The Repository Stone Record

Chester Castle's Lost Bridge Was Under the Prison

September 19, 2026 · Runtime 10:50 · Watch on YouTube

In July 2026 an archaeologist walked to the end of a dark corridor under Chester Crown Court and, by the light of his phone, found almost three metres of medieval wall — part of a thirteenth-century bridge over the castle's inner ditch, in a place every history says Thomas Harrison cleared away. Nobody dug for it. But the paper never lost that bridge.

Chapters

  1. 0:00 A wall at the end of a corridor
  2. 0:46 What Harrison built, 1788–1836
  3. 2:18 The find, and how it was read
  4. 3:01 The bridge on the Pipe Roll, 1159–60
  5. 4:18 De Lavaux's plan, 1745
  6. 5:15 How the wall got its date
  7. 6:37 One gate, two dates
  8. 7:00 The Beeston parallel
  9. 8:07 What 'demolished' meant
  10. 8:48 A correction: 1804, not the 1700s
  11. 9:54 What the stone adds

Transcript

A wall at the end of a corridor 0:00

In July this year, an archaeologist walked to the end of a dark corridor under Chester Crown Court, in what is left of an eighteenth-century prison. By the light of his phone he saw a wall that did not match the rest. Almost three metres of medieval masonry, in a place every history of the castle says was cleared away two hundred years ago.

The wall is real, it is medieval, and nobody dug for it. But it is not quite the surprise it was reported as. The paper never lost that bridge. It paid for it in eleven fifty-nine, repaired it in thirteen thirty-seven, condemned it in sixteen twenty-four, and drew it in seventeen forty-five. What the stone adds is the one thing paper cannot: it is still there.

What Harrison built, 1788–1836 0:46

The complex no longer looks like a castle. Those are the words of the county history of Cheshire. What stands on the site is a courthouse, a barracks and a gateway of Greek columns, built by Thomas Harrison between seventeen eighty-eight and eighteen twenty-two. Pevsner called it one of the most powerful monuments of the Greek Revival in the whole of England.

What Harrison found in seventeen eighty-eight was a working medieval castle. An inner ward on the mound, an outer ward with a great hall, an exchequer, a shire hall and a gaol. And between the two wards, a ditch with a bridge across it. The prison reformer John Howard had condemned the gaol. The county wanted a new one.

So the clean version reads like a demolition schedule. In seventeen eighty-eight an Act of Parliament set up the works, and Harrison began by pulling down the exchequer and the constable's house. The prison came first, finished by seventeen ninety-one. In seventeen eighty-nine came consent to remove the outer gatehouse, and by seventeen ninety-three the old shire hall was gone.

In eighteen oh-four the county got consent for a new armoury, and that meant demolishing the inner gatehouse, the Square Tower and part of the inner curtain wall. The Greek gateway went up between eighteen ten and eighteen thirteen. And the county history closes the account in eighteen thirty-six: all that remained of the ancient castle was the Agricola Tower and two much altered towers.

The find, and how it was read 2:18

Then, this summer, the University of Chester announced what Professor Stewart Ainsworth had found at the end of that corridor. The prison walls are typical late eighteenth-century masonry, he said. This was different. It turned a right angle and disappeared into a gap too narrow to squeeze into.

The team scanned it with laser scanners and set the result against old maps and plans. The wall lined up with a bridge across the deep ditch around the inner ward, the bridge that served a fortified gateway added in the thirteenth century. They put it in the twelve sixties to twelve eighties. No trench was opened. The date came from comparison.

The bridge on the Pipe Roll, 1159–60 3:01

Here is the first problem with calling it a discovery. The bridge had never gone missing on paper. In eleven fifty-nine the young earl of Chester was a minor, and the king's men rendered account for his lands. Their entry survives on the Pipe Roll, and it itemises the castle.

In the works of the castle of Chester, one hundred pounds and forty-seven shillings. And a separate line: in the works of the castle bridge, twenty pounds. The first bridge at Chester Castle is not a find. It is a payment, written down eight hundred and sixty-seven years ago.

The line keeps running. In twelve thirty-eight a record mentions the keep approached by a bridge. In thirteen thirty-seven, when over a hundred yards of wall had to be rebuilt, the repairs included the bridges leading to the two gatehouses. Two gatehouses, two bridges, one for each ward.

In sixteen twenty-four the county justices had the castle surveyed, and the survey found the bridge into the castle so dangerous as to be unusable. The Crown's officers argued that the county should pay. Then they paid themselves, reluctantly, and in sixteen twenty-seven a new bridge went in.

De Lavaux's plan, 1745 4:18

And in seventeen forty-five, with the Jacobite rising coming south, the Earl of Cholmondeley had a military engineer draw the castle. Alexander de Lavaux proposed four bastions that were never built. But he also drew what was standing, and he drew it with care.

On his plan the inner ward sits on the mound. Beside Agricola's Tower is a gate with two round towers. In front of it a hatched ditch curves round the whole ward, and across the ditch runs a bridge, with a covered way on either side. His key names it: the bridge built within the castle.

That is the plan Harrison had in front of him forty years later. The bridge was drawn, named and standing when his commission began. The county history records the gate it led to being demolished for the armoury. It never says the bridge was. Nobody asked what happened to the bottom of the ditch.

How the wall got its date 5:15

The second problem is the date, and it is worth being exact about where it comes from. A laser scanner measures shape. It does not measure age. The wall was called medieval because its stones and its coursing match the Norman parts that survive. It was called a bridge because its position matches the plans.

And it was given a date because of the gate it served. The team says that gate was added in the twelve sixties to twelve eighties, when a heavily fortified gateway replaced the entrance through Agricola's Tower. The county history dates the same gate differently.

Its account runs from the exchequer records. Between twelve eighty-four and twelve ninety-one, Edward the First spent over fourteen hundred pounds on the castle. In twelve ninety-two and ninety-three a new outer gatehouse cost over three hundred and eighteen pounds. Twin drum towers, a vaulted passage, two portcullises and a prison, under a mason named William of Marlow.

Either then, the history says, or a little earlier, a new inner gatehouse was built west of the Agricola Tower, which was blocked up and given a new staircase. In thirteen oh-one its chapel became a treasury, and the wall paintings were covered in limewash. They stayed covered until the nineteen nineties.

One gate, two dates 6:37

So the gate has two dates. The twelve sixties to eighties, from the survey team. Twelve ninety-three or a little earlier, from the county history. Around thirty years apart, and neither one was measured on the stone. The wall's date is an argument from paper, and the papers disagree. This episode does not choose.

The Beeston parallel 7:00

The team points to Beeston, fifteen kilometres away, as the surviving parallel. Beeston was built in the twelve twenties by Earl Ranulf, who most likely walled Chester's inner ward in stone as well. Its ditch is cut into the rock up to nine metres deep, and the bridge across it was probably timber, resting partly on a pillar of rock left standing when the ditch was quarried.

What visitors see under Beeston's modern bridge, English Heritage says, is a jumble of masonry from a causeway built when the entrance was remodelled in thirteen oh-three. Before it, a drawbridge that could be raised against the twin-towered gate. Something similar, the announcement says, probably once stood at Chester.

Now look at what does stand. Agricola's Tower, from about twelve ten, with its gate blocked and its chapel upstairs. The Flag Tower on the site of the first keep. The Half Moon Tower. All three altered, refaced and stripped of buttresses over three centuries. Above ground, the county history was right.

What 'demolished' meant 8:07

But demolished, in a county history, means demolished down to the ground you can see. Harrison was not clearing a park. He was building a prison and a courthouse on a site with a ditch through the middle of it. A ditch is a foundation problem. A three-metre wall in the bottom of it is not an obstacle. It is a footing.

The prison went in behind the shire hall, against the inner ward, between seventeen eighty-eight and seventeen ninety-one. The bridge wall became the end of a corridor. It was not removed, because nothing needed it removed. Then the corridor went dark for two hundred years.

A correction: 1804, not the 1700s 8:48

Last time, we said this part of the castle was believed cleared in the seventeen hundreds. That was the belief. The date on paper is later: the inner gatehouse came down for the armoury consented in eighteen oh-four, sixteen years into Harrison's rebuilding. The wall under the court outlived the gate above it by two centuries.

So what does the stone add? One thing, and it matters. The paper says a bridge was paid for, mended, condemned, replaced and drawn. The paper cannot say whether any of it survived. The wall says it did. And Ainsworth's guess is that much more survives under the car park and the castle complex.

It is worth saying why anyone was down there. The survey was not looking for a castle. It was a project to record historic stonework digitally, so that theft and damage could be prosecuted, run with Cheshire Constabulary. Seventy-two police officers were trained at the castle this year. The bridge was a by-product.

What the stone adds 9:54

A castle that becomes a prison and a courthouse is not removed. It is floored over. The paper kept Chester's bridge for six hundred years, from a line on a pipe roll to a hatched curve on an engineer's plan. The stone kept it for two hundred more, in the dark, at the end of a corridor.

Next time you read that a castle was demolished, ask two questions. What was built on top, and what did that building need for a footing? The answer is usually the bit that is still there.

Next time, the colour of a castle. Every engraving in this episode shows grey stone. In twelve forty, Henry the Third ordered the great tower of London whitened inside and out, and at Conwy the lime render still clings to the walls in patches. The bare stone castle is a ruin's colour, not a castle's.

Description and sources

The Pipe Roll of 1159–60 itemises it: £100 and 47s. on the works of Chester Castle, and a separate £20 on the works of the castle bridge. In 1337 the repairs included the bridges to the two gatehouses; a survey of 1624 found the bridge into the castle so dangerous as to be unusable, and a new one went in in 1627–8. In 1745 Alexander de Lavaux drew the inner ward, its ditch and the bridge for the Earl of Cholmondeley — and named it in his key: 'the Bridge built within the Castle'. That is the plan Harrison had in front of him in 1788.

The second problem is the date. A laser scanner measures shape, not age: the wall is 'medieval' by comparison with the Agricola Tower, 'a bridge' by comparison with the plans, and dated by the gate it served. The survey team says 1260–1280s. The county history dates the new inner gatehouse to 1292–3 'or a little earlier', the year William of Marlow built the outer gatehouse for over £318. Around thirty years apart, neither measured on the stone — this episode does not choose.

What the wall adds is the one thing paper cannot say: it survived. 'Demolished', in a county history, means demolished down to the ground you can see. Harrison was building a prison and a courthouse over a ditch, and a three-metre wall in the bottom of a ditch is not an obstacle — it is a footing. The gatehouse above came down for the armoury of 1804 — not the 1700s, as our last episode said; the wall below outlived it by two centuries.

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.

... and 16 more plates, each credited on screen in the video and listed in this project's manifest.

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

Every image is public-domain or CC BY material from Wikimedia Commons, credited on screen — nothing is AI-generated. Four frames are cropped details of the 1745 de Lavaux plan and the 1784 Grose plates, from the originals at full resolution. The diagrams and set-type quotations are code: drawings of the argument, not measured surveys.

#castles #archaeology #medievalhistory