In many ways, Norman priest Geoffrey of Monmouth is the founder of the story of King Arthur. Three centuries before Thomas Malory's Morte d'Arthur, Geoffrey's Historia Regum Britanniae (aka De Gestis Britonum) lays out the familiar story from Merlin to Uther to Arthur to Mordred, the wars against the Saxons and the brief victory he achieved - though the fact that Arthur goes off and conquers half of Europe before marching on Imperial Rome might come as something of a surprise.

But Geoffrey wasn't the first to tell tales of the British king who won a space of peace against the Saxon invasion, and I want to take a quick walk back through the earlier stories this island produced, and maybe find out where they all started.

The Life of Gildas was written by a Welsh cleric, who was apparently a friend of Geoffrey: Geoffrey actually mentions him in the Historia as writing a sequel to his work. The Life is purportedly a history of a certain 5th century monk, Gildas, of whom we will hear more later.

The Life presents Arthur as king of all Britain, much like Geoffrey's story, but does not present him sympathetically: Arthur is raided by Gildas' brother, "an active warrior and most distinguished soldier", "the high-spirited youth", "the victorious and excellent youth"; when Arthur kills him for his repeated attacks, it is "murder", and Gildas is praised for forgiving him for it. Later, "the tyrant Arthur" beseiges Glastonbury, because his queen has been carried away there; again, he is written as being completely in the wrong for this.

The Life of Saint Cadog was also by a Welsh cleric; there's some disagreement on whether this was Caradoc again, or a contemporary named Lifris. Personally, given the same town and the same weird hate shown towards Arthur, I suspect Caradoc's hand.

This Life launches straight in with the Arthur hatred: "Arthur immediately very inflamed with lust in desire for the maiden, and filled with evil thoughts, said to his companions, 'Know that I am vehemently inflamed with concupiscence for this girl, whom that soldier is carrying away on horseback.'" He doesn't actually do anything - he actually defends the maiden from attackers, and she turns out to be Saint Cadog's mother - but the writer wanted us to know that Arthur was a bad'un from the start.

Cadog himself later has an encounter with Arthur, who is now evil for trying to avenge three of his soldiers being killed. Cadog summons a court, which judges that Arthur should receive fine oxen as weregild for his soldiers, but Cadog isn't even satisfied with this: he, or God, transforms the cows into ferns to prevent Arthur having them. Seeing this wonder, Arthur repents for... some reason.

The Llancarfan Lives really stand out in Arthurian history, because they are so anti-Arthur. All the earlier sources either praise him or are neutral, and of course afterwards we get Geoffrey and eventually Malory. I think Caradoc just didn't like the guy.

The tale of Culhwch and Olwen is a Welsh prose tale. The first written version appears around 1325, and estimated dates of composition range from the 11th century to the mid-12th (ie, the same time as Geoffrey). Whatever the date, it is probably the first prose tale of Arthur.

The Arthur of Culhwch ac Olwen is a powerful British king, like Geoffrey's and Caradoc's versions, but inhabits a world of magic: all of his warriors have magical powers, such as Kai who can hold his breath for nine days, or Gilla Coes Hydd who can leap 300 acres, or Menw son of Teirgwaedd who can cast illusions over them all. Rather than fighting Saxons, Arthur's aid is sought in claiming a giant's daughter from her father.

One thing Culhwch ac Olwen seems to have given Geoffrey is the idea of Arthur going to war beyond Britain. Arthur's gatekeeper talks about his many battles at Arthur's side, "in Europe, and in Africa, and in the islands of Corsica... and when thou didst conquer Greece in the East". This idea doesn't seem to appear in other, earlier tales, so perhaps this is where Arthur's march on Rome originated.

There are a number of Welsh poems mentioning Arthur dating from the 2-3 centuries prior to Geoffrey's work. Many of them are attributed to Taliesin, who supposedly lived in the 6th century, but are currently believed to be from the 9th century onwards.

In Pa gur yv y porthaur?, quoted above, Arthur praises his warriors to a gatekeeper. All of the warriors are of superlative skill, and all have killed dozens or hundreds of enemies: "Vain was an army, compared to Cei in battle." In Preiddeu Annwfn, Arthur is "the dux bellorum, the lord of the land, the king", and sails in his ship Prydwen to attack the castles of the Otherworld. Marwnat Uthyr Pendragon (The Death-Song of Uther Pendragon) references Arthur's valour, while Kadeir Teyrnon (The Chair of the Prince) name-drops "Arthur the Blessed".

These poems are part of the same tradition as Culhwch ac Olwen, and seem to have influenced Caradoc's Lives more directly than Culhwch did. The relative dating is uncertain, but it does feel like the poems tend to be earlier: the deeds of the warriors are more grounded, and Culhwch's tale often seems to draw on, modify, and expand the poems, rather than the reverse: the extravagant battle-deeds of Arthur's warriors in Pa gur yv y porthaur? become ludicrous magical deeds and powers in Culhwch ac Olwen, which actually features the same named gatekeeper - now working for Arthur rather than challenging him.

Moving away from Welsh tales of magic, the Annales Cambriae (Welsh Annals) is a simple list of events, mostly the deaths of kings, over 500 years, from 453 to 954, which is presumed to be about the date they were originally written. I say "presumed" because the earliest manuscript dates from around 1100. Either way, it certainly predates Geoffrey's book.

The Annals represent a different tradition to the Welsh poems and tales above. Depending on the dates of the poems, the two strands may have been intertwined; but where the poems are interested in setting Arthur as a king in a world of heroes, this other tradition makes him first and foremost a war-leader.

The Annals preserve two mentions of Arthur: the first in 516, when he wins the Battle of Badon, and the second in 537, when he dies alongside Mordred (Medraut) at Camlann. There's no evidence that the older Annals preserve a written tradition actually going back that far: most likely, like Geoffrey, the annalist simply collected dates from the sources he had available.

The Historia Brittonum (History of the Britons) is dubiously attributed to Welsh monk Nennius, on the basis tha some versions have a foreword from him. For our purposes the actual author is irrelevant: we only care about the date and the text.

The Historia records Arthur as a war-leader: he is said to have won twelve battles for the Britons, with the twelfth being at "the hill of Badon. In this engagement, nine hundred and forty fell by his hand alone, no one but the Lord affording him assistance." A story which the Annales assigned to Badon Hill, where Arthur carries a Christian symbol on his shoulders, is assigned by the Historia to the eighth battle, Gurnion Castle.

The Historia also tells the story of the king who first invited the Saxons to England: Vortigern. Specifically, it gives the tale of his troubles building a certain tower; his wise men say that he needs to sacrifice a boy with no father to make the foundations secure, but when he finds such a boy, he shows that the actual problem is a pair of dragons fighting deep underground. Geoffrey of Monmouth uses this same story, naming the boy Merlin; but Nennius calls him Ambrose. Ambrose is the son of a Roman consul, and is granted the rule of western Britain and the city Cair Guorthegirn by Vortigern; later, shortly before describing Arthur's victories, the Historia adds that Vortigern's third son Pascent "reigned in the two provinces Builth and Guorthegirnaim, after the death of his father. These were granted him by Ambrosius, who was the great king among the kings of Britain."

The Historia ecclesiastica gentis Anglorum (Ecclesiastical History of the English People) was written by the Venerable Bede, an English monk from Northumbria. I think he's the only Saxon source on this list; all the others are Welsh/British (or, in Geoffrey's case, Norman).

Bede doesn't mention the name Arthur, but he does talk about Vortigern Ambrosius, who we just heard about from Nennius. There's no story about towers and dragons, but Ambrosius is still of Roman noble descent, and he is still the leader of the British. After Vortigern invites the Saxons into Britain and unleashes them on Britain, Ambrosius rises up and leads the fight back against them.

There is no string of famous victories here, like those attributed to Arthur: instead "from that day, sometimes the natives, and sometimes their enemies, prevailed". The only named battle is Badon Hill, "where they made no small slaughter of those invaders", 44 years after the Saxons first landed. After this, Bede says simply that "there was some respite from foreign, but not from civil, war."

Y Gododdin is a bit of a break from the mainline of histories: it's a poem, or rather an elegy, to a group of warriors who fell in battle. Its date of composition is widely disputed: it could be early 7th century, or 9th century, or as late as the early 11th century. It certainly predates Geoffrey of Monmouth, but could belong further up this page.

Y Gododdin contains only one reference to Arthur, in passing, while praising the warrior Gwawrddur, who while mighty, "was no Arthur". This would fit with the depiction of Arthur in Nennius, which has him killing hundreds of enemies single-handed: however good Gwawrddur was, he would never match up to Arthur, a century or more earlier.

If the early, 7th century date is right, Y Gododdin is probably the first mention of Arthur's name. If the text is 9th century or later, it's likely just a reference to Nennius. We simply don't know.

De Excidio et Conquestu Britanniae (On the Ruin and Conquest of Britain) was written by the British monk Gildas. His life story is quite vague - he may have been born anywhere between Strathclyde and Chester, and dates for De Excidio range from 483 to 545 - but we do know when he was born: at the time of the Battle of Badon Hill (here "Bath-hill").

Gildas' comments on Ambrosius Aurelianus (again, he does not mention any Arthur) are largely the same as Bede's: Ambrosius' parents are not yet "of the royal race", but more ambiguously "for their merit were adorned with the purple", and "his progeny in these our days" are still waging war against the Saxons, "although shamefully degenerated from the worthiness of our ancestors". But Ambrosius' modesty and leadership, the battle of Badon Hill, the end of foreign wars but continuing civil strife - all of these are present, in much the same words as Bede's.

There's a reason for this, and it's a fairly obvious one: Bede copied Gildas. He actually attributes some text directly to Gildas and names him, which none of these people ever do. The one big change is one of time: for Gildas, the time of peace post-Badon is still ongoing: "our foreign wars having ceased, but our civil troubles still remaining". For Bede, those days are two centuries of invasion in the past.

Was there a historical Arthur? Based on the sources, I think that's the wrong question. There does seem to have been a historical Ambrosius, of Roman heritage, who led (some of) the Britons for a while, and won a battle against the Saxons at Badon Hill, bringing peace to Britain for a time. He may have had a war-leader under him, who may have been named Arthur and been famous for his valour.

Perhaps this Arthur was even a king or prince in his own right - Nennius' claim that "there were many more noble than himself" doesn't rule out him being noble himself. Or perhaps he was only a warrior and leader of warriors, and the crowns and thrones were built up by the Welsh poets, and made text by Geoffrey of Monmouth.

Or perhaps he never existed at all.