• Dutch
  • Frisian
  • Saterfrisian
  • Afrikaans
Show all
6.2.1.Perfect auxiliaries
quickinfo

Since many aspects of the semantic function of the perfect-tense constructions are dealt with in Section 1.5, we can be relatively brief here, subsection I briefly indicates the function of the perfect auxiliaries, while Subsection II discusses the principal factors that determine whether hebben or zijn is used, subsections III and IV continue with a discussion of the form of the verb immediately governed (selected by) the auxiliary in verb clusters consisting of, respectively, two and three verbs, as well as the order of the verbs in such verb clusters, subsection V argues that perfect-tense constructions typically exhibit monoclausal behavior and that they demonstrate this by showing that the main verb and its argument can be separated by the perfect auxiliary, subsection VI summarizes the discussion by formulating a number of descriptive generalizations capturing the facts discussed in Subsections I through V, subsection VII concludes the discussion of perfect auxiliaries by showing that the perfect auxiliaries hebben and zijn can sometimes be mixed up with the (semi-)copulas hebben and zijn, and discusses how they can be kept apart.

readmore
[+]  I.  The function of the auxiliaries hebben and zijn

The perfect auxiliaries hebben and zijn are used to form perfect tenses: whereas the simple present in the primeless examples in (13) presents the eventualities of Marie walking on the moor and Jan reading a book as ongoing events in the present-tense interval, the present perfect in the primed examples presents the same eventualities as discrete units that are bounded within the present-tense interval. There are reasons, however, not to hold the auxiliary but the past participle responsible for the expression of this perfective meaning aspect; we refer the reader to Section 6.2.4 for the motivation of this claim, and to Section 1.5.1 for a more detailed discussion of the semantic interpretation of the present/past perfect tenses.

13
a. Marie wandelt op de hei.
  Marie walks  on the moor
  'Marie is walking on the moor.'
a'. Marie heeft op de hei gewandeld.
  Marie has  on the moor  walked
  'Marie has walked on the moor.'
b. Jan leest een boek.
  Jan reads  a book
  'Jan is reading a book.'
b'. Jan heeft een boek gelezen.
  Jan has  a book  read
  'Jan has read a book.'
[+]  II.  The choice between hebben and zijn

The choice between the perfect auxiliaries hebben and zijn is related to the status of the verb that they select: zijn'to be' is used with telic unaccusative verbs, whereas hebben is used in all other cases; see Table 3 in Section 2.1.2, sub V, and Table 5 in Section 2.1.3, sub IIH. In order not to have to repeat the detailed discussion of unaccusativity and its relevance for auxiliary selection in Section 2.1, we will illustrate the role of unaccusativity here by means of the monadic verbs lachen'to laugh' and vallen'to fall' only. The verb lachen is not unaccusative as is clear from the fact that it allows impersonal passivization; it therefore takes hebben as its perfect auxiliary. The verb vallen is an unaccusative verb as is clear from the fact that the participle can be used as an attributive modifier of a noun that corresponds to the subject of the corresponding active sentence; it therefore takes zijn as its perfect auxiliary.

14
a. Jan heeft gelachen.
  Jan has  laughed
  'Jan has laughed.'
a'. Er wordt gelachen.
  there  is  laughed
a''. * de gelachen man
  the laughed man
b. Jan is gevallen.
  Jan is fallen
  'Jan has fallen.'
b'. * Er wordt gevallen.
  there  is  fallen
b''. de gevallen man
  the  fallen  man

The role of telicity can be demonstrated by means of the examples in (15). The unaccusative verbs drijven'to float' and bloeden'to bleed' are atelic and therefore take hebben as their perfect auxiliary. However, when we add the complementives weg'away' and dood'dead', the constructions as a whole become telic and, as a result, the verbs take the perfect auxiliary zijn. For a more detailed and systematic discussion of unaccusativity and telicity, as well as their relevance for auxiliary selection, we refer the reader to Section 2.1.

15
a. De bal drijft (weg).
  the ball  floats  away
  'The ball is floating (away).'
a'. Jan bloedt (dood).
  Jan bleeds   dead
  'Jan is bleeding (to death).'
b. De bal heeft/*is gedreven.
  the ball  has/is  floated
b'. Jan heeft/*is gebloed.
  Jan has/is  bled
c. De bal is/*heeft weg gedreven.
  the ball  is/has  away  floated
c'. Jan is/*heeft dood gebloed.
  Jan is/has  dead  bled

Another factor that needs mentioning is that for some (especially Flemish) speakers the choice between hebben and zijn is not necessarily determined by the verb that it immediately governs but may also be determined by some more deeply embedded verb. In (16), the verb moeten selects the auxiliary hebben but nevertheless some speakers allow or even prefer zijn because the more deeply embedded unaccusative verbs komen'to come' and gaan'to go' select zijn; example (16a) is taken from Haeseryn et al. (1997:81) and example (16b) is provided by one of our own Flemish informants.

16
a. Ze hebben/%zijn niet kunnen komen.
  they  have/are  not  can  come
  'They havenʼt been able to come.'
b. Marie heeft/%is vanmorgen moeten gaan zwemmen.
  Marie has/is  this.morning  must  go  swim
  'Marie has had to go swimming this morning.'
[+]  III.  Form and placement of the governed verb in clusters of two verbs

The projection of the main verb is embedded under the finite auxiliary: the representation is [... Aux [... V[-finite] ...]]. We may therefore assume that the morphological form of the main verb is governed by the auxiliary (in the same way that a main verb may govern the case form of its nominal arguments in languages that have morphological case). The examples in the previous subsections have already shown that the non-finite main verb governed by the perfect auxiliary surfaces as a past participle if the verb cluster consists of no more that two verbs, that is, if the clause contains no other verbs than the perfect auxiliary and the main verb; if the verb surfaces as an infinitive, the resulting structure is unacceptable. This is illustrated in the examples in (17).

17
a. Jan heeft dat boek gelezen/*lezen.
  Jan has  that book  readpart/readinf
  'Jan has read that book.'
b. Marie is naar Utrecht gewandeld/*wandelen.
  Marie is to Utrecht  walkedpart/walkinf
  'Marie has walked to Utrecht.'

A phenomenon that has attracted a great deal of attention in the syntactic descriptions of Dutch is that the auxiliary and the main verb do not have a fixed place with respect to each other in clause-final position: the examples in (18) show that past participles may either precede or follow the finite auxiliary.

18
a. dat Jan dat boek <gelezen> heeft <%gelezen>.
  that  Jan that book    read  has
  'that Jan has read that book.'
b. dat Marie naar Utrecht <gewandeld> is <%gewandeld>.
  that  Marie to Utrecht    walked  is
  'that Marie has walked to Utrecht.'

When we consider the regional spread of the two word orders, it seems that the order aux–part is only found in a restricted part of the Dutch-speaking area, which happens to include the prestigious varieties of the standard language spoken in the west/middle region of this area; the maps in Pauwels (1953), Gerritsen (1991) and Barbiers et al. (2005) all show that this order is rare in the varieties of Dutch spoken in Flanders and the northern part of the Netherlands. For this reason we have marked this order with a percentage sign.
      Speakers who allow the order aux–part normally also allow the order part-aux. There is reason for assuming that the latter order (part-aux) is in fact the unmarked one for such speakers given that Barbiers et al. (2005) found that they rarely invert this order in reproduction tasks.
      It now seems generally accepted that the use of the aux–part order is characteristic for written Dutch and the more formal registers of spoken Dutch (despite that it frequently occurs in the more casual speech of many speakers); see Haeseryn (1990:ch.2) for a good review of the relevant literature on this issue. A corpus analysis by De Sutter (2005/2007) suggests that even in written Dutch the aux–part order is secondary since this order is mainly used in relatively simple sentences; there is a negative correlation between the complexity of utterances and the frequency of the aux–part order.
      The finding that the aux–part order is marked (perhaps even artificial) for most speakers of Dutch seems to be in line with the fact that this order was introduced in the 16th century and diligently promoted by normative grammarians, and that it still seems to be prescribed for journals and newspapers; see Coussé (2008:ch.10) and Van der Horst (2008:1984ff.). The attempt to promote this order has in fact been very successful since for most present-day speakers who allow this order, it simply functions as an alternative realization of the more widely accepted part-aux order.
      The factors favoring the selection of one order over the other are complex and have only been investigated for written language. The studies reviewed in Haeseryn (1990:46ff.), for example, provide evidence that the presence of a verbal particle or some other accent-bearing material preceding the verb cluster favors the use of the aux-part order, whereas the presence of material following the verb cluster disfavors it. De Sutter's (2005/2007) tested some of the more specific claims made in the literature on the basis of a more recent newspaper corpus, and found that:

19
The aux-part order is favored by:
a. the presence of a verbal particle or some other element that forms a fixed collocation with the participle;
b. a more extensive middle field (> 2 words);
c. a high information value of the word preceding the clause-final verb cluster;
d. a non-complement (adjunct) in preverbal positions.

De Sutter further found that participles with a high frequency occur more often in the aux-part order than participles that are less common, and that there is a syntactic persistency effect: the word order of a verb cluster used earlier in the discourse is likely to be repeated. Contrary to the earlier studies, De Sutter did not find a significant effect of accent; he attributes this to the fact that his corpus consists of written sources, but the same thing holds for most of the other studies, One might therefore speculate that the difference is related to the fact that the earlier studies were based on literary texts (dating from the first half of the 20th century), whereas De Sutter's data is taken from a (Flemish) newspaper.

[+]  IV.  Form and placement of the governed verb in clusters of three or more verbs

In finite monoclausal structures containing three verbs, the perfect auxiliary may be the finite, that is, structurally highest verb or a non-finite, that is, a more deeply embedded verb. Examples illustrating this are given in (20), in which the subscripts indicate the type of verb we are dealing with. We will discuss the two constructions in separate subsections; we start in Subsection A with examples such as (20a) in which the perfect auxiliary is itself governed by a finite verb and Subsection B continues with examples such as (20b) in which the perfect auxiliary is finite.

20
a. Jan moet dat boek hebben gelezen.
... Modal [... Aux [... V ...]]
  Jan mustmodal  that book  haveaux  readmain
  'Jan had to have read that book.'
b. Jan heeft dat boek moeten lezen.
... Aux [... Modal [... V ...]]
  Jan hasaux  that book  mustmodal  readmain
  'Jan has had to read that book.'
[+]  A.  Verb clusters of the form Vfinite - Auxnon-finite - Vmain

This subsection discusses finite monoclausal structures with three verbs in which the perfect auxiliary surfaces as a non-finite verb. At first sight, such structures do not seem very special: (i) the auxiliary governs the main verb, which surfaces as a past participle, and (ii) the past participle may either precede or follow the auxiliary (just as in embedded clauses with two verbs discussed in Subsection III). The first property, which implies that the main verb cannot be realized as an infinitive, is illustrated in the examples in (21).

21
a. Jan moet dat boek hebben gelezen/*lezen.
  Jan must  that book  have  readpart/readinf
  'Jan must have read that book.'
b. Marie moet vroeg zijn vertrokken/*vertrekken.
  Marie must  early be  leftpart/leaveinf
  'Marie must have left early.'

With respect to the order of the auxiliary and the past participles, the same proviso must be made as in Subsection III, namely that the aux -part order is only found in a restricted part of the Dutch-speaking area, which happens to include the prestigious varieties of the standard language spoken in the west/middle region of this area. More generally, it seems that the part-aux order is the more common one in speech (although we should mention that, to our knowledge, the variation in word order of the clause-final verbs in main clauses with three verbs has not been systematically investigated). The subscripts in (22) are added for convenience, to indicate whether the verb in question is finite, an infinitive or a past participle.

22
a. Jan moet dat boek <gelezen> hebben <%gelezen>.
  Jan mustfinite  that book    readpart  haveinf
  'Jan must have read that book.'
b. Marie moet vroeg <vertrokken> zijn <%vertrokken>.
  Marie mustfinite  early    leftpart  beinf
  'Marie must have left early.'

The examples in (23) show, however, that the placement options of the past participle in embedded clauses are somewhat surprising. Given that the participle is governed by the auxiliary we would expect these verbs to be adjacent, but as a matter of fact they can be separated by the finite modal verb.

23
a. dat Jan dat boek <gelezen> moet <gelezen> hebben <%gelezen>.
  that  Jan that book    readpart  mustfinite   haveinf
  'that Jan must have read that book.'
b. dat Marie vroeg <vertrokken> moet <vertrokken> zijn <%vertrokken>.
  that Marie early     leftpart  mustfinite  beinf
  'that Marie must have left early.'

For many speakers, the three word orders can be seen as more or less free alternates, with the Vfinaux–part order moet hebben gelezenbeing the more marked one. That this order is the more marked one seems to be confirmed by the regional distribution of these orders given in Table (24) for the sequence moet hebben gemaakt'must have made'; whereas speakers regularly indicate that they only accept one of the orders in (24b-d), there is just one speaker who indicates that (s)he only accepts (24a). Speakers who report that they only allow (24b) are mainly found in Flanders, whereas speakers who report that they only allow (24c) are spread over the Netherlands. The low frequency of order (24d) is due to the fact that it is only found in the northern parts of the Netherlands, which, in turn, may be related to the fact that this is the order normally found in Frisian (as well as Standard German). The data in (24) are taken from Barbiers et al. (2008).

24
Order of verbs in the sequence Vfinite–Aux–Part
order of verbs Total # Total # as only order
a. Vfinite–Aux–Part (moet hebben gemaakt) 91 1
b. Vfinite–Part–Aux (moet gemaakt hebben) 163 48 (Flanders)
c. Part–Vfinite–Aux (gemaakt moet hebben) 186 28 (Netherlands)
d. Part–Vfinite–Aux (gemaakt hebben moet) 48 30 (Northern Netherlands)

The literature reviewed in Haeseryn (1990:54ff.) further suggests that the order Vfin–part–aux order is especially popular in the varieties of Dutch spoken in Flanders, whereas speakers from the Netherlands generally prefer the order part–Vfin–aux; see also Stroop (2009) for the same finding on the basis of the Corpus Gesproken Nederlands. The order Vfin–aux–part is again characteristic for (but not restricted to) written and formal Dutch.
      Clusters of more than three verbs are possible but less frequent in colloquial speech. If the auxiliary immediately governs the (most deeply embedded) main verb, the principles underlying the form of the main verb and the order of the verbs are the same as in the case of three verbs: the main verb surfaces as a past participle, which may occur as the last verb of the verb cluster but may also occur more to the left. This is illustrated in (25) for the cluster zou kunnen hebben gezien'might have seen'.

25
a. dat Jan die film zou kunnen hebben gezien.
  that  Jan that movie   wouldmodal  maymodal  haveaux  seenmain
  'that Jan might have seen that movie.'
b. dat Jan die film zou kunnen gezien hebben.
c. dat Jan die film zou gezien kunnen hebben.
d. dat Jan die film gezien zou kunnen hebben.

To our knowledge, not much information is available about the spread of the orders in (25). The literature reviewed in Haeseryn (1990:70ff.) suggests that the orders in (25a&d) are the ones commonly found in the northern varieties of Standard Dutch, and that the order (25c) is more favored than (25b). In the varieties of Standard Dutch spoken in Belgium, on the other hand, the order in (25b) seems to be a common one.
      The discussion above has shown for the northern varieties of Standard Dutch that in perfect-tense constructions of the kind under discussion the past participle of the main verb may follow or precede the complete verb cluster or be placed in between any two verbs in the verb cluster. This is illustrated in (26), in which Vn stands for zero or more verbs in the verb cluster besides the auxiliary and the main verb; the angled brackets indicate the alternative placements of the participle.

26
Order in verb sequences of the form Vn - Auxperfect - Vmain
a. dat ..... <Part> auxfinite <Part>
b. dat ..... <Part> Vfinite <Part> auxinf <Part>
c. dat ..... <Part> Vfinite <Part> Vinf <Part> auxinf <Part>
d. dat ..... <Part> Vfinite <Part> Vinf <Part> Vinf <Part> auxinf <Part>
e. etc.

Although Barbiers et al. (2005) show that other orders can be found in certain dialects of Dutch, the orders in (26) exhaust the possibilities for the vast majority of Dutch speakers. Most speakers will in fact use only a subset of the word order possibilities in (26). Recall that clusters of more than three verbs are rare in everyday speech, and even in formal speech and complex written language the number of verbs will normally be limited to a maximum 4 of 5.

[+]  B.  Verb clusters of the form auxfinite - Vnon-finite - Vmain

This subsection discusses finite monoclausal structures with three verbs in which the perfect auxiliary surfaces as the finite verb. Such structures arise not only if the auxiliary governs a non-main verb like the (semi-)aspectual verbs gaan and zitten in (27a&b), but also if it governs a main verb that selects a transparent infinitival clause, like the deontic modal verb moeten'be obliged' in (27c) or the perception verb zien'to see' in (27d).

27
a. Marie is vanmorgen gaan zwemmen.
  Marie isaux  this.morning  goaspectual  swimmain
  'Marie went for a swim this morning.'
b. Jan heeft een boek zitten lezen.
  Jan hasaux  a book  sitsemi-aspectual  readmain
  'Jan has been reading a book.'
c. Jan heeft dit boek moeten lezen.
  Jan hasaux  this book  mustmodal  readmain
  'Jan has had to read this book.'
d. Jan heeft Peter dat boek zien lezen.
  Jan hasaux  Peter that book  seeperception  readmain
  'Jan has seen Peter read that book.'

The most conspicuous phenomenon in examples such as (27) is the so-called infinitivus-pro-participio (IPP) effect, that is, that the non-finite verb governed by the auxiliary does not surface as a past participle but as an infinitive: the examples in (28) illustrate this by showing that substituting a past participle for the relevant infinitival verbs in (27) leads to ungrammaticality.

28
a. Marie is vanmorgen gaan/*gegaan zwemmen.
  Marie is this.morning  goinf/gonepart  swim
b. Jan heeft een boek zitten/*gezeten lezen.
  Jan has  a book  sitinf/satpart  read
c. Jan heeft dit boek moeten/*gemoeten lezen.
  Jan hasaux  this book  mustinf/mustpart  read
d. Jan heeft Peter dat boek zien/*gezien lezen.
  Jan hasaux  Peter that book  seeinf/seenprt  readmain

Another property is that the word order of the verb cluster is very strict in most northern varieties of Dutch. In main clauses such as (27) the verb selected by the perfect auxiliary must precede the main verb: the examples in (29) show that reversing the order of the two clause-final verbs leads to degraded results.

29
a. * Marie is vanmorgen zwemmen gaan.
  Marie isaux  this.morning  swimmain  goaspectual
b. * Jan heeft een boek lezen zitten.
  Jan hasaux  a book  readmain  sitsemi-aspectual
c. * Jan heeft dit boek lezen moeten.
  Jan hasaux  this book  readmain  mustmodal
d. * Jan heeft Peter dat boek lezen zien.
  Jan hasaux  Peter that book  readmain  seeperception

In embedded clauses the word order is also very strict. This holds not only for the two non-finite verbs, which again exhibit the order in (27), but also for the finite auxiliary and the two infinitival verbs; the auxiliary must precede them.

30
a. dat Marie vanmorgen is gaan zwemmen.
  that  Marie this.morning  isaux  goaspectual  swimmain
  'that Marie went for a swim this morning.'
b. dat Jan een boek heeft zitten lezen.
  that  Jan a book  hasaux  sitsemi-aspectual  readmain
  'that Jan has been reading a book.'
c. dat Jan dit boek heeft moeten lezen.
  that  Jan this book  hasaux  mustmodal  readmain
  'that Jan has had to read this book.'
d. dat Jan Peter dat boek heeft zien lezen.
  that  Jan Peter that book  hasaux  seeperception  readmain
  'that Jan has seen Peter read that book.'

Any other order than in (30) gives rise to a severely degraded result. This implies that the perfect-tense constructions under discussion here differ markedly from the perfect-tense constructions discussed in Subsection A in that the auxiliary cannot be preceded by the verb it immediately dominates. We illustrate this in (31) for the modal construction in (30c): the auxiliary cannot be preceded by the modal regardless of the position of the more deeply embedded main verb.

31
a. dat Jan dit boek heeft moeten lezen.
  that  Jan this book  hasaux  mustmodal  readmain
b. * dat Jan dit boek moeten heeft lezen.
  that  Jan this book  mustmodal  hasaux readmain
c. * dat Jan dit boek moeten lezen heeft.
  that  Jan this book  mustmodal  readmain  hasaux
d. * dat Jan dit boek moeten lezen heeft.
  that  Jan this book  mustmodal  readmain  hasaux

In short, it seems that in the northern varieties of Standard Dutch the verb clusters can only be realized in the order in (32a), all the other logically possible orders being severely degraded. This is remarkable given that Barbiers et al (2005) show that the orders marked with a percentage sign are relatively common in specific regional varieties of Dutch: the order in (32e) can be found in Flanders, and the order in (32f) in the northern part of the Netherlands, especially Frisian. The order in (32b) is relatively rare but is reported by various speakers around the IJsselmeer; it is also the order normally found in Standard German. The orders marked with a star are rare and do certainly not occur as the dominant orders.

32
Order in verb sequences of the form: auxfinite - Vnon-finite - Vmain
a. auxfinite - Vnon-finite - Vmain (heeft moeten lezen)
b. % auxfinite - Vmain - Vnon-finite (heeft lezen moeten)
c. * Vmain - auxfinite - Vnon-finite (lezen heeft moeten)
d. * Vnon-finite - auxfinite - Vmain (moeten heeft lezen)
e. % Vnon-finite - Vmain - auxfinite (moeten lezen heeft)
f. % Vmain - Vnon-finite - auxfinite (lezen moeten heeft)

      It will not come as a surprise after the discussion above that in longer verb clusters with IPP the order of the verbs is also very strict. We illustrate this in (33) and (35) for verb clusters consisting of four verbs. The examples in (33) differ from those given in (30) in that we have added an epistemic modal verb, which surfaces as the