• Dutch
  • Frisian
  • Saterfrisian
  • Afrikaans
Show all
Replacing adjunct IPIs with second conjuncts
quickinfo

An adjunct Imperativus-pro-Infinitivo (IPI) can be replaced by a second conjunct featuring an infinitive with ordinary infinitival inflection occurring in clause-final position. It is shown that adjunct IPIs are not licensed by the non-finite complementiser. When this appears to be the case, the IPI in question must be analysed as an argument IPI.

readmore

An adjunct IPI can in many cases be replaced with a second conjunct. Consider the IPI below, with Bare Infinitival (BI) inflection:

1
Hy sil moarn wol by dy komme en nim syn eigen boek mei
he shall tomorrow DcP to you come.OI and take.BI his own book along
He will probably visit you tomorrow and bring along his own book

It can be replaced with a second conjunct featuring a phrase-final infinitive with ordinary infinitival inflection, as shown below:

2
Hy sil moarn wol by dy komme en syn eigen boek meinimme
he shall tomorrow DcP to you come and his own book along.take.OI
He will probably visit you tomorrow and bring along his own book

Not all adjunct IPIs allow this replacement. Characteristically, it is possible in case the adjunct IPI is licensed by a modal verb, as in the example above.

At first sight, it seems that an adjunct IPI can also be licensed by the non-finite complementiser om for, as in the example below:

3
In momint hie er oanstriid om de doar ticht te slaan en lit har hjir sitte mei dat naasje
a moment had he inclination for the door shut to hit and let them here sit with that gang
For a moment he was inclined to slam the door shut and leave them behind with that rabble

On this analysis, the example in (3) involves an adjunct IPI licensed by a modal element, that is, the non-finite complementiser. It is also possible to analyse the sentence above as a coordination of an infinitival clause introduced by om for and an argument IPI. This suggestion is supported by the fact that an IPI can occur as a direct complement of the noun oanstriid inclination:

4
In momint hie er oanstriid en slaan de doar ticht
a moment had he inclination and slam the door shut
For a moment he was inclined to slam the door shut

Recall that this is impossible with modals, which can only occur with adjunct IPIs:

5
*Hy woe (it) en slaan de door ticht
he wanted it and hit the door shut
He wanted to slam the door shut

Under the coordination analysis, the second conjunct is introduced by the conjunct en and, which also functions as the complementiser introducing the argument IPI. Or, as a variant of the coordination analysis, there could be two instances of en and (the conjunction and the complementiser), one of which is deleted as a result of the stuttering prohibition or a similar constraint reducing a sequence of two similar functional items to one when they are pronounced.

It is not completely acceptable to have a sequence of a two clauses of which the first is an IPI and the second a to-infinitival clause (for some, this is plainly inacceptable):

6
?In momint hie er oanstriid en slaan de doar ticht en har hjir sitte te litten mei dat naasje
a moment had he inclination and slam the door shut and them here sit to let with that scum
For a moment he was inclined to slam the door shut and leave them behind with those lowlifes

The result deteriorates in case the second clause is introduced by the complementiser om for, as shown in (7):

7
?*In momint hie er oanstriid en slaan de doar ticht en om har hjir sitte te litten mei dat naasje
a moment had he inclination and slam the door shut and for them here sit to let with that scum
For a moment he was inclined to slam the door shut and leave them behind with those lowlifes

It is not clear what prevents a coordination of an IPI as a first conjunct with a to-infinitival clause as a second conjunct from being fully acceptable. Possibly, the opacity of IPIs interacts with the hierarchical structure of coordinations so as to yield the observed pattern of grammaticality judgments. It is conceivable that an IPI is less fully specified for various features than a true infinitival clause, and it is known that second conjuncts (but not first conjuncts) may be underspecified for designated features.

In any case, there is a further argument against analysing a second conjunct in the scope of the non-finite complementiser as an adjunct IPI. Adjunct IPIs display the property that negation in the main clause must have scope over the adjunct:

8
Hy woe de door net tichtslaan en lit har sitte mei dat naasje
he wanted the door not shut.slam and let them sit with that scum
He did not want to slam the door shut and leave them behind with those lowlifes

If we consider this as a diagnostic of the adjunct IPI, let us apply it to the sentence in (7):

9
In momint hie er oanstriid om de doar net ticht te slaan en lit har hjir sitte mei dat naasje
a moment had he inclination for the door not shut to hit and let them here sit with that scum
For a moment he was inclined to not slam the door shut and leave them behind with those lowlifes

In the sentence above, the negation inside the to-infinitival clause cannot have scope over the IPI. If scope of negation is a correct diagnostic for the presence of an adjunct IPI, then it seems that no adjunct IPI is involved in (7), but that the sentence truly involves a coordination of a to-infinitival clause and an argument IPI. Correspondingly, the scope of negation inside the to-infinitival clause, which is the first conjunct, cannot have the second conjunct in its scope.

The adjunct IPI can hardly ever be replaced by an infinitival clause introduced by the complementiser om for, presumably because it is not found in the scope of elements selecting such as clause:

10
*Hy sil moarn wol by dy komme om syn eigen boek mei te nimmen
he shall tomorrow DcP to you come.OI for his own book along to take
He will probably visit you tomorrow and bring along his own book

The sentence In (10) is not grammatical in the intended interpretation. It is grammatical only in case the infinitival clause is construed as a clause of purpose.

References
    printreport errorcite