacency = no barrier between the two chains.
If VP were not a barrier, then would be no barrier between the two chains (because CP is not a barrier here since it is L-marked). In order to be correctly ruled out, we have to assume that VP is in fact a barrier.

In 1993, forgets about this reason, for he resorts to simpler reason in terms of selection, such that VP not a barrier.
]


But is there any reason to assume this intermediate adjunction? Look at (21). 21a is unambiguous. 21b is ambiguous since John may be bound by himself.
At D-structure, himself can be antecednce governed by BIll. But for (b), need to

[how many pics of himselfi/j]k Johni [CP t'k Billj tk]
is extracted successive cyclicaly

"reconstructing" [how many pic of himself] back into initial trace ti to get Bill reading, or can reconstruct into John, at which point it we can get the John reading.
do this "reconstruction" at LF, but *only* if there was ti' where ti landed during movement. So in order to ensure the ambiguity, we must have that intermediate trace. It can take the matrix subject as its antecedent.

Interestingly, sentences of this type are grammatical if the Spec CP position occupied by WH phrsae

?[how many pics of himselfi/j]k Johni [CP WH Billj tk]

which gives us a subjacency violation, just mildly degraded, plus they are *still degraded.*
In 22b, which is minimal variant of 21b, somewhat degraded because of subjacency, but importantly it is still ambiguous.
If Bill was the only possibility and not ambiguous, Syntax I would have taken care of this.

How do we get the John reading in 22b. Not from a trace in post of spec Cp. We do get it on a more densely structured tree. There is a t'l in the embedded VP and a t''k adjoined to the matrix VP.

?[how many pics of himselfi/j]k Johni [VP t''k [VP [CP WH Billj [t'k tk]

The two traces below Bill will give is the Bill reading. But to get John reading, can reconstruct to t''k. But to do this it is *essential* we do intermediate adjunction to VP.

page 8 on the handout.

this sugegcts strongly that intermediate adjunction is in factg the case.

Of course 22b does violate subjacency condition in going from tk' to t''k. IP a BC. CP is l-marked so not BC, but inherits barrierhood from IP. VP we need not worry about because t''k in VP. So get weak subjacency violation.

But, Question. Why could not intermediately adjoin to IP before extracting out of IP and then move to VP. In notes as well.

[VP t'''k [VP V [CP WH [IP t''k [IP Su I [VP t'k [VP ... tk

another possibility is to adjoin to *CP* on the way up:

[VP t'''k [VP V [CP t''k [CP WH [IP Su I [VP t'k [VP ... tk

neither of which would give us a subjacency violation. how so:

[VP t'''k [VP V [CP WH [IP t''k [IP Su I [VP t'k [VP ... tk

tk we need not worry about because only separated by segment of VP.
the IP together is BC, but t''k no problem because within IP.
t''k only dominated by segment of IP. CP is l-marked, and not BC. and CP cannot inherit barrierhood from IP in this case. and VP in total dominates, but t'''k is within VP.

If so, we are in dire straits, because we do want to show that spec, CP being occupied gives us a WH-island violation.

Therefore, we have to *declare* this derivation illigitimate, because of something else. Otherwise don;t understand why violates subjacency.
Similarly for the second derivation.

[VP t'''k [VP V [CP t''k [CP WH [IP Su I [VP t'k [VP ... tk

VP is only a segment.
IP is a BC. CP inherits barrierhood but t''k is inside of the CP.
And childs play that t''k has no division from its antecedent. VP is a segment and VP is only a segment. Must find a way from outside to block this.

Thus, make rules (27) and (28): adjunction of wh-constituents to IP is illegitimate. And adjunction to arguments is illegitimate.

I think that Maryi, Bill really loves ti.

example of topicalization.
Where could Mary be sitting. no space whatsoever except via adjunction to IP. So adjunction to IP exists.
And transformation of adjunction to IP exists.


* Who do you think that kissed Mary

standard reason bad because that-trace effect C in between t' and t

but imagine were allowed to intermediately adjoin on the way out of IP.
t' in other IP governed as *locally* as possible, and from there to t''. Thus antecedent government should work. The other traces will be eliminated at LF.

so another reason intermediate adjunction not allowed for IP extraction.

(27) is simply a blunt stipulation.

But why is it forbidden. We do not really know. And we really don't know. Luckily not just for one case, but for two. But still really bad. (Could also move out * Who do you think that kissed Mary via vacuous movement hypothesis, discussed on page 9. In which case one stone kills one bird)

If 28 *were* a brute stipulation, would be bad. But can be understood in a relatively profound way and is useful elsewhere.

Can be made to follow from theta-criterion/projection principle.
If I went to Starbucks and had ingredients of cappaccino but also whipped cream, cinnamon, etc., then said I did not order this, I order a cappaccino.
If VP ordered a CP complement, and gets back CP plus a bunch of other stuff, will not be happy. Wanted lower CP but got upper CP as well. Now Verb cannot select the lower CP segment anymore. Buried under whipped cream which it did not want.
Thus, adjunction to arguments is illegitimate.

Projection principle requires that theta-marking must engage not only at D-structure, but even at S-structure.

Not only is 28 not a stipulation that cannot be reduced to something understandable, also useful elsewhere in the theory.

* (30) Who did [NP pictures of t] cause a scandal

[CP Whoi C(did) [IP [NP pix [PP of ti]] I'

in Syntax I, subject island violation

NP is not l-marked and thus BC for ti, and thus also barrier for ti.
The IP is BC. Since NP is BC, IP is also a barrier for ti. And also on the movement path. And gives us a *heavy* subjacency violation.

Now, imagine that ti was not the only trace. Suppose allowed to adjoin to NP on the way out of the IP.

[CP Whoi C(did) [IP [NP t'i [NP pix [PP of ti]]] I'

Now, ti has antecedent in t'i. And t' separated from antecedent by IP which is BC but not a barrier. So this should be completely good.
How do we block this derivation.
Can block this via (28) which says that you cannot adjoin to arguments.
So (28) also needed for (30).

And needed also for (29), extracting out of a Noun phrase.

29a:
?? Who did John make [NP the claim [CP that Bill harassed t
=
?? Whoi ... [VP t''' [VP make [NP the claim [CP t''i [that Bill [VP t'i [VP harassed ti

ti-t'i good
t'i-t''i good
but t''i-Whoi would be problematic.
However, t'''i-t''i would make it OK!

since CP l-marked by claim, algorithm would say it is not a barrier.
Chomsky: "the CP in the complement of N is a barrier."
So CP is a barrier but not a BC. Therefore the NP cannot inherit barrierhood from CP, since CP is not a barrier.

29b is worse. here extracting from relative clause. relative clause is not l-marked.

Whoi ... [VP t''' [VP make [NP the claim [CP Opj [that Bill [VP t'i [VP V tj to ti

that Bill had denied the claim to Renquist.
the claim now is being relativized.
ti to t'i and from there to t'''i.
Here, the IP is BC, which gives barrierhood to CP. (The CP is a relative clause, which are NOT complements - they are *adjoined* to N'. That means these are not L-marked and are thus inherent barriers. CP a BC. Thus, NP inherits barrierhood from CP. Very severe violation.

But imagine could adjoin to the CP or to the NP. Would give you just one barrier crossed. Yet this is too bad for that. But NP is complement of V make, so ruled out by 28.

One more thing need rule out. Why cannot adjoin to CP on the way out of the CP? For CP is not an argument, for it is a relative clause.

Must do something to (28), broaden it, so that it covers this as well.

(28) is about banning adjunction to things that receive theta-roles. Relative clauses do not *receive* theta roles, but are standardly assumed to be *predicates* of the noun phrases connected to, and thus they *assign* theta-roles.

"Adjunction to thematically active material is illegitimate."

rationale, the projection principle, will, continue to cover this extension.
If something want to assign theta-role to a subject, ...

Syntax II - Class II

Lasnik and Saito - 84.

in case of arg extraction, need not worry about ECP, because in case of object trace, is theta-governed, or in case of subject trace, already licensed at earlier level.
So ECP must be rethought in more dynamic way in terms of condition on derivations, similar to subjacency. "minimal link condition" Chomsky and Lasnik 93. Marriage of ECP and subjacency - keep chain links minimal. does not refer to S-structure, and does not refer to projection principle. in which case, *those* particular arguments in favor of an S-structure and projection principle are obvioated. As opposed to Lasnik and Saito which is based on it. But Chomsky and Lasnik do not entirely handle everything.

this idea of unify bounding (subjacency) and government (incuding ECP). he tried doing this in monograph "Barriers." Aim of that is to unify these two theories. in standard GB (government and binding), no connection.

Binding involves 3 opacity factors: presence of closer governor (="minimality"), notion of "too much" intervening structure. (which is why PRO in "John tried to leave" not governed, because two maximal projections, Cp and IP.) and finally, non-selectedness = a notion of Barrier, where not same amount of intervening structure, e.g. leave [AP PRO nude] in (10). (perhaps say adjuncts, = "non-selected by theta-role" are opaque.

hetorgenous bunch of opacity factors. when add NP and IP in subjacency, have 5 things that restrict dependecies. sometimes movement dependencies, sometimes government dependencies. could we try and integrate these systems, and can we make it homogenous and relatively simple?

Chomsky wants to say a resounding Yes in Barriers. Comes up with such a systems, under a single algorithm for computation of what Chomsky calls "Barriers."

Page 6 in the handout.
Subjacency: movement musn't traverse *any* barrier.
def of government: a governs b iff (i) and (ii)
i) a c-commands b
ii) no barrier intervenes between a and b.

What is a Barrier?

examples 10a-d on pg 5 in handout. to show how difficult to determine what barrier might be.

10a. can just say AP is barrier because not selected.
but 10b-d different kettle of fish. no barrierhood from non-selectedness. but need PRO to be non-governed. and not because not selected. but can rule out in syntax i via (ii), that one maximal projection is OK, but passing two is not. But we would like to pinpoint one and only one. Could be CP, could be IP, or could be both.

10c. Here have PRO. for howi to govern ti (antecedent govern), IP must not be barrier. but that wll not get us PRO ungoverned. And say that CP is the barrier to take care of PRO.

In 10d, typo. insert word "to" to left of "win".
Want PRO to be ungoverned, ti to be governed, and t'i need to be governed. cannot get rid of intermediate trace since need it to govern ti. IP cannot be a barrier. CP cannot be a barrier because then t'i wont be governed. So either IP or CP must be a barrier (for PRO), but then get one of the two traces in trouble.

Solution. If identify single nodes as Barrier obstructing a specific relationship, must make designation of nodes as barriers relative to the specific relationship we are talking about. "Barrier for a specific dependencies."

In 10d, the CP can be a barrier for a government dependency between "try" and PRO, without at the same time being a barrier for a government dependency between "how" and t'i.

Trick is to build an algorithm to do precisely that.

11 reemphasizes idea of barrier being a relative notion.

in 12, a category a is a barrier for b iff (i) or (ii) or (iii)
(iii) is what most familiar with, from syntax i, - when there is a closer governor. familiar notion of minimality.
(i) if a is a blocking category for b, but a != IP.

but now need to define a "Blocking category". define it in 13. iff a dominates b and is *not* L-marked.

in 14, categrory is L marked - if is theta-governed by lexical category.

i.e. something that is not theta-governed by a lexical category and dominates, it is a blocking category, so long as is not IP. 12(i) called inherent barrierhood. (in a way, this is continuation of non-selectedness, since if not theta-governed, not selected.)

now, (ii) = even if not inherent barrier, can *inherit* barrierhood. inherit not from another barrier, but from a blocking category for b.

alpha is a barrier for beta is PI is blocking category for beta and alpha immediately dominates PI.

It is *possible* for alpha to be a Barrier for beta even though a is L-marked, because it might inherit barrierhood from PI.

blocking categories pe se do not block anything. only barriers block things.

Back to examples. In 10a, want PRO (=b) to be ungoverned.
diagrams. dont have pen. I and V cannot govern big PRO. V does not c-command. perhaps m-commands. but cannot govern. because AP is a Blocking Category from PRO. because not L-marked. since does not get theta-mark. AP != IP. Thus, is inherent Barrier for b. Could not inherit barrierhood for PRO, since it immediately dominates PRO. So have barrier between I and PRO, and between V and PRO. A (nude) cannot govern PRO because does not see outside of its A' - that is, it does not c-command it.

10b. null complementizer.

   /VP  V  /CP   C(0)  /IP    PRO  /I'

The IP is not L-marked, since C is null. so it is a BC for PRO. Cannot declare it a Barrier since it is IP.
Look at CP from outside. is CP a blocking Category? no. because it is L-marked, because teta-governed by V and V is lexical.
However, CP immediately dominated IP, and IP is BC for PRO, so CP is a Barrier.

VP governed by IP, but I does not L-mark VP. Thus VP is blocking category and thus barrier for PRO. but not relevant. since entire relationship happens or fails within VP.

10c. diagram, used diff colors for diff dependencies.

which racej ..... wonder [CP howi C0 [IP PRO ..... tj ti ] ]

so must make sure howi-ti link succeeds, wonder-PRO link fails, ad which racej-tj fails.

PRO: IP not inherent barrier. but CP is a barrier for PRO just as before.
ti: IP is BC for ti. CP is barrier, but all happening within the CP. so CP's barrierhood is irrelevant. just one BC in between, and it is not a barrier.
tj: the movement dependency, where we want subjaceny violation. Once again, CP is a barrier, and this time it matters that CP is a barrier.

10d. was the hard one. where seemed to have no ONE node that was a barrier.

howi ........ try [CP t'i C0 [IP PRO ..... ti ] ]

want try-PRO link to be bad, t'i-ti link to be good, and howi-t'i link to be good.
PRO: PRO separated from try by one barrier, CP, as a result of the IP.
ti: no particular trouble. CP is a barrier but irrelevant because happens inside CP.
hwi-t'i: IP is indeed a BC for PRO, but since it does not dominate t'i, so it is not a BC for t'i. So CP does not become an inherited barrier. and not a BC in and of itself.

Now, let us talk a bit about VP.

howi ........ try [CP t'i C0 [IP PRO I(to) [VP ..... ti ] ] ]

let us compute barriers for ti. VP dominates ti. complement of I, which is not a lexical category, so BC. VP != IP. So Barrier. Now, the IP immediately dominates the BC, so becomes a barrier. So 2 problems!

So the question is: does VP indeed dominate ti?
a dominates b only if every segment of a dominates b. so since the

 /VP/VP\  ti

lower VP does NOT dominate ti. thus the VP in general does NOT domainte ti. Therefore cannot be a BC, and thus cannot be a barrier for ti. and as a result, IP cannot inherit barrierhood for ti.

let us imagine that is was not how - which made it an adjunct, and thus do not have these 2 VP structure. What if made it *what* rather than *how*, in which case have single segment VP.

If so, then is ti dominated by VP? This time, YES!
So how do we avoid it?
Some things born on edge of VP but (like adjuncts), but others can MOVE into the edge of VP. In fact, any and all poss not theta-positions you can move into.
what about this VP adjoined position? Have evidence that can adjoin to VP. Overt evidence in sentences such as "John put on the shelf [the book he had bought at a flea-market.]"
must be result of movinng [the book he had bought at a flea-market] to the right of "on the shelf". for starts at [spec, PP] of on, and then move it to right of the PP, by adjoining to VP. (called HNPS - heavy NP shift)

next week, put forth further evidence for successive cyclic movement out of VP via adjunction to VP. next week, successive cucicity - and the thing after it.

Syntax II - Class I

7395

Modular theory of Syntax (and Linguistics)

Syntax I modular in that recognizes several levels of representation – D-structure, S-structure, PF and LF (with T-Model diagram), Levels=Modules.

Also, within theory of linguistics, all sorts of syntactic facts encountered in the pie. On pizza, are olives and salami, and not necessarily are olives and salami created by same part of the theory.

Binding by Binding Theory (Principles A, B, C),

Locality by Bounding theory (subjacency),

Control by Control Theory,

Theta theory (theta-criterion),

Case theory (case filter),

Government theory (which includes ECP)

Principles and Parameters Theory have both these types of Modules.

Works well bec takes care of a lot of data. Fairly rich. Perhaps too rich. Less than accident that takes care of the facts. Could we get by with a less rich, leaner theory, and thus be able to do more. Less is better.

Perhaps have only one level of representation and far fewer subtheories.

<<Handout: “The Program”>>

1) restrict parameters to functional elements and general properties of lexicon.

Cannot say null subject parameter, unless can specify as property of lexical category

2) cond on representations hold only at interface + motivated by properties of interface.

e.g. conds that comes from binding theory hold at “interfaces”: syntax interfaces with Phonology at PF (in diagram) and with Semantics at LF.

Economy is buzzword. Make sure linguistics satisfy conds imposed by grammar in the minimal way. Only move things around if we have to.

In Syntax I, Move α. Free as long as does not violate rules of grammar. Some said: Move! (1980’s spirit). 1990’s spirit – don’t move unless you have to. Roman Catholic church – nothing allowed unless explicitly allowed and then you have to.

Only move when have a trigger, and then you have to move.

But what about optionality, when seem to have the option to move? In past, since movement was free, optionality was expected.

Back to endpoint of Syntax I

in what you can raise.

Contrast between 1a and 1b, and 2a and 2b, follows directly from ECP.

Who did what? Can be answered by giving a list of pairs of doers and things done: Mary did this, and John did Y.

Cannot say simply Mary did. Nor is it necessary that have only a single pair. That kind of answer called a pair list. Triggered by a particular representation of the question. Constituent formed what, who. See diagrams 1 and 2.

This kind of structure will result in every LF-structure, with those pairs. The one raised overtly will give index to the top one, and the others are adjoined.

The object need not worry about, since theta-governed, so licit under ECP. Other trace, t_i, of subject, is antecedent governed by N_i. But if raised what overtly, then have NP_j at the top. t_j still OK since theta-governed. But t_i cannot be antecedent governed, since it depends upon a c-command relationship. Thus violates ECP. Same thing for 2b, what did he do why? The trace of why cannot be properly governed, since adjuncts are never theta-governed. So covered by ECP, because ECP was designed for ruling this sort of thing out.

Second Asymmetry: That-trace asymmetry.

Who do you think (*that) did it

But

What do you think that he did?

And

Why do you think that he did it

A minimality effect. Which is why 4a not OK (because intervening governor that). (see diagram 4) But 4b and 4c are theta-governed and thus do not need antecedent governed. But then, that should intervene in far distant But see diagram 5!

Traces in adjunct pos never governed by any head, thus no minimality problem. Great. So we said it. But that does not explain it.

To go back to subject case, why is extraction of trace to CP suddenly good when complementizer is suddenly null? Still there. We answered: No Arnold Shwartzeneger – no empty-headed governors. Thus empty C cannot govern. But this is an additional assumption. Would have better theory if did not have to say this.

Assymetry #3

Contrast between adjuncts vs. objects and subjects.

See 6 in handout.

a) seems like mere subjacency violation but not ECP violation. Getting out of wh-island. Same for (b). But case (c), no way in which which way modifies the fixing of the car. (so bad based on intended reading).

In (c), the t_k is antecedent government by t_k’ but t_k’ cannot, because who is a closer governor. Since t_k properly governed, peculiar thing here is that somehow (a) and (b) do not violate the ECP. But intermediate traces seem separated by intermediate other traces. These traces are all in A’ positions. Nothing in Syntax I will allow us to understand this.

But there is a contrast which we can see.

Lasnik and Soto propose something to deal with 4 and 6. Summed up in (7)a-e.

Will talk about ECP as two-pronged system. The “gamma-police” and the “judge” First get ticket from police officer. Can tear it up, appeal. In court-room, same principles applied to convict you. So everyone will appeal, and since takes a long time, entirely possible to die in between. In which case need not pay the fine. The ticket is called a gamma-mark. And judiciary system which declares convictions. Tickets come in +gamma and –gamma flavor. With +gamma mark, no tickets when come to the courtroom so never get a negative verdict passed against you.

Culprit in t_k’. In 6a and 6b does not make it into court because that trace is deleted before derivation ends at LF.

Affect Alpha splits off into Move Alpha and Delete Alpha.

Delete alpha as entering LF. LF --> court

Argue that get gamma mark as soon as arise, in (a) and (b), but lazier in 6c. Only handed out at LF for adjunct, not only for prime traces, but even for original case t_k. Since can only get properly governed by trace t_k’ at LF to save t_k, and so it itself must also get +gamma mark to be saved.

Two structures for who do you think did it

[Who_su]_i ……. [CP t_i [ C (that) t_i]]

[Why]_k ……………… [CP t_k’ [C (that) …. t_k] ] ]

in latter case, trace of adjunct extraction = +gamma at LF

[Who_su]_i ……. [CP t_i [ C that t_i]]

here in argument case, gets -gamma at S-structure, and cannot get rid trace because need it, since it is the head of the chain, and otherwise do not satisfy the case filter.

Why is there the argument /adjunct distinction? Because system was explicitly designed to deliver it. Since made distinction in timing between traces in adjunct vs. in argument positions. See 7c. Why should 7c be true. If cannot explain, just restated the facts in particularly opaque way.

Can be made to follow from Projection Principle, that args be syntactically represented and legitimate at every level of representation.

So as soon as arg traces arise, need be legitimate. And lexical predicate heads must be represented throughout. Since adjuncts do not get or assign theta-roles, the Projection principle has nothing to say about it.

Chomsky and Lasnik turn this around. Nature of chains formed by arg vs. adjunct extraction.

Lasnik and Seto said: get rid of anything do not need. Delete alpha always.

Lasnik and Chomsky. Each of these move alpha, delete alpha operations come with a price ticket.

So when should you do?

Create chain:

(Why_{k ,} t_k,, t_k,, t_k,, t_k,, t_k,, t_k,, t_k,, t_k,, t_k,, t_k,, t_k,, t_k,, t_k,, t_k,,) ------- adjunct chain

(Who_{k ,} t_k,, t_k,, t_k,, t_k,, t_k,, t_k,, t_k,, t_k,, t_k,, t_k,, t_k,, t_k,, t_k,, t_k,,) ------- argument chain

In first one, each trace in A’ position. UNIFORM.

Whereas in arg chain of A’ extraction, is non-uniform, since all t’ are in A’ position whereas first t in A position. NON-UNIFORM.

A legitmate LF object is a UNIFORM chain, as well as an OPERATOR-VARIABLE chain, with no intermediate trace: (Op_j, t_j)

So NON-UNIFORM chain is not legitimate LF object. How save? Cannot keep this chain at LF. Cannot make it uniform, since cannot change position of the original trace. But can transform into OPERATOR-VARIABLE chain, via deletion. So intermediate traces will never give you ECP violations.

Do not operate on UNIFORM chains because unnecessary, so traces will stay, and so if anything wrong with any of these, will cause an ECP violation.

Thus has nothing to do with Projection Principle, but follows from very nature of adjunct and argument extraction, that is, the very nature of the chains that they create (uniform vs. non-uniform).

This is economy of derivation (operation) as opposed to economy of representation (which would be to get rid of the unneeded traces).

A tree generator on the Web

I found a great tree generator program on the web. It takes labelled bracket notation and generates the tree, which you can see or save as an image. It runs from a web page so you need not download any program. Check out phpSyntaxTree.