Arthur's dykes

[Justin Swanton]

This thread replicates the discussion that started on the Currently Reading thread over King Arthur's Wars. I've copied the relevant posts into separate posts here and given the names of the posters (I don't think one can selectively excise posts to create a new thread).

[Justin Swanton]

Justin

Just finished King Arthur's Wars by Jim Storr. A fascinating book. As far as I know, he is the first to do a systematic analysis of the English dykes (he spent months travelling over the countryside and studying them in situ. From them he established a pattern of Romano-British defences against the Germanic invaders that is truly staggering in scale. The Wansdyke, for example, stretched from the Thames to the Severn and defended south Britain for over 60 years. Their size is also impressive - some dykes were 30 feet high. After establishing the defensive lines the Romano-British set up against the Angles, Saxons and Jutes, he deduced an approximate timeline for their gradual retreat westwards.

His study of the dykes goes with a careful use of toponomy (he is always cautious about how much (in)certitude he gives to his conclusions) which indicates where the Germanics arrivals settled and the nature of the landscape they settled in. For example, place names ending in '-ly', '-ley' and '-leigh' meant a clearing in a wood which in turn means the area of settlement was originally heavily forested. This ties in neatly with where the dykes start and end, usually against impassible forested or marshy terrain.

He also makes the point that the Rescript of Honorius rests on no firm primary source evidence and in fact the Roman army never entirely left Britain, although much of it did during the insurrection of Constantine. The dykes, at least in the earlier period up to the end of of the 6th century, are constructed with a mathematical precision possible only to trained military engineers - they run in straight lines and their dimensions are uniform.

He points out that Camelot can only be Camelodunum (Colchester) and dykes in the area plus the siting of settlements show the existence of a large fortified clearing that is too big to be a purely human habitation but is the right size for horses. In other words, he proposes that Camelodunum, situated between the Angles to the north and the Jutes to the south, was Arthur's operational headquarters.

All this goes along with a study of the primary sources that is also cautious, IMHO a little too cautious, but that does lend weight to the conclusions he reaches. The overall picture then is a Roman-British military establishment that was initially superior to the Germanic invaders in pitched battles, especially with the use of cavalry, but then gradually lost its edge and eventually could not win battles especially against the West Saxons, relying thereafter on the dykes to keep them at bay. Ultimately a losing strategy but one which bought the Romano-British a two centuries' fighting retreat.

[Justin Swanton]

Mark G

Does he mention anything of the fighting style?

I never really understood why the romano brits always seemed to be LHI hit and run hillmen types in mist lists,?when all their opponents seem to be shieldwall or warband types.

And all those dykes fit with defensive heavy infantry.

[Justin Swanton]

Justin

Does he mention anything of the fighting style?

I never really understood why the romano brits always seemed to be LHI hit and run hillmen types in mist lists,?when all their opponents seem to be shieldwall or warband types.

And all those dykes fit with defensive heavy infantry.

Not in any detail. His point is that Romano-British fought the Saxons et al. toe to toe when necessary, but once they lost their cavalry their infantry alone in a pitched battle were not quite up to the task. They would have needed fairly well organised heavy infantry to defend the dykes and could have done (and did do) so successfully with few troops. Perhaps the main problem of the Romano-British was a lack of numbers. The Angles, Saxons and Jutes came from a tribal warrior culture, where every able-bodied male was expected to fight, whereas the Roman-British had lived peacefully for centuries under the protection of a professional army and did not have the same across-the-board obligation to military service. So you have a small professional military in the 5th century that never expands into a tribal levy until it is too late.

[Justin Swanton]

Andreas Johansson

I never really understood why the romano brits always seemed to be LHI hit and run hillmen types in mist lists,?when all their opponents seem to be shieldwall or warband types.

FWIW, the DBMM list has them as mostly shieldwall types, with some regular Roman(-style) troops early on.

[Justin Swanton]

Anthony Clipsom

Is there an explanation of how the dykes were manned?  The Roman way with limes seems to have been garrisons and patrols.  Has he therefore identified where the garrison stations were?

As to Romano-British armies, aren't they usually depicted as low grade shieldwalls these days?  Aren't the "hill tribe" types reserved for early Welsh, who were a bit more prone to living in wild places?

One final point directly to Justin about the book.  If Jim Storr identifies Camelot as Camelodunum, does he explain why the place is first mentioned in a 12th century French source?  How does he explain the 600 year gap in the evidence?

[Justin Swanton]

Dave Hollin

Its worth a read but I found that the author eventually falls into the trap of restating opinions until it becomes 'fact' although not as badly as some authors on the subject. The theories are niche and very interesting and the book manages to devote a lot of time to this aspect of the landscape where others have only observed and commented on the presence of dykes but not explored the reasons for etc. Definitely worth a read

[Justin Swanton]

Stephen Brennan

Its worth a read but I found that the author eventually falls into the trap of restating opinions until it becomes 'fact' although not as badly as some authors on the subject. The theories are niche and very interesting and the book manages to devote a lot of time to this aspect of the landscape where others have only observed and commented on the presence of dykes but not explored the reasons for etc. Definitely worth a read

His stuff on dykes is really interesting and the book is certainly worth reading. 

Like many a professional soldier turned author before he likes to find evidence for regular soldiers wherever he looks.  For example he sees the cavalry of Gododdin as a Roman cavalry unit acting in the Persian cavalry tradition.  That fails to convince me because there is nothing to support such a conclusion. The Gododdin had never been part of the Roman regular army and from what we can glean their cavalry tactics were of the native tradition.  I mostly found his other historical conclusions unlikely too. 

All that said I still think Jim Storr has given us a very useful book.

[Justin Swanton]

Justin

Is there an explanation of how the dykes were manned?  The Roman way with limes seems to have been garrisons and patrols.  Has he therefore identified where the garrison stations were?

With a couple of possible exceptions (cavalry bases), there is zero archaeological evidence for where units of the Romano-British army were stationed and how they functioned, so Storr hypothesizes. Scouts stationed ahead of the dykes give ample warning of the approach of a Saxon army. This allows the local British unit to man the dyke and perhaps prepare a counterattack - say a flanking ambush to be sprung whilst the Saxons were attempting to storm the dyke. But it's a case of not knowing for sure. I find that Storr pushes speculation a bit far sometimes (whilst admitting it is speculation), but three hard facts stand out: a) there was an enormous dyke system that could only have been used as lines of defence; b) these dykes were post-Roman, from the fact that they crossed Roman roads (the easiest routes of access) with monotonous regularity, and c) the earlier dykes at least were built with an engineering precision possible only to Roman-trained military engineers.

Quote from: Erpingham on December 26, 2019, 11:16:28 AM

One final point directly to Justin about the book.  If Jim Storr identifies Camelot as Camelodunum, does he explain why the place is first mentioned in a 12th century French source?  How does he explain the 600 year gap in the evidence?

He does say that the first mention of Camelot is from Cretien de Troyes. Cretien spells other cities' names Arthur visited and which were known to have been Roman either in their original Latin or, more often, their contemporary English versions. Camelot was Camulodunum, and Cretien would not have known that it had changed to Colchester. He suggests that since 'Camelot' was spelled with a 'K' it came to Cretien from Breton sources.

Why the 600 year gap? Storr suggests that Arthur might have been expunged from British sources, a kind of damnatio memoriae, without going into the reasons why it was expunged.

[Justin Swanton]

Stephen Brennan

"c) the earlier dykes at least were built with an engineering precision possible only to Roman-trained military engineers."

Yes, so he opines, and yet the natives had been building defensive dykes forever.  I imagine they had worked out how to do it properly themselves long before the post Roman period.

The 600 year Damnatio of Arthur theory doesn't work for me.  Arthur pops up in An Gododdin and elsewhere so there was no damnatio of his name from our sources. If anything his warrior career was celebrated as an exemplar.

[Justin Swanton]

Anthony Clipsom

but three hard facts stand out:

These seem actually to be a mix of fact and speculation.  The facts appear to be there are a number of dykes and that they are post Roman.  But I leave it to the Arthurian specials like Holly and Anton to comment.

Why the 600 year gap? Storr suggests that Arthur might have been expunged from British sources, a kind of damnatio memoriae, without going into the reasons why it was expunged

How convenient if you wish to speculate wildly   You might suggest a damnatio for a couple of centuries on the basis of absence of evidence (e.g. not in Gildas) but Arthuriana is alive and well in Welsh terms long before the 12th century and they don't mention Arthur's capital as Colchester.  Chretien may have picked up a genuine tradition connecting Arthur to Colchester or he may have used names of Roman towns he found elsewhere.  Not a sound basis to build a thesis on, I fear.

[Justin Swanton]

Patrick Waterson

Yes, so he opines, and yet the natives had been building defensive dykes forever.  I imagine they had worked out how to do it properly themselves long before the post Roman period.

The one thing they would not have worked out long before the post-Roman period would be how to build them across Roman roads ...

If Jim Storr identifies Camelot as Camelodunum, does he explain why the place is first mentioned in a 12th century French source?  How does he explain the 600 year gap in the evidence?

Camelot as Camulodunum (i.e. Colchester) does not have to rely on erratic bardic nomenclature, but rather on role (a capital), location (horse country) and finally a coincidence or confluence of appellation.  Its designation is known to have have varied over time: the Historia Brittonum refers to it as 'Cair Colun'; its Saxon name was 'Colneceastre' (from which we get Colchester).  Its original (pre-Roman) name appears to have derived from 'Camulos' (war deity) plus 'dun' (fort or town).  No special wisdom here: just Wikipedia.

There is one possible, even likely, influence behind the change of name: Camulos, a Celtic deity, might have sat ill with the overtly Christian priests of later post-Roman Britain.  Arthur's less doctrinal contemporaries with their tribal traditions still flourishing would have had no such reservations.

[Justin Swanton]

Anthony Clipsom

If we are going to mention the changes in Colchester's name, we ought to explain it is on the River Colne.  Changes tend to look rather workaday rather than mystical then.

We could of course collect a selection of random facts; Colchester was a Roman capital, Camelot sounds like Camulodunum, the Romans raised horses in that part of Essex (I presume the evidence exists for this?) and create an Arthurian connection.   But, given an absence of corroboration and earlier references to Arthur being based in Caerleon or Chester, I don't know if we can put any special weight on it.  If I might speculate wildly, it is possible that Chretien has picked up the Welsh tradition of placing Arthurs capital in a legionary city but, not knowing Caerleon or Chester, he has plucked out Colchester as a possible candidate from his limited knowledge of Roman Britain.

[Justin Swanton]

Andreas Johansson

As to Romano-British armies, aren't they usually depicted as low grade shieldwalls these days?  Aren't the "hill tribe" types reserved for early Welsh, who were a bit more prone to living in wild places?

Army lists from the WRG mould don't have a separate Welsh list until AD 580, so Welsh of the Romano-British period are presumably assumed to be more-or-less indistinguishible from their British brethren in what became England.

(Perhaps surprisingly, Phil has the Dumnonians remain shieldwall types when the Welsh switch to warband. Do we, anyway, know anything about Dumnonian armies?)

I dunno if I've read enough lists to comment on "usually", but it may be noted that ADLG has a free choice of Heavy Spearmen and Medium Spearmen for the bulk of Romano-British foot, where the latter is a "hillman" sort of classification, who, if I understand, would probably be LHI or LMI in old WRG currency. They're supposed to use looser formations than proper heavy infantry, move rapidly through terrain, yet be able to bunch up to resist cavalry. Heavy Spearmen are not necessarily shieldwall types - the class includes Classical hoplites - but definitely not hit-and-run anything.

Triumph! has them as Heavy Foot, which is the class for HI who aren't good enough to be Elite Foot, bad enough to be Horde, angry enough to be Warriors, or pointy enough to be Spears or Pikes.

[Justin Swanton]

Jim Webster

Yes, so he opines, and yet the natives had been building defensive dykes forever.  I imagine they had worked out how to do it properly themselves long before the post Roman period.

The one thing they would not have worked out long before the post-Roman period would be how to build them across Roman roads ...

But cutting through a Roman road isn't high technology, it just needs a little more time, and ideally a crowbar