An important difference between picture nouns and dyadic ing-nominalizations like vernietiging'destruction' is that the former do not denote states of affairs, but concrete entities; even if a noun phrase is headed by a deverbal picture noun like schilderij'painting' in (477a), it cannot refer to the action denoted by the verb but only to the result of this action, that is, the concrete object that has been created. We may therefore say that deverbal picture nouns are lexicalized in the same sense as lexical deverbal nouns like bestrating'pavement', verzameling'collection' or uitvinding'invention'. This also accounts for the fact that picture nouns can be used quite felicitously without any complements in most cases: since they do not share the denotation of the base verb, they do not inherit the verbʼs argument structure either. Thus, a picture noun like schilderij'painting' in (477a) does not require the presence of a complement, despite the fact that its verbal counterpart schilderen'to paint' normally does; in this sense, it behaves like the non-relational noun fiets'bicycle' in (477b) rather than the ing-noun vernietiging'destruction' in (477c).

Example (478) shows, however, that leaving out the theme argument does not always yield a fully acceptable result, which suggests that picture nouns may differ with regard to their degree of lexicalization: given that afbeelding'picture' is only felicitous if the theme is expressed, it can be said to have inherited the argument structure of the verb afbeelden'to depict'; since tekening'drawing' does not require the presence of a theme, it can be said to be fully lexicalized.

Note, however, that other, more pragmatic factors may come into play. For example when we are dealing with an object of art, as in (479), felicitous use of the noun tekening'drawing' does seem to require the presence of the agent or a theme. This suggests that we are actually not dealing with inherited arguments, but with contextually evoked adjuncts. This may account for the fact that, e.g., adding the name of a style period would also make (479) perfectly natural, and that the preferred addition differs with the expertise or interest of the participants in the conversation: the layman will probably want to know what the drawing represents, whereas someone knowledgeable about art will be interested in the periodization or the maker.

      Although the arguments of a picture noun need not be overtly expressed, they may be implicitly present. This is very clear with unexpressed agent arguments, which may influence the form of referentially dependent theme arguments. In order to see this, first consider the examples in (480). If the theme of a picture noun is bound by (= coreferential with) the agent of the picture noun, as in (480a), it must have the form of a reflexive pronoun like zichzelf'himself', whereas it will normally surface as a personal pronoun if it is bound by a constituent outside the noun phrase, as in (480b). Note that coreferentiality is expressed by mean of subscripts.

The standard explanation of the examples in (480) is roughly that whereas anaphors must be bound in the smallest category containing a potential antecedent, pronouns cannot be bound in that domain; see Section 5.2.1.5, sub III, for a more extensive discussion of this version of Binding Theory. Since the possessive pronoun is a potential antecedent, the relevant binding domain is the noun phrase: the anaphor must and the pronoun cannot be bound in this domain. Now consider example (481a), which shows that the binding behavior of the pronoun remains unaffected when the agent of the noun phrase is not expressed; it can take the subject of the clause as its antecedent, just as in (480b). If we assume that the structure is as given in (481b), the clause will be the smallest category containing a potential antecedent: it therefore functions as the binding domain within which the pronoun must be free (= not bound), and we wrongly predict (481a) to be impossible. This has led to the assumption that the noun phrase contains a phonetically empty pronoun PRO, as indicated in (481b'): as a result, the noun phrase functions as the binding domain of the pronoun and (481a) is correctly predicted to be possible; cf. Chomsky (1986).

Note that (481a) cannot be interpreted with the PRO agent being construed as coreferential with Jan, that is, with Jan as the maker of the picture, since in that case the pronoun would again be incorrectly bound in its binding domain by the agent of the picture noun: Jani zag [NP een PROi foto van ’mi]. The proposal also accounts for the fact that (482a) is ungrammatical under the intended idiomatic reading “making a picture”, due to the fact that the verb nemen forces a reading in which the implied PRO argument is coreferential with the subject of the clause, as in (482b).

The claim that the agent of the picture noun can be syntactically realized by the phonetically empty pronoun PRO is consistent with the acceptability of example (483a), in which the anaphor zichzelf is coreferential with the subject of the clause: the structure in (483b) shows that the anaphor is bound by the implicit PRO agent, which in turn is bound by the subject of the clause.

There is, however, a complication here concerning the interpretation of example (484a). First assume that the implicit PRO agent is obligatory: since the anaphor zichzelf must be bound by PRO, and since zichzelf is coreferential with the subject of the clause, it would follow that PRO would also be bound by the subject of the clause. This gives rise to structure in (484b) which must be interpreted in such a way that Jan saw a picture of his own making depicting himself. Although this is certainly a possible interpretation, the sentence can also be interpreted in such a way that the picture was made by someone else. We cannot assume, however, that the PRO agent refers to someone else: the structure in (484b') is ungrammatical given that the anaphor is not bound in its binding domain. The intended interpretation can therefore only be accounted for if we assume that the PRO agent is optional: the anaphor in (484b'') has no potential binder within the noun phrase, so that it can take as its antecedent the subject of the clause, which is now the smallest category containing a potential antecedent. Given that there is no PRO agent, there is no implication concerning the identity of the photographer.

      A prediction of the claim that picture nouns may have a PRO agent is that anaphors may also be used if there is no local noun phrase that could be used as an antecedent. Consider the examples in (485), in which the subject of the matrix clause is too far away from the anaphor to serve as a local antecedent.

Example (485a) is given with a percentage mark given that Vanden Wyngaerd (1994:214ff.) claims it to be ungrammatical, whereas other speakers seem to assign it more or less the same status as example (485b), which may be judged as marked but acceptable. That the status of (485a) is unclear is furthermore evident from the fact (i) that Vanden Wyngaerd does not assign it a full star but “?*”, (ii) that he claims that its status improves if we add a restrictive clause as in een foto van zichzelf die hij onlangs had genomen'a picture of himself that he made recently', and (iii) that he goes to some length to show that the construction does not display the properties that he takes to be typical for control structures, e.g. that the presumed PRO-subject may not take a split or non-c-commanding antecedent.

It should be noted that Vanden Wyngaerdʼs rejection of a PRO agent in picture noun constructions is related to the specific theory of control he endorses, and that it is not clear whether the same conclusion follows from other proposals. It seems, however, that the acceptability of (485b) can be used in support of the postulation of a PRO agent in picture nouns. Furthermore, it might be the case that the degraded status of (485a) is due to the fact that this example competes with the fully acceptable structure in (487a) with the referential personal pronoun hen'them': given the fact established earlier that the PRO agent need not be present, this example may be more economical in the sense that co-reference can be established in a more direct way than in (485a). This preference may further be enhanced by the fact illustrated in (487b&c) that the referential personal pronouns may also be used in examples such as (486) with a split a non-c-commanding antecedent.

A similar competition is, of course, not found in the case reciprocal pronouns like elkaar'each other' in example (485b), given that referential pronouns like hen'them' cannot express the relation of reciprocity.
      It may be interesting to note that the reflexive form ’mzelf can also be used in picture noun contexts. The examples in (488) show that, like zichzelf, this reflexive form must have an antecedent, but it seems to differ from zichzelf in that it cannot be bound by an antecedent within the noun phrase. Some care is needed, however, since the binding behavior of ’mzelf has not been discussed much in the literature; the most extensive discussion can be found in Koster (1987: 344 ff.), and even this discussion is no longer than two pages. Furthermore, Koster claims there that ’mzelf can be bound by the agent in ing-nominalizations like Jansi beschrijving van ’mzelfi 'Janʼs description of himself', so it might be the case that some people have more liberal judgments concerning (488b). However, since the informants we have consulted agree that the examples in (488) contrast in the way indicated, we will adopt the judgments given there as an idealization of the data, but more research is certainly needed.

The fact that ’mzelf cannot have an antecedent within the noun phrase immediately accounts for the contrast in (489): example (489a) contains an (optional) PRO agent which is disjoint in reference from the subject of the clause and ’mzelf can therefore be correctly bound by an antecedent external to the noun phrase; the noun phrase in the idiomatic example in (489b) obligatorily contains a PRO agent that is coreferential with the subject of the clause, so that ’mzelf is incorrectly bound in the noun phrase.

For completeness’ sake, note that examples (484a) and (489a) allow an alternative reading in which the reflexive form refers not to the theme but to the agent of the picture noun. This reading is excluded with pronouns, which follows immediately from the standard Binding Theory given that the noun phrase does not contain a PRO agent (this function being performed by the pronoun itself); the clause therefore constitutes the binding domain within which the pronoun must be free. In (490) coreference is indicated by italics.

      If the line of reasoning above is correct, we may conclude from the binding facts discussed above that in many cases in which the agent of the picture noun is not visible, it can nevertheless be syntactically present as a phonetically empty PRO argument. This empty PRO agent is normally not obligatory, however.

In some cases complements of deverbal story nouns cannot felicitously be left out, whereas in other cases explicit mention of the complements does not seem to be required (although they will generally be implied). The two cases are related to the interpretation of the noun. Example (491) shows that explicit mention of at least one of the complements is preferred if the noun phrase has abstract reference, that is, refers to the contents of some object.

However, if the referent is a concrete object, as in (492), neither argument needs to be expressed. This suggests that the concrete interpretation in (492) involves a higher level of lexicalization: the examples in (491) are nominalizations, and only in (492) are we dealing with true story nouns. Recall that the discussion of the examples in (468) already led to a similar conclusion.

      Note that the implied agent in the nominalizations in the primed examples of (491) is necessarily disjoint in reference with the subject: the orator/writer cannot be Jan. These examples differ in this respect from those in (493), which also have abstract reference, where the agent is necessarily coreferential with the subject.

The fact that the theme can readily be left out in (493) can perhaps be accounted for by claiming that the agent is syntactically expressed by means of a phonetically empty pronoun PRO, given that the examples in (491) have already shown that we do not have to express both arguments; expression of either the theme or the agent is sufficient. Postulating a PRO agent for the nominalizations in (493) raises the question, however, why it is not readily possible to leave the theme argument unexpressed in the primed examples of (491); it suggests that PRO need not be present in this case, which may be independently supported by our discussion of the binding data in (496) and (497) below.
      The examples in (494) show that when the story noun is not derived from a verb, the complements need not be expressed: both the agent and the theme can readily be left out, even if it is the contents of the story noun that are relevant. This means that, like the picture nouns, story nouns behave like non-relational nouns like fiets'bicycle' rather than the ing-nominalization vernietiging'destruction'.

However, as in the case of picture nouns, unexpressed agent arguments can be implicitly present, which is shown by the fact that they may influence the form of a referentially dependent theme argument. In order to see this, first consider the examples in (495), which again show that an anaphor must be bound in the smallest domain that contains a potential antecedent, whereas pronouns must be free in that domain.

The fact that the pronoun in (496a) can be used in order to refer to the subject of the clause gives rise to the idea that the agent of the story noun is realized as a phonetically empty PRO argument. The fact that this PRO argument cannot be interpreted as coreferential with the subject of the clause without invoking a violation of the binding condition on the pronoun correctly predicts that the story was written by someone else. It also follows that examples such as (496b), where the verb forces a reading according to which PRO is interpreted as coreferential with the subject of the clause, are ungrammatical. The account of the examples in (496) is therefore completely parallel to that of (481) and (482).

The examples in (497) show that the anaphor can also be construed as coreferential with the subject of the clause. These examples can be accounted for along the lines of those in (483) and (484): the fact that (497a) is acceptable is consistent with the idea that the noun phrase contains a PRO agent, which is coreferential with the subject of the clause; the fact that (497b) is acceptable despite the fact that the agent of the story noun is not coreferential with the subject in the clause shows that the PRO agent of the story noun need not be syntactically present.

      Use of the reflexive form ’mzelf is unproblematic in constructions with the verb lezen'read' in (498a): this is as predicted given that ’mzelf can be bound by the subject of the clause, while remaining free in its noun phrase. Our discussion so far predicts that (498b) should be excluded given that it is bound by the PRO agent within its noun phrase; the fact that this example seems marginally possible may therefore be unexpected. More research on the status of examples like these is needed, however, before we can draw any serious conclusions.

      As in the case of the picture nouns, the most important conclusion for our present purpose that can be drawn from the binding facts discussed above is that in many cases in which the agent of the picture noun is not visible it can nevertheless be syntactically present as a (normally optional) PRO argument.