A “witch bottle” is not a single standardized charm. Archaeologists use the name for certain concealed glass or stoneware vessels, especially seventeenth-century examples whose contents and placement suggest a response to feared witchcraft. The Holywell find in London is valuable because its burial context was recorded.
- Evidence type
- Concealed vessel
- Period
- Mainly seventeenth century
- Anchor record
- MOLA · Holywell bottle, Shoreditch
A bottle with a context
During excavation near Shoreditch High Street, Museum of London Archaeology found a stoneware bottle beneath the floor of an eighteenth-century building. The bottle itself was dated about 1670–1710. Its placement was deliberate: tiles in a basement doorway had been lifted, a hole dug, the vessel set upright, and the floor relaid. The broken vessel held around sixty bent copper-alloy pins, rusty nails, and a possible piece of wood or bone. The surrounding floor and contents together make a stronger case for purposeful deposition than a loose bottle in an antique shop. Yet the find does not identify the person who placed it, the precise words said, or the result they expected.
Protection, cure, or something else?
MOLA’s wider Bottles Concealed and Revealed project treats “witch bottle” as a research category to be tested, not a complete explanation stamped on every vessel. Surviving examples occur in hearths, under floors, and in other archaeological settings, including ditches and churchyards. Pins and nails are common, while some cases include material such as nail clippings or liquid remains. A vessel may have been understood as protection or a prepared cure against bewitchment. Those intentions are inferred from deposits, texts, and later stories; the archaeology alone cannot tell us that every owner held identical beliefs. The project’s broad survey exists because older accounts have often simplified this varied evidence.
Keep the museum bottle separate
The Pitt Rivers Museum has a silvered glass flask registered as 1926.6.1, accompanied by a story that it held a witch who would be released if opened. Its recorded twentieth-century collection history differs from the excavated Holywell vessel. Both are popularly called witch bottles, but their forms, contexts, and evidence differ. One was recovered during an archaeological investigation; the other carries a narrated tradition recorded through a collector. A sensational label cannot replace either object’s find context and accession history. Neither record verifies that a person or spirit was trapped in a bottle.