The funerary defixio from Larzac ends with a call that Severa Tertionicna’s enemies may suffer, just as she evidently was suffering from their legal machinations. It does so using an idiosyncratic form of a ‘just as …, so too …’ formula, one that repeats several key phrases from earlier on in the find. The Larzac curse is not unique in this manner, however: similar retributory themes are just as clearly expressed in other Gaulish inscriptions that have come to light since the 1970s, not that these texts have always immediately been recognised as recording ancient imprecations. Finds of Latin binding spells from France are usually better appreciated, but not only because the language they are written in is much better understood than Gaulish is today. Latin defixiones often feature odd vocabulary, uses and wordings, and hence can sometimes be difficult to make sense of – often it is the appearance of typical magical expressions such as sympathetic rhetoric that is the most obvious feature of such texts. Other times, they are identified as curses principally because they have been found in physical circumstances that are typical of binding spells. Some Latin curse tablets found in France are occasionally so idiosyncratically Gallo-Roman, however, that they even preserve Celtic words, much as if these terms represent evidence for a native tradition of cursing that was not easily translatable into the language of the imperial conquerors.

A particularly intriguing example of such a find was unearthed in 1970 from the remains of a Gaulish hill-fort or oppidum at Montfo, some 50km south of L’Hospitalet-du-Larzac. Found in the remains of an ancient well, the mid-firstcentury Montfo defixio begins in a common-enough way for a Roman binding curse. Yet it ends in a unique manner, one which, moreover, appears to be heavily Gallified. Not only does the 100mm by 85mm lead tablet feature at least one clearly Gaulish term, it also refers to a necracantum or ‘death song’, a nonstandard, partly Greek description (cf. Greek nekros ‘corpse’) which immediately brings necromancy to mind. But not only does necromancy not make much sense in the context of a well, the term necromantia is not known in Latin until it was borrowed from Greek by early Christian writers in the third century – long after the Montfo text was deposited. Indeed, necromantic ‘death songs’ are not known from the classical cursing tradition.