Anaphora is a central topic in the study of natural language and as such the object of research in a wide range of disciplines. Yet virtually no work has been done on the early history of the notion. This paper will situate the ancient discussions of anaphora in their proper context, and expound and briefly assess certain key concepts that have been overlooked so far.

I’d love to receive more feedback before submitting the final version in June, so requests for a pre-pub PDF would be warmly welcomed…

Update 2010-10-06: After inspection of the published version, I must unfortunately say that I cannot recommend using it for more than citation purposes. Indeed, I’d almost forgotten the proofs. Some clever mind at Blackwell’s had made a great number of completely unnecessary changes to the submitted text—the aim being to help the reader, I presume. I duly pointed out all the mistakes and infelicities introduced, of course. What I now just discovered is that in many cases my remarks have not been taken into account. A few examples: on p. 504, l. 1, the reference to chapter 7 is obviously out of place; in the sections on Chrysippus and Apollonius, my definitions were mostly (though not on p. 509) incorporated into the body of the text; the grounds on which single and double quotation marks are distinguished are a mystery to me; my groupings of paragraphs were in general ignored (again, a pleasant exception on p. 510); sometimes two paragraphs were flown into one; and so on. So dear prospective reader, please use and peruse what I’d like to call the original version.

Priscianâ€s Ars is generally considered the apogee of Latin grammar. Yet he himself claims to have followed in every respect the authority of Apollonius Dyscolus (Ars 13.24.7; al.). For a proper assessment of Priscianâ€s place in the history of linguistic thought, it is thus indispensable to clarify the relation between his and Apollonius†œuvre. Focusing on books 12 and 13, I first present a list of all Apollonian passages translated or paraphrased by Priscian, and discuss one of these in more detail (Ars 13.15.9–28). I then analyze the order of Priscianâ€s exposition, and compare it with the structure of his Greek model. Finally I give two examples—the notions of figure and of reflexivity—to illustrate how, and why, their theories differ from one another. Thence I conclude that the currently fashionable view of Priscian as a foundational thinker, almost entirely detached from the Greek tradition, is mistaken, and that, on the contrary, he can only be understood against the backdrop of Apollonian grammar.