The first type of deverbal pseudo-participle consists of the past participle form of an activity verb like pratento talk, kijkento watch or vergaderento meet/deliberate, which are obligatorily preceded by the particle uit (lit.: out). The examples in (172) show that these pseudo-participles are predicative in nature: we are dealing with copular constructions, as is clear from the fact that the verb zijnto be in the primeless examples can easily be replaced by the copular verb raken in the primed examples, which typically occurs with adjectival past participles. As expected, the pseudo-participles can also be used attributively as modifiers of a head noun corresponding to the subject of the copular construction; cf. e.g. de op haar meubelen uitgekeken vrouwthe woman who is bored with her furniture.

The acceptability of the examples in (172) is remarkable, given the impossibility of the adjectival use of the bare participle forms gepraat, gekeken and vergaderd. However, one might assume that these forms are grammatical but unrealized forms, allowed by the core grammar but simply not used as lexical items. We could then assume that the internal structure of the pseudo-participles is as in (173), with the meaning associated with this form given in quotation marks; cf. Booij & Audring (2007:57).

Nevertheless, there is reason to believe that we are not dealing with a run-of-the-mill morphological derivation. For instance, we would expect that the selection restriction of the input verb would be inherited by the bare adjectival participle [[X]V-participle]A and ultimately also by the full form [[uit]prt. [[X]V-participle]A ]A. This seems to be the case for the intransitive verb praten and the pseudo-participle uitgepraat in the (a)-examples in (174), which both take a PP-complement headed by the preposition over, but not for the intransitive verb kijken and the pseudo-participle uitgekeken in the (b)-examples, since the first takes a naar-PP, while the second takes an op-PP as its complement.

Note that the structure proposed in (173) is not consistent with deriving the pseudo-participle directly from the past participle form of a particle verb (if it exists at all), since the particle would then be part of (or adjoined to) the verbal stem X. This also explains why the pseudo-participles uitgepraat and uitgekeken in the examples above have meanings that are quite different from the particle verbs uitpratento resolve a conflict by talking and uitkijkento take care/look forward. This means that the “done with” reading of the pseudo-participles under discussion must be attributed to the specific morphological configuration in (173).

Note that there are also some formations which seem to be based on a nominal stem like uit-gekleuterd and uit-gebodemd based on kleutertoddler and bodembottom. Since the verbs bodemen and kleuteren are possible (but non-existent) denominal verbs, we may assume that these verbs can still be the input for the formation of pseudo-participles; concretely, Booij & Audring (2007:57) proposes a process of template conflation as depicted in (176).

For completeness’ sake, note that there are various unsolved problems with pseudo-participles like uitgepraat and uitgekeken. As stated in (173), we are dealing with adjectives and would therefore expect them to have the distribution of adjectives. This means that they should precede copular verbs in clause-final position, while the primeless examples show that they can occur on both sides. This may not be very disturbing, as this is a quite common performance error with pseudo-participles; cf. the discussion below example (112c) in Section 31.3.1, sub I.

A more disturbing problem for the assumption that pseudo-participles like uitgepraat and uitgekeken are complex adjectives is that it is possible to split them, as in (178); the judgments given here are based on Booij & Audring (2007:55-6); examples such as these are easy to find on the internet.

If such examples are indeed acceptable, they violate the lexical integrity constraint; this would raise the question as to whether pseudo-participles like uitgepraat and uitgekeken are really morphologically complex words. We leave this question for future discussion.

The discussion of constructions such as Jan komt aangewandeldJan comes strolling up in (179a) will be relatively short, since we have already discussed this construction in Section V6.3.2, sub IV. It was shown there that the participle alternates with an infinitive without a clear difference in meaning. Rather, according to Haeseryn et al. (1997: 964-5), the difference is geographical: the use of participles is preferred by speakers of the southern varieties, while speakers of the northern varieties prefer the use of infinitives (without rejecting participles).

There are in principle three possible analyses for examples such as (179a), The first is that we are dealing with verbal complexes with komen as non-main verb and the particle verb as main verb. This might be a plausible analysis for the case with an infinitive, but not for the case with a participle, as participles usually only occur with perfect and passive auxiliaries; cf. Section V6.3.2, sub IV for further discussion. This analysis is also problematic in light of the fact that the main verb *aanwandelen cannot be used as finite verb; cf. example (179b).

An alternative analysis would be to assume that aangewandeld/aanwandelen are complementive (or perhaps supplementive) phrases. This seems to be the analysis which Booij & Audring (2007: §3) proposes; it is pointed out that komen can be used as a copular verb in e.g. Het komt wel goed/in ordeIt will be fine/alright. The article also shows that the particle aan in (180a) can be replaced by the predicative PP de hoek om in (180b); it is therefore proposed that both the particle aan and the postpositional phrase de hoek om function as directional PPs (i.e. complementives).

However, this does not tell us what the syntactic function of the infinitive/participle in (179) and (180a) is. The structure proposed in Booij & Audring (2007) for the examples with the particle aan is given in (181); the (somewhat idiosyncratic) labeling of the brackets suggests that we are dealing with a construction in which aan-VINF/VParticiple is a main verb.

The proposal in (181) does not do justice to the earlier suggestion that komen functions as a copular verb, nor does it take into account that the examples in (180) are acceptable without the infinitive/participle, in which case we have the particle verb aankomento arrive and the complex verb phrase de hoek om komento come round the corner. This optionality of aan-VINF/VParticiple may be compatible with the fact noted in Duinhoven (1997:281ff) that in medieval Dutch the co-occurrence of komen and the past participle was very common and did not require the addition of a directional phrase or the verbal particle aan. This leads to a third possible analysis with the participle acting as a modifier (i.e. a manner adverb or a supplementive) of the verb phrase; this would also be consistent with the meaning assigned to the komen + participle construction in (181). It will be clear that the properties of the construction(s) discussed in this section require further research in order to evaluate the three proposals in more detail. We will not digress any further here, while referring the reader to Section V6.3.2, sub IV, for a somewhat more detailed discussion of the third analysis.

The third and final case of deverbal pseudo-participles discussed here is geblazenblown in the examples in (182), which seems to be a relatively new construction in the language. The first attestations date from the early 20th century and the construction became widespread in the second half of the 20th century. The construction is also special in that it expresses a modal meaning of necessity or desirability, which makes the construction as a whole well-suited for use in adhortative contexts. The construction was first discussed in detail in Bossuyt (2019); cf. the references given there for earlier discussions.

The constructions may have originated in the military custom of giving orders by blowing trumpets or beating drums. Typical cases are de aftocht blazento beat the retreat and alarm slaanto sound the alarm, where the verbs are ordinary transitive verbs, as shown by the fact that they can be passivized, as in (183).

However, the participle form geblazen in (182) is clearly not verbal, as can be seen from the fact that these examples do not seem to be related to the primeless examples in (184), which are in actual fact quite unlikely to crop up in common discourse, but rather to the copular construction in the primed examples with a nominal infinitive as predicate and more or less the same meaning as the examples in (182).

Bossuyt (2019) therefore tentatively suggests that we are dealing with what is known in construction grammar as a multiple source construction (i.e. based on both the original transitive construction and the copular construction). Be that as it may, the construction remains interesting from a syntactic point of view because it is in some respects quite puzzling. Let us start by adopting Bossuyt’s conclusion that we are dealing with a copular construction, just like the primed constructions in (184). A good reason for assuming this is that the verb zijn can easily be replaced by the copular verb blijvento stay/remain; note that the corresponding constructions without geblazen would also be perfectly acceptable.

The subject of the copular construction is typically the (impersonal) pronoun hetit, although Bossuyt found some cases with other subjects, e.g. demonstrative pronouns or clauses. An attested example is given in (186b), which seems to alternate with (186a); again, the construction without geblazen would also be perfectly acceptable (and perhaps even more common).

Bossuyt’s proposal that we are dealing with a copular construction seems uncontroversial so far, but problems arise when we consider the predicative part of the construction. The central question seems to be: what is the predicate of the construction, the nominal infinitive or the participle geblazen? Bossuyt (2019) does not explicitly address this question, but offers an important clue in his sketch of the diachronic development of the construction, in which he (correctly) assigns an identical meaning description to the copular constructions in the primed examples of (184) and to the geblazen-constructions in (182). The syntactic and semantic similarities between the two constructions seem to support the proposal that the nominal infinitive is the predicate, not only of the ordinary copular construction, but also of the geblazen construction. Another argument that may support this conclusion is that both constructions seem to reject the use of sentence negation.

This leaves the question of what the syntactic function of the pseudo-participle geblazen is: since we do not have a clear answer to this question, we have to leave it for further investigation. Another question we have to leave unanswered is related to the fact that the results of the corpus research reported in Bossuyt (2019: Table 1) show that the number of attested nominal infinitives in the geblazen-construction is quite high, namely 172. According to our own intuition the list includes cases that cannot easily be used in the copular construction without geblazen (with the intended modal interpretation). Some examples with infinitives with a token frequency of 5 or more in the set of attested geblazen-constructions are given in (188).