The Witch's Familiar: A History

The word “familiar” covers at least three distinct things in English history, and modern usage tends to collapse all three into one. There is the animal imp of the witch trials — the cat, toad, or mouse that fed from a mark on the witch’s body and worked harm on command. There is the fairy helper of the cunning folk — the spirit guide encountered in a trance or at a liminal threshold, who taught remedies and located stolen goods. And there is Emma Wilby’s more unsettling proposal: that the second of these was the original, and that the first was what the law made of it when it passed through the machinery of prosecution.

Keeping these three separate is the first task of anyone who wants to understand what they are actually working with.

An English Peculiarity

The animal familiar is overwhelmingly an English legal and folkloric phenomenon. Continental European witchcraft prosecutions — from the Alpine panics of the fifteenth century through the German trials of the late sixteenth — were organized around the sabbat: the nocturnal gathering at which witches renounced their baptism, submitted to the devil, and performed ritual inversion of Christian practice. The animal helper appeared in some continental accounts but was peripheral to the prosecutorial template. In England and lowland Scotland it was central.

Owen Davies, surveying the regional distribution of familiar belief across English trial records, found the concept almost entirely absent from Wales, sparse in the north, and densest in East Anglia — Essex, Suffolk, Norfolk, Cambridgeshire. The geographic concentration matters. It suggests that the animal familiar was not a mechanical byproduct of any witchcraft trial wherever it occurred, but something rooted in specific regional folk belief that then became legible to the law.

Why East Anglia specifically is debated. One explanation locates it in pre-existing belief about household spirits — brownies, hobs, and similar creatures that could be propitiated and set to work. Another connects it to the character of English legal procedure: unlike the Scottish and continental courts, English common law did not routinely use judicial torture. Confessions therefore grew from what accused and accusers actually believed, rather than being systematically forced into elite demonological templates.

The 1563 and 1604 Witchcraft Acts

The legal framework shapes what appears in the records. The Elizabethan Witchcraft Act of 1563 (5 Eliz. I c.16) made invocations and conjurations of evil spirits a felony. The 1604 Act (1 Jac. I c.12), passed at the accession of James VI of Scotland as James I of England — a king who had himself written Daemonologie — significantly expanded the criminal list to include “consulting, covenanting with, entertaining, employing, feeding or rewarding any evil and wicked spirit.”

The word “feeding” is not accidental. By 1604, the animal familiar as a creature that required nourishment from the witch’s body was sufficiently established in English legal culture that parliament wrote it into statute. The law was, in this sense, confirming something that had already organized popular accusation for forty years.

Chelmsford, 1566

The first printed account of an English witch trial is a pamphlet published in London in 1566, describing the cases of three women from Chelmsford, Essex: Elizabeth Francis, Agnes Waterhouse, and Joan Waterhouse. Its central figure is a creature that appears at different moments as a white spotted cat, then a toad, then a black dog with an ape’s face. His name is Sathan.

According to Elizabeth Francis’s confession, she received the cat from her grandmother, who taught her to “feed the said cat with bread and milk.” When she wished harm on a neighbor, the cat required “a drop of [her] blood” in return. Sathan passed from Francis to Agnes Waterhouse, and from Agnes to her daughter Joan — a chain of inheritance presented in the pamphlet as a routine mechanism, as if it were a known property of such creatures.

This 1566 pamphlet is the earliest clear printed record of the English animal familiar, and it already contains all the structural features that would recur across the next century of trials: the animal form, the blood-feeding, the inherited transmission, the contractual exchange of harm worked in return for nourishment given. Elizabeth Francis was tried three times — 1566, 1573, and 1579 — before being executed at the final hearing. She is the first named individual in the English trial record specifically identified as keeping a familiar.

The Witch’s Mark

If familiars fed on blood, the body of the witch had to bear the evidence. English witch-finders, from local magistrates to the professional witch-hunter Matthew Hopkins, searched accused bodies for the “witch’s mark” — a blemish, spot, or supernumerary nipple from which imps supposedly suckled.

Hopkins, working in East Anglia during the 1640s at the height of the Civil War disruption, was the most systematic practitioner of this search. Along with his associate John Stearne, he presided over the largest single witch-hunting campaign in English history, executing over two hundred people across Essex, Suffolk, Norfolk, Cambridgeshire, Bedfordshire, and Huntingdonshire. His 1647 pamphlet, The Discovery of Witches, defends his methods and describes his procedure for detecting familiars: keeping the accused awake through the night, since only when at rest could a witch summon her imps, who would come to feed at the marks and make themselves visible to observers.

Hopkins pricked any skin deformity on the accused with a pin; a mark that proved insensible was taken as proof of a devil’s brand. In the 1644 Manningtree cases — his earliest prosecutions — he documented the named familiars of the accused, among them a creature called Pyewacket. Hopkins’ remark about this name is telling: it was, he noted, a name “no mortal could invent.” By 1644, the named familiar had become a category of evidence in its own right.

The witch’s mark appears throughout the trial records at scale. In the 1582 St. Osyth trials in Essex, Ursley Kempe’s four familiars were said to suck blood directly; Margerie Sammon’s took the form of toads. Agnes Glascock’s body was found to bear “a few marks like well-sucked spots.” The examination of these marks was a formal step in the prosecutorial process, carried out by female searchers appointed by the court.

Cats, Toads, Dogs, and Mice

The creatures named in English trial records are notably ordinary. Cats dominate — not the sleek black familiar of later popular imagery, but spotted cats, grey cats, white cats, cats with human voices. Toads appear throughout; they were associated across European folk tradition with poison, with transformation, and with the underworld. Dogs, rabbits, ferrets, mice, and in several cases creatures described as “things like chickens” complete the bestiary.

What makes these creatures peculiar in the trial context is not their bodies but their names. English familiars had names: Tyffin, Pygine, Jacke, Margot, Grissell, Greedigut, Vinegar Tom. The register is intimate — the same register as domestic animals. A cat that answered to a name, that was fed and sheltered, that was companion as much as creature, became, under prosecutorial pressure, a diabolical agent. The feeding relationship was what the 1604 statute specifically sought to criminalize: not the animal as such, but the ongoing covenant of nourishment that implied a binding between witch and devil.

Familiars were also, in the trial record, regularly inherited rather than simply acquired. Most of the time they were described as initiating contact themselves: choosing a person, appearing at a moment of distress or need, and offering a bargain. In the Chelmsford trials Sathan approached Joan Waterhouse and offered his services unprompted. All the familiars in the 1612 Lancaster trials similarly “appeared” before the person of their choosing. This pattern — the spirit that seeks out the practitioner rather than being summoned — recurs in the cunning-folk accounts, but in very different terms.

Cunning Folk and Fairy Familiars

The animal imps of the English witch trials are one strand. Beneath them, and probably prior to them, runs a different tradition: the fairy familiar of the cunning folk.

Cunning folk — healers, diviners, finders of lost property and strayed cattle — were the ordinary magical service providers of early modern Britain. They were not typically prosecuted for harm-working. They were consulted. Their magic was practical: locate the stolen horse, identify the person who cursed the cattle, prescribe the herb for the sick child. A significant number of them, when questioned under trial or examination, described working with familiar spirits they identified not as devils but as fairies, or as the dead.

Bessie Dunlop of Ayrshire, tried in Edinburgh in 1576, described her familiar as Thom Reid — the ghost of a man who had died at the Battle of Pinkie in 1547. Thom appeared to her, instructed her in healing remedies and in the locations of lost property, and introduced her to “the good neighbours” — the fairy court. She insisted, consistently under examination, that she had never harmed anyone and that her knowledge came from a benevolent source. She was found guilty of sorcery and executed by strangulation. Wilby opens Cunning Folk and Familiar Spirits with Dunlop’s trial transcript, and the choice is deliberate: Dunlop is not a case of the diabolical familiar, but of the fairy helper that predates and underlies it.

Alison Pearson of Byrehill, tried in 1588, described being led into the fairy court by a mysterious kinsman and receiving medical knowledge there. Joan Tyrrye of Somerset maintained that the fairies had given her only good and godly powers and instructed her in which herbs would counter witchcraft. Andro Man of Aberdeen, whose trial records date to around 1597, claimed a decades-long relationship with the Queen of Elfame and the fairy court — he spoke of them with easy familiarity, as neighbors rather than supernatural beings.

Isobel Gowdie of Nairn, whose four voluntary confessions in 1662 are among the most detailed documents of Scottish folk belief ever produced, described the Queen of Elfame as a fine woman dressed in white linen who provided more food than Gowdie could eat, and claimed to have entered the fairy hills and received instruction there. By Gowdie’s time, the fairy familiar had been largely absorbed into diabolical frameworks under prosecutorial pressure: her confessions weave the fairy court together with explicit devil-worship and sabbat attendance in ways that probably reflect a century of accumulated compression between popular fairy belief and elite demonological expectation.

Wilby’s Argument

The conventional historical reading of familiar-spirit confessions, until the early twenty-first century, was broadly dismissive: these were prosecutorial fictions, naive attributions of personality to ordinary animals, or products of coercion and sleep deprivation. Emma Wilby’s Cunning Folk and Familiar Spirits (2005) argued for something more substantial, and her argument has reshaped the scholarship.

Wilby’s first point is distributional. The cunning folk — who were typically being consulted rather than prosecuted for crime — described familiars spontaneously and in consistent terms across decades and widely separated geographies. The detail and internal coherence of accounts like Bessie Dunlop’s cannot be explained as a prosecutorial template, because Dunlop’s interrogators were not following a template for fairy familiars; they were trying to force her experience into the category of devil-worship, which she resisted. The fairy-familiar tradition was something she brought to the trial, not something the trial imposed on her.

Wilby’s second point is comparative. The descriptions of fairy familiars map closely onto the spirit helper found across shamanic traditions in tribal societies worldwide: a distinct personality with its own agenda, encountered in visionary or liminal states, capable of supplying practical knowledge — healing, divination, recovery of lost things. The encounter typically happens at a moment of personal crisis; the spirit initiates contact; the relationship involves ongoing reciprocity. These features appear in Dunlop, in Pearson, in Gowdie, and in dozens of other cases Wilby examines.

Her third point concerns what happened to this tradition under prosecution. The fairy familiar was not invented by witch-trial interrogators. It was transformed by them. The spirit helper became the devil’s imp. The food shared with a familiar became the witch’s blood given to a diabolical creature. The cunning woman’s helper became the witch’s servant in a pact. The legal category of “feeding and rewarding evil spirits” gave the transformation statutory force.

This is not an argument that all familiar-spirit confessions record genuine supernatural experience. It is an argument that the raw material from which those confessions were constructed was real popular belief — not invented by interrogators, but processed, pressured, and reinterpreted by them into something their legal and theological frameworks could handle.

What Made It English

The confluence of factors that made the animal familiar specifically English — as opposed to a feature of continental witch prosecution — is worth setting out plainly.

Continental demonology was primarily an elite intellectual project that worked downward: the learned sabbat template was imposed on accused persons, often under torture, producing confessions that matched the template. The animal familiar, with its domestic intimacy and its named, individual personality, fits poorly into the sabbat schema. It belongs to a different register of belief entirely — local, practical, relational, rooted in folklore about household spirits and fairy neighbors.

English common-law procedure, without systematic judicial torture, allowed popular belief to surface in confessions more or less on its own terms. The result was a distinctive trial record: full of animals with names, covenants sealed with drops of blood, and creatures that chose their witches rather than being summoned. This record looks strange against the continental background because it is drawing on a different source.

Ronald Hutton, placing English witchcraft in global comparative context, notes that the English trial record shows features paralleled in shamanic traditions far outside Europe — not because of direct connection, but because both are drawing on broadly similar patterns of folk belief about spirit helpers, altered states, and practical magical knowledge. The English familiar is less a variant of continental demonology than a local crystallization of something much older.

Reading Across the Layers

A practitioner working with the concept of the familiar today is working with at least three layers of historical sediment, and clarity about which layer is which changes how the concept operates.

The oldest and least fully documented layer is the fairy helper of the cunning-folk tradition — a spirit encountered in visionary or liminal states, associated with the fairy world and with the dead, capable of practical assistance with healing and divination. This is Wilby’s recovered tradition; it is the one with the deepest folkloric roots and the one most thoroughly dismantled by the witch trials.

Above it sits the animal familiar of the English witch trials: the creature in a pot under the bed, the toad fed milk in a bowl, the cat with a human voice and a name already given. This layer is heavily shaped by legal pressure. It is also the layer that produced the most vivid textual record, because trials generated documents in ways that folk practice did not.

At the top is the modern practitioner’s familiar: the spirit companion in meditation, the animal totem, the beloved pet reimagined as a co-worker in ritual. This layer draws on both earlier traditions and on Romantic-era and twentieth-century occultism, and it is largely free of the blood-covenant that made the earlier concept legally dangerous.

All three layers are available for study and for practice. None of them is the whole truth about what a familiar is. The most honest starting point is to know which one you are holding.

Cross-references on this site: cunning folk (forthcoming), cat correspondence (forthcoming), toad correspondence (forthcoming).