The Search for Brunanburh: Sailing to Battle

I still get the occasional missive insisting that Brunanburh should be located off the Humber — a tradition that begins with a man named John of Worcester nearly two centuries after the event.

I applaud the idea of going against the grain a bit — I was raised in the American West, after all — but it seems to me that holding on to John of Worcester’s Brunanburh-off-the-Humber scenario is starting to look like desperation rather than consideration. Objectivity seems to go out the window as emotions are weighed against the preponderance of evidence in the Casebook that rationally argues otherwise (see especially Cavill’s conclusions on p. 339). One doesn’t need to look much further than one of the men using his celebrity status to drive much of the effort to retain John of Worcester’s account: Michael Wood, a historian journalist for the BBC who insists that the great battle happened off the Humber. Why? Because as a school-boy he was moved by the view from White Hill and liked to imagine the battle happening there (“Brunanburh Revisited,” Saga Book of the Viking Society 20.3 [1980], 200-17).

Me? As I’ve repeatedly said, I have nothing but a desire to get us as close to the truth as possible. I just want a rational consideration of the sources. I like to think, in fact, that it is this aim that makes the Battle of Brunanburh: A Casebook such a success.

It was in that same spirit that I posted, a little over a year ago, a summary of the requirements for the battle site, followed by a more informal rebuttal of the Brunanburh-off-the-Humber theory. One of my observations there was that it would be logistically improbable to make a Humber assault given the timeline involved.

Today I received from a reader an objection to this point. The reader commented:

I have recently finished reading a study of King Alfred, written by John Peddie (OBE, MC) The book is called Alfred – Warrior King. Peddie discusses at length the capabilities of the Viking longships – based on the Gokstad discovery in 1880.

“Under sail the vessels were reputed to be able to achieve maximum speeds of 10-11 knots and to travel some 150 miles per day. They had a high rating for seaworthiness. In 1893, an exact replica of the Gokstad ship was sailed across the North Atlantic by Captain Magnus Andersen and despite encountering very heavy seas, he completed the journey between Norway and Newfoundland in a remarkable twenty-seven days”

By my calculations Michael, the journey from Dublin to the Humber could have been done in less than a week. Maybe we should not overlook this very important factor when considering the validity of JoW account.

As it happens, I’m working on an article going into these matters in more detail — one ending with the driving of a stake through the Humber theory — so I should be able to cobble together a reply pretty quickly.

Here goes.

1. We have no real idea about the commonality of ships like the remarkable Gokstad recovery. Still, let’s assume that every ship in this invading fleet was its equivalent. I think that’s doubtful in the extreme, but I’ll play along…

2. There are essentially two ways to get from Dublin to the Humber by sea. The southern route through the English Channel is about 1400 km. The northern route around Scotland is about 1550 km. Of the two routes I think we can safely assume Anlaf would have taken the northern in this hypothetical. He needed to rendezvous with the Scots, the Northumbrians, and the other Norse kings up there. Plus, he wanted to surprise his enemy, which would be nearly impossible via the southern route. Indeed, the need for surprise would probably mean keeping off Anglo-Saxon shores until the Humber push, which would add a fair bit to the 1550 km “straight shot” — but we’ll ignore that for now to help get the best case possible.

(For comparison, the direct shot from Dublin to Bromborough is only about 200 km.)

3. So let’s assume this Gokstad-style ship managed a little over half of its maximum speed over the course of the journey. If the max speed is 18 km/h — pretty standard numbers for the Gokstad — then let’s call the managed speed in reality something closer to 10 km/h. I still think that’s pushing it, quite frankly, but I’ll run with it to give the Humber theory its best face. That means 155 hours of sailing time.

(Bromborough, by contrast, is going to be a fair bit less than 20 hours of sailing time, only half that on open waters, and surprise almost assured.)

4. So if you assume a “straight-through” sail — going 24/7, no stops, no hang-ups — you have almost exactly a calendar week pushing through the waves.

(For Bromborough it’d be less than a day through very familiar waters.)

Piece of cake, right? If there’s an 8-week window between the battles of Lough Ree and Brunanburh, you’re only losing 12.5% of it to the sailing.

True. But look at the line of assumptions:

• 10 km/h with no stops for 155 straight hours. You’re going to have some tired and miserable and likely sick fighters when they finally land after this week fighting the sea. Then you’re going to assault York? Not likely, I think. Plus, the need to rendezvous with the Scots and everyone else is going to slow things down quite considerably. You’d probably lose at least a day with every anchorage, and going north you’d have several. It’s probably more like a 2-week journey if all goes pretty smoothly.
• An entire fleet of Gokstad equivalents. We don’t know the numbers of the armies involved, much less the number of ships. If I recall correctly, the earliest source to give a number on the latter is Symeon of Durham’s Libellus de exordio, written in the early twelfth century, about 175 years after the battle, which says Anlaf’s fleet was 615 strong (see Casebook, pp. 54-55). John of Worcester omits this detail in his later Humber identification, but Symeon of Durham combines the traditions in his subsequent Historia regum (Casebook, pp. 64-65), and the number sticks when the Chronicle of Melrose repeats this tradition around 1173 (Casebook, pp. 66-67). Peter of Langtoft, writing around 1300, ups the count to 715 (Casebook, pp. 94-95), but let’s assume the earlier information is to be preferred. (We’re throwing that basic principle out the window to follow John of Worcester’s Humber at all, but we’ll continue to ignore that for the moment.) So 175 years after the battle the claimed number was 615 ships. That’s surely an exaggeration of time, of course, so let’s just assume it’s off by a factor of ten. A fleet of 60 is quite a bit more likely, I think. By comparison, the allied fleet that ambushed Olaf Tryggvason at the Battle of Svolder in the year 1000 probably numbered in the 70s. That was a planned naval engagement close to shorelines back home in Norway, but at least it’s a point of reference. 60 ships of the Gokstad type (or its equivalent) would be rather costly both in coin and in time to construct and outfit for this hypothesized week-long voyage.

• 60 Gokstad ships arriving at the Humber means a lot more leaving Dublin. I honestly don’t know what the on-water rate of loss would be for such a voyage in the tenth century, but let’s call it 25%. That means Anlaf would have left Dublin with 80 of them in order to arrive with 60.
• 80 Gokstad ships means 80 able crews and 80 excellent captains to man them. Whereas the wind-at-your-back Dublin-to-Bromborough shot is so simple that I suspect just about any Norseman could sail it, the route through the treacherous isles and seas up north and than back down again to the east would require excellent navigation skills even under perfect conditions (which are statistically unlikely!). Sailing across the North Atlantic is no run-of-the-mill easy task. Getting around the north shores of Scotland would be even harder.

Add it all up, and I just don’t see it as logistically feasible to get it all done.

Again, Anlaf would have to get his army from Lough Ree down to Dublin, then get confirmation and reconfirmation with the Scots and the Northumbrians and everybody else that it was “game on” to go get Athelstan. Then he’d have to put together a fleet of maybe 80 Gokstad-style ships. He’d have to outfit his army and all these boats. Then he’d sail to the north, struggling around the difficulties there, connecting up with allies and figuring out how they were going to get to their destination and how to time the entire endeavor. Then he’d sail on to Humber and attack the Anglo-Saxons in just the sort of place they’d expect to be attacked.

And he’d do it all in less than two months?

Sorry, but I’m not buying it. We already knew that the Humber entry point doesn’t fit the linguistic evidence or the primary literary record. I think it is likewise true that it doesn’t fit the logistic possibilities, either.

For that matter, the Brunanburh-off-the-Humber theory doesn’t even fit the common sense question of why all these groups would form the alliance in the first place. No doubt Anlaf wanted York, but why would everyone else go to such effort to help him to that end? Because for the alliance it wasn’t about York. It was about the destruction of Athelstan. And victory at the Humber doesn’t get you that, while victory at Bromborough just might. And while we can easily explain how John of Worcester got the detail confused if Bromborough is Brunanburh (Cavill nails this in the Casebook), we have no way of explaining the contrary case.

Sorry, folks. I still think John of Worcester is wrong.