Witchcraft: what it is, and the history that gets rewritten
Two claims travel with this subject constantly, and both are false: that it is an unbroken ancient religion, and that nine million women were burned.
9 min read · Published 16 August 2026
The answer
Witchcraft today covers a wide range of practices, from Wicca as an organised religion to eclectic solitary practice with no doctrine at all. The common thread is ritual intended to produce an effect, usually involving correspondence between objects, timing and intention. Two widespread historical claims don't hold. Modern Wicca isn't a surviving pre-Christian religion: it was assembled in England in the 1940s and 1950s by Gerald Gardner from ceremonial magic, folklore and Margaret Murray's theory of a witch-cult, which historians have since rejected. And the European witch trials did not kill nine million people; the figure accepted by historians is roughly 40,000 to 60,000 executions across about three centuries, with perhaps 80,000 tried. Both corrections matter, because the real history is grimmer and more specific than the myth that replaced it.
What people mean by the word
The term covers several quite different things, and conversations go wrong when they're conflated.
Wicca is a modern religion with initiatory lineages, a duotheistic structure of God and Goddess, seasonal festivals and, in traditional forms, oath-bound material. It has clergy, covens and a body of practice.
Eclectic or solitary witchcraft is a practice instead of a religion (spellwork, herbs, candles, moon timing) with no required beliefs and often no deities at all. Many practitioners are atheists who describe what they do as psychological instead of supernatural.
Traditional and folk practices (cunning folk, brujería, hoodoo, Appalachian granny magic) are older, culturally specific, and often held by their practitioners to be quite distinct from anything called witchcraft.
Witch as accusation is different from all of these, and historically the most consequential. It's something done to a person, not something they claim.
Wicca is younger than it says
Gerald Gardner published *Witchcraft Today* in 1954 and presented Wicca as the survival of an ancient pagan religion that had gone underground. The claim was central to its appeal, and it doesn't survive scrutiny.
Gardner drew on Margaret Murray's thesis that the witch trials had persecuted a real organised fertility cult. Murray's argument was influential in the first half of the twentieth century and is now rejected by historians, who found she had relied on confessions extracted under torture and had omitted the parts that contradicted her. The rituals themselves show the borrowing plainly, the Hermetic Order of the Golden Dawn, Aleister Crowley's writing, Freemasonry, and Charles Leland's *Aradia*.
This isn't a debunking that should end anyone's practice, and many Wiccans accept it comfortably. A religion isn't invalidated by having a known founder. Every religion had one, and most obscured him. But an invented tradition presented as an ancient survival is making a historical claim, and that claim is checkable and false.
Wicca is younger than my grandmother, and I don't think that makes it less of a religion.
The numbers, and why the myth persists
| Claim | What the record supports |
|---|---|
| Nine million killed | Roughly 40,000–60,000 executions; around 80,000 tried |
| All women | Around 75–80% women, varying sharply by region |
| Burned at the stake | Common in continental Europe; in England, hanging |
| The Middle Ages | Mostly 1450–1750, peaking in the early modern period |
| The Church driving it | Largely secular courts; the Spanish Inquisition was notably sceptical |
| Suppressing pagan survivals | No evidence the accused practised any organised pre-Christian religion |
The nine million figure traces to an eighteenth-century German estimate, extrapolated from one town's records and repeated ever since. It gained new life through twentieth-century feminist and neopagan writing, where it functioned as a founding atrocity.
The correction isn't a defence of the trials. Sixty thousand judicial killings, overwhelmingly of women, on evidence that couldn't possibly be true, is an atrocity that needs no inflation. And inflation costs something: a figure that's easy to refute makes the real history easier to wave away.
Who was accused
The pattern in the records is consistent, and it's a pattern about power instead of about belief.
The accused were disproportionately poor, old, widowed or unmarried women, people without a household to protect them and without economic value to the community. Midwives and healers appear, though less than folklore suggests. Quarrelsome neighbours appear often. So do women who had inherited property.
Accusations clustered in periods of stress: failed harvests, epidemics, war, and religious conflict where communities were already suspicious of each other. They spread through legal machinery that used torture and rewarded naming others, so trials arrived in waves rather than singly.
This is the part that stays relevant. Witch accusation is what happens when a community has a problem, a mechanism for blame, and someone with no protection.
What modern practice involves
Contemporary practice is largely correspondence and timing. Herbs, colours, stones, days of the week and moon phases are assigned meanings, and a working combines them towards an intention: a candle of a particular colour, dressed with a particular oil, lit at a particular phase.
The moon timing is the piece with a real structure underneath: waxing for building, waning for removing, full for culmination. The correspondence tables are largely inherited from Renaissance ceremonial magic and vary between sources, which practitioners generally treat as flexibility rather than as a problem.
A substantial share of practitioners describe the mechanism in psychological terms, ritual as a way of committing to an intention, focusing attention, and marking a decision with an act. That's a defensible account of what a ritual does, and it's compatible with the evidence in a way that claims of external causation aren't. It's also close to what this site says about vibrational therapy and the evil eye: keep the practice, decline the mechanism.
Five dates
| Year | Event | Why it counts |
|---|---|---|
| 1486 | The Malleus Maleficarum is printed | The handbook that made witchcraft a heresy to be prosecuted, reprinted for two centuries |
| 1563 | Johann Weyer publishes De praestigiis daemonum | A physician argues the accused are deluded or ill, the first sustained case against the trials |
| 1612 | The Pendle trials in Lancashire | Ten hanged on the testimony of a nine-year-old; the best documented English case |
| 1692 | Salem | Twenty killed, around 200 accused, in a colony that then repented in writing |
| 1782 | Anna Göldi executed in Glarus | The last legal execution for witchcraft in Europe; formally exonerated in 2008 |
Salem gets most of the attention in English, and it was small. The Würzburg and Bamberg trials in the 1620s each killed hundreds, including children and clergy, and the Trier trials of the 1580s and 1590s several hundred more. The American case is famous because it was written about, not because it was large.
The eight festivals, and where the names came from
Wicca and much eclectic practice keep a wheel of the year with eight festivals: the two solstices, the two equinoxes and the four cross-quarter days between them. Samhain at the end of October, Yule at midwinter, Imbolc at the start of February, Ostara at the spring equinox, Beltane at the start of May, Litha at midsummer, Lammas or Lughnasadh at the start of August, and Mabon at the autumn equinox.
Four of those names are old Irish or English festivals. Three are not. Ostara, Litha and Mabon were assigned to the equinoxes and midsummer by Aidan Kelly in 1974, who wanted names that sounded like the others. Ostara is from Bede's guess at an Anglo-Saxon goddess, Litha from Bede's name for the summer months, and Mabon from a figure in Welsh legend with no known connection to the autumn equinox. None of this stops the festivals being kept sincerely. It is another case of a recent structure presented as an old one, which is the theme of this entire subject.
What a beginner's practice usually involves
- An altar, which is a shelf with a candle, a cup, a dish of salt and whatever else has meaning. Its function is to make a place where the practice happens.
- Moon timing, waxing for bringing in, waning for clearing, dark for rest. The phases are real and the schedule gives a rhythm.
- Correspondences: a table of which herb, colour, stone and day go with which intention. The tables descend from Renaissance magic by way of twentieth-century compilations, and they disagree with each other freely.
- A written record, often called a book of shadows or a grimoire, in which workings and results are noted. This is the one item with an obvious benefit whatever you believe: people who write down what they intended and what happened learn something about themselves.
- Cleansing and grounding: salt water, smoke, a walk. Mostly ways of starting and finishing.
A practitioner who keeps the record faithfully will notice, over a year, which workings 'worked' and what else they did at the same time. That is the same test this site recommends for a birth chart, and it costs nothing.
Where it turns costly
Most of this is harmless and some of it's good for people, structure, agency, community, and a way to act when there's nothing practical to do.
The costs come in familiar shapes. Curse and hex work invites the fear discussed in hexes and curses, and the person most affected by believing in curses is usually the one who believes. Spellwork sold as a service, particularly love spells aimed at a specific named person, is one of the oldest fraud formats there is. And any practice that replaces medical treatment has crossed a line that no correspondence table justifies.
Then there's consent. Working on a named person who hasn't asked (to bind them, attract them, or change their mind) is doing something to someone rather than for yourself, whatever one believes about whether it functions.
The follow-ups
- Is Wicca an ancient religion?
- No. It was assembled in England in the 1940s and 1950s by Gerald Gardner, drawing on ceremonial magic, Freemasonry, Crowley and Charles Leland, together with Margaret Murray's theory of a surviving witch cult that historians have since rejected. Having a known founder doesn't invalidate a religion, but the claim of ancient survival is false.
- How many people died in the witch trials?
- Historians estimate roughly 40,000 to 60,000 executions across Europe between about 1450 and 1750, with around 80,000 people tried. The often repeated figure of nine million comes from an eighteenth century extrapolation from a single town's records and has no support in the archives.
- Were all the accused witches women?
- Around 75 to 80% were women, though this varied sharply by region, in Iceland and Estonia most of the accused were men. Those targeted were disproportionately poor, old, widowed or otherwise without household protection, and accusations clustered during harvest failures, epidemics and religious conflict.
- What is the difference between Wicca and witchcraft?
- Wicca is a specific modern religion with deities, festivals and initiatory lineages. Witchcraft is a broader practice that may involve no religion at all, and many practitioners are atheists who describe what they do in psychological terms. All Wiccans practise witchcraft in some form; most witches aren't Wiccan.
- Do witchcraft accusations still happen?
- Yes, and they still cause serious harm. Killings, mutilation and expulsion following witchcraft accusations are documented in parts of sub-Saharan Africa, Papua New Guinea and rural India, with elderly women and children most at risk. Several governments and UN bodies treat it as an active human rights issue.
- Are the Wiccan sabbat names ancient?
- Samhain, Imbolc, Beltane and Lughnasadh are old Irish festivals and Yule is old English. Ostara, Litha and Mabon were assigned to the equinoxes and midsummer by Aidan Kelly in 1974 to match the others.
- How big were the Salem witch trials?
- Twenty people were executed and around two hundred accused in 1692 and 1693. That is small beside the German trials of the 1620s at Würzburg and Bamberg, which each killed hundreds. Salem is famous because it was documented and repented in writing.
Salimah Ibrahim