• Dutch
  • Frisian
  • Saterfrisian
  • Afrikaans
Show all
14.3.Right dislocation
quickinfo

This section discusses two types of right dislocation (henceforth: RD), which are illustrated in example (93). Semantically, RD is characterized by the fact that the dislocated phrase adds more specific information to what is said in the preceding clause: in (93), the right-dislocated phrases provide more information about the function of Mr Jansen: he is a manager in some presupposed organization. The two types of RD differ in that afterthoughts provide additional information that is new for the hearer while backgrounded phrases provide information already familiar to the hearer but which may help him to identify the intended reference; cf. Ott & De Vries (2015). Furthermore, afterthoughts but not backgrounded phrases can often be preceded by a special marker such as je weet wel'you probably know who': Ik heb dhr. Jansen gesproken, je weet wel, de directeur.

93
a. Ik heb dhr. Jànsen gesproken, de directeur.
afterthought RD
  have  Mr Jansen  spoken  the manager
  'I have spoken to Mr Jansen, the manager.'
b. Ik heb dhr. Jansen gespròken, de directeur.
backgrounding RD
  have  Mr Jansen  spoken  the manager
  'I have spoken Mr Jansen, the manager.'

Speakers’ judgments seem to differ with respect to the question as to whether the clause-internal correlate of an afterthought can be a weak proform. For some speakers (including the second author of this work) substituting the direct object in the afterthought construction (93a) results in a degraded result, while it is easily possible in the backgrounding construction in (93b). In the remainder of this work we will follow the more permissive variety, according to which examples in (94) are both fully acceptable. We leave it to future research to investigate the differences in speakers’ judgment in more detail.

94
a. % Ik heb ʼm gespròken, de directeur.
afterthought RD
  have  him  spoken  the manager
  'I have spoken him, the manager.'
b. Ik heb ʼm gespròken, de directeur.
backgrounding RD
  have  him  spoken  the manager
  'I have spoken him, the manager.'

Phonetically, RD constructions are characterized by the fact that the right-dislocated phrase cannot receive sentence accent (indicated by a grave accent in the examples above); this accent is always located on some element in the preceding clause. That sentence accent cannot be placed on the right-dislocated phrase is related to the fact that the latter can be preceded by an intonation break: in the case of afterthoughts this break is normally distinctly present while, at least in casual speech, it is often less prominent in the case of backgrounding. The two types of RD also differ in that afterthoughts are assigned contrastive accent (indicated by small caps), while backgrounded phrases are normally pronounced with a flat intonation contour (that is, without a prominent accent).
      This section is organized as follows. Subsection I starts by showing that RD resembles left dislocation (LD) in various respects. Subsection II continues by briefly reviewing a number of differences between RD and extraposition; this partly repeats information which was discussed in more detail in Chapter 12, to which we refer the reader for more information. Subsection III discusses a number of restrictions on right-dislocated phrases and their clause-internal correlate (if present). Subsection IV continues by discussing a number of differences between afterthought RD and backgrounding RD; we will see that these can be traced back to the fact that afterthoughts provide discourse-new while backgrounded phrases provide discourse-old information. Subsection V shows that RD resembles hanging-topic LD in that it is not sensitive to various islands for wh-extraction and Subsection VI discusses a number of word order issues related to RD. Subsection VII, finally, discusses some possible theoretical approaches to RD.

readmore
[+]  I.  A brief comparison between left and right dislocation

The two types of RD constructions, illustrated again in the (a)-examples in (95), resemble in various respects the two types of LD constructions discussed in Section 14.2, which are illustrated in the (b)-examples.

95
a. Ik heb hem gespròken, de directeur.
afterthought RD
  have  him spoken  the manager
  'I spoke to him, the manager.'
a'. Ik heb hem gespròken, de directeur.
backgrounding RD
  have  him spoken  the manager
  'I spoke to him, the manager.'
b. De directeur, die heb ik gesproken.
contrastive LD
  the manager  dem  have  spoken
  'The manager, I have spoken to him.'
b'. De directeur, ik heb hem gesproken.
hanging-topic LD
  the manager  have  him  spoken
  'The manager, I have spoken to him.'

First, all four types of dislocated phrases in (95) seem to be clause-external, as all of them have a clause-internal pronominal associate, namely the pronouns hem'him' and die'that'. The fact that the thematic role of the verb is assigned to the pronoun suggests that the dislocated phrases are not licensed within the clause but in some other way. That the dislocated phrases are not part of the clause is also supported by the fact that (like parentheticals) they are separated from the intonation contour of the clause, which, at least in the case of afterthoughts and contrastive left-dislocated phrases, goes hand-in-hand with a distinct intonation break. More reasons for assuming that dislocated phrases are clause-external are that left-dislocated phrases precede the sentence-initial position and that right-dislocated phrases cannot be assigned sentence accent, which is again indicated by a grave accent in the (a)-examples.
      Secondly, all four types of dislocated phrases provide more specific information than their clause-internal associate: in (95), they all provide information about the function of the person referred to by the pronoun.
      Thirdly, LD and RD both come in two types: one type in which the dislocated phrase is typically accented and one in which the dislocated phrase is normally pronounced with a flat intonation contour. It should be noted, however, that contrastive LD and afterthought RD differ in that the former invites a set of alternative propositions, while the latter simply provides discourse-new information. This can be brought to light by the examples in (96): while (96a) is fully compatible with a contrastive maar-phrase, the use of a contrastive maar-phrase gives rise to a somewhat marked result in (96b). Note in passing that this example becomes acceptable if the weak referential pronoun is replaced by a contrastively stressed pronoun, but in that case the contrastive phrase is licensed by the pronoun and not by the afterthought.

96
a. Jan, die heb ik niet gezien (maar Marie wel).
contrastive LD
  Jan  dem  have  not  seen  but  Marie  aff
  'Jan, I havenʼt seen him, but I did see Marie.'
b. Ik heb ʼm niet gezien, Jan, (??maar Marie wel).
afterthought RD
  have  him  not  seen  Jan     but  Marie  aff
  'I havenʼt seen him, Jan.'

      Finally, example (97a) shows that right-dislocated phrases may provide information that may help the speaker to identify the intended reference of the clause-internal nominal correlate, but example (97b) shows that the right-dislocated phrase may also be an epithet. If the nominal correlate is predicative, as in (97c), the right-dislocated phrase provides a more precise qualification.

97
a. dat hij te laat kwam, mijn broer.
  that  he  too late  came my brother
  'that he arrived too late, my brother.'
b. dat Peter te laat kwam, de sukkel.
  that  Peter  too late  came  the twerp
  'that Peter came too late, the twerp.'
c. dat Jan een groot kunstenaar is, de beste schilder die ik ken.
  that  Jan a great artist is  the best painter that I know
  'that Jan is a great artist, the best painter I know.'

The fact that left-dislocated phrases do not seem to have such “modifying” function is the principal reason why we will use different notions for the relation between the left and right-dislocated phrases and their clause-internal associate: resumption versus correlation. Left-dislocated phrases provide information that is simply taken up again by their clause-internal associate, while right-dislocated phrases provide more specific information than their clause-internal associate (or about it).

[+]  II.  Right dislocation versus extraposition

Right-dislocated and extraposed phrases both follow the verbs in clause-final position; consequently, it may sometimes be difficult to distinguish the two cases. In the case of nominal arguments, confusion will not easily arise because extraposition of such arguments is normally not possible. One example is given in (98); for more examples, see Section 12.2, sub I.

98
Ik heb <de directeur > gesproken <*de directeur >.
extraposition
  have     the manager  spoken
'I have spoken to the manager.'

Furthermore, nominal arguments are generally obligatorily present, as a result of which right-dislocated nominal phrases will typically have an overt clause-internal correlate, as in (93) and the (a)-examples in (95). That means that right-dislocated nominal phrases without a correlate are only expected in the case of pseudo-intransitive verbs and (optional) indirect objects. Example (99) shows that the former case indeed occurs, but only if the right-dislocated phrase functions as an afterthought, that is, if it is accented and provides new information.

99
Jan heeft altijd graag gerookt, sigaren/*sigaren.
RD
  Jan has  always  gladly  smoked  cigars/cigars
'Jan has always liked to smoke, cigars.'

That we are dealing with a right-dislocated phrase in (99) is immediately clear from its position after the clause-final verb gerookt'smoked'. However, if no clause-final verb is present, as in the examples in (100), confusion could arise with cases in which the object occupies its regular position in the middle field of the clause, but the intonation pattern normally provides sufficient information to distinguish the two: the direct object in (100a) is integrated in the intonation contour of the clause and can carry sentence accent; the afterthought in (100b) is preceded by an intonation break and assigned contrastive accent while the sentence accent is assigned to some other element in the preceding clause.

100
a. Jan rookt graag (*vooral) sigàren.
object occupies the middle field
  Jan smokes  gladly  especially  cigars
  'Jan likes to smoke cigars.'
b. Jan ròòkt graag, (vooral) sigaren.
RD
  Jan smokes  gladly  especially  cigars
  'Jan likes to smoke, (especially) cigars.'

The distribution of the focus particle vooral'especially' can also be used as a test for recognizing RD in examples such as (100). The examples in (101) show that afterthoughts can easily be preceded by this element if the clause-internal correlate is indefinite or not present.

101
a. Jan heeft veel boeken gekocht, vooral romans.
  Jan has  many books bought especially novels
  'Jan has bought many books, especially novels.'
b. Jan heeft altijd graag gerookt, vooral sigaren.
  Jan has  always  gladly  smoked  especially cigars
  'Jan has always liked to smoke, especially cigars.'

Clause-internal phrases, on the other hand, can only be preceded by a focus particle if they are scrambled into a designated focus position, which precedes manner adverbs such as graag'gladly'; cf. (102). The fact that vooral can immediately precede the direct object sigaren in (100b) but not in (100a) therefore supports the proposed analysis.

102
a. dat Jan <vooral romans> graag <*vooral romans> leest.
  that  Jan   especially novels  gladly  reads
  'that Jan especially likes to read novels.'
b. dat Jan <vooral sigaren> graag <*vooral sigaren> rookt.
  that  Jan especially cigars  gladly  smokes
  'that Jan especially likes to smoke cigars.'

The second case in which a right-dislocated nominal phrase may be expected to occur without an overt correlate pertains to ditransitive constructions without an (overt) indirect object, but it seems that such cases do not occur. Example (103b) shows that they are degraded regardless of whether the right-dislocated phrase expresses new or old information; this may be due to the fact that the alternative with a prepositional indirect object in (103b') is preferred.

103
a. dat Jan (zijn vrouw) graag bloemen geeft.
  that  Jan   his wife  gladly  flowers  gives
  'that Jan likes to give (his wife) flowers.'
b. dat Jan graag bloemen geeft, (vooral) zijn ??vrouw/*vrouw.
RD
  that  Jan gladly  flowers  gives  especially  his wife/wife
b'. dat Jan graag bloemen geeft, (vooral) aan zijn vrouw.
RD
  that  Jan gladly  flowers  gives  especially  to his wife
  'Jan likes to give flowers, especially to his wife.'

Because prepositional indirect objects can be extraposed, confusion between extraposition and RD may arise in such cases, but again the intonation contour will generally provide sufficient information to identify the two cases (cf. Ott & De Vries 2015): the extraposed prepositional indirect object in (104a) is integrated in the intonation contour of the clause and can even carry sentence accent; the afterthought in (104b) is separated from the preceding clause by a distinct intonation break and is assigned contrastive accent. The two cases again differ in that only the latter can be preceded by the marker vooral.

104
a. dat Jan graag bloemen geeft (*vooral) aan zijn vròuw.
extraposition
  that  Jan gladly  flowers  gives  especially  to his wife
  'Jan likes to give flowers to his wife.'
b. dat Jan graag blòemen geeft, (vooral) aan zijn vrouw.
RD
  that  Jan gladly  flowers  gives  especially  to his wife
  'Jan likes to give flowers, especially to his wife.'

Prepositional objects that are obligatorily realized will not pose any problems either. The examples in (105) show that in such cases right-dislocated PPs typically have an overt clause-internal correlate, while extraposed PPs cannot be combined with such correlates. Observe that in the case of prepositional objects, the right-dislocated phrase need not be an afterthought but can also be backgrounded.

105
a. dat Jan (*ernaar) verlangt naar vakantie.
extraposition
  that  Jan     for.it  longs  for vacation
  'that Jan is longing for a vacation.'
b. dat Jan *(ernaar) verlangt, naar vakantie/vakantie.
RD
  that  Jan     for.it  longs  for vacation/vacation
  'that Jan is longing for.it, for a vacation.'

If the prepositional object is optional, as in the case of wachten (op)'to wait for', similar problems may arise as with pseudo-intransitive verbs in that we mainly have to rely on the intonation pattern of the construction if the correlate of the right-dislocated PP is not overtly expressed.

106
a. dat Jan al weken (*erop) wacht op zijn bòeken.
extraposition
  that  Jan  already  weeks     for.it  waits  for his books
  'that Jan is already waiting for weeks for his books.'
b. dat Jan al weken (erop) wacht, op zijn boeken/boeken.
RD
  that  Jan  already  weeks   for.it  waits  for his books/books
  'that Jan has already been waiting for weeks, for his books.'

Fortunately, there is an additional syntactic test that may help us distinguish extraposition from RD, namely VP-topicalization. While extraposed phrases can be pied piped under VP-topicalization, right-dislocated phrases are generally stranded.

107
a. [VP Wachten op zijn boeken] doet hij al weken.
extraposition
  wait for his books does  he  already  weeks
b. [VP Wachten] doet hij al weken, op zijn boeken/boeken.
RD
  wait does  he  already  weeks  for his books/books
b'. * [VP Wachten, op zijn boeken/boeken] doet hij al weken.
RD
  wait  for his books  does  he  already  weeks

Adverbial PPs pose similar problems as optional prepositional objects because they are normally optional, as shown for a comitative met-PP in (108a) and a locative PP in (109a). Consequently, right-dislocated adverbial PPs without a correlate in the preceding clause could in principle be confused with their extraposed counterparts, but the (b)- and (c)-examples show that the two diagnostics used above, intonation and VP-topicalization, may help us make the correct distinction.

108
a. dat Jan graag (met Peter) schaakt.
  that  Jan gladly   with Peter  plays.chess
  'that Jan likes to play chess (with Peter).'
b. dat Jan graag (*met hem) schaakt met Peter.
extraposition
  that Jan  gladly     with him  plays.chess  with Peter
b'. [VP Schaken met Peter] doet Jan graag.
  play.chess  with Peter  does  Jan  gladly
c. dat Jan graag (met hem) schaakt, met Peter/Peter.
RD
  that  Jan gladly  with him  plays.chess  with Peter/Peter
c'. [VP Schaken] doet Jan graag, met Peter/Peter.
  play.chess  does  Jan gladly  with Peter/Peter
c''. * [VP Schaken, met Peter/Peter] doet Jan graag.
  play.chess  with Peter/Peter  does  Jan gladly
109
a. dat Jan graag (op zijn club) schaakt.
  that  Jan gladly   at his club  plays.chess
  'That Jan likes to play chess (at his club).'
b. dat Jan (*daar) graag schaakt op zijn club.
extraposition
  that Jan     there  gladly  plays.chess  at his club
b'. [VP Schaken op zijn club] doet Jan graag.
  play.chess  at his club  does  Jan  gladly
c. dat Jan (daar) graag schaakt, op zijn club/club.
RD
  that  Jan   there  gladly  plays.chess  at his club/club
c'. [VP Schaken] doet Jan graag, op zijn club/club.
  play.chess  does  Jan gladly  at his club/club
c''. * [VP Schaken, op zijn club/club] doet Jan graag.
  play.chess  at his club/club  does  Jan gladly

Some clausal constituents such as supplementives cannot be extraposed while they can be right-dislocated. This can also be brought to light more clearly by means of VP-topicalization as they cannot be pied-piped if they are in postverbal position.

110
a. Jan is daarnet <kwaad> weggelopen <*kwaad>.
no extraposition
  Jan is just.now    angry  away-walked
  'Jan walked away angry just now.'
a'. [VP <kwaad> weggelopen <*kwaad>] is Jan daarnet.
  angry  away-walked  is Jan just.now
b. Jan is daarnet weggelopen, kwaad.
RD
  Jan is just.now  away-walked angry
  'Jan walked away angry just now.'
b'. [VP Weggelopen] is Jan daarnet, kwaad.
  away-walked  is Jan just.now  angry
b''. * [VP Weggelopen, kwaad] is Jan daarnet.
  away-walked  angry  is Jan just.now

The same pattern can be seen in various types of (especially non-prepositional) adverbial phrases headed by a manner adverb such as zorgvuldig'carefully', a temporal adverb such as morgen'tomorrow', or a modal adverb such as misschien'maybe'. It can again be brought to light by means of VP-topicalization as these modifiers must be stranded, as illustrated in (111) for the manner adverb zorgvuldig'carefully'.

111
a. Jan heeft het boek <zorgvuldig> gelezen <*zorgvuldig>.
no extraposition
  Jan has  the book    carefully  read
  'Jan has read the book carefully.'
a'. [VP zorgvuldig gelezen] heeft Jan het boek.
  carefully  read  has  Jan the book
b. Jan heeft het boek gelezen*(,) zorgvuldig.
RD
  Jan has  the book  read  carefully
b'. [VP gelezen] heeft Jan het boek, zorgvuldig.
  read  has  Jan the book  carefully
b''. * [VP gelezen, zorgvuldig] heeft Jan het boek.
  read  carefully  has  Jan the book

Because the two diagnostics have been more systematically applied to a wider range of constructions in our discussion of extraposition, we will not digress on this issue any further here but refer the reader to Chapter 12 for more discussion.

[+]  III.  Restrictions on right-dislocated phrases and their clause-internal correlates

Backgrounded phrases resemble hanging topics in that they are pronounced with a flat intonation contour, but the examples in (112) show that they are more flexible with respect to their categorial status; while hanging topics are typically nominal in nature, backgrounded phrases can be nominal, clausal, adjectival, or adpositional. The examples in (112) also show that the clause-internal correlate of the backgrounded phrase may perform various syntactic functions: it can be an argument, as in the (a)-examples, a complementive (112b), or an adverbial phrase (112c). The correlate is typically a phonetically light element, like the pronouns ʼm'him' and het'it' or the R-word er'there', although the phonetically heavier demonstrative forms like die/dat'that' and daar'there' are occasionally found as well.

112
a. Ik heb ʼm niet meer gezien, Peter.
noun phrase
  have  him  not  anymore  seen  Peter
  'I havenʼt seen him anymore, Peter.'
a'. Hij heeft ʼt me gisteren verteld, dat hij vertrekt.
clause
  he  has  it  me yesterday  told  that he leaves
  'He told it to me yesterday, that he is leaving.'
b. Ik ben ʼt mijn hele leven geweest, gelukkig.
AP
  am  it  my whole life  been happy
  'I have been it my whole life, happy.'
c. Ik ben er gisteren nog geweest, in Utrecht.
PP
  am  there  yesterday  prt  been  in Utrecht
  'I was there yesterday, in Utrecht.'

Judgments on the examples in (112) do not seem to change if we assign contrastive accent to the right-dislocated phrase, that is, afterthoughts have the same properties as backgrounded phrases, but the correlate can more easily be heavy or phrasal. This is illustrated in (113).

113
a. Ik heb die jongen niet meer gezien, Peter.
noun phrase
  have  that boy  not  anymore  seen  Peter
  'I havenʼt seen that boy anymore, Peter.'
a'. Hij heeft me dat gisteren verteld, dat hij vertrekt.
clause
  he  has  me that  yesterday  told  that he leaves
  'He told me that yesterday, that he is leaving.'
b. Ik ben dat eigenlijk mijn hele leven geweest, gelukkig.
AP
  am  it  in.fact  my whole life  been happy
  'I have in fact been that my whole life, happy.'
c. Ik ben daar gisteren nog geweest, in Utrecht.
PP
  have  there  yesterday  prt  been  in Utrecht
  'I have been there yesterday, in Utrecht.'

      Right-dislocated phrases add to the information expressed by their correlates: (114a) presupposes that the hearer does not know that Mr Jansen is the manager, and (114b) suggests that the hearer may confuse the intended referent with someone who is not the manager. The right-dislocated phrase and its correlate can be interchanged but then it is presupposed that the hearer does not know the name of the manager or may confuse the intended referent with someone who is not called Jansen.

114
a. Ik heb dhr. Jànsen gesproken, de directeur.
afterthought RD
  have  Mr Jansen  spoken  the manager
  'I have spoken to Mr Jansen, the manager.'
b. Ik heb dhr. Jànsen gesproken, de directeur.
backgrounding RD
  have  Mr Jansen  spoken  the manager
  'I have spoken to Mr Jansen, the manager.'

The examples in (115) show that afterthoughts of the type in (114a) can surface in German either as an accusative or as a nominative noun phrase. The two cases have a slightly different meaning, which Ott & De Vries (2015: Section 6) try to clarify by means of the paraphrases which are given here as translations: in the first case, the referent of the correlate is contextually given and the afterthought simply provides more specific information about this referent; in the second case, the correlate may introduce a new referent into the discourse, and the afterthought is used to identify this referent as the speaker’s neighbor. Because of this difference in meaning, Ott & De Vries refer to these cases as, respectively, specificational and predicative afterthoughts, and provide different analyses for the two cases. The same meaning difference is found in Dutch, but since Dutch has no morphological case we will largely ignore predicative afterthoughts in this section and refer the reader to Ott & De Vries’ article for more discussion of this type. Note in passing that Van Riemsdijk (1997) and Van Riemsdijk & Zwart (1997:fn.5) observe a similar optionality in case agreement in German LD constructions, although in such constructions a mismatch in case assignment does not seem to trigger a similar predicative reading.

115
a. Ich habe denacc Jan getroffen, meinenacc Nachbar.
specificational
  have the  Jan met  my neighbor
  'I have met Jan, that is, I have met my neighbor.'
b. Ich habe denacc Jan getroffen, meinnom Nachbar.
predicative
  have the  Jan met  my  neighbor
  'I have met Jan, who is my neighbor.'

That right-dislocated phrases must provide more specific information than their clause-internal correlates also accounts for the acceptability contrast between the (a)-examples in (116): since referential pronouns and definite noun phrases both presuppose that the hearer is able to identify the intended referent, definite noun phrases are more informative due to their descriptive content, and consequently the pronoun cannot be the right-dislocated phrase. Note that we mark (116b) with a dollar sign because the construction is certainly not ungrammatical, as is clear from the fact that it becomes felicitous if the right-dislocated pronoun is accompanied by specific extra-linguistic information, such as a pointing gesture. The acceptability of (b)-examples is also expected because the two coordinated phrases provide more precise information than their clause-internal pronominal correlates.

116
a. Ik heb hem gesproken, de directeur.
  have  him  spoken  the manager
a'. $ Ik heb de directeur gesproken, hem.
  have  the manager  spoken  him
b. Jan heeft ons uitgenodigd, jou and mij.
  Jan has  us  prt.-invited  you and me
b'. Jan heeft jullie uitgenodigd, jou en haar.
  Jan has  you  prt.-invited  you and her

      Backgrounded noun phrases are generally definite noun phrases, due to the fact that they express discourse-old information. Afterthoughts, on the other hand can be indefinite provided they are more informative then their clause-internal correlates: this implies that the correlate must be indefinite as well.

117
a. Ik heb iets/*hem gekocht, een rode vaas.
afterthought RD
  have  something/him  bought  a red vase
  'I have bought something, a red vase.'
b. * Ik heb iets/hem gekocht, een rode vaas.
backgrounding RD
  have  something/him  bought  a red vase

      The (a)-examples in (118) show that RD is like LD in that it cannot be applied to non-referential expressions; while definite noun phrases can easily be right-dislocated, quantified noun phrases cannot. The (b)-examples illustrate the same by showing that non-referential parts of idiomatic expressions resist right dislocation.

118
a. Ik heb de/iedere kandidaat gesproken.
  have  the/every candidate  spoken
  'I haven spoken to the/every candidate.'
a'. Ik heb hem gesproken, de/*iedere kandidaat.
  have  him  spoken  the/every candidate
b. Ik geloof er de ballen van.
  believe  there  the balls  of
  'I donʼt believe any of it.'
b'. * Ik geloof ze er van, de ballen.
  believe  them  there  of  the balls

      De Vries (2009) claims that nominal right-dislocated phrases do not exhibit connectivity effects for binding and, at first sight, this appears indeed to be borne out by the unacceptability of example (119a): RD of reflexive/reciprocal pronouns leads to unacceptability because the pronoun is not bound by a local antecedent. However, it is not clear whether the unacceptability of (119a) is really the result of the lack of connectivity; it may also be due to the fact that under the given coindexation the referential pronoun ze'them' is incorrectly bound within its local domain, the clause. This means that in order to investigate whether connectivity effects occur we have to consider more complex examples like (119b), in which the anaphor is embedded in a right-dislocated phrase. The acceptability status of such examples is somewhat unclear: while De Vries (2009) gives similar examples as unacceptable, we find them acceptable and certainly much better than examples such as (119a).

119
a. * [Jan en Peter]i vinden zei erg knap, zichzelfi/elkaari.
  Jan and Peter  consider  them  very bright  themselves/each other
b. % [Jan en Peter]i vinden zej erg spannend, [elkaarsi boeken]j.
  Jan and Peter  consider  them  very exciting  each.other’s books
  'Jan and Peter consider them very exciting, each others books.'

That connectivity effects with reflexive pronouns do occur is also clear from the fact that they may occur as an afterthought in examples such as (120a), taken from Ott & De Vries (2015); example (120b) provides a similar case with a reciprocal.

120
a. Jan zag iemand in de spiegel, zichzelf.
  Jan saw  someone  in the mirror  himself
b. Jan en Peter beschuldigen alletwee iemand van fraude, elkaar.
  Jan and Peter  accuse  both  someone  of fraud  each.other
  'Jan and Peter both accuse someone of fraud: each other.'

Example (121) shows that connectivity effects can also be demonstrated by means of bound variable reading: De Vries (2009) claims this reading to be unavailable, but similar examples are given as fully acceptable in Ott & De Vries (2015); we agree with them.

121
Elke schrijveri is er trots op, [zʼni debuut].
  every writer  is  there  proud  of   his debut
'Every writer is very proud of it, his debut.'

      Example (122a) suggests that connectivity effects also occur with referential expressions: the proper noun Peter cannot be construed as coreferential with the subject pronoun hij'he'. De Vries (2009) correctly points out, however, that this is not a telling fact because coreferentiality is also blocked if the pronoun and the proper noun occur in two subsequent clauses; linear order may be the crucial factor here.

122
a. * Hiji heeft hetj gelezen, [dat boek van Peteri]j.
  he  has  it  read   that book by Peter
  Intended reading: 'Peter has read it, his own book.'
b. * Hiji heeft de boeken ontvangen, maar het boek van Peteri ontbrak.
  he  has  the book  received  but  the book by Peter  was.missing
  Intended reading: 'Peter has received the books, but his own book was missing.'

      If the correlate of a right-dislocated phrase is a non-obligatory clausal constituent, it can be omitted. Various cases were already given in Subsection II, but we illustrate this again in (123a) for a temporal adverbial phrase. Some right-dislocated phrases never have a correlate, simply because there is no proform available; this holds for modal adverbs like misschien'maybe' in (123b).

123
a. Ik was (toen) erg moe, na die lange wandeling.
  was   then  very tired  after that long walk
  'I was very tired then, after that long walk.'
b. Hij komt morgen, misschien.
  he  comes  tomorrow  maybe
  'He will come tomorrow, maybe.'
[+]  IV.  Differences between afterthoughts and backgrounded phrases

Although the previous subsection has shown that afterthoughts and backgrounded phrases as well as their correlates exhibit similar behavior in various respects, there are also a number of differences; see Ott & De Vries (2015) and the references cited there. We have already mentioned that afterthoughts provide new information while backgrounded phrases express information already known to the hearer. This can easily be shown by the question-answer pair in (124): RD of the noun phrase Marie is possible in the answer, but only if pronounced with a flat intonation.

124
a. Ken jij Marie?
  know  you  Marie
  'Do you know Marie?'
b. Ja, ik ken haar goed, Marie/*Marie.
  yes  know  her  well  Marie/Marie

Related to this difference in information load is that afterthoughts can be preceded by epistemic modal adverbs such as waarschijnlijk'probably' if their correlates are indefinite; this option does not arise with backgrounded phrases as their correlates normally refer to entities known both to the speaker and to the hearer. For the same reason, afterthoughts but not backgrounded phrases can be combined with a modality marker like wellicht'maybe' or an hesitation marker like toch.

125
a. Jan heeft iemand bezocht, waarschijnlijk Marie.
afterthought
  Jan has  someone  visited  probably  Marie
  'Jan has visited someone, probably Marie.'
a'. * Jan heeft haar bezocht, waarschijnlijk Marie.
backgrounded
  Jan has  her  visited  probably  Marie
b. Jan heeft iemand bezocht, Marie wellicht/toch?
afterthought
  Jan has  someone  visited  Marie  perhaps/prt
  'Jan has visited someone; Marie perhaps/it was Marie, wasnʼt it?'
b'. * Jan heeft haar bezocht, Marie wellicht/toch?
backgrounded
  Jan has  her  visited  Marie  perhaps/prt

As afterthoughts add more specific information to the assertion in the preceding clause, the hearer can negate the added information independently of the clause. For the same reason the afterthought can be provided by the hearer. Note that examples like (126b) show that afterthoughts can at least sometimes be independent from the clause containing their correlate; Subsection VI will provide more evidence in favor of this conclusion.

126
a. Jan heeft iemand bezocht, Marie.
speaker A
  Jan has  someone  visited  Marie
  'Jan has visited someone, Marie.'
a'. Niet waar: hij was de hele dag thuis/het was Els.
speaker B
  not  true  he  was  the whole day  home/it  was  Els
  'Thatʼs not true: he has been at home all day/it was Els.'
b. Jan heeft iemand bezocht. Ja, Marie.
speaker A & B
  Jan has  someone  visited  yes  Marie
  'Jan has visited someone. Yes, Marie.'

This subsection has shown that there are a number of differences in use between afterthoughts and backgrounded phrases, which can be traced back to the role they play in the information structure of the discourse; afterthoughts provide discourse-new, while backgrounded phrase provide discourse-old information.

[+]  V.  Island-sensitivity

Section 14.2, sub V, has shown that hanging-topic LD is not island-sensitive, due to the resumptive pronoun hem'him' remaining in situ. Contrastive LD is different in this respect as the resumptive demonstrative pronoun die must be moved into sentence-initial position; the contrast in (127) can be attributed to the fact that the demonstrative die is extracted from an interrogative clause.

127
a. Jani, [Ik weet niet [wie (of) hemi geholpen heeft]].
hanging-topic LD
  Jan   I  know  not  who comp  him  helped  has
  'Jan, I donʼt know who has helped him.'
b. * Jani, [diei weet ik niet [wie (of) ti geholpen heeft]].
contrastive LD
  Jan  dem  know  not   who  comp  helped  has

      Because RD is like hanging-topic LD in that it does not involve movement of the correlate of the right-dislocated phrase, we expect it not to be island-sensitive either. The examples in (128) show that this is indeed borne out for interrogative and adjunct clauses, for afterthoughts as well as backgrounded phrases.

128
a. Ik weet niet [wie (of) hemi geholpen heeft], Jani/Jani.
  know  not  who comp  him  helped  has  Jan/Jan
  'I donʼt know who has helped him, Jan.'
b. Ik ben bedroefd [omdat ik hemi niet gezien heb], Jani/Jani.
  am  sad  because  him  not  seen  have  Jan/Jan
  'I am sad because I havenʼt seen him, Jan.'

      The situation is less clear for non-clausal islands, an issue to which we will return in Subsection VII. The examples in (129) show that RD seems to be like LD in that it is not sensitive to the islandhood of PPs: the preposition can be used but is not needed. We added a percentage sign to (129b), however, because Zwart (2011:78) as well as Ott & De Vries (2015:40ff.) have claimed that the preposition must be realized; for them RD may therefore be sensitive to the islandhood of PPs. Note in passing that an anonymous reviewer of De Ott & De Vries (2015) indicated that not all Dutch speakers require a preposition to be present.

129
a. Jani, ik wil op hemi niet langer wachten.
hanging-topic LD
  Jan  want  for him  no longer  wait
  'Jan, I donʼt want to wait for him any longer.'
b. Ik wil op hemi niet langer wachten, %(op)Jani/Jani.
RD
  want  for him  no longer  wait  for Jan/Jan
  'I donʼt want to wait for him any longer, Jan.'

      The primeless examples in (130) show that there a sharp acceptability contrast between hanging-topic LD and RD if the correlate is embedded in a coordinate structure; this may suggest that RD is sensitive to the islandhood of those structures. It should be noted, however, that the primed examples show that a similar contrast is found if the correlate is simply embedded in, e.g., a direct object, which indicates that we cannot attribute the ungrammaticality to the presence of the coordinate structure.

130
a. Jani, ik heb [hemi en zijni vrouw] niet gezien.
hanging-topic LD
  Jan,  have   him and his wife  not seen
  'Jan, I havenʼt seen him and his wife.'
a'. Jani, ik heb [zijni vrouw] niet gezien.
  Jan,  have   his wife  not seen
  'Jan, I havenʼt seen his wife.'
b. * Ik heb [hemi en zijni vrouw] niet gezien, Jani/Jani.
RD
  have   him and his wife not seen  Jan/Jan
b'. * Ik heb [zijni vrouw] niet gezien, Jani/Jani.
  have   his wife not seen  Jan/Jan

Finally, we need to point out that on the basis of example (131a) De Vries (2002) suggests that attributively used APs can be used as afterthoughts, which would of course be another example of island-insensitivity. However, Veld (1993:132ff.) already pointed out that this is only apparent: we are dealing with reduced noun phrases, as is also clear from the fact that the indefinite article een'a' shows up obligatorily if the correlate is singular, as in (131b). Like (130b), this shows that RD is sensitive to certain non-clausal islands.

131
a. Jan heeft druiven gekocht, witte.
  Jan has  grapes bought,  white
  'Jan has bought grapes, white ones.'
b. Jan heeft een auto gekocht, een witte.
  Jan has  a car  bought,  a white
  'Jan has bought a car, a white one.'
[+]  VI.  Word order restrictions

Right-dislocated phrases normally follow extraposed phrases such as the obligatory prepositional object naar meer informatie in (132): placement of the modal adverb in a position between the clause-final verb and the extraposed PP gives rise to a degraded result. For completeness’ sake, we contrast example (132b) with example (132b'), in which the PP is right-dislocated.

132
a. dat Jan verlangt naar meer informatie, waarschijnlijk.
  that  Jan longs  for more information  probably
  'that Jan probably wishes more information.'
b. * dat Jan verlangt, waarschijnlijk, naar meer informatie.
  that  Jan longs  probably  for more information
b'. dat Jan ernaar verlangt, waarschijnlijk, naar meer informatie.
  that  Jan for.it  longs  probably  for more information

Right-dislocated phrases also obey certain order restrictions: cf. Ott & De Vries (2015). Before illustrating this, we want to observe that backgrounded phrases and afterthoughts can be reiterated (although some speakers have difficulty with this). Note in passing that the first afterthought in (133b) is not only marked by means of accent but also by the fact that it can be preceded by althans; we will use this as a diagnostic in (134).

133
a. Jan heeft haar gezien, gisteren, die vrouw.
backgrounding RD
  Jan has  her  seen  yesterday  that woman
  'Jan saw her yesterday that woman.'
b. Jan gaat wintersporten, althans volgend jaar, skiën.
afterthought RD
  Jan goes winter.sport  at.least  next year  skiing
  'Jan will go on winter sports, at any rate next year: skiing.'

Backgrounded phrases and afterthoughts can also be combined but then the former must precede the latter; in tandem with our finding in Subsection IV that afterthoughts may occasionally occur as separate utterances, this shows that backgrounded phrases are more closely related with the preceding clause than afterthoughts.

134
a. dat ik hem morgen ontmoet, Peter, (althans) waarschijnlijk.
  that  him  tomorrow  meet  Peter   at.least  probably
  'that I will meet him tomorrow, Peter, at least probably.'
b. * dat ik hem morgen ontmoet, (althans) waarschijnlijk, Peter.
  that  him  tomorrow  meet   at.least  probably  Peter

That backgrounded phrases are more closely related with the preceding clause is also suggested by the fact, illustrated in (135), that they must be adjacent to the minimal clause containing their correlate, while afterthoughts follow the complete sentence containing their correlate; we return to these instances in Subsection VII.

135
a. Dat hij weg was, Peter/??Peter, was vervelend.
  that  he  away  was  Peter/Peter  was annoying
  'that he was away, Peter, was annoying.'
b. Dat hij weg was, was vervelend, Peter/*Peter.
  that  he  away  was  was annoying Peter/Peter
  'That he was away, was annoying, Peter.'

      Although the examples in (133) have shown that backgrounded phrases and afterthoughts can be reiterated, this does not always give rise to a felicitous result. The examples in (136b&c) show that modal and temporal adverbs can easily be right-dislocated, but for unknown reasons the (d)-examples show that it is difficult to have them simultaneously in right-dislocated position.

136
a. dat Jan morgen waarschijnlijk vertrekt.
  that  Jan tomorrow  probably  leaves
  'that Jan will probably leave tomorrow.'
b. dat Jan morgen vertrekt waarschijnlijk.
c. dat Jan waarschijnlijk vertrekt morgen.
d. *? dat Jan vertrekt morgen waarschijnlijk.
d'. *? dat Jan vertrekt waarschijnlijk morgen.
[+]  VII.  Analyses of RD

While LD has received relatively much attention in the theoretical literature, this is much less so for RD. Because of the similarities between LD and RD constructions discussed in Subsection III, it seems preferable for the two types of dislocation to receive a similar analysis. Consequently, as RD does not involve movement of the correlate of the right-dislocated phrase, this may be a good reason for dismissing the various movement approaches to contrastive LD, discussed in Section 14.2, sub VIII. This would leave us with Ott’s (2014) hypothesis that contrastive LD constructions consist of two juxtaposed clauses, the first of which is reduced under identity with the second clause. Ott & De Vries (2015) go on to provide a similar analysis for RD; they argue that the derivation of RD differs from that of LD only in that the reduction does not apply to the first but to the second clause of the juxtaposition. They analyze the right-dislocated phrase in (137a) in the same way as the fragment answer in (137b), that is, the juxtaposition analysis of RD appeals to an independently motivated deletion operation.

137
a. Ik heb het gelezen, Hersenschimmen van J.Bernlef.
  have  it   read  Hersenschimmen by J.Bernlef
  'I have read it, Hersenschimmen by J.Bernlef.'
a'. [Ik heb het gelezen] &: [Hersenschimmen van J.Bernlefiheb ik ti gelezen].
b. Welk boek heb je gelezen?
  which book  have  you  read
  'Which book have you read?'
b'. Hersenschimmen van J.Bernlefi heb ik ti gelezen.
  Hersenschimmen by J.Bernlef   have  read

A potential problem for the analysis in (137a') is that it is not very clear what would trigger topicalization in the derivation of RD. Nevertheless, Ott & De Vries (2015) claim that examples of the type in (138) provide independent evidence for topicalization. Zwart (2011:79) has noted that the two examples in (138a&b) differ in their relative scope of the indefinite subject twee mensen'two persons' and the epistemic modal vermoedelijk: the most prominent reading of example (138a) is that the modal is in the scope of the numeral, that is, two people have seen something that is presumably a wolf. The most prominent reading of the RD construction in (138b) is that the modal takes scope over the complete proposition including the numerals, that is, it is presumably the case that two people have seen a wolf. This would follow immediately under the proposed analysis of RD because the wide scope reading of the modal is also the most prominent one for the topicalization construction in (138c).

138
a. Twee mensen hebben vermoedelijk een wolf gezien.
numeral > modal
  two persons  have  presumably  a wolf  seen
  'Two people have presumably seen a wolf.'
b. Twee mensen hebben een wolf gezien, vermoedelijk.
modal > numeral
  two persons have  a wolf  seen  presumably
  'Two people have seen a wolf, presumably.'
c. Vermoedelijk hebben twee mensen een wolf gezien.
modal > numeral
  presumably have  two persons  a wolf  seen
  'Presumably, two people saw a wolf.'

Another potential problem for assuming topicalization in the second conjunct is the fact established in Subsection V that RD is not sensitive to, e.g., interrogative and adjunct islands. Section 14.2, sub VIII, has shown, however, that this also holds for fragment clauses and LD constructions. This means that we are dealing with the more general fact discussed in Section 5.1.5, sub IB, that the ellipsis operation found in fragment clauses in one way or another cancels island violations; we refer the reader to Merchant (2001/2006), and references cited there for possible explanations of this fact.
      Ott & De Vries (2015) develop an alternative to the island-insensitivity of RD by assuming that the juxtaposition involves the minimal clause of the associate of the dislocated phrase; we illustrate this here for RD only. The minimal clause restriction requires that we analyze example (139a) as in (139b); the analysis in (139b') is not available.

139
a. Els zei dat hij weg was, Peter.
  Els said  that he away was  Peter
  'Els said that he was away, Peter.'
b. Els zei [[dat hij weg was] &: [Peter was weg]].
b'. * [[Els zei [dat hij weg was]] &: [Peterizei Els [dat tiweg was]]].

A potential drawback of relying on the minimal clause restriction is that we have to allow for coordination of embedded and main clauses; cf. the structure in (139b), in which the first conjunct is a non-main while the second conjunct is a main clause. Although this kind of unbalanced coordination is normally not possible, the minimal clause restriction is supported empirically by the fact illustrated in the primeless examples in (140) that backgrounded phrases must be adjacent to their minimal clauses; the structures in the corresponding primed examples show that this can only be derived by assuming the minimal clause restriction, otherwise, (140a) could not be derived and example (140b) would be incorrectly predicted to be acceptable.

140
a. Dat hij weg was, Peter, was vervelend.
  that  he  away  was  Peter  was annoying
  'that he was away, Peter, was annoying.'
a'. [dat hij weg was] &: [Peter was weg] was vervelend.
b. * Dat hij weg was, was vervelend, Peter.
  that  he  away  was  was annoying Peter
b'. * [[dat hij weg was] was vervelend] &: [Peteriwas [dat ti weg was] vervelend].

Example (140a) again illustrates the island-insensitivity of RD, given that wh-movement out of subject clauses is normally impossible; cf. *Peteri was [dat ti weg was] vervelend. The lack of island-sensitivity follows immediately from the minimal clause restriction. We illustrate the same island-insensitivity again in (141) for an interrogative island. For completeness’ sake observe that (141b') would also be unacceptable because it violates the complementizer-trace filter; cf.Ik weet dat Els al gegeten heeft'I know that Els has already eaten' versus *Elsi weet ik dat ti al gegeten heeft.

141
a. Ik weet niet of zij gegeten heeft, Els.
  know  not  whether  she  eaten  has  Els
  'I do not know whether she has eaten, Els.'
b. Ik weet niet [[of zij gegeten heeft] &: [Els heeft gegeten]].
b'. * [[Ik weet niet [of zij gegeten heeft]] &: [Elsiweet ik niet [of ti gegeten heeft]]].

Recall from Subsection VI that afterthoughts can be detached from the clause containing their correlate, that is, example (140b) becomes fully acceptable for at least some speakers if the right-dislocated phrase is contrastively stressed. This seems to be related to the fact that afterthoughts can be used as independent expressions; if afterthoughts are independent of the sentences preceding them, the minimal clause restriction cannot hold for them by definition.
      The minimal clause restriction seems superior to Merchant’s analysis according to which ellipsis cancels island violations because it accounts for the fact that RD is normally sensitive to certain non-clausal islands, such as the coordinate structure in (142a): this follows from the fact that topicalization in the second conjunct in (142b) violates the coordinate structure constraint (but see Ott & De Vries, 2015:fn.50, for a potential problem). The disadvantageous aspect of the minimal clause restriction is that LD does not seem to be sensitive to the coordinate structure constraint (cf. Section 14.2, sub V), which is a problem for Ott & De Vries in light of their claim that LD and RD should be analyzed in essentially the same way.

142
a. * Ik heb [hem en Marie] niet gezien, Jan.
  have   him and Marie  not seen  Jan
b. [Ik heb [hem en Marie] niet gezien] &: [Jan heb ik [ti en Marie] niet gezien].

The fact that for at least some speakers RD of a prepositional object does not require the preposition to be present is another potential problem for the minimal clause restriction: because preposition stranding is not possible in the case of topicalization (cf. *Mijn vaderi wacht ik op ti) example (143a) is predicted to be unacceptable without the preposition. But this problem is not new, as Merchant (2001:ch.3, fn.6) already found that Dutch speakers exhibit a great deal of variation with respect to preposition stranding in ellipsis constructions. The fact that we find this variation in the case of RD as well can therefore be construed as an argument in favor of unifying the analyses of fragment clauses, LD and RD. We refer the reader to Section 5.1.5, sub IB, and Section 14.2, sub VIII, for more relevant discussion.

143
a. % Ik wacht op hem, mijn vader.
  wait  for him  my father
  'I am waiting for him, my father.'
b. [Ik wacht op hem] &: [mijn vaderiwacht ik op ti].

Recent research has made great progress in describing the properties of RD by attempting to develop a unifying account of fragment clauses, LD and RD. We have also seen that there are still a number of questions to be answered, but we have to leave these to future research.

References:
  • Merchant, Jason2001The syntax of silence: sluicing, islands and the theory of ellipsisOxford/New YorkOxford University Press
  • Merchant, Jason2001The syntax of silence: sluicing, islands and the theory of ellipsisOxford/New YorkOxford University Press
  • Merchant, Jason2006SluicingEveraert, Martin & Riemsdijk, Henk van (eds.)The Blackwell companion to syntax4Malden, Ma/OxfordBlackwell Publishing271-291
  • Ott, Dennis2014An ellipsis approach to contrastive left dislocationLinguistic Inquiry45269-303
  • Riemsdijk, Henk van1997Left dislocationAnagnostopoulou, Elena, Riemsdijk, Henk van & Zwarts, Frans (eds.)Materials on left dislocationAmsterdam/PhiladephiaJohn Benjamins Publishing Company1-10
  • Riemsdijk, Henk van & Zwarts, Frans1997Left dislocation in Dutch and the status of copying rules [originally written in 1974]Anagnostopoulou, Elena, Riemsdijk, Henk van & Zwarts, Frans (eds.)Materials on left dislocationAmsterdam/PhiladephiaJohn Benjamins Publishing Company13-54
  • Veld, Joop1993Postverbal constituents in Dutch and TurkishUniversity of AmsterdamThesis
  • Vries, Mark de2002The syntax of relativizationAmsterdamUniversity of AmsterdamThesis
  • Vries, Mark de2009The right and left periphery in DutchThe Linguistic Review26291-327
  • Vries, Mark de2009The right and left periphery in DutchThe Linguistic Review26291-327
  • Vries, Mark de2009The right and left periphery in DutchThe Linguistic Review26291-327
  • Vries, Mark de2009The right and left periphery in DutchThe Linguistic Review26291-327
  • Vries, Mark de & Ott, Dennis2015Right dislocation as deletionNatural Language & Linguistic Theory33
  • Vries, Mark de & Ott, Dennis2015Right dislocation as deletionNatural Language & Linguistic Theory33
  • Vries, Mark de & Ott, Dennis2015Right dislocation as deletionNatural Language & Linguistic Theory33
  • Vries, Mark de & Ott, Dennis2015Right dislocation as deletionNatural Language & Linguistic Theory33
  • Vries, Mark de & Ott, Dennis2015Right dislocation as deletionNatural Language & Linguistic Theory33
  • Vries, Mark de & Ott, Dennis2015Right dislocation as deletionNatural Language & Linguistic Theory33
  • Vries, Mark de & Ott, Dennis2015Right dislocation as deletionNatural Language & Linguistic Theory33
  • Vries, Mark de & Ott, Dennis2015Right dislocation as deletionNatural Language & Linguistic Theory33
  • Vries, Mark de & Ott, Dennis2015Right dislocation as deletionNatural Language & Linguistic Theory33
  • Vries, Mark de & Ott, Dennis2015Right dislocation as deletionNatural Language & Linguistic Theory33
  • Vries, Mark de & Ott, Dennis2015Right dislocation as deletionNatural Language & Linguistic Theory33
  • Vries, Mark de & Ott, Dennis2015Right dislocation as deletionNatural Language & Linguistic Theory33
  • Vries, Mark de & Ott, Dennis2015Right dislocation as deletionNatural Language & Linguistic Theory33
  • Zwart, Jan-Wouter2011The syntax of DutchCambridgeCambridge University Press
  • Zwart, Jan-Wouter2011The syntax of DutchCambridgeCambridge University Press
report errorprintcite