The class of monadic intransitive psych-verbs is very small; the only clear candidate we are aware of is the archaic verb versagento despond, which is mainly used in combination with the negative adverb nietnot.

The fact that monadic intransitive verbs are practically non-existent suggests that psych-verbs normally require the presence of an additional argument besides the obligatory experiencer. This additional argument can take the form of a PP-complement, i.e. the psych-verbs then appear as intransitive PO-verbs. A sample of such verbs is presented in (451); the PP-complement expresses the object (target/subject matter) of emotion.

In cases like rouwento mourn and treurento grieve in (452a), the PP-complement is optional, but if it is omitted, the object of the emotion is semantically implied. In most cases, however, the PP-complement is usually present, as shown for the PO-verbs hunkerento hanker, smachtento yearn and verlangento desire in (452b).

Psych-verbs that normally require a PP-complement may/must sometimes occur without a PP-complement if they appear with an adjunct-PP or an als-clause. This is illustrated in (453) for the verb gruwento abhor; the implied object (subject matter/target) of emotion in these examples can be recovered from the content of the adjunct, viz. al die ellendeall that misery.

The complement of the PP can sometimes be a clause, in which case the PP is realized as an anticipatory pronominal PP. This PP can be obligatory or optional, depending on the properties of the verb; cf. Section 2.3.1, sub VI.

The syntactic behavior of intransitive psychological PO-verbs seems to be on a par with that of non-psychological ones. The subjects in (452), for example, are external arguments, which is clear from the fact that these experiencer subject constructions can be passivized; cf. (455).

A possible problem is that there are a number of reasons to assume that intransitive psychological PO-verbs are not agentive. First, these verbs cannot be input for agentive er-nominalization, although one can easily counter this by saying that er-nominalization is rare with PO-verbs in general; cf. Section 2.3.2.

Another argument for claiming that subjects of intransitive psych-verbs are non-agentive is that psych-verbs often denote involuntary activities; the subjects of these verbs seem incapable of controlling the event. This can be shown by embedding these intransitive psych-verbs under the causative verb latento make; while this is perfectly acceptable with regular intransitive PO-verbs, it is usually impossible with intransitive psych-verbs. See Section 5.2.3.4 for a more detailed discussion.

Note in passing that embedding an intransitive psych-verb under causative laten is possible if the subject of the latter functions as a cause, as in (458a); this does not affect the argument above, since such examples do not imply that the experiencer is able to control the state of affairs denoted by the psych-verb. Embedding intransitive psych-verbs is also possible if laten has a permissive reading corresponding to “let” or “to not hamper”, as in (458b).

That the subject of an intransitive psychological PO-verb cannot control the event is also suggested by the fact, illustrated by the examples in (459), that psych-verbs cannot co-occur with agent-oriented adverbial phrases like opzettelijkdeliberately. Nor can they normally be in the scope of the volitional verb willento want; this is only possible if willen is contrastively stressed, as in Ik wil wel van je houden, maar ik kan het niet I do want to love you, but I cannot.

An important argument against the claim that subjects of intransitive psych-verbs are (necessarily) non-agentive is that there are a number of cases in which they seem to be able to control the event. A clear example is the verb genieten vanto enjoy: we see in (460) that this verb can be the input of er-nominalization (provided that the object of emotion is also incorporated), and that it can be embedded easily under the volitional verb willen.

The discussion above has shown that intransitive psychological PO-verbs behave more or less like regular PO-verbs. This suggests that they are simply agentive PO-verbs, so that nothing special needs to be said about them from a syntactic point of view. However, much may depend on the weight one wants to attribute to the semantic property of controllability of the event; if the feature [±control] is not a defining property of agentivity, but is simply superimposed on subjects of different types, as argued in Section 1.2.3, sub IIIB, we can dismiss the data in (457) to (459) as irrelevant for the issue at hand.

Direct objects of transitive subject-experiencer psych-verbs always function as the target of emotion, i.e. the entity toward which the positive or negative emotion of the subject experiencer is directed. Two examples with a negative and a positive emotion, respectively, are given in (461).

A sample of the transitive subject-experiencer verbs is given in (462).

The set in (462) should probably also include fixed collocations like hoogachtento have esteem for. Although hoogachten is special in that it incorporates a predicative adjective (i.e. hoog) and probably originated as a vinden-construction comparable to Jan vindt Peter aardigJan considers Peter nice, it seems to be on the verge of becoming a complex (separable) verb. That hoogachten may be halfway through the process of becoming a complex verb is clear from the fact that its antonym minachtento despise has already been fully reanalyzed as a verb: the fact that min is pied-piped under verb-second shows that it has become part of the verb. Another example involving a predicative adjective is the (separable) collocation liefhebbento love.

As in the case of intransitive PO-verbs, there seems to be hardly any reason to distinguish transitive psych-verbs from the non-psychological ones syntactically. For example, passivization of psych-verbs gives rise to a fully grammatical result.

The transitive subject-experiencer verb mogento like in (465a) may be special in that it does not seem to allow passivization to us, although this may vary from speaker to speaker, as a Google search (March 16, 2024) on the string [gemogen worden/wordt] yielded many relevant cases.

It is also easy to find transitive psych-verbs that can be the input for er-nominalization. The examples in (466a&b) are acceptable if the target of emotion is incorporated or expressed by a van-PP. However, the examples in (466c) show that there are also psych-verbs that do not allow er-nominalization (the result improves somewhat if an adverb like echttruly precedes the noun phrase: ??Dat is echt een sportwaardeerderthat is truly someone who appreciates sports).

The acceptability of the er-nominalizations in (466a&b) suggests that the external argument is a true agent. However, like intransitive psych-verbs, the transitive psych-verbs in (462) cannot normally be embedded under the causative verb latento make with an external causer argument, which suggest that these verbs also denote involuntary/uncontrollable activities; we return to this issue in Section 5.2.3.4.

That the subject of a transitive psych-verb cannot control the event is further suggested by the fact, illustrated by the examples in (468), that psych-verbs cannot easily be in the scope of the volitional verb willento want, and cannot co-occur with agent-oriented adverbial phrases like opzettelijkdeliberately.

We conclude that transitive subject-experiencer psych-verbs are just regular transitive verbs; however, the experiencer subject is special in that it is often incapable of controlling the activity denoted by the verb (which is, of course, consistent with our knowledge of the world).

There are only a small number of unaccusative subject-experiencer verbs. Some examples are the simplex verbs bedarento calm down, kalmerento calm down and schrikkento be frightened in the primeless, and the particle verbs opmonterento cheer up, opfleurento cheer up and opkikkerento cheer up in the primed examples of (469).

That the verbs in (469) are unaccusative is clear from the following facts: they take the auxiliary zijnto be in the perfect tense; the past/passive participle of the verbs can be used attributively to modify a noun corresponding to the experiencer subject; impersonal passivization of these verbs is excluded. This is illustrated for the verb schrikken by (470); the facts in (470a&b) are sufficient to assume unaccusative status.

Note in passing that er-nominalization is never possible in the intended reading, in which the noun denotes the experiencer; cf. the unacceptability of #bedaarder, *kalmeerder, *schrikker, *opmonterder, *opfleurder, and *opkikkerder.

The examples in (471) show that clauses containing an unaccusative psych-verb can contain an adverbial door-PP expressing the cause of the emotion. However, the referent of the cause must be inanimate; if it refers to an animate entity, the sentence is degraded. Example (471c') shows that the cause can sometimes also be expressed by a van-PP; the complement of this PP can be either animate or inanimate. We conclude from this that the door-PP invariably refers to a cause, whereas the van-PP may also refer to a causer.

With particle verbs, a van-PP can also be used to refer to a cause of emotion, but, in such cases, the complement of the PP is typically inanimate; the use of the proper names requires special circumstances in which the person can be seen as a cause (e.g. when we are dealing with adorable babies).

Like the subjects of other subject-experiencer verbs, the subjects of unaccusative psych-verbs usually do not control the event denoted by the verb. This is not surprising, since this is also the case with other unaccusative verbs, but for the sake of completeness we will show here that a verb like schrikkento be frightened cannot normally be embedded under volitional willento want, nor can it license agent-oriented adverbial phrases such as opzettelijkintentionally. Note that schrikken cannot occur in positive imperatives either; in this respect it differs from bedaren and kalmeren, which do allow imperative forms: Bedaar/Kalmeer!Calm down!. Note that the negative imperative Schrik niet!Do not be alarmed!, which is not normally used as an order but as a warning or reassurance, is easily possible.

Finally, note that the verb bedarento calm down is special in that it can occur as the object of a predicative tot-phrase; this is illustrated in the examples in (474).

As far as we know, there are no clear cases of undative psych-verbs (although it might be interesting to investigate whether some of the putative intransitive psych-verbs discussed in Subsection I could be candidates for such an analysis). It should be noted, however, that the verbs hebbento have, krijgento get and houdento keep can be part of fixed collocations with certain nouns denoting a psychological state; some examples are given in (475). Since we have argued in Section 2.1.4 that hebben, krijgen and houden are undative, we are probably dealing with constructions in which the experiencer is an internal argument that is promoted to subject.

The objects of emotion in these constructions can be analyzed as part of the noun phrase, as is clear from the fact illustrated in (476) that they can (optionally) be pied-piped under topicalization.

Undative psych-constructions of the type in (475) are also sometimes formed with taboo words like de pestthe plague, and they can also be completely idiomatic; this is shown in example (477).

Finally, note that example (475c) is also possible with the verb voelento feel: Jan voelt berouw over zijn laffe daadJan feels remorse over his cowardly deed. It is tempting to take this as evidence for the assumption that this verb is also undative, especially since it also behaves like an undative verb in that it does not allow passivization and er-nominalization.

The previous subsections have discussed three types of subject-experiencer psych-verbs and have shown that, from a syntactic point of view, these verbs can simply be considered as regular intransitive, transitive and unaccusative verbs. However, these psych-verbs are of a special semantic subtype in that they usually seem to denote involuntary activities.

That there are intransitive and transitive subject-experiencer psych-verbs raises the question as to whether we should assume two types of external arguments with, respectively, the thematic role of agent and the thematic role of experiencer. The choice seems to depend on whether the semantic property of controllability of the event is relevant for distinguishing between thematic roles. Since Section 1.2.3, sub IIIB, argues that the answer to this question is negative, we provisionally conclude that there is no need to postulate external arguments with the thematic role of experiencer.

The fact that there are unaccusative subject-experiencer psych-verbs shows that the experiencer need not be an external argument of the verb, but can also be an internal argument. This conclusion seems to be confirmed by the fact that there also seem to be psych-constructions based on the undative verbs hebbento have, krijgento get, and houdento keep in combination with a psychological noun such as een hekel hebben aanto dislike. From this perspective, it is not surprising that experiencers do not have to appear as subjects, but can also be realized as (dative or accusative) objects. We will discuss such cases in Section 2.5.1.3.