Dating Arthur's Battles

Let's conclude this short run of posts about Arthur's final conflict with a consideration of the date of that battle.

To recap: in my more recent posts, I've suggested that "Camlann" was not a place-name.  Rather, it meant "Broken Sword".  The battle was remembered, not for where it was fought, but for its relevance: when Arthur's sword - and Arthur himself - was broken, the world of the Britons collapsed.

The place of the battle was, however, recalled in the popular memory.  Aneirin, writing about the battle almost immediately after it had happened, referred to a specific landmark - "the Alledd", or the Hill of Alyth in Perthshire.

Separately, the British refugees who fled to Brittany remembered that the battle had been fought around "Kerlouan"- the "Castle of Luan" - and they named a whole district of Brittany in honour of that terrible disaster.  St Luan, as we noted yesterday, is commemorated solely at Alyth in Perthshire, where this contemporary of Artuir mac Aedain was the local patron saint.

There is also the ancient tradition that Arthur's wife was held captive by the "Pictish" king Mordred at Barry Hill (above), which is adjacent to the Hill of Alyth, and that she was punished for her treachery and buried at Meigle, four miles to the south (and next to the spot where Arthur was betrayed).

Arthur hacked his way across the Mains (farmland) of Camno to Arthurbank, just south of the River Isla.  The place where he suffered his mortal wound was, until the 1790s, marked by a standing stone known as the Arthurstone.

So - when did this happen?

The Irish Annals of Tigernach indicate that "Artur" was one of four sons of Aedan mac Gabrain, King of the Scots, who died in a battle in Angus in AD 594 (another source, written a hundred years later, noted that the battle was fought against the southern Picts).  The Annals of Ulster date the same battle to 596 and tells us that it was not Arthur but Bran who died there (Bran is, of course, the name of the stricken lord and knight in the Breton poem, analysed in yesterday's blogpost; he was, in reality, Arthur).

Ah, but ... those who like to pretend that they know all that there is to know about Arthur invariably protest that Artuir mac Aedain (the first Arthur to appear in any historical records) can't have been the historical Arthur.  There must have been another Arthur, earlier and decidedly more "English", after whom the northern Arthur was named.

Going back a few blogposts, I endeavoured to explain why the entries in the Annals of Wales (Annales Cambriae) pertaining to Arthur were misleading.  The dates they give for his battles are wrong.  Or perhaps it would be more accurate to say that the dates appear to be wrong (they are, in fact, fairly accurate, but we need to adjust them to account for variations in the dating system being used).

The Annals of Wales include an entry for circa AD 664:

The first celebration of Easter among the Saxons.  The second battle of Badon.  Morgan dies.

As previously explained, the battle of Badon is one of the main sticking-points when it comes to identifying the real Arthur.  This "second" battle of Badon is seldom taken into consideration.  But it's important, because (as I show in The King Arthur Conspiracy) Arthur's main rival at his final battle was Morgan the Wealthy, who had been accepted by the southern Picts as their "Chief-Boar" or pseudo-king.

The entry in the Welsh Annals for the "second battle of Badon" and the death of Morgan appears to relate to Arthur's last battle.  But the given date is wrong.

What I suggested in a recent post is that the date given in the Welsh Annals for this long-forgotten battle was the result of confusion over the Easter cycle used by the early Church.  The calculations for when Easter would be celebrated each year were complex, but there was an 84-year cycle which the early Christians were able to follow.  In other words, rather than trying to figure out each year when Easter should fall, the early monasteries kept a table of the 84-year Easter cycle.  As each year passed, they just moved to the next date in their table.  The various annals of Wales, Ireland and elsewhere were eventually compiled using notes made in the margins of these Easter tables.  As each year passed, any major event of the previous year was entered alongside the relevant date in the Easter cycle.

But this system was prone to confusion.  The date of AD 664 for Arthur's "second battle of Badon" is wrong.  It was linked to the wrong Easter cycle.  Take away 84 years, and you arrive at AD 580: not the date of Arthur's last battle (at which Morgan died), but the date of another decisive battle fought by the real Arthur - his first battle of Badon.

Now, it's one thing to claim that a date relating to the military career of Artuir mac Aedain - the original "King Arthur" - got muddled thanks to confusion over the 84-year Easter cycle.  It's another thing altogether to prove that this sort of mistake actually happened.

Well, hold onto your hats, because I'm about to prove that this sort of blunder did take place.

Bruide son of Maelgwn was the High-King of the Picts from about AD 553.  He was contemporary with Artuir mac Aedain.  They were, in fact, related: Bruide's daughter, Domelch, had married Arthur's father, Aedan.  Domelch ferch Bruide was the mother of Arthur's half-sister, Muirgein, and his half-brother, Gartnait, who became King of the Picts after the death of his uncle Bruide.

The Annals of Ulster indicate that Bruide, King of the Picts, died in AD 584, which fits in with the known facts.  The Annals of Tigernach, however, tell a different story.  They state that the year 752 saw a "Battle of Asreth in the land of Circinn between Picts on both sides, and in it Bruide son of Maelchon fell."

So, two sets of Irish annals (both ultimately deriving from notes made in the Easter tables maintained on the Isle of Iona) contain information regarding the same battle, fought in Angus between warring Pictish factions, which resulted in the death of King Bruide.  But one - the Annals of Ulster - give the date of that battle as 584, which is correct.  The other - the Annals of Tigernach - give the date as 752, which is way out.

The difference between the two dates is 168 years.  That is, of course, 84 years times 2.  The Tigernach annalist based his record on the same Easter tables entry as the Ulster annalist did, but he mistook the Easter cycle.  In fact, he skipped two full Easter cycles.  Bruide son of Maelgwn died in AD 584, but one annalist mistakenly ascribed this to a date which was two full Easter cycles later.

Evidently, then, it could happen.  The same note in the margins of an Easter table could be interpreted accurately, or it could be attached to another date altogether - 84 or 168 years later, depending on which Easter cycle the annalist mistakenly plumped for.

The date of circa AD 664 given in the Welsh Annals for the "second battle of Badon" and the death of Arthur's treacherous antagonist, Morgan the Wealthy, was the victim of a similar error.  It was out to the tune of 84 years.  Artuir mac Aedain fought his first "battle of Badon" at Badandun Hill in Glen Isla in 580, overcoming the "Chief-Boar" of the southern Picts, Galam Cennaleth (whose epithet meant "Chief of Alyth").

The so-called "second battle of Badon" was fought nearby, in the valley of the same River Isla, against a new "Chief-Boar" of the southern Picts.  His name was Morgan Mwynfawr: "Morgan the Wealthy".

If you look at the map, you'll see that seven miles from the Arthurstone on the south bank of the River Isla in Perthshire, there is a Morganstone, a few miles west of the Hill of Alyth.  The cataclysmic Battle of the Broken Sword (Camlann) was fought between those two landmarks.

The year was AD 594.  For the Britons, it was the end of the world as they knew it.

The Second Battle of Badon

Even though the first Arthur to appear in the historical records was Artuir mac Aedain (Arthur son of Aedan), there seem to be two main reasons why some people refuse to accept that he was the original Arthur.

The first reason is emotional.  Quite simply, a lot of people want Arthur to fit in with some imperialistic paradigm.  We know he can't have been English (although it could be said that plenty of Arthur enthusiasts would like to imagine that he was), so we'll plump for the next best thing: he was Roman.  Certainly southern British.  And absolutely NOT a Scot of Irish heritage.  No, anything but.

However, it's fairly obvious that this emotional attachment to a sort of prototype-Englishman Arthur has no historical support.  It's little more than a nationalistic impulse, insisting that Arthur was anything but Scottish.  So, in order to advance the claim that Arthur son of Aedan must have been named after an unknown earlier Arthur, scholars point to the Annales Cambriae or Annals of Wales.

The Annals of Wales were compiled in the 10th century.  They start in about the year AD 447, which in the Annales Cambriae is designated "Year 1".

The entry for "Year 72" reads "The battle of Badon in which Arthur bore the cross of our Lord Jesus Christ on his shoulders for three days and three nights, and the Britons were the victors."

The entry for "Year 93" reads "Gueith [Battle of] Camlann in which Arthur and Medraut perished; and there was plague in Britain and Ireland."

Taking the year 447 as the starting point, these entries are usually adjusted to read AD 518 for "Year 72", the "battle of Badon", and AD 539 for "Year 93", the "Battle of Camlann".

Arthur son of Aedan was, by my reckoning, born in 559.  He fought his last battle in 594.  So, evidently, he can't have been the original Arthur, right?  Because the Annals of Wales clearly date Arthur's battles to AD 518 and 539.

But here's the thing.  My researches, published as The King Arthur Conspiracy, led me to the conclusion that Arthur's first battle was fought in AD 573.  The gap between his first battle and his last, fought in 594, was 21 years - which is exactly the same as the gap between the two battles ascribed to Arthur in the Annals of Wales.

In fact, the year of Arthur's first battle (573) is given as "Year 72" in the Annales Cambriae, while the date of Arthur's last battle (594) is given as "Year 93").  It's as if the chroniclers of Wales were 501 years out.  Indeed, just place the digit 5 before both years given in the Annals of Wales, and you arrive at pretty much the exact dates of Arthur's first and last battles.

Still, there's a discrepancy.  For Arthur's battle of Badon, the Welsh annalists indicate the year 518; for his catastrophic Camlan conflict, they indicate 539.  The dates, according to my scheme, were actually 573 and 594 respectively.  The actual difference between my dates and those given in the Welsh Annals is 55 years.

Now, we know that the Welsh annalists were not working with the Anno Domini system, although that dating system was already in existence.  But they did draw much of their information from the work of the Anglian historian, Bede, whose Ecclesiastical History of the English People was written in AD 731.  And Bede did use a version of the Anno Domini dating system.

The problem with the Anno Domini dating system is that you have to agree where to start.  Let us suppose that a Welsh annalist, many years after the events, wished to record the dates of Arthur's first and last battles, which he knew had been fought in 573 and 594 (i.e., 21 years apart).  The said annalist is working with a chronicle which actually uses the year 447 as its starting point, probably because that was the year in which the Anglo-Saxons first invaded Britain.  However, the annalist also knows that Bede used a different dating system, and so he wishes to convert the dates for Arthur's battles into something which fits both Bede's Anno Domini system and the system used by the Welsh annalists, working forwards from AD 447.

The first chapter of Bede's magnum opus comprises a geographical description of the Island of Britain.  The second chapter provides the first date:

Britain remained unknown and unvisited by the Romans until the time of Gaius Julius Caesar ...

Julius Caesar led the first abortive Roman invasion of Britain in 55 BC.  This - as far as Bede, the Church, and many others since were concerned - was the beginning of British history.  Nothing really happened before that date.  55 BC was Britain's Year Zero.

So let's say that the Welsh annalist, working in the tenth century, chose the year 55 BC as the start of British history.  He knew that the Annals of Wales began in 447 AD ("Year 1"), but if he was starting his count from 55 BC, that would actually be designated 502.

Following the same logic, the annalist worked forwards from his revised starting date, adjusted to account for the beginnings of British history in 55 BC.  "Year 72" would therefore be AD 573 - the date of the historical Arthur's first battle.  And his last battle would have been fought in AD 594 - or "Year 93" in the annalist's system.

Only by assuming that all of the dates given in the Annals of Wales should be dated from AD 447 - the year of the Saxon invasion - do we arrive at the familiar dates of 518 and 539 for Arthur's battles.  But if the interpolations, made more than 300 years after those battles were fought, were based on a misunderstanding of Bede's AD dating system (the mistaken belief that 55 BC was the start of British history) then the dates given in the Annales Cambriae perfectly match my dates for Arthur's first and last battles, which were (as the Annals of Wales indicate) fought 21 years apart, in 573 and 594.

The Annals of Wales also indicate that there was a second battle of Badon.  This is dated to about the year AD 666.  The entry reads:

The first celebration of Easter among the Saxons.  The second battle of Badon.  Morgan dies.

The important point here is the last statement: "Morgan dies".  As I show in The King Arthur Conspiracy, Arthur's main rival in his last battle was Morgan Mwynfawr - "Morgan the Wealthy".  He was a Man of the North and is also mentioned in a medieval list of the "Four-and-Twenty Horsemen at the Court of Arthur".

But two things stand out, here.  The first is the date.  AD 666, or thereabouts, is way out.  What went wrong, though, is suggested by the first part of the entry ("The first celebration of Easter among the Saxons.")  In the Celtic Church, Easter was calculated on the basis of an 84-year cycle.

At the Synod of Whitby (AD 664), the Northumbrian (English) Church chose to abandon the Celtic dating system for Easter in favour of the Roman system.  The reference in the Annals of Wales, then, is to the adoption of the Roman dating system by the English Church ("the Saxons").  But at this point, a mistake seems to have crept in, probably as a result of confusion over the Easter Annals used by the British (Celtic) Church.  The 84-year Easter cycle, as used by the Celtic Church, indicates that AD 666 was the 33rd year of its cycle.  The 33rd year of the previous cycle was AD 582 - which is the date I give for Arthur's battle of Badon.  The reference in the Annals of Wales to the "second battle of Badon" at which "Morgan dies" would appear to be a mistake, based on a misreading of the 84-year Easter cycle and the first battle of Badon fought by Arthur (at which Morgan didn't die).

So what was this "second battle of Badon" - which was, in fact, a mis-remembered reference to Arthur's last battle?

In The King Arthur Conspiracy, I locate Arthur's battle of Badon (fought in AD 582) at Badandun Hill in Angus, on the edge of the Cairngorm mountains.  Arthur and his men attacked a patrol of Pictish warriors in the valley of the River Isla, in the shadow of Badandun Hill.

Arthur's last battle - commonly known as Camlan - was also fought in Angus, and in the valley of the River Isla.  It was, arguably, the "second battle of Badon", and Arthur's main rival at this battle was Morgan.

The "Badon" term survives to this day in the Highland region of Badenoch.  This area, the boundaries of which have always been rather unspecific, is thought to take its name from the Gaelic Baideanach, meaning "drowned land" (figuratively, "overwhelmed") - from the Old Celtic badio, a "bath".  But I suspect that this derivation is wrong.  "Badon" here actually derives from the Welsh (i.e. British) word baedd, meaning a "boar".  The Pictish warriors against whom Arthur was pitted in both of his "Badon" battles (Badandun Hill in 582 and Strathmore/Arthurbank in 594) were known as "boars".  The Pictish region in which these battles were fought was known as Circenn - from cir, a boar's crest or comb.  In the ancient Welsh tale of Culhwch ac Olwen, Arthur has to hunt down two terrible boars, and these two boar-hunts correspond to his first and second "Badon" battles, the one against Galam Cennaleth of the southern Picts in 582, the second against Morgan the Wealthy, nominal leader of the southern Picts, in 584.

Both of these battles were fought in Circenn, the "land of the boars", which was remembered as "Badon" (from the Welsh baedd, which became the Gaelic Baideanach - "Place of Boars").

So - up to a point, the Annals of Wales are right.  Arthur's first and last battles were fought 21 years apart, in 573 and 594.  And two "Badon" battles were fought - the latter seemingly resulting in the death of Arthur's enemy, Morgan.  Admittedly, we have to adjust the dates given in the Annals of Wales - the first two to account for a misunderstanding of Bede's Anno Domini dating system and Julius Caesar's first incursion into Britain as the start of British history, and the "second battle of Badon" date to account for the 84-year Easter cycle.

Once we've done that, we find that the Annals of Wales actually square with the dating of Arthur's battles in The King Arthur Conspiracy and the locations of the two major battles fought by Arthur in the Pictish province of Circenn - the Boars' Land, or "Badon".

There was no earlier Arthur.  The real "King Arthur" was Artuir mac Aedain.