THE TERRIBLE ONE'S HORSE: CHAPTER 31

A NEW IDENTIFICATION OF THE BATTLE OF BRUNANBURH

Several places have been identified over the years with the famous Brunanburh battle first found mentioned in the Anglo-Saxon Chronicle (with accompanying heroic poem) in the year AD 937. To quote from Wikipedia on this battle:

‘The Battle of Brunanburh was an English victory in 937 by the army of Æthelstan, King of England, and his brother Edmund over the combined armies of Olaf III Guthfrithson, the Norse–Gael King of Dublin; Constantine II, King of Scots; and Owen I, King of Strathclyde. Though relatively little known today, it was called “the greatest single battle in Anglo-Saxon history before the Battle of Hastings.”’

For additional details on the battle, and a splendid listing of all the relevant sources, please see http://www.brunanburh.org.uk/.

To the various candidates for Brunanburh we may add the most recent by Dr. Andrew Breeze, who opts for the Lanchester Roman fort on the River Browney in Durham (http://www.thenorthernecho.co.uk/news/10717666.Lanchester_____birthplace_of_a_unified_kingdom_/).

Currently, the most favored site is Bromborough on the Wirral Peninsula (http://www.nottingham.ac.uk/ncmh/dna/Brunanburh.htm). The case for Bromborough is dependent on two identifications, one of which may well be right, and the other of which is certainly wrong.  It is true that Bromborough would to be the only perfect match etymologically, as its 12th century form Brunburg aptly demonstrates.  The error in the selection of Bromborough has to do with the imaginative creation of ‘Sea of the Thing’ out of the Anglo-Saxon Chronicle’s Dingesmere, presumably because of the presence on the Wirral Peninsula of the Norse place-name Thingwall.  The latter is an important part of the puzzle, for the surviving Norsemen fled by or on this “mere” from Brunanburh on their way back to Dublin in Ireland.

What is or, rather, where is Dingesmere? The line immediately following reference to this “sea” describes a passage over “deop waeter” (http://www8.georgetown.edu/departments/medieval/labyrinth/library/oe/texts/a10.1.html; see lines 54-55). The Otho MS of the ASC has Dinnesmere, with din meaning noise.  The related word dynge is defined as ‘a noise, dashing, storm’.

From the authoritative Bosworth and Toller Old English Dictionary:

dynge, dinge, dynige, es; m? A noise, dashing, storm; sonus, strepĭtus, procella :– On dynges mere on the sea of noise, Gst. Rthm. ii. 66, 20; Chr. 937; Th. 206, 12, col. 2. v. dyne

Dingesmere therefore means the noisy or stormy sea, undoubtedly in this case the Irish Sea.  It is NOT an actual place-name, but a poetic descriptor no different than the following ‘deep water’.  Similar poetic descriptions of the sea are found throughout OE poetry.  In works like Beowulf, kennings for the sea abound.

Andrew Breeze (private communication) prefers to see in Dingesmere the ‘Dingle-mere’, despite dingle not being found in the language until Middle English.  He cites J. A. W. Bennett, _Middle English Literature_ (Oxford, 1986), 280, which states of God that ‘his domes the derne [that more secret] beoth ant deopre then eni sea-dingle’; which shows ‘dingle’ used of the abyss of the sea, after the psalm.  However, here the term is again poetic, meaning something like ‘sea-valley’ (as dingle is ‘a deep dell or hollow’, used in place-names of certain valleys).

We then have Florence of Worcester’s Humber reference to wrestle with. Florence is dateable to the late 12th century and he claimed that Anlaf (Olaf) entered the mouth of the Humber.  When I went to the maps and scoured the Humber region, I noticed Anlaby near Kingston-Upon-Hull.  Eilert Ekwall (The Concise Oxford Dictionary of English Place-Names), drawing upon early forms (spellings) long ago showed this place to be Anlaf’s By.  We have no idea who this Anlaf was, but he COULD have been the Anlaf of the Brunanburh battle.  More likely he was a Danelaw settler.  But the mere presence of his name here on the Humber could have led Florence or his source to situate the Brunanburh battle on the Humber.  There is nothing noteworthy about the vicinity around Anlaby, and no reason to think that Brunanburh was once located there.

The Chronicle of Aethelweard was written within living memory of the battle, possible around 975.  It is thus a far more valuable source than that of his successors.  Aethelweard calls the site Brunandun:

“In the year in which the very mighty king Æthelstan enjoyed the crown of empire, 926 years were passed from the glorious incarnation of our Saviour. [939 = 937]  After thirteen years a huge battle was fought against the barbarians at Brunandun, wherefore it is still called the ` great battle’ by the common people. Then the barbarian forces were overcome on all sides, and held the superiority no more. Afterwards he [i.e. the king] drove them off from the shores of the ocean, and the Scots and Picts both submitted.  The fields of Britain were consolidated into one, there was peace everywhere, and abundance of all things, and [since then] no fleet has remained here, having advanced against these shores, except under treaty with the English. [941 = 939] After a period of two years Æthelstan, a king worthy of honour, left the world.”

The Vikings, especially early on, liked to restrict their attacks to coastal districts.  Aethelweard here says they were driven off “from the shores of the ocean”, which implies a coastal or near-coastal location of Brunandun (or Brunanburh).  Dun in OE is a hill, and by extension sometimes (and certainly for the Welsh with their similar word) a fortified hill.  Dr. Andrew Breeze’s Lanchester IS on a ridge, so dun might work in this context.

But why wasn’t Lanchester used?  It was Langecestr in 1196, and while this may show an English ‘long’ plus the usual name used for a old Roman fort, there is little doubt that he English word would have substituted for the first element of Longovicium, the name of the Roman fort at this location.  Hence the name Lanchester is by definition very old and, indeed, must have existed from the sub-Roman period.  So why Brunanburh or Brunandun?

I note from the ASC that in 933/4, just before the Brunanburh battle, Aethelstan is invading Scotland itself by land AND sea, and is said to have be harrying much of the country.  This allows for the possibility, therefore, that Brunanburh may be fairly far north.  It also almost certainly means that trying to place Brunanburh on theWirral Peninsula makes little or no sense.  Aethelstan had seized Northumberland, remember, not the northwestern part of Mercia! Nor does it make any sense for the Scots or the Strathclyde Britons to have participated in a battle on the Wirral.

If we allow for Aethelweard’s early Brunandun, a good candidate emerges BY NAME:

Brandon, Northumberland. This has early forms Bremdona, Bremdon, Bromdun, and such names from ‘broom’ (although note this is on the Breamish, a Celtic river-name containing the same root as Roman Bremenium).  In looking at Ekwall, many of these broom names can occur as Brune, Brun and the like.

http://streetmap.co.uk/map.srf?x=405500&y=617500&z=120&sv=brandon&st=3&tl=Map+of+Brandon+Hillhead,+Northumberland+&searchp=ids.srf&mapp=map.srf

At Brandon (Hillhead) there is an Iron Age hillfort.  This fortified hill is situated less than a mile west of the Devil’s Causeway Roman road.  The area around the Brandon hillfort is full of other forts, some of a major nature.  Several kilometers to the WSW is the impressive Brough Law complex with its satellite forts and settlements, while a few kilometers to the NNE is the Old Bewick hillfort (http://www.northumberlandnationalpark.org.uk/__data/assets/pdf_file/0008/188522/Breamish-Valley-leaflet.pdf).

For the Brandon hillfort itself, here is the entry at English Heritage’s Pastscape:

http://www.pastscape.org.uk/hob.aspx?hob_id=4993&sort=2&type=&typeselect=c&rational=a&class1=None&period=None&county=1306807&district=None&parish=None&place=Brandon&recordsperpage=10&source=text&rtype=&rnumber=

[NU 04371783] Camp [OE] (1)

On Brandon Hill in a plantation about 350 yards south-west of the top of the hll are the remains of a camp. Though situated below the summit it is on a promontory.

(2) Type B2 (49): “Forts on high ground less dependent upon natural slopes for protection”.

(3) The ramparts are destroyed to the east, but elsewhere preserved to the extent of half an acre from which it is conjectured that the original figure was oval of about 65 yards by 45 yards interior.

The earthwork is situated in a cleared plantation on a nearly level shelf on the south west slope of a hill at about 550 feet above sea level. Immediately to the south west are the steep natural slopes of the Brandon Dean (gradient c 1:4), but the work is overlooked at close quarters from the north-east.

Only fragments of the west side of the earthwork now remain, the rest having been completely ploughed out. There are three concentric ramparts partly formed by scarping the natural slope, with the spoil placed on top of the scarp to form a slight bank. In the more level places however there are two medial ditches between the ramparts. A gap to the Northwest, although badly mutilated seems to be the original entrance. There are no traces of internal habitation.

A stream to the west is the nearest source of water.

Although the position is not ideal for defensive purposes it is well concealed – and the number and construction of the banks indicate the work to have been defensive.

The proximity of a cairn (NU 01 NW 11) and finds of bronze weapons (NU 01 NW 13) as well as the construction of the earthwork itself suggests pre-Roman date.

(4) Listed as pre-Roman IA multivallate (forts settlements and enclosures).

(5) Re-surveyed at 1:2500.

(6) NU 044 178. Brandon Hill. Listed in gazetteer as a multivallate hillfort covering (?)0.20ha.

(7) NU 043 179. Brandon Dean camp. Scheduled No ND/258.

(8) An Iron Age multivallate hillfort, centred at NU 0437 1784, is visible as an earthwork on air photographs. Of this feature only the western third survives. The entire east side of the hillfort has been ploughed out. From the remaining portion we can see that the structure consisted of three ramparts and at least two ditches (only two short lengths – to the north – are visible on the air photographs available). The rampart banks are all approximately 2.5m wide. The ditches are almost 3m wide and visible only on the external side of the two outermost ramparts. There appears to be a berm of 4m between a ditch and the next rampart. Towards the north-west there is a gap in the inner two ramparts of 2.5m which may be that recorded as an entrance by authority 4. No internal features are visible on the air photographs available.

(9) For an aerial photograph of the fort, see here:

http://www.pastscape.org.uk/maps.aspx?a=0&hob_id=1381932&criteria=Brandon%20%20Hill%20Head&search=all&rational=q&recordsperpage=10&sort=4&mv=s

Brandon Hillhead fort, then, would be my best guess for the location of Brunandun/Brunanburh.  This location is not far from the coast, and it is in the southern part of Scotland, which was probably that region Aethelstan had invaded a couple years before.  The site is very close to a Roman road as well.