Poppets and Image Magic: The Documented History

The assumption embedded in the term “voodoo doll” is that a tradition of stabbing pins into effigies belongs to Haiti. The historical record says otherwise. Image magic — the use of a figure to stand in for a person, so that what is done to the figure affects that person — appears in Egyptian cultic papyri, Greek curse deposits, Roman sanctuary finds, and printed English witch-trial accounts. It is one of the most extensively documented magical practices in the archaeological and textual record. Haiti had nothing to do with it.

What follows is that record: what survives, where it survives, and what it actually shows.

The operating logic

Image magic runs on a single principle: a figure that resembles a person, or is linked to them through material contact (hair, nail clippings, clothing), participates in the reality of that person. What happens to the figure affects the person. The technical term is sympathetic magic, from the Greek sympatheia — feeling or suffering with.

The application varied by intent. A wax figure could be melted to waste a person away, bent or bound to constrain them, pierced to cause pain at a specific location, or buried near a grave to deliver the figure — and its named target — to underworld powers. The same instrument ran in the other direction: wax images of diseased limbs were left at healing shrines, so the site’s power would continue to attend the afflicted spot after the worshiper had gone home.

Sympathetic logic makes no distinction between cursing and healing. The tradition does not either.

Egypt: the wax figure in cultic texts

The oldest surviving textual evidence for figure magic in the Egyptian tradition is Papyrus Salt 825, now British Museum BM 10051. This is a Ptolemaic-period cultic text whose contents may preserve older material. It contains explicit directions for modeling wax figures of enemies in order to destroy the name of Seth — the figures are named, treated, struck, spat upon, and burned. The procedure is not folk magic operating at the margins of Egyptian religion; it sits inside a formal cultic context.

The UCL Digital Egypt database notes that all ancient Egyptian wax used in such contexts was beeswax, either pure or mixed with resin, oil, or pigments. The material’s malleability and combustibility made it apt for transformation magic: a figure is easily shaped, and what is shaped can be deformed or destroyed. Maarten Raven’s “Wax in Egyptian Magic and Symbolism” (1983) traces the use of beeswax in ritual contexts from the Middle Kingdom onward and documents the consistent association between wax and the ability to inflict or reverse change on another person.

The literary tradition reinforces the technical one. The late-antique Alexander romance attributed to the pseudo-Callisthenes describes Nectanebo II — the last pharaoh of independent Egypt — working wax figures of enemy armies in secret chambers of his palace, defeating them at a distance. This is not a historical account of what Nectanebo actually did; it is a literary reflection of what Egyptians in the Ptolemaic and Roman periods understood a powerful magician to be capable of.

One significant constraint: the UCL database notes explicitly that no wax figure certainly associated with an incantation has survived intact from Egypt. The physical objects, being wax, are gone. What survives is the textual and administrative evidence that they existed and were made.

Greece: kolossoi and the curse deposit

In the Greek world the standard instrument is the kolossos (plural kolossoi) — a small figure of wax or lead associated with binding magic, typically deposited alongside inscribed curse tablets (defixiones) in graves, wells, or sanctuaries where underworld powers were understood to reside. The earliest examples date to the fourth century BCE; they are attested archaeologically from Athens, Delos, and other sites across the Aegean.

Daniel Ogden’s Magic, Witchcraft, and Ghosts in the Greek and Roman Worlds (Oxford University Press, 2002), the standard English-language sourcebook for ancient magic, presents the Greek figurine tradition across roughly three hundred translated texts from literary and documentary sources. Ogden traces the procedures: a figure was named for its target, sometimes twisted, sometimes had its limbs bent backward (an enactment of constraint), and was deposited where chthonic powers could be petitioned. The operation was not always destructive. A large proportion of surviving defixiones are binding texts — the goal is to restrain a rival in a lawsuit, to hold a lover in place, to silence a business competitor. The ambition is control, not death.

This distinction between killing magic and binding magic tends to get collapsed in modern retellings, where any pin or figurine reads as a murder attempt. The ancient record is more varied.

Rome: the Anna Perenna find and the Leiden papyrus

The Roman tradition absorbed the Greek defixio and extended it across the empire. Curse tablets — thin lead sheets inscribed with binding formulas, rolled or folded and deposited in graves or wells — are among the most numerous single category of magical artifact to survive from antiquity, with major concentrations in Roman Britain, Gaul, the Rhineland, and North Africa.

Some deposits included figurines alongside the tablets. A well-documented archaeological example is the figure found at the sanctuary of Anna Perenna near the Tiber in Rome: a second-century CE female figurine deposited prone, hands bound behind her back, accompanied by a lead tablet inscribed with a binding formula. The bound posture is not incidental — it enacts the restraint the tablet requests. This is image magic in a stratified archaeological context, not a text describing what someone might do, but an object recovered from where it was placed.

The Greek Magical Papyri (Papyri Graecae Magicae, PGM) — the body of papyri from Graeco-Roman Egypt, dated from the second century BCE to the fifth century CE — give procedural detail for constructing such figures. The Leiden papyrus (inv. AMS. 75, fourth century CE) describes the making of wax and herb figures representing Eros and Psyche, explicitly intended to gain power over men and women. The PGM recipes are technical documents: they specify materials, spoken words, timing, and expected effects. They are not mythology. They are instructions.

Medieval Europe: learned tradition and folk practice

The late-antique tradition passed into the medieval learned world through Arabic translations of Greek texts and through Latin compilations on natural philosophy and ceremonial magic. Medieval church authorities attacked image magic consistently — not because they regarded it as harmless nonsense but because they understood it to operate through illegitimate means. The conceptual framework assumes efficacy; the theological objection is to the source of that efficacy.

At the folk level, materials shifted from wax and lead to cloth, clay, straw, and root. The technique did not. In English, the dominant term became poppet, from Middle English popet — a small child, or a small doll. By the sixteenth century the word covered both a child’s plaything and a figure used in magic; that dual sense reflects how unremarkable the practice was as a category of object, whatever the specific use.

The word survives in Northern English speech today as a term of endearment for a small child, entirely detached from any magical sense. The continuity is linguistic, not functional.

The trial record: North Berwick, Pendle, and Salem

Image magic enters the English-language witch-trial record in three major clusters. What the record preserves is accusation and testimony. The question of what the accused were actually doing with the objects described — or whether the objects were as described at all — is one the trial record alone cannot answer.

North Berwick, Scotland, 1590. The North Berwick trials produced some of the most dramatic testimony in early modern Scottish legal history, partly because the alleged target was King James VI. Confessions extracted under torture described the accused making wax effigies of the king and passing them hand to hand while reciting curses, on the understanding that the figure would waste as the wax was abraded. Whether the confessions reflect actual practice or the interrogators’ expectations is irresolvable from the surviving sources.

Pendle, Lancashire, 1612. Thomas Potts’s The Wonderfull Discoverie of Witches in the Countie of Lancaster (1613), the official printed account compiled from court records, records that poppets were allegedly found among the belongings of the accused and cited as physical evidence of malevolent intent alongside witness testimony. Potts’s account is unusually detailed relative to most early modern English trials; it remains the primary documentary source for the Pendle case.

Boston and Salem, 1688–1692. Cotton Mather’s Memorable Providences (1689) describes the Goody Glover trial in Boston, 1688: “several small images, or poppets, or babies, made of rags and stuffed with goat’s hair and other ingredients” were found in her home, and Glover acknowledged that wetting her finger with spittle and stroking the figures was her method of tormenting her targets. Glover was hanged that year. Her trial set the template the Salem proceedings followed four years later.

At Salem itself, poppet evidence appeared most explicitly in the case of Bridget Bishop. Two men testified that while dismantling a wall in her cellar they found “several poppets made up of rags and hogs’ bristles with headless pins in them with the points turned outward.” Bishop was executed on June 10, 1692 — the first person hanged in the Salem trials. The Salem records as a whole list the discovery of poppets among the categories of material evidence alongside confessions and testimony about spectral appearances.

What museum collections show

The Museum of Witchcraft and Magic in Boscastle, Cornwall, holds one of the most significant collections of English poppets extant — cloth figures from the seventeenth to early twentieth centuries, some incorporating inserted pins, tied cords, or encapsulated personal items. These were collected from across Britain. They are the objects the trial testimony describes, in their actual surviving form: modest, often clumsily made, clearly constructed from available materials.

The British Museum holds Papyrus Salt 825, the Bath curse tablets from the spring of Sulis Minerva (Roman Britain), and related amulet material from Egypt and the Roman world. The Leiden collection holds the papyrus already cited. The physical evidence is distributed across half a dozen major research institutions and treated in those contexts as routine archaeological and archival material — not as sensational, not as aberrant.

The consistent finding across museum collections is that image magic was not confined to any single tradition, was not always malevolent in intent, and left material traces recoverable over millennia.

The voodoo doll: how the stereotype was built

The term “voodoo doll” is not from Haitian Vodou. The Wikipedia article on the subject states unequivocally: “Despite its name, the voodoo doll is not prominent in the African diaspora religions of Haitian Vodou nor Louisiana Voodoo.” The association was constructed in early twentieth-century American popular culture and served a specific set of cultural and political purposes in that context.

The traceable dates are specific. John Houston Craige’s 1933 book Black Bagdad: The Arabian Nights Adventures of a Marine Captain in Haiti — written by a U.S. Marine officer during the American occupation of Haiti — describes a Haitian prisoner inserting pins into an effigy to induce illness. The book’s framing contributed to the association between pin-stuck effigies and Haitian practice. Victor Halperin’s film White Zombie (1932) and Jacques Tourneur’s I Walked with a Zombie (1943) refined and extended the image for cinema audiences. By mid-century the “voodoo doll” as a commercial object — crudely made, pin-studded, sold as tourist kitsch — was an established pop artifact.

Haitian Vodou is a sophisticated creolized religious system focused on maintaining relationships with ancestor spirits and lwa. Its actual ritual objects serve purposes of healing, spirit communication, and protective work, not the pins-in-effigy cursing of the Hollywood template. What commercial vendors in Haiti sell today under that name is largely a response to external tourist demand, not a survival of traditional practice.

Calling the practice “voodoo” mislabels it, misrepresents a living religion, and collapses a history that is considerably more interesting than the shorthand. “Poppet,” “kolossos,” “defixio with figurine,” “wax image” — all of these accurately name what they describe. The pin-stuck effigy labeled “voodoo” is, structurally and historically, a European poppet: the tradition documented in Pendle and Salem, relabeled in the early twentieth century through a process that served racialized anxieties about Haiti and Afro-Caribbean religion.

Where that leaves the record

Image magic is attested from Egyptian papyri through Greek curse deposits, Roman sanctuary finds, medieval learned texts, early modern folk practice, and printed trial records. The thread is long, geographically broad, and documented in primary sources accessible in major research collections. The practice was not secret, not exotic, and not confined to any single tradition or culture.

A practitioner working with figure magic today is working within one of the oldest recorded streams of sympathetic technique in the human record. The specific intent — healing, binding, love magic, protection, cursing — and the specific materials determine the ethical weight of any given working. The historical record does not adjudicate that; it establishes the practice’s antiquity and range.

Further reading

For the ancient record, Ogden’s sourcebook is the starting point, available from Oxford University Press and via Internet Archive. Robert Kriech Ritner’s The Mechanics of Ancient Egyptian Magical Practice (Oriental Institute, 1993) treats the Egyptian material in depth. For the English trial record, Thomas Potts’s Wonderfull Discoverie (1613) is available in full from various digital archives. For the material culture of British folk magic, the Museum of Witchcraft and Magic’s online catalogue is an accessible starting point. Owen Davies’s work on cunning-folk in English history addresses figure magic in practice, distinct from the prosecutorial frame of the trial record.