Celtic Curses (review)

Bernard Mees brings together a considerable body of material from Continental antiquity to sub-Roman Britain, medieval Ireland, England, Scotland, and Wales in Celtic Curses. Some fundamental questions are raised by the book. Can a culturally specific model for the use of curse be formulated from the limited ancient and medieval textual remains of a particular Indo-European linguistic group? Can sources and analogues for these textual remains be recovered to the extent that they may be used to reveal the true "Celtic" nature of any one so-called "Celtic curse"? Can a provable relationship between the cursing language and strategies of the Continental Celts and their descendants in Ireland and Britain be established on this basis? That is, are we at all justified in devising a category called "Celtic curses"?

There is much to recommend this book, particularly its collection of a number of Continental curse tablets and its specialist considerations of a few of these in great detail. As with Runic Amulets and Magic Objects, jointly authored with Mindy MacLeod (2006), Mees is in his element when discussing the world of individual artifacts and makes genuine contributions to their understanding, particularly by supplying proposed translations for difficult texts. His even-handedness in dealing with readers' biases is appreciated. Criticism, however, may be directed to the relationship between methodology, data, and global commentary. Unfortunately, what seems to be the cornerstone discussion of Celtic Curses reveals, in this reviewer's opinion, the most problematic aspects of Mees's self-stated goal to "breach" the divide between Continental Celts and their "insular cousins" on the subject of cursing (p. 9).

The impression that methods are unevenly applied and the suspicion that, at times, Mees desires conformity of the evidence to a presupposed argumentative trajectory may well be substantiated in the following example concerning the problem of "binding" curses, or defixiones, as raised by the author, particularly in Chapter 7, "Breastplates and Clamours." It would seem that most of the (sometimes convoluted) argumentation of this chapter and the remainder of the book might be cleared up by asking the simple question, "are the lorica prayers, such as the Lorica of Laidcenn/Gildas, defixiones?" No. But the idea that they may be like them in many ways (as proposed by scholars before Mees) is curiously unacceptable to him on a number of counts. Mees's argument stresses that because we do not have the exact replication in word or form of the defixiones in the evidence from "medieval Celts" (p. 9), no continuity, and only questionable influence if any from these older strategies may be claimed. Mees charges that modern commentators who make the comparison are missing the obvious, that is, that the language of cursing among these "Celts" is overwhelmingly Latinate, "too biblical," and too Christian. "Instead of obviously continuing an older Celtic tradition, in early medieval times Insular Celtic curses were rather more clearly both named for and modelled on those which featured in the literature of the Latin-speaking Roman church, not an inherited pre-Christian tradition" (p. 116). His observations are underpinned by the conviction that "there does not appear to be any reflection of the ancient notion of binding in any of the medieval Celtic words for 'curse' or in such accounts of early insular imprecation as have survived," and that there is no anecdotal evidence for such practices from medieval Irish saints' lives (p. 119); the logic is, for example, that Latin maledictus, whence Irish mallacht/maldacht, Welsh mellith, and Breton malloc'h, has nothing to do with defixiones nor native Irish áer (satire) in thought, word, or deed (p. 116). Mees is determined to show that one very well known apotropaic text from medieval Ireland and Anglo-Saxon England, which he refers to as the Lorica of Gildas, is not, as argued by Michael Herren et al., the closest analogue for ancient binding or defixing curses in that milieu and cannot be considered to belong to the category of "anti-defixiones." Leaving aside the obvious overdetermination of the lorica...