Geoffrey of Monmouth claimed his History of the Kings of Britain was simply a translation of a much older book. No version of an older book with the same title or contents has survived, and for a variety of reasons, most historians agree Geoffrey was lying. However, it is clear that he had access to older, Welsh or Breton sources which he drew on when composing his book. He gets too many of his details right, or at least, close enough to other written sources for him to have concocted it entirely on his own.

There is no "ancient book" with the title The History of the Kings of Britain. But we do have a book titled The History of the Britons, or the "Historia Brittonum." Several versions of it have survived, but its full text is mostly contained in a collection in the British Library labeled Harleian 3859. It is usually attributed to a ninth–century monk named Nennius. Nennius' writing is wildly chaotic, with little to no organization whatsoever, and trying to piece together the information contained in its pages is a project of its own. Yet, this unrestrained neglect to orderliness lends an air of credibility. Nennius clearly has not altered any of his material, because he has made no attempt to compose it into any kind of orderly narrative. In his own words, he has simply taken a heap of everything he has found, and put it all together.

Different versions of the Historia Brittonum are wildly different from one another, which indicates that some of them may have been altered by later copyists. Even if they were not, the book contains far too many accounts of miraculous events to be taken seriously. It is not a reliable book as far as genuine history is concerned, but it is at least different enough from Geoffrey that it can be taken as some kind of representation of older, Welsh folklore.

In Nennius, we find the account of Vortigern's tumbling fortress and of the battling dragons which is featured in Geoffrey. This time, however, the boy prophet is named Ambrosius. In Nennius' version, Ambrosius' mother does deny ever having had intercourse with a mortal man, but there is no mention of an incubus. In the end, it is revealed that Ambrosius' father had, in fact, been a Roman consul.

Unlike his predecessors, Gildas and Bede, Nennius does, however, mention Arthur as being a separate figure from Ambrosius Aurelianus. About the Saxons, he writes:

"Then Arthur fought against them in those days, together with the kings of the British; but he was their leader in battle.

The first battle was at the mouth of the river called Glein. The second, the third, the fourth, and the fifth were on another river called the Douglas, which is in the country of Lindsey. The sixth battle was on the river called Bassas. The seventh battle was in Celyddon Forest, that is, the Battle of Celyddon Coed. The eighth battle was in Guinnion fort, and in it Arthur carried the image of the holy Mary, the everlasting Virgin, on his [shield,] and the heathen were put to flight on that day, and there was a great slaughter upon them, through the power of Our Lord Jesus Christ and the power of the holy Virgin Mary, his mother. The ninth battle was fought in the city of the Legion. The tenth battle was fought on the bank of the river called Tryfrwyd. The eleventh battle was on the hill called Agned. The twelfth battle was on Badon Hill and in it nine hundred and sixty men fell in one day, from a single charge of Arthur's, and no one laid them low save he alone; and he was victorious in all his campaigns."
(Nennius, 1980, pp. 35)

Clearly, much about this record is fanciful. It is difficult to believe that a single charge from one person could kill 960 men. Yet, this is one of the few records we have of Arthur which gives us something to work with. Unlike other early writings about him, here, we have a record which tells us specific battles that, according to Welsh tradition, Arthur is said to have fought.

For this reason, historians have done their best to work with this passage to see what information can be taken from it. Another interesting point of note is that the words which have been translated as leader in battle are dux bellorum in the original Latin. This literally means "duke of waging war," and more closely resembles titles like "commander" or "general" than "king."

Another interesting point of note is that the part about carrying the Virgin Mary's image on his shield is a correction by Nennius' modern translator, John Morris. The original text reads shoulders. The Welsh word for "shoulders", iscuid,, is only one letter off from iscuit, the word for "shield." What's more, these two words exchanged meaning in different localities. If a medieval scribe accidentally copied the word incorrectly, or if Nennius was writing in a region in which the word iscuid, took on the meaning "shield," then this may only mean that Arthur had an image of the Virgin Mary painted onto his shield.

Similarly, the part about slaying 960 men might only be a reference to the number of men killed by Arthur's army. Military leaders were often credited with the deeds of their armies.

But this may just be wishful thinking, more than scholarly conjecture. Even if we find a way to explain how the passages about Arthur could be believable, the fact remains that the Historia Brittonum is far too fanciful to ever be taken as genuine history.

Different historians have tried to locate the sites of the battles mentioned by Nennius. As we have already seen, Badon is usually thought to have been in Badbury or Bath, but other suggestions have been offered as well, such as Buxton or Bardon Hill. In the 1950s, Professor Kenneth Jackson used linguistic evidence to identify four more of the battle sites: the river Glein is believed to be a river which is now called Glen. One river in present–day Lincolnshire has this name, and there is another in Northumberland. "Lindsey" in the translation above has been inserted by the translator; the original Latin uses the name Linnuis, which is probably the Lindsey district of Lincolnshire. Another candidate is the region now known as Lennox, which contains a river called Douglas. "Celyyddon Coed" was the Caledonian Wood. This forest once covered vast regions of what is now Scotland. The city of the Legion was once thought to have been Caerleon, but is now believed to have been Chester. "Tryfrwyd" has also been inserted by the translator. Jackson believed the original text, Tribruit, meant a stretch of beach beside the sea where three rivers come together. But it may also come from the Battle of Tryfrwyd, which an ancient Welsh poem associates with Arthur. This is usually placed near the Firth of Forth. Bassas, Guinnion fort, and Agned have all defied identification. But Geoffrey of Monmouth placed Agned in Edinburgh.

The fact that these battle sites are difficult to identify works in favor of the belief that Arthur was a real historical figure. If Nennius had concocted this tale on his own, he would have chosen locations that were known in his time. The fact that historians have difficulty identifying them indicates that he was probably making a faithful copy of an older, Welsh tradition, as he claims.

However, not all historians agree with this view. N. J. Higham believes that the Historia Brittonum was more carefully constructed than has generally been believed. In Higham's view, the Historia was written as both a British history, and a Christian one. When the book was written, Anglo–Saxon kings had already come to dominate much of the British Isle, and they were still continuing to encroach on more and more British land. Higham believes the Historia was written as a moral response to this, and that Arthur was created to fulfill a specific role within this conflict between British and Saxon. A number of historians have noted that Nennius' battle list does resemble a poem, and the lines in the original Welsh do rhyme. This does seem to indicate that the passage was artfully constructed, rather than written to be a genuine account of history.