The Archaeology of Curse Tablets: Katadesmoi & Defixios

I want to talk about curse tablets & how they were supposed to work. Curse tablets were the most popular and common type of magic used in the Graeco Roman world, attested by the sheer number of them that have been found, and copious references in ancient literature, and as we can tell were used irrespective of social class.

In total we have about 1600 tablets currently on record, 220 of which have been found in Attica, but these things have turned up all over the roman world. They therefore represent a tremendously large number of instances of a single genre and style of inscription and we therefore know a lot about them. 

They’re called Katadesmoi in Greek, which is derived from the verb Katadeõ, the literal meaning of which is “to bind on or to”, and has clear connotations of constraint and coercion. Remember this, we’re gonna come back to it later. This idea is also echoed in their Latin name, defixio, which is derived from the Latin verb defigere, which means literally "to pin down," but also has a secondary connotation with the idea of delivering someone to the powers of the underworld.

The usual process you may be familiar with involved writing the victim's name on a thin sheet of material, usually lead, wax or ostraca, along with varying magical formulas or charaterēs (although only about 25% of magical artefacts from Late Antiquity are also inscribed with magical signs), then burying the tablet in or near a tomb, a place of execution, or a battlefield, to give spirits of the dead power over the victim and transmit your intention to the restless dead who were believed to carry out the instructions. 

There is a problem here though… our definitions inevitably fall short. In the ancient world, not all lead tablets are curses, and not all curses are written on lead. There is a serious lack of delineation over how we actually define the methodology and efficacy of these things. How was inscribing someone’s name on a lead tablet supposed to bring them misfortune? Or bind their anger? Equally, if we look at cases like the Bath Curse Tablets, these are not strictly curses. 

They are appeals to the Romano Celtic deity Sulis Minerva for justice against thieves. In fact, the inscribed prayer is clearly not intended for the culprit to be actually punished. Instead, the tablet acted as a threat whereby the thief would be punished only if the items were not returned. The efficacy here in the case of the Bath tablets is fairly simple, it is through divine mediation as opposed to by daemons or the restless dead. First the author “gifts” the stolen items to the goddess so that they become her property, and then prays for her to intervene to retrieve “her” goods back, so they may be deposited in the temple. In other words, because of the transfer of ownership, the items the thieves stole now in fact belong to the goddess, not the original author.

All of this speaks to the wider issue of a lack of classification in how we approach magic in the ancient world. Ancient Magic was far more concerned with keeping order and dispensing justice.

The use of the word “curse” in our vocabulary naturally introduces a bias that makes us assume defixios are harmful or always negative magic. However, if we look at examples from the PGM, there are numerous cases in which lead lamella are used as a means to restrain anger from another person against you, which inherently implies a protective function to some of them, they act as protective charms, even though they are using the same verbal formulas.

Of course, the PGM also contains a decent amount of the “traditional” curse tablets, either of an erotic nature or a harmful nature. But still, we can therefore assume the inscription of names and intentions on materia was simply a standard practice of magical operation, rather than something unique to harmful or cursing style magic. In order to define the practice of defixios and how they work then, including the metaphysical ideas and theology behind them we have to look at the specific formulas, words and themes used to refer to acts of binding in literature and used on the tablets themselves.

First try to do away with thinking of magic in terms of “curses' ' and “blessings' ', it isn't helpful for ancient stuff, the lines are far too loose. Understand we usually have two or more people involved in a ritual, the operator themselves & the client. Sometimes these two people can be identical, but it still needs to be clarified.

Defixios do not seem to have made use of magical signs or characterēs. In context, very few of the tablets we have contain any trace of sigils, signs or astrological symbols. The procedure is almost always mediated by either the material itself and its wider correspondences or by the verbal formulas employed whether to invoke supernatural assistance or bind through your own will.

One of the most detailed literary examples of defixios comes from Plato, both in the Republic & Laws. He questions what the legal ramifications of “binding spells” should be, which for him are defined as wax images deposited in doorways, crossroads or graveyards. All of these are liminal places, the other common location for deposit being the sanctuary of a chthonic deity. 

As far as material goes, lead is by far the most common inscription medium. This is partly due to it being cheap and very easily available and as a soft metal, easy to engrave on. However there are cultural and theological reasons for choosing lead. Graf & Faraone both note that Greek perceptions of lead held it was a material without value, since it lacked lustre and was characterised by its coldness and flimsy nature. Ovid refers to wax defixios also being made, however, the Aegean climate does not preserve them. Wax was likely another common material used alongside lead.

How did they function?

Defixios contain a formal wish or intent to bring other people or animals under a client or operator's power, not just curse or harm. 

Christopher Faraone & Eugen Kagarow have proposed four categories for modes of operation when it comes to how the tablets were supposed to work, most of which are defined by verb forms, which tell us a great deal about the assumptions made by the engravers.

The first way in which Defixios were thought to operate was actually through unmediated action and the engraver’s will alone. This form uses first person singular verbs, probably in an attempt to establish a link between the spell caster and the target. A good example of this form is tablet DTA 87a, designed to restrain business competitors. It was found in Attica and dated to the fourth century BC. One by one, the engraver names local business owners who he was to constraint and stop competing against and writes “I bind you”. In this Unmediated Type, there is no clear supernatural aid invoked, so it seems the author is banking on the nature of the verbal form as a metaphor of declaring his will.

The next way in which Defixios were thought to work is kind of the “classic” way, in which a supernatural intermediary or assistant is invoked in the inscribed formula to enact the operators will on the target. The spiritual creature invoked can be a god, a daemon or one of the auroi, the restless dead. Interestingly though, any notion of supplication or asking for help from these creatues is rare in defixios. The invocations of this type are phrased in the second person imperative (commanding) tone. A good example of this type of tablet  is DT 52, again found in Attica and dating to the second century BC. The tablet invokes and commands Chthonic Hermes to bind the targets.This type of coercive formula is typical of older defixios, as time goes on, the writers appear to have shifted to more supplicatory formulas.

John Gager notes in his 1992 book “Curse Tablets & Binding Spells from the Ancient World” that another typical form of curse tablet invocation takes on a more judicial form where the inscription reads like a legal contract. In this case, the formula is similar to the typical “I bind” formula, but instead replaces “bind” with the term “register”, in the sense that the operator hands over jurisdiction of the target’s fate to the invoked deity. Gager gives the example here of a defixio found in Delos, dating from the 1st century BC-AD in which the operator lists the various parts of the target’s body and registers them under the control of a deity.

The final cartridges in Faraone’s discussion of Defixios are what he dubs Wish Formulas and Similia Similibus. These invocation typically involve making a desire or wish for the target to take on certain traits associated with something else, usually part of the tablet material such as lead or wax, or qualities of the place where the tablet is deposited. For example, Tablet DTA 67 asks that the “words” of the carver be as “cold” to the target as the lead upon which he engraves. The thought here is not necessarily sympathetic magic, but persuasive. The thinking is that the tablet itself will somehow pursue the target to take on and transfer its characteristics to themselves.

The most common deities invoked in Defixios are chthonic or erotic and otherwise hold sway over the underworld. Usually it is Hermes, Hekate, Perspehone, Hades or Demeter & Gaia, however as the tablets spread out of Greece, we also have invocation to Osiris, Ereshkigal and even Iao. Eventually even intermediary daemons are employed through the use of Voces Magicae and Onomata.

Sources:

Audollent, A. 1904: Prooemium. In Defixionum tabellae, xvii-cxxviii. Paris: Fontemoing. (See here for a comprehensive list of curse tablets mentioned in the video under the DTA & DA labels)

Faraone, C. 1999: Ancient Greek Love Magic. Cambridge: Harvard University Press

Faraone, C. 1991: The Agnostic Context of Early Greek Binding Spells. In Magika Heira: Ancient Greek Magic & Religion. 3-32. Oxford: Oxford University Press

Gager, R, J. 1992. "Introudction". In Curse Tablets and Binding Spells from the Ancient World, edited by Gager, J, 3-41. Oxford: Oxford University Press

Kagarow, E. 1929: Griechische Fluchtafeln. Eus Supplementa 4. Leopoli: Societatem philologan polonorum.