This section deals with the distribution of the morphologically simplex reflexive personal pronouns. Table 11 in Section 18.2.1.5 has shown that there is only one specialized form, the third-person pronoun zich. All other forms in the paradigm of simplex reflexives are identical to the weak object forms of the referential personal pronouns; this somewhat strange fact can probably be explained by appealing to the hierarchy (147) discussed in Section 22.3, sub IB. For the sake of clarity, the binding behavior of the simplex reflexives will be demonstrated by means of the specialized form zich. We will see that the binding behavior of zich cannot be easily explained by classical binding theory, which is repeated here in a slightly updated formulation as (164).

According to Everaert (1981/1986), the form zich is typically found in inherently reflexive constructions such as (165); this holds for approximately 95% of the occurrences of zich in his corpus of (written) texts. Since the simplex reflexive, by definition, cannot be replaced by a referential expression such as the proper noun Marie, it is often assumed that zich is not an argument of the verb in such constructions.

We will see, however, that there are also cases in which zich occurs in an unambiguous argument position. But first, note that zich differs from the complex reflexive zichzelfhimself in that it cannot occur in simple sentences as a nominal/prepositional object; the same normally holds for the nominal part of an adverbial PP.

According to Everaert (1986: §3) and Rooryck & Vanden Wyngaerd (2011: §5.4), simplex reflexives acting as arguments typically have one of the three syntactic functions in (167), which we will discuss in more detail in 22.4.1.

The three cases are illustrated by the examples in (168). Note that Rooryck & Vanden Wyngaerd (2011: §5.4) do not commit to the small-clause analysis of (168a), but treat the PP as an adverbial phrase; this contradicts the earlier observation in (166c) that zich cannot normally be used in adverbial PPs in simple sentences (but see Section 22.4.2, sub IIID, for other cases that might be at odds with this generalization).

As stated above, the main goal of this section is to examine the distribution of the simplex reflexive zich in more detail. We have already pointed out that its binding behavior cannot be easily explained by classical binding theory (164). That binding condition A in (164a) is not sufficient to account for the binding behavior of zich in argument position will be clear from the fact that zich cannot be bound by a co-argument, which is the prototypical case for anaphors (i.e. complex reflexives and reciprocals): zich behaves in this respect more like referential pronouns such as haarher. This is illustrated in (169) for cases with a subject antecedent.

The examples in (170) show that zich also behaves differently from anaphors in examples with a prepositional small clause: in (170a) the reciprocal must be bound by the logical subject of the PP-predicate, de hondenthe dogs, which is impossible for zich; (170b) shows that zich must instead be bound by the subject of the matrix clause. Again, zich seems to behave like a referential pronoun, since the weak pronoun ’rher can also be bound by the subject of the clause in this construction.

Before continuing, note that zichzelf in (170b) is more or less acceptable with emphatic accent on zelf; in such cases we are not dealing with the complex reflexive zichzelf. but with the simplex reflexive zich followed by the modifier zelfhimself/herself/themselves, which we also find with other noun phrases; cf. Jan heeft met Marie zelf gesproken Jan has spoken to Marie herself. We will ignore such cases in the following and refer the reader to Section 18.2.3.2, sub V, for a more detailed discussion of the emphatic modifier zelf.

The examples in (169) and (170) make it clear that the binding behavior of zich is not in keeping with binding condition A, but resembles that of the referential personal pronouns. However, it cannot be explained by binding condition B in (164b) either, since its distribution is much more restricted than that of the referential personal pronouns. Example (171a) shows that zich behaves like an anaphor in that it must be bound within its minimal clause. Example (171b) shows that this holds not only for finite clauses, but also for (om +) te-infinitival clauses: that the simplex reflexive zich cannot be interpreted as coreferential with Marie is of course to be expected, since this would lead to it being bound by a co-argument (viz. the implicit subject PRO), but it cannot be bound by the subject of the matrix clause, Jan, either. The examples in (171) thus show that the simplex reflexive zich behaves in this respect like the complex reflexive zichzelf.

Note that the fact that zich cannot be bound by an antecedent external to its minimal clause shows that it also differs from Icelandic sig, which clearly takes an antecedent external to its minimal clause in Jóni sagði [að ég hefði svikið sigi] Jón said that I had betrayed him; cf. Thráinsson (1991/2007: §9.1.2-3) and De Vries (1999) for further discussion.

The above discussion suggests that zich has mixed binding behavior: it sometimes patterns with anaphors and sometimes with referential pronouns. Section 22.4.1 will take a closer look at the similarities and differences in the binding behavior of the three types of referential elements. We will see that it is possible to partially explain the binding behavior of zich by reformulating the binding conditions in (164), but this leaves several problems unresolved. Section 22.4.2 will therefore explore an entirely different approach to simplex reflexives, according to which they are not anaphoric elements but unaccusativity markers, as first proposed in Everaert (1986) and revived in Rooryck & Vanden Wyngaerd (2011).