A fresh look at the debate on floating quantifiers

Linguists have for a long time been in two opposing camps regarding how to explain the phenomenon of floating quantifiers, exemplified in the following two sentences: (i) All the students have read the book. (ii) The students have all read the book. The quantifier all in (i) modifies the phrase the students. The question is what enables the quantifier to float away from the phrase that it modifies in (ii). For a long time, many linguists have believed that (i) and (ii) are unrelated. They have assumed that the subject in (i) is "all the students" and that the subject in (ii) is "the students," with "all" being a kind of adverb located in a typical adverbial position, right next to the verb phrase. This approach is usually called the Adverbial Approach to floating quantifiers. In 1988 a theory was developed whereby (i) and (ii) are the same sentence. Under this approach the subject of both sentences is "all the students," but in (ii) the phrase "the students" has moved to a higher position in the sentence and has stranded the quantifier next to the verb. This approach, called the Stranding Analysis, generated enthusiasm when it was first presented in 1988, but its success was short-lived. It was found to have weaknesses, and this caused many linguists to return to various adverbial approaches. Linguists are still very much divided on the subject. The purpose of this article is first of all to explain what the debate on floating quantifiers is all about and secondly to argue that if one updates the Stranding Analysis for more recent developments in generative linguistic theory, and if one considers more data in more languages, one will find that it is too early to write off the Stranding Analysis and declare the Adverbial Approach the winner