Where Was Reading's Lost Castle?

In Reading, we have Castle Street and Castle Hill - we even have a Castle Crescent. But we don't have a castle. 

At least not one we can see. 

Did one ever exist in the town, and if so, where was it? 

Photo by Linus Sandvide on Unsplash

Full disclaimer: I burned my dinner thinking about the mystery of Reading's lost castle, it will definitely melt your brain and leave you wondering about this backwards and forwards for days. 
You're welcome.  

We have a few pieces of evidence that there was once a castle in Reading. One is quoted by the early nineteenth century historian John Man (1816), who cites Robert Beccensis, a manuscript in the Bodleian Library: 

"... in the preceding year, the followers of duke Henry, who resided at Wallingford, had not only destroyed the camp at Britwell, which had long infested their quarters, but also the castle, which King Stephen, against all right, had erected near the abbey of Reading."

Asser, in his Life of King Alfred (893), tells us a castle existed here in Saxon times, which may indicate that we're actually looking for two castles:

"...King Aethelred and his brother Alfred, uniting their forces and assembling an army, marched to Reading, where, on their arrival at the castle gate, they cut to pieces and overthrew the heathen whom they found outside the fortifications." 

And John Leland, writing in 1535-1543, tells us that:

"There was a castelle in the Saxons tyme in this towne: and the name of Castelle-Streat yet remaynithe, lying from est to west to passe to Newbyri: but I could not perceive or clerely lerne wher it stoode. But by al lykelihod at the west-ende of the Castelle-Streat: and, as sum think, about the place of execution. It is very likely that a peace of the abbay was buildid of the mines of it. Peraventure it stoode whe thabbay was."

So. If there was a castle, where is it now? And if there was a castle in Alfred's time and one built by King Stephen in the 1100s, are we talking about two castles? With either of them leaving no trace? We know from John Man that Henry II's men demolished at least one of these castles in around 1153, but you'd at least expect some consistency about where it stood. Surely? 

Welcome to the mind-bending mystery of the site of Reading's castle. Or castles. 

Here are the contenders. 

The Reading Abbey Grounds

In 1816 historian John Man said he had no doubt that the castle once stood within or just outside the grounds of the abbey. He even identifies its remains: a tall ruin, with battlements and arrow-holes cut into it, next to the river. He offers us a drawing, showing it as one tall square building, set away from the rest of the abbey which he tells us - somewhat imaginatively - once contained towers, arrow holes that point upwards and battlements on top - although he admits he can't actually see the battlements for the ivy that has grown over it. (1) His only further proof is that castles were usually built near rivers and on higher ground. He also concludes that the stones from the castle and the abbey ruins looked different in his view and were therefore originally completely different structures. 

And then it dawned on me.  Could this drawing in fact depict the abbey's Rere Dorter, or Neccessarium? In other words, the monks' toilet block? This was traditionally built next to, or even straddling a river to ensure waste was flushed away. Other buildings had tall, square Rere Dorters, such as Castle Acre in Norfolk. Southampton Friary had a similar one, described on a city plaque as 'a tall square tower, mounting two guns in time of war' built in the fourteenth century. They might have been dual purpose, accounting for Man's arrow holes, but in any case there's a big leap from a toilet block to a fully defendable, military castle.  

Jamieson B Hurry's plan of the abbey from 1906 does show the Rere Dorter right next to the river and away from the rest of the abbey. (2) Interestingly, Hurry worked on the abbey's history extensively, even mapping out its plans - and never mentions the ruins of a castle. If not the Rere Dorter, Man's 'castle' could have been part of the Dormitory, where the monks slept, which is shown on the plan near the spot. And call me cynical but I'm not convinced by the drawing. The hand sketches what the artist wants us to see, and perspective can be easily distorted to fit an argument.   

Some excitement about the castle's site has been centred around the mound in The Forbury. That this was the site of a twelfth century castle was was debunked in 2017 when it was investigated and concluded that, due to archaeological evidence, it could not have dated from before the thirteenth and fourteenth centuries and was most likely a rubbish pit dating from the sixteenth century or even part of the Civil War defences of the seventeenth. 

In my view, the evidence just doesn't support a castle here near the abbey. John Man's ruin wasn't properly identified, the evidence is ambiguous and at the mound in particular the archaeology just doesn't fit. There are also no other precedents for a defensive castle being built inside or immediately next to a busy,  thriving abbey (if you do know of one from the 12th century, then let me know) and actually seems illogical when you consider their very different uses. 

And anyway, why name two adjoining streets after castles, when the castle was nowhere near it? 

Castle Street

Back in the sixteenth century, it seems as if the traveller John Leland pulled up on a horse and asked a few locals where the castle used to be. They likely then shrugged and pointed, vaguely, towards Castle Street. The name is first recorded in the mid-1200s, placing it well after the accepted date of the castle's destruction in 1152, but could have come from living memory of the event. There are also traces of Medieval buildings in this area. 'Lancet windows and round arches' were found when pulling down the old gaol in Castle Street in 1798, leaving the authors of Magna Brittanica in 1806 to conclude that there was 'little doubt' that these were the remains of the castle. (3) But were they? It makes a lot of sense from a security perspective to convert a secure, defensive building into a gaol. But perhaps these are the remains of another early building. Medieval archaeology, including a floor, ditch and pit, were found in a survey of Castle Street in 2001, with some remains possibly from the eleventh to the twelfth centuries which is exciting, but still doesn't confirm the definite existence of our castle, just some early Medieval activity here, which we already know about from historical sources. It just seems strange to me that we know about more mundane things in the centre of Reading like the bridge being built over the Hallowed Brook by 1186 and that there was a 'New Street' exiting the market place by 1165 - but there was nothing written down about something so important as a large defensive castle here in the heart of Medieval Reading, amidst the markets, church, shops and houses. (4) 

Castle Hill

Walk up Castle Street, past the police station roundabout (which didn't exist of course in the 1200s) and you'll end up at Castle Hill. On place name basis, this makes a lot more sense. Castles were built on hills and the site does stand on higher ground. And other similarly named streets lead to castles, and don't necessarily mark their actual site, like Castle Lane in Donnington, near Newbury. This site would fit Leland's description of being out towards the west of the town, on the road to Newbury (today's modern A4). However, Leland was researching the castle around four hundred years after its demolition. No local would have had living memory of the structure, but this place name evidence is encouraging. 

What is disappointing then is that an archaeological survey carried out in part of Castle Hill in 1998 found no evidence of Medieval activity. But is this really such a surprise? In 2001 Slade put forward the building of Stephen's castle (although says it was in the Forbury) at 1150, and its demolition in 1152. (3) If the castle was here and not next to the abbey, would we really expect to find much archaeological evidence of its use, being away from the thriving Medieval town centre and only standing for a maximum two years? Slade does agree that an earlier castle probably did exist on Castle Hill - but  doesn't it make sense that if you're hurriedly building a fortification in the twelfth-century that the site of an existing castle would be a good place to build it? I would argue that the two castles - Saxon and Medieval - were built on the same spot, one over the other - and that the second castle was never finished. This accounts for the lack of archaeological and written evidence left behind and explains how two castles named in Reading have left no trace: one was destroyed and built on by the other, and the new castle (and any surviving Saxon work) was demolished before the new building (or, in this case, renovation) was finished. And finally - if John Man's Medieval scribe located the castle 'near the abbey of Reading', wouldn't that still fit, relatively, for Castle Hill? It would only have been a 10-15 minute walk away. In Medieval terms, that's pretty much next door. 

So. Where was the castle? 

I would love the idea that there was a twelfth century castle, with flagged Medieval turrets, and battlements rising up in the mist, alongside the towering stone abbey. But I just can't find the evidence for it. Stephen had nothing  to gain by locating it here in the centre of the town, next to its most important and prestigious religious building and there's nothing similar elsewhere. And the abbey ruins survive today. Isn't it strange that Stephen's castle, built for defence and a show of strength and intimidation during the Anarchy, hasn't left a single definite trace, on ground or on paper?

Castle Street's evidence, especially of the lancet windows and arches being found at the old goal, is encouraging but not definitive evidence of an actual castle here. And if the castle was on Castle Street, then why didn't Stephen make use of the more advantageous higher ground of what is now Castle Hill, just yards away?  

Castle Hill, however, sounds like it would make more sense. It would have been sited on the entrance to the town, from Newbury for defence. It was on higher ground, and just far enough away from the Medieval town to be an effective military barrier without disrupting the homes and livelihoods of the townspeople. There's archaeological evidence for twelfth and thirteenth century activity  a few minutes' walk, down the road, in Castle Street, and back then, geographically the two roads would have in fact been one and not separate as they are today. The location given by the writer of the Beccensis manuscript also fits. The lack of Medieval archaeology here could just be down to the site being used only very temporarily - and maybe not even finished. A Medieval castle built on the remains of an older Saxon castle here fits both Asser's description and Leland's. Stephen's castle in Reading just simply never got the chance to grow into the sprawling hive of buildings, apartments and inns that we see at Windsor or Stansted Mountfitchet in Essex. No pottery, no coins or rubbish pits - and no archives relating to castle deeds, royal visits, grants, employees and expenditure. 

But all isn't lost. It's perfectly possible that the secrets of Reading Castle still lie silently underneath our shops, pubs and houses in the town and the archaeologists just haven't found them yet. What do you think? 

What do you think? Do you reckon you know where it was? Or do you think Reading never had a castle in the first place? Let me know in the comments below. 

Reading's castle was long gone by the time the Wars of the the Roses erupted in the fifteenth century. I explore the women impacted by this conflict in my book Forgotten Women of the Wars of the Roses, published by Pen and Sword. I also look at some local women with links to the wars. Order your copy here. 

Never want to miss a post? Subscribe to my newsletter here: 

Notes and Sources:

(1) Man, J. (1816). The history and antiquities, ancient and modern, of the Borough of Reading, in the County of Berks. Reading: Printed by Snare and Man. Accessed 4 November 2020. page 176. 
(2) Hurry, J. B. (Jamieson Boyd). (1906). The rise and fall of Reading Abbey. London: E. Stock. pp88-89. Accessed 4 November 2020. 
(3) Gatehouse Gazetteer. Accessed 4 November 2020. 
(4) *affiliate link*: Dils, Joan. Reading: A History. Carnegie Publishing, 2019. p14-15. 



  1. A great article and absolutely fascinating. Saxon castles were rare - mostly the only stone buildings they built were churches, although there were a few examples of stone-built fortifications. So if there had been one here, it might have been a fortified wooden construction, easily built over/demolished. Either way, I'm intrigued by the later one - and as you say, so many unanswered questions!!

    1. Thanks Annie! Interesting about the wooden castles, thanks for that bit of info - adds a little more context to the mystery :)

  2. Just musing over your interesting article - I think a possible site for the / a castle would be ironically at the current Berkshire records office site.
    It is sited at the top of Castle hill, was previously used as a training facility for the Berkshire Yeomanry the previous building being called Yeomany House - originally the house was called Castle Hill house though. The road curving behind it is called Castle crescent. It would seem to me worth further investigation- although I am sure others must have explored this before. Regards Andy Q

    1. Hi Andy, thanks, this makes so much sense. It's a really good theory, and yes that would be a good site for a castle. Wouldn't it be great if we could find some archaeological evidence for it - maybe one day!

  3. I found your article fascinating! I'm afraid I don't have any ideas to offer but just wanted to tell you how impressive I found your arguments. Do you know if there are any other cases like this, where there are certain things that suggest there was once a castle in an area but no one has ever managed to provide conclusive proof of this? I was just thinking that if there are other cases then we could maybe argue that it perhaps isn't that surprising if physical evidence of a castle's existence has disappeared over the centuries. It seems that one of the questions we are interested in is how likely is that if there was a castle in a particular location there would still be physical evidence of that fact today? Please forgive me if I'm completely wrong about this but I guess to me it seems that if a stone building from, say, 800 years ago had just been left to deteriorate naturally then parts of it would still remain today unless people had deliberately tried to destroy it (for example to build something else in place of it). I don't know how easy it would have been to remove all trace of a castle! I guess that if something else now occupies the same land then archaeologists wouldn't be able to investigate properly so maybe a castle could remain hidden even if highly skilled archaeologists and historians were looking for it! It would be interesting to know if there have been any discoveries of previously unknown castles - castles that no one knew had ever existed. I live quite near London and I sometimes wonder how much history has been lost through building work and redevelopment. I guess that some parts of London have changed so much that people who lived only a century ago and knew the city as it was back then simply wouldn't recognise it today.

    Thank you for firing my historical imagination!


Post a Comment