Roman Poetry as Evidence for Ancient Magic
For evidence, I will draw on longer and more complex texts than the usual curse tablets; I again turn to Propertius and his four books of elegy.
Propertius’s first publication, the so-called Monobiblos, was composed around 29/28 BC. As discussed in the previous chapter, his poetic production continued until ca. 16 BC, when his last (fourth) book was finished and probably published. Thus he was a contemporary of the elegist Tibullus and of the epic and bucolic poet Vergil, both of whom died shortly before Propertius completed his fourth book, and of the lyric and iambic poet Horace, who died shortly thereafter. The theme of magic appears in various contexts in the poetry of all four of these authors. In some instances, characters in poems have quite extensive recourse to magic.
Georg Luck and Anne-Marie Tupet have dealt thoroughly with the descriptions of magic practices in these and other texts. Whereas the research of the 1960s and 1970s was directed toward compiling examples of magical practices and understanding the techniques and logic of these, more recent philological treatments have concentrated on the poetic function of the passages concerned with magic—how a reference to magic, for example, may serve as a metaphor for the binding qualities of love in relationships, or how the formulation of claims and expressions of skepticism regarding magical practices are informed by the techniques of rhetoric. In these discussions, however, the magic per se is usually heavily downplayed.
Other work on ancient magic has been engaged in historicizing the cultural techniques classified by this ancient term. There is unambiguous evidence from as early as the sixth century BC for practices of binding (katadein, defixiones) in different social relationships, in particular in the context of lawsuits. Likewise, examples of erotic magic can be found from the fifth century onward. Hellenistic literature offered full-fledged literary models of erotic magic and witches: Apollonius of Rhodes’s Ar- gonautica gives a detailed account of Medea in books 3 and 4, and Theocritus’s Idyll 2 portrays a young woman engaged in magical incantations and rituals directed toward her former lover. Thus, from the third
Snake on the exterior south wall of a latrine, Pompeii, caupona of Euxinus, I, 11, 11. Photo by J. Rupke, used by permission of Ministero dei Beni e delle Attivita Culturali e del Turismo, Soprintendenza Pompei.
century BC onward, Greek-reading Romans could draw on such texts in addition to Italic ritual traditions. The latter, however, are difficult to trace. By the end of the republic the practice (attributed to the Marsians) of appropriating a neighbor’s crop by means of incantation and snake charms had been subsumed into the conceptual sphere that encompassed binding spells, the use of which is attested in southern Italy in Oscan examples from the fourth century onward.  “Killing by poison,” venefi- cium, must have encompassed practices addressed by terms derived from Greekpharmakon.1() In Plautus, the term venefica refers both to the sorceress and to the poisoner.
The extensive treatment of magic in poems of the 30s and 20s BC is contemporary with (as far as we can see) sudden harsh policing moves. Agrippa had astrologers and sorcerers driven out of the city in 33 BC; Augustus banned the sorcerer Anaxilaos of Larissa in 28 BC. Against this background, the case for historic reference versus the purely literary value of poetic descriptions need not to be argued solely on the analogy of smoke and fire: much talk, therefore real magic. Obviously, poetic treatments constituted part of a wider public discourse that was engaged in cultivating a negative image of professional practitioners of magic.
The image of the old, drunk, and savagely cruel witch, so prominently developed in the early Augustan love poems, seems to be an innovation in the ancient discourse about magic. Given the growing danger of criminalization, as pointed out by Richard Gordon, the poets would not have had much interest in denying the fictitious character of their magic figures. Textual analyses should, therefore, pay special attention to the linguistic cues that signal the reality or implausibility of the characters and practices described, but even more so—as indicated above—to the more general framing of references to magic.
-  Luck 1962, 1992, 2000; Tupet 1976.
-  Cairns 1979, 140; Zetzel 1996 (hence Reinhard 2006, 208—9). Despite its prominence inthe poems (see below), the topic of magic is absent from many monographs on Propertius. I willrestrict myself to citing bibliography on points of controversy or to provide suggestions for furtherreading.
-  See, e.g., Hubbard 1974, 17—18. For a larger overview of the research on magic and its intellectual frameworks, see Gordon 2013.
-  In particular Ap. Rhod. 3.1026ff., 1191ff.; 4.123ff., 445ff., 1636ff. Dickie 2001, 99-104.
-  See Papanghelis 1987, 48, on the importance of both strands.
-  Dickie 2001, 128-44.
-  Briefly discussed by Graf and Johnston 1999, 662-70, particularly 669.
-  Dickie 2001, 131.
-  Graf and Johnston 1999; Dio Cassius 49.43.5 (the date being confirmed by Broughton,MRR 2:415); Jerome Chron. a. 735 auc.
-  Cf. Dickie 2001, 178, who concentrates on the sheer number of literary and specificallydeclamatory instances.
-  Simon 2001.
-  Ogden 2008, 39—76, esp. 75—76. For the resulting portraits, particularly in the case of prostitutes, see Dickie 2001, 175—91.