The Repository Stone Record

Murder Holes Were Not for Boiling Oil

August 26, 2026 · Runtime 9:18 · Watch on YouTube

There is a row of holes cut straight through the vaulted stone ceiling of a gate passage in this video, and every guidebook tells you the same thing about what went through them: boiling oil, poured onto whoever was trapped below. Almost none of that is what the sources actually say — oil was expensive, needed twice the heat of water to pour, and the documentary record has exactly one confirmed case of it ever being used in a real siege. And the castle in this video may be the best proof of all: it had exactly one moment where its own murder holes might have mattered, and it is not even recorded whether that moment happened.

Chapters

  1. 0:00 The hole everyone thinks they understand
  2. 0:55 What a murder hole actually is
  3. 2:21 The scene from every film
  4. 3:14 Water boils at 100°C. Oil needs double that.
  5. 4:16 What the opening was actually for
  6. 4:41 Back to Bodiam's real portcullis
  7. 5:10 1483: the one close call
  8. 6:01 Never used, not even once
  9. 7:21 The transferable rule
  10. 8:50 Next: Château Gaillard, 1204

Transcript

The hole everyone thinks they understand 0:00

There is a stone ceiling in this video with a row of round holes cut straight through it, and it belongs to a castle this channel has already shown you once, for a completely different argument. Every guidebook, every documentary, every film with a castle under attack tells the exact same story about what went through holes like these: defenders standing over them with a cauldron, pouring boiling oil onto whoever was trapped in the passage below. It is one of the most repeated images in all of medieval architecture. And almost none of it is what the sources actually say.

By the end of this video, you will know what genuinely went through openings like this one, why the oil version barely holds up under basic physics, and — more specifically — you will have a real, sourced reason to suspect that the holes in this particular castle may never have poured anything on anyone at all, not even once, in the six hundred and forty years this building has stood.

What a murder hole actually is 0:55

Rule one, same as it always is on this channel: locate yourself before you trust the legend. A murder hole is not just any hole in a ceiling. It sits in the vaulted stone roof of a gate passage, at the exact chokepoint between the outer gate and whatever defence comes next — the one narrow stretch every attacker is physically forced to stand under, whether they like it or not, to get anywhere further into the building.

Quick reminder, if this is the first video of the channel you are watching: this is Bodiam Castle, in Sussex, licensed in thirteen eighty-five. It was built by Edward Dallingridge, the king's own coastal defence commissioner — a man who spent the same years professionally deciding which walls would actually stop a French landing. Not a fantasy new-money landlord playing at war. Someone who understood siege architecture for a living.

So where does the boiling oil version even come from? Not from the fourteenth century. Most of what the public imagines about castle combat was fixed later — nineteenth-century Gothic revival illustration, then twentieth-century film, both of which needed a single dramatic image that reads instantly on a page or a screen. A steaming cauldron and a scream is cinema. A cold calculation about the price of oil is not. The legend survived because it is a better story, not because anyone checked it against a siege account.

The scene from every film 2:21

Here is the scene everyone already has in their head, whether they have ever visited a castle or not. A defender stands over the hole with a cauldron, timed perfectly to the attacker's position, and tips boiling oil straight down. The attacker screams, falls back, and the siege is broken by one well-placed pour. It is a genuinely great scene. It is also, according to the actual documentary record, almost entirely fiction.

Problem one is money, and it is the simplest problem of all. Oil, in the fourteenth century, was an expensive commodity, not a battlefield consumable. No castle garrison kept barrels of it sitting in reserve for a siege that might never come — that is capital doing absolutely nothing, for years at a stretch, on the off chance that an army eventually shows up outside the gate.

Water boils at 100°C. Oil needs double that. 3:14

Problem two is the one that actually closes the case, and it is pure physics. Water boils at one hundred degrees Celsius. Oil needs roughly double that heat — around two hundred and four degrees — to reach the same pouring point. That difference means more fuel burned, more time spent waiting over a fire during an active attack, and the genuinely dangerous job of carrying a near-boiling cauldron of oil up a narrow, uneven stone stair while people are trying to kill you at the other end of it.

And the written record backs this up directly, not just in theory. Historians researching medieval sieges have found exactly one confirmed reference to boiling oil actually being used in combat — at the siege of Orleans, in fourteen twenty-eight. One. Set against that single case, the sources mention boiling water, molten lead, and heated sand far more often — all of them cheaper, all of them already sitting in a castle kitchen or forge, needing no special stockpile at all.

What the opening was actually for 4:16

And the hole itself was never only a weapon in the first place. It doubled as an observation point, letting a defender look straight down and see exactly who was standing in the passage below before committing to anything. And if the gate itself ever caught fire — wood does, especially under a determined attack — the same opening became the fastest way to pour water straight down onto the flames from directly above them.

Back to Bodiam's real portcullis 4:41

Now come back to Bodiam's own gatehouse, because this is where the argument stops being general. The holes are real, cut into the ribbed stone vault of the passage, exactly where the architecture predicts an attack would be stopped. And the gate underneath them is not a Victorian replica built for atmosphere. It is the original portcullis, iron-clad oak, from thirteen eighty-five — one of only a small handful of medieval portcullises still surviving anywhere in Britain.

1483: the one close call 5:10

This castle had exactly one moment in its entire history where any of that engineering might have mattered for real. November, fourteen eighty-three: Buckingham's rebellion against Richard the Third collapses across the south of England, and the owner of Bodiam, Sir Thomas Lewknor, takes refuge behind his own walls while royal forces — his own uncle among them, on the other side — are authorised to raise men and besiege him.

Except it is not recorded whether that siege actually went ahead. The best-supported historical reading, from the castle's own documented history, is that Lewknor surrendered with little resistance. No chronicle describes fighting in the gate passage. No account mentions the murder holes at all. Not recorded means exactly that, and nothing more — not recorded.

Never used, not even once 6:01

Which means it is entirely plausible — not proven, but plausible on the actual evidence — that the murder holes in this specific castle never dropped anything on anyone. Not oil, not water, not sand, not a single stone, across the full six hundred and forty years this building has stood on its island in the moat. The whole mechanism stayed a threat. It was never once used as a tool.

And that has to be said with the right amount of care, because this channel does not turn silence into proof. Absence of a record is not proof of zero use across six centuries — it is only absence of a record. Nobody in the fifteenth century bothered to write down a day when nothing happened at the gate. The only thing genuinely missing here is someone writing down a day when something did.

And the position of the hole is not automatic, either — which is the detail most guidebooks skip entirely. At other documented sites, like Audley's Castle and Tully Castle in Northern Ireland, openings very similar to this one sit not over a main gate at all, but over an ordinary interior doorway, leading to an ordinary interior room. Nobody was fighting a war through that one. It was almost certainly just a way to talk, or pass something, between one floor and the next.

The transferable rule 7:21

Here, then, is the transferable rule, and it is the whole point of this episode. To tell whether a murder hole was a real weapon or something far gentler, do not trust the shape of the hole at all. Look at where it sits. Over the chokepoint of a main gatehouse passage, with a portcullis and a defensible approach beneath it, it is genuine military architecture, full stop. Over an ordinary interior doorway, it is probably just a floor talking to a floor. Position decides the verdict. The hole itself decides nothing.

None of this makes the mechanism a joke, or pure decoration, for what it's worth. It was engineered by people planning seriously for a real war — the very same coastal commissioner this channel already showed you fortifying Rye in exactly these years. The engineering behind these holes was completely sound and completely intentional, even though, at this particular gate, it was apparently never actually put to the test.

The film version of this story is simpler than the real one, and considerably less interesting. What sits above your head in that passage was never built for a horror movie. It was built to drown a rush, to smother a fire, to watch a chokepoint in silence, and only rarely, and expensively, to burn — a complete and carefully costed mechanism, engineered for a war that, at this specific gate, never actually arrived to test it.

Next: Château Gaillard, 1204 8:50

Next time, a different castle, and a different argument about what the stone actually shows versus what the written record claims. In twelve-oh-four, a chronicle insists the attackers of Chateau Gaillard broke in through a chapel window, high above the ground. The modern archaeological reading points somewhere far less dignified — a latrine shaft, cut low through the rock — and the argument over which route is actually correct is still not settled.

Description and sources

This channel already showed you Bodiam Castle once, in the first episode, to argue its weak defences were partly theatre. This time the target is one detail in its gatehouse: the murder holes in the vaulted ceiling of the gate passage, exactly where an attacker would have to stand.

The popular story is boiling oil. The sources tell a cheaper, less cinematic one: oil was too expensive to stockpile, needed roughly double the heat of water to pour, and the record has exactly one confirmed case of it used in a real siege, at Orleans in 1428 — against far more mentions of boiling water, molten lead and heated sand. The openings doubled as observation points and as a way to douse a burning gate, which most retellings skip.

Then the episode turns back to Bodiam. Its portcullis is not a replica — it is the original iron-clad oak gate from 1385, one of only a handful surviving in Britain — and the murder holes above it are real. But Bodiam's one brush with a real siege, in November 1483 during Buckingham's rebellion, is not even recorded as having gone ahead; the best-supported reading is that its owner surrendered without much resistance. This specific, real, engineered killing mechanism may never have been used, not once, in six hundred and forty years.

PRINT-READY, FROM THIS CHANNEL

How to Read a Castle, Vol. 01 — the method, the measured drawings and the documents behind the first two episodes

https://therepository.gumroad.com/l/read-a-castle

PRIMARY SOURCES

ARCHIVE PLATES — public-domain documents, with author, date, archive and licence.

  • VanBeem — 2008 | Wikimedia Commons | Public domain

... and 5 more plates, each credited on screen in the video and listed in this project's manifest.

Stone Record reads castles as documents — what the surviving paperwork and the standing masonry actually say, against what gets repeated about them. Primary sources for this episode are linked above.

The historical images in this video are public-domain or CC0 material, reused from the first episode of this channel: an 1737 Buck engraving of Pevensey Castle, an 1737 Buck/Yale engraving of Herstmonceux Castle, a public-domain photograph of Bodiam Castle (VanBeem, 2008), a CC0 interior photograph of Bodiam's tower (Poliphilo, 2015), and a public-domain 1905 ground plan of the castle. Four illustrations — the south elevation, the distant approach, the port gate, and the hearth scene that opens the myth — are AI-generated with FLUX.1-schnell in this channel's survey-drawing style; three are reused from the first episode and one, the hearth, is new for this episode. The passage sequence, the two branching diagrams and the two timelines are original technical drawings made for this channel from the sources cited above — not photographs, and not AI-generated images.

#castles #medievalhistory #bodiamcastle #archaeology #mythbusting