Love Magic: The Documented History

The word love is doing a lot of work in the phrase love magic. It suggests mutuality, warmth, the hope that someone feels the same. The surviving ancient record uses that word too — and means something considerably harder.

The erotic magic of the Greek and Roman world is among the best-documented magical practice in any pre-modern corpus. Hundreds of lead tablets, dozens of papyrus formularies, scattered literary references from Plato to Lucian, and a small but striking number of physical figurines survive. When scholars examine them as a whole — as Christopher Faraone did in Ancient Greek Love Magic (Harvard University Press, 1999), and as Daniel Ogden assembles in his sourcebook — the picture that emerges is of a practice organized not around mutual desire but around compulsion, and frequently around the suffering of an unwilling target.

This page works through what that corpus actually contains, what categories the ancient practitioners themselves recognized, and what gets flattened when the tradition is repackaged as “attraction magic.”

The Corpus

The primary source is the Greek Magical Papyri (abbreviated PGM from the Latin Papyri Graecae Magicae), a collection of magical texts discovered in Egypt and dating from roughly the second century BCE to the fifth century CE. The standard English translation is Betz’s 1992 University of Chicago edition. The papyri are not a single document — they are a set of practitioner’s handbooks, formularies written in Greek with significant Demotic and Coptic sections, reflecting the syncretic milieu of Greco-Roman Egypt. Greek spell-logic runs alongside Egyptian divine names; Jewish voces magicae appear throughout.

PGM IV is the longest single text in the corpus — a 72-page codex probably copied in fourth-century CE Thebes. It alone contains over fifty recipes covering love-binding, divination, cursing, and plant-gathering. Its lines 296–466 preserve what is often called a “wondrous spell for binding a lover,” one of the most complete surviving instructions for creating the inscribed lead tablets that survive in the hundreds from the ancient Greco-Roman world.

Alongside the papyri, the second major evidence base is the katadesmoi — the lead curse tablets, also called defixiones. Inscribed, rolled or folded, and deposited in graves, wells, or chthonic sanctuaries, these tablets have been recovered across the Mediterranean, from Athens to Sicily to Roman Britain. They range in date from the sixth century BCE onward, giving the tradition a documented lifespan of at least a thousand years.

Together the papyri and tablets form the backbone of any serious study. Literary sources — Theocritus’s Idylls, Lucian’s Dialogues of the Courtesans, Apuleius’s Golden Ass — provide cultural context and attitude but must be handled carefully: they are literary representations, not procedural instructions.

Two Spells, Not One

Faraone’s most durable contribution is the argument that “love magic” in the ancient world is not a single category. He identifies two structurally distinct types:

Agoge spells — from the Greek agōgē, “leading” or “bringing” — are spells of compulsion. Their grammar is consistently violent: the target is to be tormented, burned, sleepless, driven mad with desire until she (the target is almost always female) submits and comes to the practitioner (almost always male). The PGM formularies are explicit. A typical agoge calls on the demon of the dead, on Hekate, on underworld powers, to heat the target’s heart, burn her liver, and give her no rest until she arrives. The deina formula — ὁ δεῖνα (“so-and-so,” male) commanding action upon ἡ δεῖνα (“so-and-so,” female) — runs through the Formulary erotic spells with near-total consistency. This is not a reading-into-the-text; the asymmetry is structural.

Philia spells — from philia, affection or attachment — operate on a different logic. These are binding and retention spells, aimed not at bringing a new person to the caster’s bed but at preventing an existing partner from leaving or straying. Faraone’s analysis shows that philia magic was more commonly employed by women — wives anxious about philandering husbands, women trying to stabilize a relationship under social and legal pressure. The philia idiom is less overtly violent than the agoge idiom; the goal is sedation of conflict and attachment, not pursuit and capture.

These two types are often lumped together under “love magic” in modern popular treatments. Keeping them apart matters: they reflect different social positions, different gender dynamics, and different magical technologies.

The Material Technology

The PGM spells are not purely verbal. They specify materials with precision.

Lead tablets. The inscribed lead defixio is the characteristic physical artifact of binding magic. Lead is a chthonic metal — heavy, cold, a material associated with the underworld and with permanence. A practitioner inscribing a love-binding would write the target’s name, the operator’s name and mother’s name (maternal lineage was the identifying formula), the command, and often a string of divine names. The tablet was then rolled or pierced with a nail — the piercing is both symbolic and attested in the physical finds — and deposited in a liminal location: a grave, a well, a sanctuary of the dead. The goal was to engage the spirit of the dead person at whose side the tablet was placed, enlisting chthonic power as the engine of compulsion.

Figurines. The papyri are also the source of instruction for what scholarship calls “voodoo dolls” — more precisely, kolossoi or bound figurines. PGM IV.2943–57 directs the practitioner to take unbaked dough or unmelted wax and mold a small dog; to take the eyes of a bat and set them in the dog’s eye sockets; to thread a needle through the eyes. PGM IV.296–466 describes a human figure with limbs and sensory organs named and dedicated to specific torments — the hands bound behind, the face turned downward. Excavated examples from Egypt and elsewhere confirm these instructions were executed: twisted lead or clay figures with pins inserted have been recovered from ritual deposit contexts. These are not metaphors. Ogden’s sourcebook reproduces several in his chapter on magical images and their use alongside erotic binding.

Gemstones and amulets. A third material register — engraved gemstones with divine images and magical inscriptions — appears in the archaeological record, though the erotic type is harder to assign clearly to one spell-category or another.

The Social Frame

Ancient love magic in the agoge mode is organized around a presumption of unwillingness. The spell is necessary precisely because the target does not want to come. This is not a nuance or a modern over-reading — the formulas say so directly. The agoge calls for the target to be given no peace until she crosses the threshold. The spell imagines the erotic relation as a siege.

This stands in stark contrast to how the same tradition gets presented in many modern contexts, where the same Greek and Egyptian precedent is invoked to justify working “attraction spells” for a willing or ambiguously-interested target, conceived as helping along a relationship that wants to happen. The two framings are not compatible. One can choose the modern frame — that is a legitimate contemporary practice decision — but one cannot claim the ancient precedent honestly and sanitize what that precedent actually describes.

Faraone’s analysis also complicates the question of who was harmed and who was protected. Philia spells, deployed by women, were themselves a form of control — but they operated inside the relational structure that already confined women, and their target was typically a husband whose departure would mean social and economic disaster for the woman concerned. The asymmetry of power runs through both spell types, in different directions.

What Survives into Modern Practice

The direct line between Greco-Roman erotic magic and contemporary Western magical practice is broken, not continuous. Modern “love spells” — candle workings, honey jars, petition papers — descend through early modern grimoire traditions, hoodoo and rootwork, and the mid-twentieth century revival, not through unbroken practice from ancient Egypt. This matters because it means practitioners drawing on the PGM for historical grounding are doing scholarship, not tradition. Which is fine, and more honest, than claiming the thread is unbroken.

What the ancient corpus provides for a modern practitioner is a documented vocabulary: the distinction between compulsion and retention, the material logic of binding (metal, wax, earth, the dead), and a clear-eyed account of what erotic magic was understood to be in the world where it is best attested. It was not soft. It was not romantic. The agoge worked by making someone suffer until they gave in.

Whether that premise belongs in contemporary practice is a question each practitioner answers for themselves. The record has the right to be read as it is.

Cross-References

Related entries on this site: curse tablets and defixiones (forthcoming); poppets and image magic (forthcoming); the Greek Magical Papyri as a corpus (forthcoming).