The Hebrew Object Marker as a Type-shifting Operator Gabi Danon, Tel-Aviv University danon@post.tau.ac.il Abstract The Hebrew object marker, et , is often taken to be a marker of accusative case. But in addition to its syntactic properties, et also seems to have semantic content. When the presence of et is optional, it can be seen that et affects the interpretation of the object in various ways that might include specificity, definiteness or distributivity. I propose that all these semantic effects can be given a uniform account by assuming that et is the overt realization of a lifting operator from entities to generalized quantifiers. 1 Introduction It is well known that the object marker et usually precedes definite objects only: (1) a. dan kara et ha-sefer. Dan read et the-book 'Dan read the book.' b. dan kara sefer. Dan read book 'Dan read a book.' However, it has also often been observed (cf. Glinert 1989, Wintner to appear, and many others) that the distribution of et cannot be accounted for in purely semantic terms, and that there are two notions of definiteness in Hebrew, semantic and syntactic. This can be seen by looking at the contrast in the following sentences: (2) a. dan kara et ha-sefer ha-ze. Dan read et the-book the-this 'Dan read this book.' b. dan kara sefer ze. Dan read book this 'Dan read this book.' Semantically the object is definite in both sentences; but only in the first sentence there is formal marking of definiteness, and this is what et is sensitive to. Another example is the following pair: (3) a. dan kara et axad ha-sfarim. Dan read et one the-books 'Dan read one of the books.' b. dan kara exad me-ha-sfarim. Dan read one of-the-books 'Dan read one of the books.' Again, despite the semantic equivalence of the objects in these two sentences, which are both partitives, et is used by most speakers only in sentence (a) 1 . Clearly, this is the result of the different syntactic structures of the two partitives. For the vast majority of DPs in Hebrew, the presence or absence of et can be predicted from the syntactic structure of the object and from the presence of formal definiteness marking (Danon 2000). There is, however, a small number of cases where the use of et is optional. Consider, first, the partitives given in (3); for many speakers, et is optional in both cases: 2 1 Prescriptive grammars allow only the possibilities in (3), but for many speakers et is optional in both structures. There is a great degree of variation among speakers with respect to these cases. 2 The object in (3a) is a kind of cross-categorial genitive construction known as the Construct State (see for instance Siloni 1997). In (3b) the object is a ‘regular’ prepositional partitive construction.