Taalportaal - the digital language portal

Dutch
Frisian
Saterfrisian
Afrikaans

Show all

5.5.2.Co-occurrence of R-words

quickinfo

It is possible to have more than one R-word in a single clause, but there are several restrictions on their co-occurrence, subsections I and II show that, in the general case, a clause can never contain more than one weak or more than one strong R-word, whereas Subsection III shows that a weak and a strong form can co-occur. Before we start a caveat is in order. The data in this section are very complex, and it is hard sometimes for native speakers to give their judgments on the examples discussed; it is not surprising, therefore, that conflicting judgments can be found in the literature. Moreover, Subsections to VI will show that additional factors seem to play a role in the acceptability of certain sentences, some of which have scarcely been investigated. It is therefore virtually impossible to give an exhaustive review of this topic. We therefore recommend that the reader also consult the existing literature (e.g., Bech 1952, Van Riemsdijk 1978, Bennis 1980/1986, Huybregts 1991, Odijk 1993 and the references cited therein) if one wants to investigate this topic more thoroughly.

I. Co-occurrence of multiple weak R-words
II. Co-occurrence of multiple strong R-words
III. Co-occurrence of weak and strong R-words
IV. Wh-movement and topicalization in multiple R-word constructions
- A. Wh-movement and topicalization in double R-word constructions
- B. Wh-movement and topicalization of -R phrases across R-words
V. Co-occurrence of multiple pronominal PPs
VI. Co-occurrence of multiple locational pro-forms
VII. Summary

[+] I. Co-occurrence of multiple weak R-words

In the general case, a clause can contain at most one weak R-word, which will be demonstrated below by means of examples of clauses with two weak R-words with distinct functions. We start with a discussion of the distribution of weak R-words in embedded clauses, which is followed by a discussion of their distribution in main clauses.

[+] A. Embedded clauses

Consider the examples in (124). Since the (a)- and the (b)-example are impersonal passives, the first occurrence must be expletive er; and the same thing holds for the (c)-example since the associate noun phrase [twee [e]] of quantitative er is an indefinite noun phrase. As is shown in these three examples, expletive er cannot co-occur with the other uses of er. The function of the occurrences of er are indicated by the following abbreviations: expl = expletive, loc = locational, quant = quantitative, and pron = pronominal.

124

a.	*	dat	er	er	gedanst	wordt.	.. expl .. loc ..
		that	there	there	danced	is
		Intended reading: 'People are dancing there.'

b.	*	dat	er	er	over	gesproken	wordt.	.. expl .. pron ..
		that	there	there	about	spoken	is
		Intended reading: 'People are talking about it.'

c.	*	dat	er	er [NP	twee [e]]	gestolen	zijn.	.. expl .. quant ..
		that	there	there	two	stolen	have.been
		Intended reading: 'Two [e.g., computers] have been stolen.'

The pattern in (124) has given rise to the idea that er must be placed into a unique, designated [+R]-position in the middle field of the clause, that is, the [+R]-position in (121). The postulation that this position is unique accounts for the fact that only one occurrence of er is allowed; [+R] can be occupied by one R-word only, so that the other occurrence(s) would violate Axiom I.

125

Axiom I: Er must be moved into a unique [+R]-position in the middle field of the clause

Axiom I also gives the correct predictions for the examples in (126): the examples in (126a&b) show that an R-pronoun cannot be combined with a locational pro-form or quantitative, and (126c) shows that the co-occurrence of quantitative er and a locational pro-form is also correctly excluded. Since the grammaticality judgments on the examples in (126) remain the same if we reverse the interpretations (e.g., if we interpret the first occurrence of er in (126c) as locational and the second as quantitative), we have discussed all possible combinations.

126

a.	*	dat	Jan er	er over praatte.	.. pron .. loc ..
		that	Jan there	there about talked
		'that Jan talked about it there.'

b.	*	dat	Jan er	er	drie	in	stopte.	.. pron .. quant ..
		that	Jan there	there	three	into	put
		'that Jan put three [e.g., cigars] in it.'

c.	*	dat	Jan er	er [NP	twee [e]]	gezien	heeft.	.. quant .. loc ..
		that	Jan there	there	two	seen	has
		'that Jan saw two [e.g., rats] there.'

Note that axiom I is overruled for occurrences of er that must be adjacent to the preposition, like er in the temporal phrase er voor/na'before/after it'; cf. (127). This exception to Axiom I follows from the fact, discussed in Section 5.3.1, that these temporal PPs do not allow R-extraction; since er remains PP-internal, it is correctly predicted that the clause may contain two occurrences of er in such cases.

127

a.	dat	Jan er	misschien	[er voor]	iets	over	zal	zeggen.
	that	Jan there	maybe	there before	something	about	will	say
	'that maybe Jan will say something about it before it.'

b.	dat	er	waarschijnlijk	[er na]	nog	iets	werd	gedronken.
	that	there	probably	there after	prt.	something	was	drunk
	'that people probably drank something afterwards.'

[+] B. Main clauses

The pattern of grammaticality judgments is slightly different if we are dealing with main clauses. If expletive er occupies the regular subject position after the finite verb, as in (128), the pattern we find is identical to the one in (124).

128

a.	*	Morgen	wordt	er	er	gedanst.	.. expl .. loc ..
		tomorrow	is	there	there	danced

b.	*	Morgen	wordt	er	er	over	gesproken.	.. expl .. pron ..
		tomorrow	is	there	there	about	spoken

c.	*	Gisteren	zijn	er	er [NP	twee [e]]	gestolen.	.. expl .. quant ..
		yesterday	have.been	there	there	two	stolen

However, if expletive er occupies the clause-initial position, as in (129), it can co-occur with quantitative er (but not with pronominal or locational er). Whereas the pattern in (128) follows directly from axiom I, the pattern in (129) is somewhat mysterious; we will not provide an explanation here for this deviant pattern.

129

a.	*	Er	wordt	er	morgen	gedanst.	.. expl .. loc ..
		there	is	there	tomorrow	danced

b.	*	Er	wordt	er	morgen	over	gesproken.	.. expl .. pron ..
		there	is	there	tomorrow	about	spoken

c.		Er	zijn	er	gisteren [NP	twee [e]]	gestolen.	.. expl .. quant ..
		there	have.been	there	yesterday	two	stolen

Similar problems do not arise with the other types of er given that these never occur in sentence-initial position; they therefore invariably exhibit the pattern in (126) in main clauses as well.

[+] C. Conclusion

We conclude from the discussion in the previous subsections that Axiom I provides an apt description of the attested facts; the only problem is the grammaticality judgment on example (129c). In order not to confuse the reader at a later stage in the discussion, we want to note here that the ungrammaticality of the examples in (124), (126), (128), and (129a&b) does not necessarily mean that the intended meanings cannot be expressed. Some of them can, due to the fact that er is able to perform more than one function at the same time. Discussion of this is postponed to Section 5.5.3.

[+] II. Co-occurrence of multiple strong R-words

This subsection discusses the co-occurrence restrictions on strong R-forms like hier and daar. In what follows we can ignore the expletive or quantitative uses of R-words, given that these uses always involve the weak form er, which leaves us with the locational and pronominal R-words. The examples in (130) show that strong demonstrative pronominal and locational R-forms cannot co-occur. Example (130a) is the reference sentence. The examples in (130b) and (130c) illustrate, respectively, that it is possible to replace the adverbial phrase op dit congres by the locational pro-form daar and to pronominalize the prepositional complement of the verb over de oorlog. The (d)-examples in (130) show, however, that it is impossible for a clause to simultaneously contain a locational pro-form and a pronominalized PP; this holds regardless of the order of the R-words.

130

a.	Zij	heeft	op dit congres	vaak	over de oorlog	gesproken.
	she	has	at this conference	often	about the war	spoken
	'She spoke often about the war at this conference.'

Zij heeft daar vaak over de oorlog gesproken.

.. loc ..

Zij heeft hier op dit congres vaak over gesproken.

.. pron ..

Zij heeft hier daar vaak over gesproken.

.. pron .. loc ..

d'.

Zij heeft daar hier vaak over gesproken.

.. loc .. pron ..

The pattern in (124) has given rise to the idea that the strong demonstrative R-words can also be placed into a unique, designated [+R]-position in the middle field of the clause. The postulation that the position is unique again accounts for the fact that only one occurrence of a strong demonstrative R-word is allowed.

131

Axiom II: A strong demonstrative R-word may be moved into a unique [+R]-position in the middle field of the clause

Axiom II is phrased slightly more weakly than Axiom I in (125) in that it does not require that a strong demonstrative R-word be moved into the [+R]-position. This is needed to account for examples such as (132). Given the fact that the pronominal R-word follows the (optional) clausal adverb vaak, we may claim that it occupies its base-position within PP (an option independently argued for in Section 5.5.1), so that we correctly predict this example to be grammatical: the unique [+R]-position is occupied by the locational R-word daar only, while the pronominal R-word hier is still PP-internal.

132

Zij	heeft	daar (vaak) [PP	hier over]	gesproken.	.. loc .. pron ..
she	has	there often	here about	spoken
'She often spoke about this there.'

It should be noted, however, that example (133), provided by Hans Bennis (p.c.), is a potential problem for axiom II; presumably, the locational pro-form hier occupies the [+R]-position, but even so the pronominal R-word daar can be moved out of its PP. Our informants provide somewhat diverging judgments on this example, which seems best if the pronominal R-word is assigned contrastive accent; theories that postulate a designated focus position in the clause may perhaps account for the relative acceptability of (133) by claiming that the pronominal R-word occupies this focus position.

133

%	Jan heeft	hier	met mij	daar	vaak	over	gesproken.	.. loc .. pron ..
	Jan has	here	with me	there	often	about	spoken
	'Jan often talked with me about that here.'

Note that Axiom II is restricted to demonstrative R-words and is not concerned with quantified R-words. Examples that contain both a demonstrative and a quantified R-word are ambiguous: the demonstrative pronoun daar in (134a-c) can either be interpreted as a locational pro-form or as a pronominal R-word (with perhaps a slight preference for the latter). This suggests that quantified R-words need not move into the [+R]-position, which is confirmed by the fact that an adverbial phrase may occur between the two R-words.

134

a.	Jan heeft	daar	(gisteren)	ergens	over	gesproken.
	Jan has	there	yesterday	somewhere	about	spoken
	'Jan spoke there about something yesterday.'
	'Jan spoke about that somewhere yesterday.'

b.	Jan heeft	daar	(gisteren)	nergens	over	gesproken.
	Jan has	there	yesterday	nowhere	about	spoken
	'Jan spoke there about nothing yesterday.'
	'Jan spoke nowhere about that yesterday.'

c.	Jan heeft	daar	(gisteren)	overal	over	gesproken.
	Jan has	there	yesterday	everywhere	about	spoken
	'Jan spoke there about everything yesterday.'
	'Jan spoke about that everywhere yesterday.'

[+] III. Co-occurrence of weak and strong R-words

Axioms I and II give rise to the expectation that weak and strong R-words cannot co-occur either. This expectation is not borne out, however, as can be seen in the examples in (135), which correspond to the examples in (124a&b).

135

a.	dat	er	hier	gedanst	wordt.	.. expl .. loc ..
	that	there	here	danced	is
	Intended reading: 'People are dancing here.'

b.	dat	er	hier	vaak	over	gesproken	wordt.	.. expl .. pron ..
	that	there	here	often	about	spoken	is
	Intended reading: 'People are talking about this.'

The examples in (136) show that strong R-words can also co-occur with quantitative er.

136

a.	dat	Jan er	hier [NP	twee [e]]	gezien	heeft.	.. quant .. loc ..
	that	Jan there	here	two	seen	has
	'that Jan saw two [e.g., rats] here.'

b.	dat	Jan er	hier [NP	drie [e]]	in	stopte.	.. quant .. pron ..
	that	Jan there	here	three	into	put
	'that Jan put three [e.g., cigars] in this.'

Example (137a) shows that a weak pronominal R-word can also be combined with a strong locational R-word. A strong pronominal R-word, on the other hand, cannot be combined with a weak locational R-word, as is shown in (137b).

137

a.	dat	Jan er	hier	vaak	over	praatte.	.. pron .. loc ..
	that	Jan there	here	often	about	talked
	'that Jan often talked about it here.'

dat Jan er hier vaak over praatte.

.. loc .. pron ..

Note that the examples in (135b) and (136b) are actually ambiguous and also allow a reading in which hier acts as a locational pro-form. In (135b) er then simultaneously performs the functions of expletive and pronominal R-word, and in (136b) it functions then both as licenser of the nominal gap [e] and as part of the pronominal PP; cf. Section 5.5.3, sub I.
In view of the data in (135) to (137) it seems clear that we cannot maintain the idea that the designated [+R]-positions in Axiom I and II can be identified, but that we need two distinct [+R]-positions. Given the fact that the weak R-pronoun always precedes the strong one, we postulate that the weak [+R]-position precedes the strong one. This gives rise to the clause structure in (138). Axioms I and II are now rephrased as in (138a&b).

138

XP V+fin (Subject) [+R]weak ... [+R]strong ... CLAUSAL ADVERB ... V-fin

Axiom I: Er must be moved into the unique weak [+R]-position.

Axiom II: A strong demonstrative R-word may be moved into the unique strong [+R]-position.

The axioms in (138) still leave the ungrammaticality of (137b) unexplained. The fact that the locational phrase is always the second (strong) R-word in the examples above has given rise to the idea that the second [+R]-position in (138) must always be used as a landing site for a demonstrative locational R-word, if there is one. From this two predictions follow. First, it is correctly predicted that only weak R-pronouns can be used in the presence of a locational R-word (but see example (133) for a potential problem); the strong R-position is occupied by the locational R-word and hence inaccessible to other strong R-words. Second, it is predicted that the locational R-word must be strong in the presence of another R-word; if the locational R-word were weak, it would have to move to the weak [+R]-position via the strong [+R]-position so that all landing sites for R-pronouns are occupied, the weak one by the phonetically realized locational R-word and the strong one by a trace of it. This gives an exhaustive account of the data discussed so far with the exception of (129c), which we have put aside.

[+] IV. Wh-movement and topicalization in multiple R-word constructions

This subsection discusses wh-movement and topicalization in multiple R-word constructions and show that there is a restriction on the co-occurrence of demonstrative and interrogative/topicalized R-words, subsection A will show that wh-movement and topicalization seem to be blocked by the presence of certain demonstrative R-words; we will account for this in terms of the two [+R]-positions in (138), subsection B will discuss an example in which wh-movement and topicalization of a -R word seems to be blocked by the presence of an R-word.

[+] A. Wh-movement and topicalization in double R-word constructions

In some cases, wh-movement of an R-word is blocked by the presence of another R-word. Consider examples (139a), which is ungrammatical under the interpretation that waar is part of the pronominal PP. This has been accounted for by assuming that the wh-word waar cannot be moved into clause-initial position in one fell swoop, but must first be placed into one of the [+R]-positions in (138). However, since we have seen that the locational R-word er is moved via the strong [+R]-position into the weak one, these positions are no longer available for the wh-word, so that movement of waar into the clause-initial position is blocked. If the locational R-word is strong, as in (139b), it does not move into the weak [+R]-position. Hence, this weak position is accessible to the wh-phrase and wh-movement is predicted to be possible.

139

a.	*	Waar	heeft	Jan er	vaak	over	gepraat?	.. pronwh .. loc ..
		where	has	Jan there	often	about	talked

a'.

.... V+fin ... [+R eri] ... [+Rti] ti vaak [waar over] ...

b.	Waar	heeft	Jan	hier	vaak	over	gepraat?	.. pronwh .. loc ..
	where	has	Jan	here	often	about	talked
	'What did Jan talk often about here?'

b'.

waarj V+fin ... [+Rtj] ... [+Rhieri]ti vaak [tj over] ...

The examples in (140) show that the same contrast can be found in the case of topicalization of demonstrative R-words.

140

a.	*	Daar	heeft	Jan er	vaak	over	gepraat.	.. prondemonstrative .. loc ..
		there	has	Jan there	often	about	talked

b.	Daar	heeft	Jan hier	vaak	over	gepraat.	.. prondemonstrative .. loc ..
	there	has	Jan here	often	about	talked
	'Jan talked often about that here.'

The judgments on the examples in (139) change under the reverse interpretation, that is, with the preposed interrogative R-word as locational and the second R-word as pronominal. In (141a), pronominal er can be placed in the weak [+R]-position, and the locational phrase can be moved via the strong [+R]-position into the clause-initial position. In (141b), however, the locational R-word has been moved via the strong R-position into clause-initial-position, so there is no landing position for the demonstrative R-word hier (the empty weak [+R]-position is of course not accessible to hier since it is not a weak R-word).

141

a.	Waar	heeft	Jan er	vaak	over	gepraat?	.. locwh .. pron ..
	where	has	Jan there	often	about	talked
	'Where did Jan talk about it often?'

a'.

waari V+fin

... [+R erj] ... [+Rti] ti vaak [tj over] ...

b.	??	Waar	heeft	Jan	hier	vaak	over	gepraat?	.. locwh .. pron ..
		where	has	Jan	here	often	about	talked

b'.

waari V+fin

... [+R ..] ... [+Rti] ti vaak [hier over] ...

The examples in (142) show that the same thing seems to hold for demonstrative pronominal R-words.

142

a.	Daar	heeft	Jan er	vaak	over	gepraat.	.. prondemonstrative .. loc ..
	there	has	Jan there	often	about	talked
	'Jan talked about it often there.'

b.	??	Daar	heeft	Jan hier	vaak	over	gepraat.	.. prondemonstrative .. loc ..
		there	has	Jan here	often	about	talked

The judgments on the examples in (141b) and (142b), which are the ones given in Huybregts (1991), are disputed by Bennis (p.c), who considers (141b) and (142b) fully acceptable and suggests that the relevant reading can be forced in relative clauses like (143). Although a locational interpretation of the relative pro-form waar seems more or less acceptable to us, the pronominal reading of waar remains the more prominent one, which would be in line with Huybregts' judgments on (141b) and (142b).

143

de universiteit	waar	Jan	hier	vaak	over	gesproken	heeft
the university	where	Jan	here	often	about	spoken	has
preferred reading: 'the university that John spoke often about here'
possible reading: 'the university where John spoke a lot about this'

The suggested analysis correctly predicts that expletive and quantitative er never block wh-movement of a locational or pronominal R-word: the latter can be moved via the strong [+R]-position into the clause-initial position. The examples in (144) and (145), which are the wh-movement counterparts of (135) and (136), show that this prediction is indeed borne out. Note in passing that, like (135b) and (136b), (144b) and (145b) also allow a reading in which waar'where' acts as a locational pro-form; the weak R-word er then simultaneously performs the functions of expletive and pronominal R-word; cf. Section 5.5.3, sub Section 5.5.3, sub I.

144

a.	Waar	wordt	er	gedanst?	.. locwh .. expl ..
	where	is	there	danced
	'Where do people dance?'

b.	Waar	wordt	er	vaak	over	gesproken?	.. pronwh .. expl ..
	where	is	there	often	about	talked
	'What are people often talking about?'

145

a.	Waar	heeft	Jan er	twee	gezien?	.. locwh .. quant ..
	where	has	Jan there	two	seen
	'Where did Jan see two [e.g., rats]?'

b.	Waar	stopte	Jan er	drie	in?	.. pronwh .. quant ..
	where	put	Jan there	three	into
	'Where did Jan put three [e.g., cigars] into?'

It has been reported, however, that topicalization differs from wh-movement; examples such as (146) with topicalized demonstrative daar differ sharply from the examples in (144), which involve interrogative waar.

146

a.	*	Daar	wordt	er	gedanst.	.. locdemonstrative .. expl ..
		there	is	there	danced
		'People are dancing there.'

b.	*	Daar	wordt	er	vaak	over	gesproken.	.. prondemonstrative .. expl ..
		there	is	there	often	about	talked
		'People are talking about that?'

The unacceptability of the examples in (146) is probably not due to the fact that the expletive blocks movement of the topicalized R-word, but to the fact that the presence of expletive er depends not only on whether a (definite) subject is present but also on whether the clause contains material that belongs to the presupposition of the clause; see Section N8.1.4 and Bennis (1986) for more discussion on this restriction on expletive er. For instance, example (147a), in which both the subject and the direct object are (nonspecific) indefinite, must contain the expletive, whereas (147b), which contains a definite direct object, is marginal at best if expletive er is present; if the direct object is a pronoun, as in (147c), er must definitely be absent. The contrast between (144) and (146) can therefore be traced back to the fact that interrogative elements like waar are indefinite by definition, whereas demonstrative elements like daar are definite.

147

a.	dat	*(er)	iemand	gisteren	iets	vertelde.
	that	there	someone	yesterday	something	told
	'that someone told a story yesterday.'

b.		dat	(??er)	iemand	het verhaal	gisteren	vertelde.
		that	there	someone	the story	yesterday	told

c.		dat	(*er)	iemand	het	gisteren	vertelde.
		that	there	someone	it	yesterday	told

The examples in (148) show that replacement of waar by daar does not affect the grammaticality judgments when we are dealing with quantitative er; the judgments on the examples in (148) are similar to those in (145).

148

a.	Daar	heeft	Jan er	twee	gezien?	.. locdemonstrative .. quant ..
	there	has	Jan there	two	seen
	'Jan saw two [e.g., rats] there?'

b.	Daar	stopte	Jan er	drie	in?	.. prondemonstrative .. quant ..
	there	put	Jan there	three	into
	'Jan put three [e.g., cigars] into that ?'

That the judgments on (145) and (148) are similar is consistent with the fact, illustrated in (149), that quantitative er must be realized irrespective of the presence of presuppositional material.

149

a.	Jan heeft	er	[NP	twee e]	een verhaal	verteld.
	Jan has	there	[NP	two	a story	told
	'Jan told a story to two [e.g., children].'

Jan heeft

[