Taalportaal - the digital language portal

Dutch
Frisian
Saterfrisian
Afrikaans

Show all

11.3.1.2.Wh-extraction from embedded clauses (long wh-movement)

quickinfo

This section discusses a special case of wh-extraction, which we will refer to as long wh-movement. This type of wh-movement is special in that it is apparently unbounded: it may cross an in principle indefinite number of clausal boundaries (although in actual fact the number is of course limited for practical reasons). We illustrate this in (181): in (181a) and (181b) wh-movement crosses, respectively, one and two clausal boundaries.

181

a.	Wati	zegt	Marie	[dat	Peter ti	gekocht	heeft]?
	what	says	Marie	that	Peter	bought	has
	'What does Marie say that Peter has bought?'

b.	Wati	denkt	Jan	[dat	Marie zegt	[dat	Peter ti	gekocht	heeft]]?
	what	think	Jan	that	Marie says	that	Peter	bought	has
	'What do you think that Marie says that Peter has bought?'

Long-distance dependencies of the type in (181) apparently go against the general finding in generative grammar that syntactic dependencies are local, which can normally be taken to mean at least "clause-bound". There is, however, reason for assuming that wh-movement in (181) does not apply in one fell swoop, but in a so-called cyclic fashion; see Chomsky (1973), and Boeckx (2008) for a more recent discussion. The derivation thus proceeds as indicated in (182): the wh-phrase wat is first moved into the initial position of its own clause (the first cycle), from where is it subsequently moved on into the clause-initial position of the next higher clause (the second cycle), etc. The primed traces in (182) indicate all intermediate landing sites of the wh-phrase and show that all individual movements are local, provided that we assume that the initial position of a clause functions as an "escape hatch" for the wh-phrase.

182

a.	Wati	zegt	Marie [t'i	dat	Peter ti	gekocht	heeft]?
	what	says	Marie	that	Peter	bought	has
	'What does Marie say that Peter has bought?'

b.	Wati	denkt	Jan [t''i	dat	Marie zegt [t'i	dat	Peter ti	gekocht	heeft]]?
	what	thinks	Jan	that	Marie says	that	Peter	bought	has
	'What do you think that Marie says that Peter has bought?'

Despite the fact that long wh-movement can be broken up in smaller, local movement steps, we will follow general practice in using the notion of long wh-movement as a convenient descriptive term for wh-extraction from embedded clauses. For convenience, we will often omit the intermediate (primed) traces from our structural representations if they are not relevant for our discussion.
Long wh-movement is a severely restricted phenomenon subject to various stringent conditions, subsection I starts by showing that this does not hold for the wh-moved phrase itself: the same set of elements allowing local wh-movement may undergo long wh-movement. If long wh-movement consists of a sequence of local movement steps, this is of course expected, subsection II will show, however, that there are some more or less concealed issues with long wh-movement of subjects, which are related to the so-called complementizer-trace filter, subsections III and IV continue to show that there are constraints on the embedded clause from which wh-movement takes place as well as the matrix verb, subsection V provides a brief comparison of long wh-movement with other strategies to establish "long" wh-dependencies, which can be found in certain dialects of Dutch as well as German. Since Subsections I-V are only concerned with finite clauses, Subsection VI concludes with a discussion of long wh-movement from infinitival clauses; such cases have received much less attention in the literature, but are interesting in their own right because they have a number of special properties.

I. Restrictions on the moved element
II. Complementizer-trace effects
III. Restrictions on the syntactic function of the embedded clause
IV. Bridge verbs
V. Long wh-movement is obligatory and leaves an intermediate trace
VI. Long wh-movement from infinitival clauses

[+] I. Restrictions on the moved element

Long wh-movement does not seem to differ from local wh-movement when it comes to the syntactic functions of the moved elements. The examples in (183) show that it may affect clausal constituents of all types: argument, complementive and adjunct. Just as in the case of local wh-movement, the only requirement seems to be that an interrogative form is available. Recall that we will leave out the intermediate trace in the clause-initial position of the embedded clause if this is not immediately relevant for our discussion. Note that the wh-phrase wanneer'when' in (183d) can also be construed as a modifier of the matrix-clause, but this is of course not the reading intended here.

183

a.	Wiei	zei/dacht	je	[dat ti	dat boek	gekocht	had]?	subject
	who	said/thought	you	that	that book	bought	has
	'Who did you say/think had bought that book.'

b.	Wati	zei/dacht	je	[dat	Peter ti	gekocht	heeft]?	object
	what	said/thought	you	that	Peter	bought	has
	'What did you say/think that Peter has bought?'

c.	Hoe oudi	zei/dacht	je	[dat	dit fossiel ti	was]?	complementive
	how old	said/thought	you	that	this fossil	was
	'How old did you say/think that this fossil was?'

d.	Wanneeri	zei/dacht	je	[dat	Peter ti	vertrokken	was]?	adjunct
	when	said/thought	you	that	Peter	left	had
	'When do you say/think that Peter had left?'

The examples in (184) further show that long wh-movement is not confined to clausal constituents but may also be applied to wh-elements embedded in clausal constituents (provided that local wh-movement also allows stranding). We illustrate this in (184) by means of, respectively, an interrogative modifier of an adjectival complementive and a split wat voor-phrase in (184b).

184

a.	Hoe zwaari	denk je	[dat	Jan [AP ti	verslaafd]	is]?
	how heavily	think you	that	Jan	addicted	is
	'How severely addicted do you think that Jan is?'

b.	Wati	denk	je	[dat	Peter [NPti	voor een boeken]	gekocht	heeft]?
	what	think	you	that	Peter	for a books	bought	has
	'What kind of books do you think that Peter has bought?'

[+] II. Complementizer-trace effects

The examples discussed in the previous subsection suggest that long wh-movement does not impose any special conditions on the syntactic function of the moved element. It is nevertheless necessary to say more about long wh-movement of subjects given that it triggers special effects in various languages. This is illustrated for English in the examples in (185), which show that long wh-movement of subjects but not of objects requires omission of the complementizer that. Chomsky & Lasnik (1977) exclude the configuration [... [C that] ti ...] by means of the so-called that-trace filter, but since it is possible to find similar facts in languages other than English, we will use the more general term complementizer-trace filter.

185

Whoi do you think [(*that) tiwill read the letter]?

subject

Whati do you think [(that) John will do ti]?

object

Long wh-movement of subjects also triggers a special effect in French. The translation of (185a) in (186a) shows that the subject trace cannot occur if the declarative complementizer appears in its regular form, que, but requires it to surface as qui; cf. Kayne (1976). Example (186b) further shows that this que/qui alternation does not apply in the case of long wh-movement of, e.g., an object.

186

a.	Quii crois-tu	[qui/*que ti	lira	la lettre]?	subject
	who think-you	that/that	readfuture	the letter
	'Who do you think will read the letter?'

b.	Quei	crois-tu	[que/*qui	Jean	fera ti]?	object
	what	think-you	that/that	Jean	dofuture
	'What do you think that Jean will do?'

The Dutch translations of the examples in (185)/(186) in the primeless examples in (187) suggest that the subject-object asymmetry found in English and French does not occur in Standard Dutch, as they are both fully acceptable; cf. Dekkers (1999). It seems that for at least some speakers the question as to whether the subject-object asymmetry shows up depends on the type of interrogative noun phrase: while non-D-linked subject pronouns such as wie'who' in (187a) easily allow long wh-movement without any special ado, D-linked subjects such as welke jongen'which boy' in (187a') are marked (but certainly not ungrammatical) for such speakers.

187

a.	Wiei	denk	je	[dat ti	de brief	zal	lezen]?	subject
	who	think	you	that	the letter	will	read
	'Who do you think will read the letter?'

a'.	?	Welke jongeni	denk je	[dat ti	de brief	zal	lezen]?	subject
		which boy	think you	that	the letter	will	read
		'Which boy do you think will read the letter?'

b.	Wati	denk	je	[dat	Jan ti	zal	doen]?	object
	what	think	you	that	Jan	will	do
	'What do you think that Jan will do?'

A possible reason for the difference in acceptability of the two (a)-examples may be that, despite appearances, the traces of the two wh-phrases do not occupy the same position in the clause. We will first illustrate the difference in location by means of the examples in (188) and (189) without a definite object. The examples in (188) show that long wh-movement of wie requires the presence of the expletive er'there'; long wh-movement of welke jongen is severely degraded if er is not present and again marked for some speakers with er present.

188

a.	Wiei	denk	je	[dat	*(er) ti	gelogen	heeft]?
	who	think	you	that	there	lied	has
	'Who do you think has lied?'

b.	Welke jongeni	denk je	[dat	*(?er) ti	gelogen	heeft]?
	which boy	think you	that	there	lied	has
	'Which boy do you think has lied?'

What is crucial for our argument is not so much the admittedly subtle effect of D-linking on the acceptability of an overt expletive in the two examples in (188), but the contrast between the examples in (188) and those in (189); while omission of the expletive is completely excluded in (188), it is (at least marginally) allowed in (189).

189

a.	Wie	heeft	?(er)	gelogen?
	who	has	there	lied
	'Who has lied?'

b.	Welke jongen	heeft	(?er)	gelogen?
	which boy	has	there	lied
	'Which boy has lied?'

The acceptability contrasts indicated in (188) and (189) can be accounted for by appealing to the complementizer-trace filter. First consider the two (a)-examples. Since the expletive er is right-adjacent to the declarative complementizer dat in (188) or the finite verb in non-subject-initial clauses such as Gisteren heeft er iemand gelogen'Yesterday someone lied', it can be assumed to occupy the regular subject position. If we further assume that absence of the expletive indicates that the subject has been moved into the regular subject position (a marked option for non-D-linked wie), the acceptability difference between the two (a)-examples follows from the complementizer-trace filter: if the expletive er is not present, the C-position is immediately followed by a wh-trace, which is prohibited if the C-position is filled by the complementizer dat but allowed if it is filled by the finite verb. The contrast is even clearer in the case of the two (b)-examples, due to the fact signaled by the markedness of er that D-linked wh-phrases are preferably wh-moved via the regular subject position.
Let us now return to the contrast between (187a) and (187a'). At first sight, the proposal above does not seem to help much to account for this, as these examples do not contain the expletive er. If this indicates, as suggested above, that the subject has been wh-moved via the regular subject position, we would predict these examples to be both unacceptable, contrary to fact. However, the fact that the expletive is not realized is not due to the position of the subject but to yet another factor, which was discussed in Section N8.1.4, namely that the realization of expletives is not only sensitive to the (in)definiteness of the subject, but also depends on the presence of presuppositional material in the clause. Consider the examples in (190), in which the subjects are all interpreted as non-specific indefinites, and in which er should not be construed spatially ("there") but as a pure expletive.

190

a.		dat	?(er)	iemand	een boek	gekocht	heeft.
		that	there	someone	a book	bought	has

b.		dat	(?er)	iemand	het boek	gekocht	heeft.
		that	there	someone	the book	bought	has

c.		dat	(*er)	iemand	het	gekocht	heeft.
		that	there	someone	it	bought	has

The contrast between the two examples in (190a&b) shows that the definiteness of the object may affect the distribution of the expletive er. This is even clearer in (190c), where the referential personal pronoun het blocks realization of the expletive. Consequently, in order to show that the acceptability of long wh-movement of the subject depends on D-linking, we also have to control for the definiteness of the object. This has been done in the examples in (191), which show that with an indefinite object omission of the expletive again has a severely degrading effect in the case of long but not in the case of local wh-movement. The contrast between the primeless and primed examples (191) thus shows again that wh-movement of subjects is sensitive to the complementizer-trace filter.

191

a.	Wiei	denk	je	[dat	*?(er) ti	een boek	gekocht heeft]?
	who	think	you	that	there	a book	bought has
	'Who do you think has bought a book?'

a'.	Wiei	heeft	?(er) ti	een boek	gekocht?
	who	has	there	a book	bought
	'Who has bought a book?'

b.	Welke jongeni	denk	je	[dat	??(?er) ti	een boek	gekocht heeft]?
	which boy	think	you	that	there	a book	bought has
	'Which boy do you think has bought a book?'

b'.	Welke jongeni	heeft	(?er) ti	een boek	gekocht?
	which boy	has	there	a book	bought
	'Which boy has bought a book?'

It is important to note that the complementizer-trace filter crucially involves a phonetically realized complementizer. This is clear from the examples in (192), which show that local wh-movement into the clause-initial position of the embedded clause does not require the presence of the expletive er, that is, that the empty complementizer Ø does not trigger the complementizer-trace effect. The primed examples in (191) have already shown that the complementizer-trace filter crucially involves a phonetically realized complementizer, not just a phonetically filled C-position, as finite verbs in second position do not evoke this effect.

192

a.	Ik	vraag	me	af	[wiei	Ø	?(er) ti	gelogen	heeft]?
	I	wonder	refl	prt.	who	comp	there	lied	has
	'I wonder who has lied.'

b.	Ik	vraag	me	af	[welke jongeni	Ø	(?er) ti	gelogen	heeft]?
	I	wonder	refl	prt.	which boy	comp	there	lied	has
	'I wonder which boy has lied.'

We conclude this discussion of complementizer-trace effects by raising a warning flag related to the fact that Maling & Zaenen (1978) have suggested that there are regional varieties of Dutch in which the expletive er can be freely omitted. Although this claim is controversial, there may indeed be a certain amount of individual variation in speaker judgments when it comes to dropping the expletive in the examples discussed in this subsection. For more detailed discussion, we refer the reader to Bennis (1986:section 3.6.1).

[+] III. Restrictions on the syntactic function of the embedded clause

The acceptability of long wh-movement depends on properties of the embedded clause from which the wh-phrase is extracted. The examples in (193) show that the embedded verb must be an argument of its matrix clause; long wh-movement from complementive or adverbial clauses is prohibited.

193

a.	De directeur	had verwacht	[dat	hij	een bonus	zou	krijgen].	direct object
	the manager	had expected	that	he	a bonus	would	receive
	'The manager had expected that he would receive a bonus.'

a'.	Wati	had	de directeur	verwacht	[dat	hij	zou ti	krijgen]?
	what	had	the manager	expected	that	he	would	receive
	'What had the manager expected that he would receive?'

b.	Het probleem	is	[dat	de directeur	een te grote bonus	krijgt].	complementive
	the problem	is	that	the manager	a too big bonus	receives
	'The problem is that the manager receives a big bonus.'

b'.	*	Wati	is	het probleem	[dat	de directeur ti	krijgt]?
		what	is	the problem	that	the manager	receives

c.	De directeur	juichte	[toen	hij	een vette bonus	kreeg].	adverbial
	the manager	cheered	when	he	a fat bonus	received
	'The manager shouted with joy when he received a fat bonus.'

c'.	*	Wati	juichte	de directeur	[toen	hij ti	kreeg]?
		what	cheered	the manager	when	he	received

The examples in (194) show that long wh-movement is also blocked from argument clauses if they are introduced by the anticipatory pronoun het'it'. This would follow immediately from the observation above if we assume that the anticipatory pronoun is the "true" argument of the verb while the clause is an adjunct or a right-dislocated (that is, clause-external) element.

194

a.	De directeur	had het	verwacht	[dat	hij	een bonus	zou	krijgen].
	the manager	had it	expected	that	he	a bonus	would	receive
	'The manager had expected it that he would receive a bonus.'

b.	*	Wati	had	de directeur	het	verwacht	[dat	hij ti	zou	krijgen]?
		what	had	the manager	it	expected	that	he	would	receive

Long wh-movement is not only possible from object clauses but also from subject clauses, as is clear from the fact that impersonal passivization of (193a') gives rise to a fully acceptable result; this is shown in (195a'). Use of an anticipatory pronoun again blocks long wh-movement, as is clear from the fact that the passivized counterpart of example (194b) is unacceptable; this is shown in (195b'). For completeness' sake, the primeless examples show that the corresponding cases without long wh-movement are both acceptable.

195

a.	Er	werd	verwacht	[dat	hij	een bonus	zou	krijgen].
	there	was	expected	that	he	a bonus	would	receive
	'It was expected that he would receive a big bonus.'

a'.		Wati	werd	er	verwacht	[dat	hij	zou ti	krijgen]?
		what	was	there	expected	that	he	would	receive

b.	Het	werd	verwacht	[dat	hij	een bonus	zou	krijgen].
	it	was	expected	that	he	a bonus	would	receive
	'It was expected that he would receive a big bonus.'

b'.	*	Wati	werd	het	verwacht	[dat	hij	zou ti	krijgen]?
		what	was	it	expected	that	he	would	receive

It should be noted, however. that at least some speakers perceive an argument-adjunct asymmetry in the case of subject clauses. So, while all speakers accept argument extraction both from object and subject clauses, some speakers consider adjunct extraction from subject clauses to yield a worse result than from object clauses; this is illustrated by, respectively, the (a)- and (b)- examples in (196). It suggests that subject but not object clauses are weak islands for wh-movement.

196

a.	Wati	verwacht	Peter	[dat	Marie morgen ti	zal	kopen]?
	what	expects	Peter	that	Marie tomorrow	will	buy
	'What does Peter expect that Marie will buy tomorrow?'

a'.		Wati	wordt	er	verwacht	[dat	Marie morgen ti	zal	kopen]?
		what	is	there	expected	that	Marie tomorrow	will	buy

b.	Wanneeri	verwacht	Peter	[dat	Marie een nieuwe auto ti	zal	kopen]?
	when	expects	Peter	that	Marie a new car	will	buy
	'When does Peter expect that Marie will buy a new car?'

b'.	%	Wanneeri	wordt	er	verwacht	[dat	Marie een nieuwe auto ti	zal	kopen]?
		when	is	there	expected	that	Marie a new car	will	buy

The acceptability of the passive example in (195a') raises the expectation that long wh-movement is also possible from subject clauses in unaccusative constructions. This seems to be borne out by the fact that the modal verb blijken'to turn out' licenses long wh-movement provided the anticipatory pronoun het'it' is not present; cf. Bennis (1986:ch.2). Even speakers who consider wh-extraction in example (197b) marked with the expletive er will agree that there is a sharp contrast in acceptability with regard to the version with the anticipatory pronoun het.

197

a.	Er/Het	is	gebleken	[dat	Jan	staatsgeheimen	verkocht	heeft].
	there/it	is	appeared	that	Jan	secrets.of.state	sold	has
	'It has turned out that Jan has sold official secrets.'

b.		Wati	is	er/*het	gebleken	[dat Jan ti	verkocht	heeft]?
		what	is	there/it	appeared	that Jan	sold	has

Long wh-movement from subject clauses is nevertheless quite rare due to the fact that subject clauses are normally obligatorily introduced by the anticipatory pronoun het. For example, the modal verb schijnen'to seem' differs from blijken in that it does not allow the impersonal construction with the expletive er'there', so that long wh-movement is categorically excluded.

198

a.	Het/*Er	schijnt	[dat	Jan	staatsgeheimen	verkocht	heeft].
	it/there	seems	that	Jan	secrets.of.state	sold	has
	'It seems that Jan has sold official secrets.'

b.	*	Wati	schijnt	het/er	[dat Jan ti	verkocht	heeft]?
		what	seems	it/there	that Jan	sold	has

The primeless examples in (199) show that the anticipatory pronoun het cannot appear if the subject clause is in sentence-initial position, while the primed examples show that long wh-movement is nevertheless impossible. This shows that long wh-movement is only possible from subject clauses in clause-final position, although it is not clear whether this should be considered a restriction on wh-movement, as subject clauses are never possible in the middle field of the clause: cf. Koster (1978).

199

a.	[Dat	Jan staatsgeheimen	verkocht	had]	bleek	al	snel.
	that	Jan secrets.of.state	sold	had	turned.out	prt	quickly
	'It turned out quickly that Jan had sold official secrets.'

a'.	*	Wati	bleek	[dat	Jan ti	verkocht	had]	al	snel?
		what	appeared	that	Jan	sold	had	prt	quickly

b.	[Dat	Jan	staatsgeheimen	verkocht	had]	was duidelijk.
	that	Jan	secrets.of.state	sold	had	was clear
	'It was clear that Jan had sold official secrets.'

b'.	*	Wati	was	[dat	Jan ti	verkocht	had]	duidelijk?
		what	was	that	Jan	sold	had	clear

We conclude with a brief digression on matrix verbs that normally select a prepositional object such as klagen (over)'to complain about'. Although Section 2.3.1, sub VI, has shown that many of these verbs allow the anticipatory pronominal PP to be omitted if the prepositional object is clausal, long wh-movement is normally excluded.

200

a.	Jan klaagt	(erover)	[dat	Marie zijn aantekeningen	weg	gegooid	heeft].
	Jan complains	about.it	that	Marie his notes	away	thrown	has
	'Jan complains (about it) that Marie has thrown away his notes.'

b.	*	Wati	klaagt	Jan (erover)	[dat	Marie ti	weg	gegooid	heeft]?
		what	complains	Jan about.it	that	Marie	away	thrown	has

The verb hopen (op)'to hope for' appears to be an exceptional case. Example (201a) first shows that this verb selects a prepositional object; the use of a nominal object (without op) leads to an unacceptable result. Example (201b) shows that the anticipatory pronominal PP erop can easily be dropped if the object is clausal; it is in fact the preferred option. Example (201c) finally shows that long wh-movement is acceptable if the pronominal PP is not present.

201

a.	De directeur	hoopt *(op)	een grote bonus.
	the manager	hopes for	a big bonus
	'The manager is hoping for a big bonus.'