Curse Tablets: The Documented Root of Binding Magic

More than a thousand lead sheets survive from antiquity with curses still readable on them. They were inscribed, folded or rolled tight, pierced with iron nails, and dropped into graves, wells, and sacred springs. Some were buried with the newly dead. Others sank into the thermal waters of Roman spa-shrines. A few were deposited at crossroads or in the foundations of buildings. The people who made them were ordinary: bath-house visitors who had lost a cloak, charioteers who wanted a rival’s horses to stumble, women trying to bind men to them, litigants terrified of an upcoming trial.

These are defixiones — the Latin term, from defigere, to fasten down or fix. In Greek the equivalent word is katadesmoi, from katadein, to bind. Modern witchcraft calls what descends from them “binding magic.” The ancestor is considerably rougher than the descendant.

What survives

According to the archaeological record, curse tablets were in wide circulation for roughly one thousand years, from 500 BCE to 500 CE. Surviving examples have been found in Athens, Rome, Spain, Syria, and as far away as Britain, where a cache of more than 130 tablets was recovered from the Roman baths at Bath.

Called katadesmoi in Greek and defixiones in Latin, over 1,600 of these survive throughout the Greco-Roman world from about 500 BCE onward. Gager’s 1992 corpus — still the standard English-language edition — draws on tablets from North Africa to Britain, Syria to Spain, and treats them as primary documents rather than curiosities. They are primary documents. Unlike the magical papyri, which are recipe books compiled by professionals, defixiones are the spells that were actually used: deposited, nailed shut, left to work.

Lead was the typical medium. Lead was readily available and inexpensive, a byproduct of silver mining, making it a practical writing medium for ritual purposes. Its malleability meant the sheet could be inscribed with a stylus without specialist tools, then folded or rolled to conceal the text and physically enact the binding — the spell was folded in on itself, sealed, fixed. Some tablets were then transfixed with one or more iron nails: the gesture reinforced the defigere logic, pinning the curse as one pins a living thing.

Who commissioned them, and why

The four main categories that recur across the corpus are: judicial curses (targeting opponents in litigation), competition curses (targeting rival athletes, charioteers, theatrical performers), commercial curses (targeting business rivals), and erotic curses (binding a specific person to the curser, or separating a rival from a desired partner). The majority of defixiones fall into one of these four categories: legal/justice; choral or athletic competitions; trade; and love and sex.

The judicial category is especially well attested at Athens, where the fifth and fourth centuries BCE produced a significant corpus of Attic curse tablets. A litigant facing trial in the dikasteria might curse the tongue, hands, and feet of opposing witnesses — body-part by body-part, listed with legalistic precision — asking that their speech be bound and their arguments fail. The logic was coercive rather than supplicatory: the curser attempted to exert compulsion over persons and outcomes that formal institutions could not guarantee.

Competition curses, particularly those targeting chariot racing, are among the most detailed texts in the corpus. Circus factions hired professional curse-writers (magoi) to compose elaborate defixiones against rival teams. These tablets name specific drivers and horses, list the body parts to be bound, and invoke chthonic deities and the spirits of the untimely dead — aōroi, those who had died too young and were therefore believed to be restless, still close to the living world, and available for magical errands.

Delivery: chthonic logic

The depositional choices were not arbitrary. Graves, wells, springs, and crossroads are all liminal or chthonic locations — thresholds between the living world and the realm of the dead. Defixiones typically sought to bind or restrain their targets, often utilizing reversed, twisted, or jumbled forms of writing, conventions, or imagery, and had much in common with figurines used as voodoo-style dolls, which appear as early as the archaic period.

The dead were the delivery mechanism. A curse deposited with a fresh corpse was understood to travel with it; the restless dead — the aōroi, those killed violently or before their time — were the most effective couriers and the most susceptible to magical compulsion, not yet settled enough to ignore the living world. Ogden’s sourcebook documents how the same logic runs through literary descriptions of sorcery and through the actual physical tablets: you gave the curse to the dead, and the dead carried it below.

Water made a different kind of sense. The Bath tablets are thin sheets of lead or pewter inscribed with Latin texts, primarily directed to the goddess Sulis Minerva for justice against thieves, deposited in the sacred spring at the Roman baths in Bath, Somerset, dating from the 2nd to the late 4th centuries AD. The sacred spring at Aquae Sulis had its own chthonic valence — water rising from underground carried petitions downward to divine attention. Over 130 such defixiones have been recovered, folded or rolled after inscription to activate their ritual power, deposited in the thermal waters as offerings to invoke divine intervention in everyday disputes.

The Bath cache as case study

The tablets from Bath are particularly important for any practitioner trying to understand what binding magic looked like in actual practice, because they survive in unusually large numbers and their texts are largely legible. Typically, texts from Bath and Uley relate to theft — at Bath, for example, of small amounts of money or clothing from the bath-house, at Uley of animals or farm implements. In formulaic, often legalistic language, tablets appeal to a deity, for example Sulis at Bath or Mercury at Uley, to punish the known or unknown perpetrators until reparation is made.

One Bath tablet, quoted in the secondary literature, reads in paraphrase: “Docilianus asks the most holy goddess Sulis to curse him who has stolen my hooded cloak, whether man or woman, whether slave or free.” The relationship between petitioner and deity was framed as a contract: in exchange for divine powers delivering justice, the devotee would provide the god with an offering or sacrifice. In order to help the invoked god track down the guilty party, the petitioner tried to mention every possible type of individual the villain could be.

Henk Versnel (cited in the scholarly literature on Bath) distinguishes this category — the “prayer for justice” — from offensive cursing: the petitioner is wronged, names the wrong, and invokes divine redress rather than personal vengeance. The distinction is not always clean in practice, since the requested punishment could include denial of sleep, cessation of bodily functions, and death. But the framing matters: these are people with no access to formal legal remedy, using the available supernatural institution.

The Pella tablet and erotic binding

The erotic defixiones raise questions about agency and coercion that modern discussions of “love magic” often soften. The Pella curse tablet, dating to around 380–350 BCE, is a love charm written by a woman, possibly named Dagina, whose lover Dionysophon is apparently about to marry a woman named Thetima. She invokes “Makron and the demons” to cause Dionysophon to marry her instead, and never to marry another woman unless she herself recovers and unrolls the scroll. The purpose is not to attract a stranger but to block a rival and compel a specific outcome: the target is to be bound from marriage to anyone else.

Katadesmoi — the Greek term the Pella tablet exemplifies — were “spells written on non-perishable material, such as lead, stone, or baked clay, and were secretly buried to ensure their physical integrity, which would then guarantee the permanence of their intended effects.” The spell’s physical integrity and the spell’s operative duration were understood as linked: roll the tablet back out and the binding lifts. The tablets were, in this logic, contracts with the dead rather than one-time commands.

What modern binding is not

Contemporary witchcraft discourse frames “binding” as primarily defensive — you bind a harmful person to prevent further harm, typically to yourself or someone you are protecting. The procedure is presented as a last resort, less serious than a curse, protective in intent. None of this framing appears in the defixiones corpus.

The ancient tablets are overwhelmingly offensive and coercive. They target opponents, rivals, and desired partners. They specify body parts to be incapacitated. They invoke the dead to carry the compulsion. They are not restrained, reluctant, or defensive. The category “prayer for justice” comes closest to the modern defensive frame, but even there the requested punishments are severe.

This is not an argument against modern binding practice. It is an argument for clarity about what the historical record shows. Practitioners who want to understand where binding magic comes from will find in defixiones a set of texts that are specific, operational, and morally blunt — and that look very little like the sanitised contemporary version. Gager’s corpus translates these texts directly. Ogden puts them in context. Both books are primary reading for anyone who wants to work with the tradition honestly rather than reconstruct it from inference.

A note on consultation and further reading

The defixiones corpus is one of the best-documented archives in ancient magic studies precisely because the objects survived: lead does not rot, and dry graves and mineral springs preserve inscription. Gager (1992) collects English translations with facing commentary. Ogden (2002) situates the tablets within broader Greco-Roman magical practice and provides translations of related literary sources — Plato’s Laws, Lucan’s Pharsalia, the Greek Magical Papyri — that illuminate how the tablets were understood by contemporaries.

For British tablets specifically, the Corpus of Writing Tablets from Roman Britain maintained by the Oxford Centre for the Study of Ancient Documents provides access to the Bath and Uley material with scholarly annotation.

Cross-references on this site: (forthcoming — Aquae Sulis sacred spring rite, the Greek Magical Papyri, knotting and cord magic in European folk tradition.)