Something that many Arthurian researchers are interested in is discovering where it was that Arthur fought his battles. Normally, his twelve battles against the Saxons (as recorded in the Historia Brittonum) are the subject of investigation, but there is another major battle that Arthur is recorded as having. This is the Battle of Camlann, first recorded in the Annales Cambriae. Although not stated clearly in this earliest mention, later tradition claims that it was the climactic battle of a civil war between Arthur and his nephew Mordred.
Of course, just as with the twelve battles against the Saxons, Arthurian enthusiasts are very keen to discover where this final battle took place. There have been numerous suggestions over the years, but one of them is particularly popular: Camboglanna. Let us have a look at what we know about this site historically, and what reasons there are for believing that it might have been the true site of Camlann, as well as potential flaws in the theory.
The fort is about 400 by 400 feet, unsurprisingly being a square shape. But what is quite surprising is its position relative to Hadrian’s Wall. While most of the other forts are positioned right up against the Wall, the fort of Camboglanna is over 1000 feet south of the Wall. This reason for this is believed to be that the Wall was built to run along the easiest route across the Cam Beck, while the fort was built where the terrain would offer the greatest strategic advantage.
There is not much of the fort left to see now, because in the 18th century much of it was destroyed to make way for landscaping being done around Castlesteads House. Ornamental gardens now cover most of it. In fact, in addition to being levelled to make way for these ornamental gardens, much of the fort was torn apart to be used as material to build a wall around the grounds. Furthermore, the north west side of the fort has been eroded by the Cam Beck gorge. So all things considered, there is really very little left to see of this fort.
Fortunately, during the landscaping, many inscribed and sculptured stones were found and preserved, many of which are now kept in the summer house on the grounds.
History of the Fort
Not much is known about the history of the site, since there are not many written records describing it. But obviously it must have been constructed around the time Hadrian’s Wall was built, since it was one of the forts devoted to defending that border. Hadrian’s Wall was built in the 120s, and the forts were added a few years later, so that indicates that Camboglanna was constructed at some point around the year 130 C.E.
The inscribed stones discovered at the site, mentioned earlier, also enlighten us about the history of the fort. In the second century, evidently quite soon if not immediately after the fort was completed, it was occupied by the Cohors I Batavorum – a cohort that had been involved in the initial invasion of Britain and had most likely been involved in the famous Battle of Mons Graupius in 83 C.E. And in fact, beyond just occupying this site, it seems that this cohort was involved in the actual construction of the fort.
After this cohort left, Camboglanna was occupied by the Cohors IV Gallorum equitata, which had been transferred all the way from Thrace to Britain. This cohort was later moved to Vindolanda, another fort along Hadrian’s Wall. Another inscription was found among the Roman ruins of Camboglanna which testifies to the Cohors II Tungrarum being stationed at the fort in the third century. It is impossible to know if this cohort directly succeeded the Cohors IV Gallorum at the site or whether there were other cohorts stationed at Camboglanna between them, because the history of the site is so fragmentary.
Very little is known about the subsequent history of this fort. It is unknown when it was finally abandoned by the Romans, but it seems that at the latest, it would have been around the beginning of the fifth century, when Constantine III withdrew the majority of the Roman troops from the island.
However, despite being abandoned by the Romans, there is very slight evidence for it still being occupied in some sense in the Arthurian period. An inscribed stone was found at Camboglanna that is believed to date from some time between the mid-fifth century to the end of the sixth century. The interpretation of the stone has proved difficult, but many scholars consider it to be a reference to the Christian God, thus indicating not only a presence of some kind at the site in the Arthurian period, but specifically a Christian presence.
Was it Camlann?
Now that we have examined the state and history of this Roman fort, lets us examine the question of whether or not Arthur’s final battle of Camlann took place here.
It is easy to see why this is an appealing theory. Firstly, the name is an obvious reason to associate this fort with the Arthurian site. Consider how similar the two names are:
Granted, the first name is longer than the second, but the first name is much older, allowing it to have developed over time into the shorter ‘Camlann’. For example, consider how the personal name ‘Urbgen’ evolved into ‘Urien’, with the ‘b’ disappearing completely and the ‘g’ evolving into an ‘i’.
Similarly, the Roman name ‘Ambrosius’ evolved into ‘Emrys’, again providing an example of the ‘b’ disappearing (this was a regular occurrence whenever a ‘b’ came after an ‘m’). So when the ‘b’ is removed from ‘Camboglanna’ (which results in ‘Camoglanna’), the similarities are more evident.
Additional shortenings and distortions over time could have further evolved the name into something like ‘Camglanna’ or ‘Camlanna’, or perhaps even ‘Camlann’.
In fact, the meaning of ‘Camboglanna’ is identical to one suggested etymology for the Arthurian ‘Camlann’. The etymology of the Roman fort is quite simple, being composed of the elements ‘cambo’ (meaning ‘crooked’ or ‘bent’) and ‘glanna’ (meaning ‘bank’, possibly in the sense of a river bank). This exact etymology has been regularly proposed for the Arthurian ‘Camlann’, meaning that we can say with certainty that the name of this Roman fort appears to work very well.
And what about the location? Is the location of Camboglanna compatible with what the sources tell us about Camlann? Well, in some ways it is, but in other ways it isn’t. Let us first consider how it does not fit the sources.
Geoffrey of Monmouth provides us with the earliest surviving description of the Battle of Camlann. He placed it in Cornwall, and it is generally believed that the location he was describing was Camelford. Most subsequent sources place Camlann in that general area, the south west of Britain. Needless to say, Camboglanna does not fit this information. So that is one indication that it is not the right place.
However, it must be acknowledged that Geoffrey of Monmouth’s geography was not always accurate. In fact, there are many examples of it being inaccurate throughout the Historia Regum Britanniae. Therefore, just because he placed the Battle of Camlann at Camelford, this does not mean that we must believe this. In fact, it could easily be the case that there was never any tradition of the battle taking place there, and Geoffrey simply chose that location because of the similarity between the word ‘Camlann’ and the name of the River Camel that runs by Camelford.
Regarding the fact that most subsequent sources also place the Battle of Camlann in the same general part of the country, this is worth taking into consideration, but it is not conclusive. After all, Geoffrey of Monmouth’s HRB is widely recognised as having been incredibly influential. Many of the Arthurian texts written after the HRB clearly contain details and even entire stories that are not seen in previous works, prior to the HRB.
Thus, it could easily be the case that the general placement of Camlann in the south west in most subsequent tales is simply due to most people following on the heels of Geoffrey of Monmouth, as we see with numerous other details and stories.
On the other hand, is there anything to actively support Camboglanna’s northern location for Camlann? There certainly is. Recall that Arthur’s opponent at the battle was said to have been Mordred (admittedly, this is not clearly stated in the earliest surviving source, but it is certainly what all later tradition claims and it is compatible with the earliest surviving source). Mordred came from a dynasty based in the north of Britain. His father was Lleudun, king of Lothian (a neighbouring kingdom to the powerful northern kingdom of Rheged).
Therefore, Mordred would have lived in the north of Britain, possibly in what is now Scotland. We do not know exactly where he lived, but suffice it to say that it was in the same general part of the country as Camboglanna, although probably somewhat further north.
Consider the events that led up to the battle. According to Geoffrey of Monmouth, Arthur was away from Britain when Mordred decided to rebel. He then gathered his forces and attempted to land on the shores of the island, but he was opposed by Mordred and his army. Nonetheless, Arthur overcame his enemy’s forces and caused him to flee. According to the HRB, after fleeing quite some distance, Arthur eventually caught up to Mordred at Camlann in Cornwall (but, as we have seen, this placement could easily be Geoffrey’s own mistake).
While the details in the account are by no means trustworthy, the general overview of what allegedly happened is worth considering. Supposedly, after unsuccessfully attempting to prevent Arthur getting back on the island, Mordred retreated until Arthur finally caught up with him, and then the Battle of Camlann ensued. This is the general outline of events. It is very simple and plausible. If the rightful king had returned and already beaten Mordred in battle, it would only be logical for him to attempt to flee back to his own land.
With this in mind, it would not at all be surprising to find Camlann somewhere in the vicinity of Lothian, Mordred’s home territory. More specifically, it would be logical for Camlann to be somewhere south of Lothian, if Arthur had arrived in Britain somewhere in the south (where he then fought Mordred). Thus, Mordred would have been travelling on a route that would have taken him from that southern location back to Lothian, meaning that the site of his final battle with Arthur should be somewhere along that route.
Camboglanna fits that information perfectly. Lothian was situated north of Hadrian’s Wall, meaning that it was, of course, north of Camboglanna. It cannot be definitively stated that Mordred’s route would have taken him past this particular Roman fort, because we simply do not know where exactly his shoreline battle with Arthur took place. Nonetheless, using this basic information and logic, it does seem like a very attractive location for the battle.
In summary, the name of this fort seems to be a good fit for ‘Camlann’, and the location also seems to match the available information.
However, that is not all. There is another piece of evidence that this might be the true location of Arthur’s final battle. Starting from Geoffrey of Monmouth’s account and continuing into subsequent works, Arthur is described as being taken away to Avalon after being wounded at the Battle of Camlann. There have been countless attempts to identify the location of Avalon, just as there have been for Camlann itself.
A seemingly logical assumption is that Avalon, where Arthur was taken after his last battle, would have been fairly near the site of said battle. The logic is simple and attractive. Therefore, if there is a location near Camboglanna that could be convincingly identified as Avalon, this would seriously strengthen the case for Camboglanna being Camlann.
Further along Hadrian’s Wall is another fort with a very interesting name. Just two forts along from Camboglanna, we find a fort called ‘Aballava’. The similarity to ‘Avalon’ is already evident, but in the seventh century Ravenna Cosmography, the name of this fort is spelt ‘Avalana’ – almost completely identical to ‘Avalon’.
For many researchers, this fort of Aballava, or Avalana, is the obvious option for Avalon. This being the case, a location nearby being Camlann would make a good deal of sense. Therefore, Camboglanna would seem like a very sensible suggestion for Camlann, in conjunction with all the reasons already considered.
Flaws in the Theory
However, that is not to say that there are not problems with this identification. There are in fact a number of problems, which we will now consider.
Firstly, the name is not a perfect fit for ‘Camlann’. It is true, as stated above, that it could have produced ‘Camlann’ (and probably would have eventually) and some scholars do indeed derive the name of the Arthurian battle site from the elements ‘cambo’ and ‘glanna’. But, nonetheless, it is recognised that there are problems with this etymology.
Primarily, this etymology does not seem to be compatible with the fact that the 10th century Annales Cambriae spells the name ‘Camlann’. By this time, the ‘g’ would not have disappeared. ‘Camboglanna’ would have evolved into ‘Camglann’, not ‘Camlann’. Yet no source ever records it as having a ‘g’. It is true that this might be the case that it was simply a scribal error or a case of a scribe modernising the spelling (for no original 10th century manuscript of the Annales Cambriae survives). But, nonetheless, the most logical conclusion is that the name did not come from ‘Camboglanna’.
The favoured explanation among scholars is that ‘Camlann’ comes from the elements ‘cambo’ and ‘landa’ (meaning ‘enclosure’). The ‘b’ would disappear as in the ‘Camboglanna’ etymology, and it was also common for a ‘d’ to disappear when it came after an ‘n’. Thus, ‘Cambolanda’ would evolve into ‘Camlann’ without any difficulties. This origin best fits the evidence from the Annales Cambriae and all the other sources.
In addition, one line of evidence regarding where the Battle of Camlann took place involves Welsh records about certain individuals who survived the battle. These individuals, where they can be identified from other sources, were all associated with locations in north west Wales, not the north of Britain. Most intriguingly, there is an ‘Afon Gamlan’ in that very area of Wales (as well as a few other spots known by the name ‘Camlan’ in the same vicinity).
Given the fact that the individuals said to have survived the Battle of Camlann were all associated with locations surrounding Afon Gamlan in north west Wales, this strongly suggests that the Arthurian Camlann was there, not anywhere along Hadrian’s Wall.
Granted, this evidence uses Welsh tradition rather than more firmly historical sources. One might argue that this evidence can be dismissed, but it does not seem very wise to just dismiss tradition like that when it seems to converge on one particular point.
Finally, the connection between Avalon and Aballava is very tempting, but the available sources are quite certain that Avalon was an island. It is always described as such, even from its first appearance in the HRB. Thus, identifying it as a Roman fort along Hadrian’s Wall is not a very satisfactory explanation, despite the similarities of the names.
In conclusion, we can see that the Roman fort of Camboglanna has a reasonable amount going for it as an Arthurian site. It was built in the early Roman period of Britain, so it definitely existed in the Arthurian period. There is also some evidence that it continued to be inhabited and even functioned as a Christian site in Arthur’s time. In addition, we know that it had the intriguing name ‘Camboglanna’ since Roman times, so it is definitely not a post-Arthurian place name. But was it Camlann, the site of Arthur’s infamous final battle?
As we have seen, there is some limited evidence that it was. It was in the right general vicinity according to what we would expect from Mordred if he was fleeing a southern Arthur and heading back to his home territory. It is also next to a site which bore a name that was incredibly similar to ‘Avalon’, which ties in with the logic that Avalon would have been close to the location of Arthur’s final battle.
And, finally, the name of the site is also a reasonably good fit. It could have eventually become ‘Camlann’. Yet, it is not a perfect fit. The name ‘Camlann’ is much better explained with a different etymology, and the name of this Roman fort does not seem to be compatible with the evidence from the earliest source mentioning Camlann.
In addition, we have seen that the nearby fort which supposedly matches Avalon was probably unrelated, since it was not an island and it is difficult to see how a Roman fort could have evolved into an island in the legends. Furthermore, evidence from Welsh tradition points toward a location in north west Wales, not the north of England, as the true location of Camlann.
So, in summary, the Roman fort of Camboglanna is not totally excluded from potential candidates for Camlann. But, at the same time, it is important to recognise the significant problems with the theory, and the fact that the case for this being Camlann has been greatly exaggerated. In reality, it is a possibility that it was Camlann, but not a very likely one.