THE DUBIOUS READING OF 'ARMATOS' FOR THE LUCIUS ARTORIUS CASTUS 'ARM[...]S' LACUNA

For years now I've sought to demonstrate, along various lines, that Dr. Linda A. Malcor's proposed reading for the ARM[...]S lacuna of the Lucius Artorius Castus stone simply does not work. This whole time, however, it never occurred to me that the primary reason it doesn't work is simply because it doesn't fit the space.

Ironically, one of Dr. Malcor's most frequently expressed refrains is "Only ARMATOS fits" the lacuna. This is, of course, not true. At least three other proposed readings fit: ARMENIOS (with a N/I ligature), ARMORICOS (with a couple of ligatures, including a C/O ligature - something I was the first to do) and ARM.GENTES (for armatas gentes, with a common abbreviation and a ligature already found elsewhere on the stone).

But her insistence that only ARMATOS fits the space prompted me to look again at the reconstruction of the stone prepared by Dr. Malcor's colleague and co-author of MISSINGS PIECES, Alessandro Faggiani.  

I was immediately struck by something odd that I hadn't noticed before. To make sure I wasn't seeing things (or seeing only what I wanted to see!), I asked my wife to look at it and let me know if she saw anything unusual about ARMATOS relative to all the other words on the inscription.  

It shouldn't surprise me (as she is an accomplished artist with an excellent pattern recognition ability) that she noticed quite quickly the same thing I had noticed: ARMATOS appears to have been unusually widely spaced compared to other words in the lower lines. The line containing the lacuna was otherwise notably compressed/compacted. So much space was being saved on the lacuna line that 3 ligatures are present and one word is abbreviated. Another word is divided in half, its second part being shoved down to the beginning of the next line.

Now, it is a known characteristic of this memorial stone that the lines at the top of the stone are taller than lines near and at the bottom of the stone. In other words, the lines diminish in height by stages as we drop down the surface of the stone.

This being true, I decided to compare the distance between three identically paired sets of letters found above the line of the lacuna - lines that are taller than that of the lacuna. I've circled the letter pairs in question on this image. They are AT, TO and OS. I've also underlined the corresponding letters of the proposed lacuna reading.

Although, obviously, I've captured this image on a regular sized sheet of paper and we are not here dealing with the actual stone, we are still comparing the relative scale of lines that are to be found on the stone itself.

When I measured the distance between the larger circled letters and compared the measurements from those with the measurements found between corresponding letters on the lacuna line, guess what I found?

The measurements between the two SO sets of letters were roughly the same.  In the case of the TO in ARMATOS, the space is slightly wider than that between the TO of PRAEPOSITO. The same is true of the AT in ARMATOS when compared with the AT in MISENATIUM. 

What does this mean?

Well, the letters of ARMATOS, being considerably smaller letters (that is, they are not as tall as the letters in the lines above the lacuna line), should have proportionately smaller spaces between them. In other words, the distances between the letters in the higher lines and those of the lacuna line should be scaled appropriately.  

THEY ARE NOT.

And merely making sure proportional spacing is correct does not also account for any extra compression of the lacuna line.

Unfortunately, this tells us that Mr. Faggiani has, willingly or otherwise, perpetuated a gross inaccuracy on the academic communitiy (or, at least, on the Arthurian community).  

And it also means that Dr. Malcor would appear to be wrong when she authoritatively asserts that only ARMATOS fits the lacuna.

ARMATOS does not, in fact, fit the lacuna of the Lucius Artorius Castus stone. Rather, it has been made to fit by stretching out the word.

A Selection from My Revised "MAKING SENSE OF CAMLAN: MY FINAL TREATMENT OF THE PROBLEM"

https://mistshadows.blogspot.com/2026/05/making-sense-of-camlan-my-final.html

West and East Wittering

***

Chichester, Cissa and Cymenesora

The Chichester Noviomagus should perhaps be discussed a bit more in the context of Arthur's Camlan.

According to English tradition, Chichester was named for the eponymous Cissa.  Cissa is introduced to us in the ASC as a warrior who fights at a place called Cymenesora near Chichester:

477

Her cuom Ęlle on Bretenlond 7 his .iii. suna, Cymen 7 Wlencing 7 Cissa, mid .iii. scipum on þa stowe þe is nemned Cymenesora, 7 þær ofslogon monige Wealas 7 sume on fleame bedrifon on þone wudu þe is genemned Andredesleage.

[https://asc.jebbo.co.uk/a/a-L.html]

I had once considered Cymenesora as a candidate for Camlan. But I dispensed with it in favor of other sites in southern England, like The Cams in Hampshire. I had come up with another etymology for Cymenesora: 

https://mistshadows.blogspot.com/2018/07/a-new-identification-for-location-of.html

But as Medard of Noviomagus in Gaul was related to Camlan in the Welsh Annals, and Noviomagus was the ancient name for Chichester, and Cissa of Chichester fought at nearby Cymenesora, we might be tempted to favor Cymenesora, i.e. Wittering, on Selsey Bill, as the Camlan being referred to in the Welsh Annals. Whatever the origin of Cymen-, the Welsh may have taken it for their own camen -

[cam2+-en] 

eb. ll. camennau.

1.  Camdra, gwyredd, tro, plyg, dolen:

crookedness, curvature, turn, bend, loop. 

Submit

14g. GDG 194, A ganodd neu â genau / O fawl o’r twrf meistrawl tau, / Gymar hwyl, gem yr heli, / Gamen môr, gymain’ â mi?

c. 1624 CRC 133, ai Daü forddwyd a bortreüwyd / fal dwy gamen kar yr ychen.

1794 P.

Digwydd hefyd mewn e. lleoedd, e.e. Ffridd Camen (Llandrillo) a Camen Fawr (Llanfyllin).

- plus AS ora, 'shore, bank.'

If Cymenesora is the Camlan of the Welsh Annals, does this mean that a real Arthur perished there in 545?

No. As we have actual Camlan names in NW Wales (where the later Welsh tradition places Arthur's death site) and on Hadrian's Wall (the Camboglanna Roman fort), the Welsh Annals entry which employs St. Medard (= Medrad/Medrawd) of Noviomagus to fix Camlan at Cymenesora near Chichester has all the hallmarks of a cleric's Isidorian-style etymological identification.  Such an identification, placed as it is directly across from Cerdic's Isle of Wight, the Wessex founder's last conquest before his death, may well have been intentional.  I have suggested that the chronological bracketing of the Arthur of the HISTORIA BRITTONUM corresponds perfectly with the floruit of Cerdic and that, therefore, Arthur may have been meant to be a sort of counter to the founder of Wessex.

***

MAKING SENSE OF CAMLAN: MY FINAL TREATMENT OF THE PROBLEM (REVISED VERSION 5/14/2026)

During my recent YouTube talk with Richard Johnson of O'r Golwg (https://youtu.be/ICS0cyB8bqg?si=lxWaM02MwhQlPXJW), we briefly discussed Arthur's Camlan battle.  Time would not permit us to delve into the subject very deeply, and we may dedicated an entire show to that next time around.  But for now I would like to summarize here my current thinking on Camlan - especially in the context of my belief that the original and most famous Arthur was the Roman dux Lucius Artorius Castus.  

I'd like to start off with my treatment of Medrawd, whom I recently (at least to my satisfaction) have proven to be an intrusion into the Camlan entry of the Welsh Annals.  I will then progress to consideration of the localization of the Welsh Camlan in NW Wales, and will finish off by taking a look at Camboglanna on Hadrian's Wall and how - if at all - it may factor into the equation.  

***

It was during my comparison of the Annals Cambriae entry for the year 537 with entries in the Irish Annals that I noticed something peculiar. Where the AC has Arthur and Medraut perish in Camlan during a year that saw plague in Britain and Ireland, the corresponding year entry in the Ulster Annals had the death of Comgall son of Domangart, King of Dalriada.  Yes, the same Dalriada that later saw an Arthur in its royal family.  A later Domangart, billed as this Arthur's brother, dies with him.

That got me sidetracked for awhile.  But only the other day I more thoroughly investigated the strange fact that there is no plague mentioned in the Irish Annal for 537.  Instead, that source duplicates the death of Comgall son of Domangart in 545, and it is there that we are told about the plague. 

T537.2 [538 in Ulster]

Comgall son of Domongort, king of Scotland died in the 35th year of his reign.

U545.1

The first mortality called bléfed, in which Mo-Bí Clárainech died.

U545.2

Death of Comgall son of Domangart, as some say.

I asked myself this question: what if we run with 545 instead of 537 and see what happens?

What happened is something so strange I find it difficult to assign merely to coincidence.

In 545, in Gaul, the famous St. Medard died. He belonged to Noviomagus and, as it happens, there were two places of this exact same name in Britain.  One was at Crayford and one was at Chichester. Allowing for a simple and perfectly allowable metathesis, Medard becomes Medrad and hence Medrawd/Medraut in Welsh.  His folk relocation to one of the British Noviomagus sites would have been an easy matter. 

The mixing in of Medard as Medrad may well have been accidental - indeed, I think it unnecessary to propose intentional creativity in this instance.  Once Medard's Noviomagus was wronly associated with Chichesteror Crayford, it would have been an easy matter to combine annal entries. We might imagine original multiple entries for the same year, something like -

Year X Arthur perishes at Camlan.  Blessed Medard rests.  And there was plague in Britain and Ireland.

A copyist at some point merely assumed Medard died with Arthur at Camlan and shortened the entry.

Or sucessive entries -

Year Xa Arthur perishes at Camlan.

Year Xb Blessed Medard rests.

Year Xc Plague in Britain and Ireland.

Again, combined to read like the AC entry we now possess.

The Domangart bit is interesting, as it hints at a connection of the Dalriadan Arthur's death with the perishing of the Welsh Annals' Arthur at Camlan.  Of course, the Dalriadan Arthur died much later than the Welsh Annal date for Camlan.  But we can't rule out that the Domangart name in the Irish Annals has something to do with the Welsh Camlan entry.  

Chichester, Cissa and Cymenesora

The Chichester Noviomagus should perhaps be discussed a bit more in the context of Arthur's Camlan.

According to English tradition, Chichester was named for the eponymous Cissa.  Cissa is introduced to us in the ASC as a warrior who fights at a place called Cymenesora near Chichester:

477

Her cuom Ęlle on Bretenlond 7 his .iii. suna, Cymen 7 Wlencing 7 Cissa, mid .iii. scipum on þa stowe þe is nemned Cymenesora, 7 þær ofslogon monige Wealas 7 sume on fleame bedrifon on þone wudu þe is genemned Andredesleage.

[https://asc.jebbo.co.uk/a/a-L.html]

I had once considered Cymenesora as a candidate for Camlan. But I dispensed with it in favor of other sites in southern England, like The Cams in Hampshire. I had come up with another etymology for Cymenesora: 

https://mistshadows.blogspot.com/2018/07/a-new-identification-for-location-of.html

But as Medard of Noviomagus in Gaul was related to Camlan, and Noviomagus was the ancient name for Chichester, and Cissa of Chichester fought at Cymenesora, we might be tempted to favor Cymenesora, i.e. Wittering, on Selsey Bill, as the Camlan being referred to in the Welsh Annals. Whatever the origin of Cymen-, the Welsh may have taken it for their own camen -

[cam2+-en] 

eb. ll. camennau.

1.  Camdra, gwyredd, tro, plyg, dolen:

crookedness, curvature, turn, bend, loop. 

Submit

14g. GDG 194, A ganodd neu â genau / O fawl o’r twrf meistrawl tau, / Gymar hwyl, gem yr heli, / Gamen môr, gymain’ â mi?

c. 1624 CRC 133, ai Daü forddwyd a bortreüwyd / fal dwy gamen kar yr ychen.

1794 P.

Digwydd hefyd mewn e. lleoedd, e.e. Ffridd Camen (Llandrillo) a Camen Fawr (Llanfyllin).

plus AS ora, 'shore, bank.'

If Cymenesora is the Camlan of the Welsh Annals, does this mean that a real Arthur perished there in 545?

No. As we have actual Camlan names in NW Wales (where the later Welsh tradition places Arthur's death site) and on Hadrian's Wall (the Camboglanna Roman fort), the Welsh Annals entry which employs St. Medard (= Medrad/Medrawd) of Noviomagus to fix Camlan at Cymenesora near Chichester has all the hallmarks of a cleric's Isidorian-style etymological identification.  Such an identification, placed as it is directly across from Cerdic's Isle of Wight, the Wessex founder's last conquest before his death, may well have been intentional. I have suggested that the chronological bracketing of the Arthur of the HISTORIA BRITTONUM corresponds perfectly with the floruit of Cerdic and that, therefore, Arthur may have been meant to be a sort of counter to the founder of Wessex

***

Camlann and the Grave of Osfran’s Son

The purpose of this essay is to prove, once and for all, where Arthur’s Camlann battle site was located. Or, more accurately, where Welsh tradition happen to place it!

It is fairly well known that the Welsh record seven survivors of Camlann. Yet, to my knowledge, no one has sought to plot these personages out on a map. To do so may help us pinpoint a geographical region in which Camlann was believed to be situated.

One of the seven – Geneid Hir – it a difficult and otherwise unknown name. P.C. Bartram (in “A Welsh Classical Dictionary: People in History and Legend up to about A.D. 1000) suggests the name may be corrupt and offers an unlikely identification with a personage named Eueyd or Euehyd Hir (often rendered Hefeydd). However, I would see in Geneid ‘Cannaid’, “white, bright, shining, pure, clean, radiant,” an epithet substituted for the original title Ceimiad, ‘Pilgrim’, of St. Elian. Elian had churches on Mon/Anglesey and in Rhos, Gwynedd.

Sandde Bryd Angel looks to be a pun for the Afon Angell, Aberangell, etc., places immediately to the south of the Camlan on the Afon Dyfi in Merionethshire.

Morfran son of Tegid is from Llyn Tegid, now Bala Lake in Gwynedd.

St. Cynfelyn is of Llancynfelyn in Ceredigion just below the Afon Dyfi.

St. Cedwyn of Llangedwyn in Powys, while somewhat further removed than the rest, is still in NW Wales.

St. Pedrog of Llanbedrog is on the Lleyn Peninsula in Gwynedd, just opposite the three Camlans in Merionethshire.

St. Derfel Gadarn is at Llandderfel near Bala Lake in Gwynedd.

Needless to say, if we “triangulate” with all these names/places, we find at the center the three

Merionethshire Camlans.

So which one is the right one?

Only one way to know for sure: we must find the Camlann that is claimed as the gravesite of Osfran’s son. This reference comes from the ‘Stanzas of the Graves:’

Bet mab Ossvran yg Camlan,

Gvydi llauer kywlavan…

The grave of Osfran’s son is at Camlan,

After many a slaughter…

[“The Black Books of Carmarthen ‘Stanzas of the

Graves’, Thomas Jones, Sir John Rhys Memorial

Lecture, 1967, Critical Text and Translation.]

While –fran of Osfran looks like Bran or ‘Raven’, the Os- does not look at all right for a Welsh name. I suspected Ys- and after a first search failed, I defaulted to bryn or ‘hill’ as the original of –bran. Thus I was looking for an Ysbryn.

And I actually found him – or, rather, it! [See “An

Inventory of the Ancient Monuments in Wales  and Monmouthshire: VI – County of Merioneth”, p. 98, RCAHMW, 1921.]

On the Mawddach River in Merionethshire there is a Foel Ispri. It used to be Moel Ysbryn and was the legendary residence of Ysbryn Gawr or Ysbryn the Giant. If we go north on the Mawddach we run into its tributary the Afon Gamlan, i.e. the Water of the Crooked Bank.

Conclusion

If an Arthur fought and died at the Afon Camlan, the only candidate worth considering is Arthur son of Pedr of Dyfed.  Yes, once again, the dates are wrong.  Arthur son of Pedr lived well after the floruit of the more famous Arthur of the 6th century.  But geopolitically speaking, we know that Dyfed was often in conflict with its northern neighbor.  A traditional MABINOGION tale, "Pwyll Prince of Dyfed", has Dyfed fighting Gwynedd, and we know historically that Dyfed took Ceredigion (later called Seisllwg) and surrounding kingdoms, combining them all into the Kingdom of Deheubarth. It would thus be logical to see in the death of an Arthur in what had been Meirionydd (the nucleus of Merionethshire) the result of a war between Dyfed and the Merionydd sub-kingdom of Gwynedd.

***

But if Camlan belongs properly to Arthur son of Pedr, albeit in an anachronistic fashion (remembering that the goal of the bracketing of Arthur in the Welsh Annals was to provide a counter for Cerdic of Wessex, whose floruit in the ANGLO-SAXON CHRONICLE perfectly that given in the Welsh Annals for Arthur), what are we to make - if anything - out of the Roman fort of Camboglanna, modern Castlesteads, on Hadrian's Wall?  This latter site of the only ancient Camlan we know of in Britain, and while we can't prove it, there is adequate justification for suggesting that L. Artorius Castus may well have been at Camboglanna.  Let's take a look at the evidence for Castus at the Castlesteads Roman fort, assuming, for the sake of argument, that I'm right about this officer serving under the Emperor Septimius Severus during the great Roman invasion of northern Britain.

Camboglanna

 Castlesteads Stone to the Emperors Severus, Caracalla and Geta

https://romaninscriptionsofbritain.org/inscriptions/1978 

Alas, the Castlesteads House estate early on, in order to construct their manor and garden, systematically destroyed the Camboglanna fort.  We have precious little remaining from it. One of the stones we do have is pasted above.  Its inscription is described as follows at the RIB page cited:

Primary

Discipu-

[l]inae

[A]ug(ustorum) (trium)

Secondary

Discipu-

[l]inae

[A]ug-

usti

Apparatus

3, 4.  avg …|vsti, Carlisle, &c.; avg⟦gg⟧|vsti, Lys., F.H.

Translation

Primary

To the Discipline of the (three) August (Emperors).

Secondary

To the Discipline of the August (Emperor).

Commentary and notes

Primary: a.d. 209-11 Severus, Caracalla, Geta; secondary: a.d. 212-17 Caracalla.

For Septimius Severus as vindex et conditor Romanae disciplinae see EE vii 353, CIL viii 17870 (ILS 446) Thamugadi.

Fortunately, we do know that substantial work was done by Severus and Caracalla at the neighboring Hadrian's Wall fort of Birdoswald/Banna.  Here are some pages from Tony Wilmott's archaeaological report on the Birdsowald excavations  of 1987-72:

Simon Elliott in SEPTIMIUS SEVERUS IN SCOTLAND also discusses the major rebuilding on both walls - Hadrian's and the Antonine - under Severus:

It does not seem reasonable to object to the notion that similar rebuilding was undertaken at Camboglanna by both Severus and Caracalla.

So might Castus have been at Camboglanna?

Absolutely.  Furthermore, he could even have fought there! From Anthony R. Birley's SEPTIMIUS SEVERUS: THE AFRICAN EMPEROR on the possible rebellion of the Brigantes under Severus:

First, urgent measures were taken in the north-west. Marius Maximus was

made governor of Belgica and Valerius Pudens of Lower Germany, replacing

Virius Lupus, who became governor of Britain. The defeated British legions

were sent back, no doubt heavily reinforced with new manpower, necessary

to restore losses—and their loyalty. Lupus faced an appalling situation. In the

absence of the Roman garrison, the northern part of the province had been

plundered by the Maeatae, probably joined by other tribes closer to Hadrian’s

Wall and some of the Brigantes within the province.

Dio reports how Lupus had to buy off the Maeatae with ‘large sums of money’, receiving ‘a few prisoners back’ in return. Clearly, in the absence of most of the British garrison in 196–7,

Rome’s northern enemies had invaded the province, and the Brigantes of

the Pennines had perhaps risen in rebellion. The Maeatae, whose home

was north of the Antonine Wall, were on the point of being joined by the

Caledonians, from the Highlands beyond, when Lupus arrived.

Thus, there is every possibility that if Castus were serving under Severus he may well have been fighting at Camboglanna or was perhaps involved in the rebuilding process.  Needless to say, the same may have been true at Burgh-By-Sands not far to the west on the Wall, which was where stood the Aballava (variant Avalana, i.e. 'Avalon') Roman fort.  I elsewhere discussed the goddess Dea Latis or 'Lake Goddess' (= the Arthurian 'Lady of the Lake'?) found at Birdoswald near Castlesteads and at Burgh-By-Sands.  As with Camboglanna as Camlan, the Aballava name is the earliest "apple place" name we encounter in the history of Britain.  Its proximity to Camboglanna is difficult to account for merely as a coincidence.

I'd even made a somewhat more fanciful case for a prototypical Grail Castle at Drumburgh just a little west of Burgh-By-Sands (https://mistshadows.blogspot.com/2024/12/a-new-theory-on-concavata-name-for.html). While I wouldn't push this idea too far, its presence in the region is interesting.

While sheer speculation, might we wonder if L. Artorius Castus had been wounded in a battle at Camboglanna, then carried clear of the sphere of action to Aballava to be healed?  

Yes, that is quite a stretch.  And not helpful in trying to determine where the original Camlan episode belongs, in terms of time, place and the personage involved.  The most that we can say, I think, is that a heroic tale featuring Castus fighting rebelling Brigantes at Camboglanna could later have been repurposed to provide a death for the Arthur of the Historia Brittonum via the Welsh Annals' entry. During the usual legend building process, Camboglanna was relocated to NW Wales, where the same place-name existed and where another known historical Arthur, viz. Arthur son of Pedr of Dyfed, could conceivably have fallen.  

THE NON-MYSTERY OF ARTHUR'S SWORD IN THE ANVIL (NOT SWORD IN THE STONE!)

The Glastonbury Cross

A great deal of ink has been spilt over the years regarding the famous Glastonbury Cross of King Arthur. I myself have contributed to that staining puddle.  But the account of the discovery of the cross at Glastonbury in the works of Gerald of Wales, combined with a couple of other medieval literary works, quite adequately explains where Robert de Boron's motif of the Sword in the Anvil (not Stone!) comes from.

In brief, other than the fraudulent discovery of Arthur's grave recorded by Gerald, we need plug in only two other sources to be able to demonstrate how the Sword in the Anvil story was created.  These sources are Geoffrey of Monmouth's HISTORY OF THE KINGS OF BRITAIN and the LIFE OF ST. EDWARD.

We may begin with the critical line from Geoffrey. It is there that the author tells us King Arthur's sword was forged on the Isle of Avalon:

Accinctus etiam Caliburno gladio optimo, et in insula Avallonis fabricato

https://la.wikisource.org/wiki/Historia_Regum_Britanniae/Liber_IX

 

The whole business of the difficult extraction of the sword from the anvil was taken from an early saint's life (1138), as described in the article The Staff in the Stone: Finding Arthur's Sword in the "Vita Sancti Edwardi" of Aelred of Rievaulx by MARSHA L. DUTTON in Arthuriana, Vol. 17, No. 3 (FALL 2007), pp. 3-30. 

https://www.jstor.org/stable/27870843

[I have suggested that Kay's involvement in the story of the extraction of the sword from the anvil may come not from a literary source, but from a folktale embedded in an ancient ritual landscape.  For more on this, please see 

https://mistshadows.blogspot.com/2021/10/the-anvil-stone-of-sir-kay-discovered.html.]

Bear in mind Geoffrey of Monmouth's work was compelted ca. 1138. Robert de Boron and Gerald of Wales have the following dates (drawn from THE NEW ARTHURIAN ENCYCLOPEDIA, ed. by Norris J. Lacy):

I think what happened (and this idea first appears in my piece 

http://www.facesofarthur.org.uk/articles/guestdan11.htm in 2008) is this:

Geoffrey of Monmouth said Caliburnus was forged on the Isle of Avalon.  Medieval tradition identified Avalon with Glastonbury.  Robert de Boron places the “Sword in the Stone” in a churchyard, and Arthur’s grave was supposedly discovered in the yard of St. Dunstan’s church.  It is the inscribed lead cross of this grave that holds the clue to unraveling the mystery of the “Sword in the Stone”.  From the account of the exhumation of Arthur at Glastonbury, by Gerald of Wales, c. 1193 (144):

Unde et crux plumbea lapide supposito, non superius ut [nostris] solet diebus, [sed] inferiori potius ex parte infixa, quam nos quoque vidimus, namque tractavimus litteras has insculptas et non eminentes et exstantes, sed magis interius ad lapidem versas, continebat

As this passage has frequented been mistranslated, I enlisted the help of Dr. David Howlett, MA, DPhil, editor of the Medieval Latin Dictionary and author of the “Dictionary of Medieval Latin from British Sources, Fasicule 5”, University of Oxford:

“Whence also a lead cross with a stone placed beneath, not further above, as is customary in [our] days, [but] rather infixed [the antecedent is feminine, so 'cross', not 'stone'] from the lower part, which we also have seen, for we have passed hands over these letters, ensculpted and not raised and outstanding, but rather turned inward toward the stone, it contained ...

There is no way one could construe this as implying that the cross was under the stone. Instead, we are to envisage an inscribed lead cross whose lower portion is infixed, i.e. thrust into, a stone.”

We thus have, in St. Dunstan's churchyard at Glastonbury/"Avalon", where according to Geoffrey of Monmouth the sword Caliburnus was forged, an inscribed cross driven into/piercing a stone - a stone which was found above the supposed tomb of King Arthur. 

Although Howlett is authoritative, I aslo checked with the Classical Latinist Dr. Roger Tomlin, himself a notable epigrapher. His take on the passage from Gerald:

I have no difficulty with this meaning of tractare. The Oxford Latin Dictionary (s.v. tracto, 2) translates it as 'To subject to the action of the hands, to handle ... to pass one's hand over, rub, stroke'.

And I agree with Howlett's translation, although I would probably write 'incised' for insculptas and 'standing out' for exstantes. The lettering was cut into the lead, not moulded in relief.

Now, this is okay so far as it goes.  And, regardless how we interpret the description of the cross in the stone, there is ample cause for us to view this account as the root source for the Arthurian sword story.  But let us take my discussion of the sword with Tomlin and Howlett to its conclusion:

Me:

How is the cross attached to the stone? I'm trying to envision a cross thrust into a stone, but with its letters incised inwards towards the stone. Doesn't much sense unless the base of the cross was stuck into the stone to fasten it, then bent over so the rest of the cross was resting, as it were, on the stone.

Tomlin:

The Latin only says that the stone was 'underneath' the cross. A freestanding cross made of lead would not be all that strong or stable, and I wonder whether it was actually set into the stone, i.e. part of the surface, but easier to inscribe than stone.

I think the cross was simply laid flat on the stone – I expect it could have been attached by solder. All this business about 'above' and 'below' refers only to its place on the stone. This would have been a long narrow rectangle, like a coffin slab or the flat top of a sarcophagus. He is saying that we would have put a cross at the top of a memorial stone, over the head, but this cross was at the bottom, over the feet.

Me:

Infix works for something soldered?

Tomlin:

It is the 'crux plumbea' which is 'infixa', which means 'fixed in'. I think 'inferius' refers to the lower part of the stone (don't say 'rock'). We might say 'set into' the stone, meaning that it was firmly attached to it. 'in' may well mean that the forger hollowed out a recess for it.

Howlett:

Shall we suppose that molten lead had been poured into the stone, in which sideways indentations ensured that the metal would not easily be prised from the stone?

Me:

Now that makes perfect sense! But this was done on the lower, rather than on the upper portion of the stone, right? 

Howlett:

When I wrote earlier of the antecedent as cross, not stone, I was thinking of petrus as masculine, but of course in Greek and Classical Latin petra is feminine, so either cross or stone could grammatically have been the antecedent. I think now the latter.

Tomlin:

Yes, it's possible that the stone itself was incised with the outline of a cross which was filled with melted lead. When this had set, level with the surface of the stone, it could have been inscribed with the epitaph.

Swords had cruciform hilts, so it was an easy matter to substitute Caliburn for the Glastonbury Cross. The extraction of the sword from the anvil (of course an anvil, as the sword had been forged in Avalon, after all!) was then concocted by borrowing the staff motif from the Life of St. Edward.  

The staff story takes place at Westminster, the site of the coronation of English kings. Its use, therefore, had implications for the crowning of Arthur as king of all Britain. In the Galfridian tradition, Arthur is crowned at Silchester.

The only other possible influence on Robert de Boron's tale is the St. Galgano sword. In 1180, the medieval Italian knight Galgano Guidotti plunged his sword into a rock when he renounced war and worldly goods to become a hermit. The abbey at Montesiepi near Siena preserves the sword in its chapel. There the hilt and some of the blade protrude from the rock in the shape of a cross. For many years the sword was thought to be a fake, but recent metal testing has determined that the alloys and style of the sword are consistent with a genuine 12th century weapon.  In addition, ground penetrating radar has shown that beneath the sword is a six and a half foot by three foot room, which is quite possibly St. Galgano’s tomb. 

If St. Galgano really dates to the 12th century, this would place a ‘Sword in the Stone” story just prior to Robert de Boron’s Arthurian version, which is dated c. 1200 A.D.  

However, an important detail is missing from the St. Galgano legend: the Italian knight’s sword does not bear an inscription, which is true of Arthur's Caliburn. And, as it all too obvious, Arthur's sword is in an anvil, not in a stone.

NOTE:

When it comes to the Arthurian Sword in the Anvil story, I have not found evidence of influence from the Alanic practice of thrusting the war-god's sword into the earth or into heaps of brushwood, an idea put forward by Dr. Linda Malcor.  Instead, all sources of Robert de Boron's story plainly derive from Britain or the Continent.  

SCYTHIANS AND THE SWORD OF MARS

Dr. Linda Malcor has long sought to associate the Scythian Sword of Area of Herodotus 4:62 with Arthur's sword. Unfortunately, there is no relationship between a sword stuck in variously a pile of sticks or the ground and the taking of Arthur's sword from an anvil on a stone. In the story of Attila and the sword, the great king of the Huns is gien the sword by a shepherd; he does not draw it forth from the cow pasture himself.  

Here is the account of Herodotus:

Such is their way of sacrificing to all other gods and such are the beasts offered; but their sacrifices to Ares are on this wise. Every district in each of the governments has in it a building sacred to Ares, to wit, a pile of fagots of sticks three furlongs broad and long, but of a less height, on the top of which there is a flattened four-sided surface; three of its sides are sheer, but the fourth can be ascended. In every year an hundred and fifty wagon-loads of sticks are heaped upon this; for the storms of winter ever make it sink down. On this pile there is set for each people an ancient scimitar of iron, which is their image of Ares; to this scimitar they bring yearly sacrifice of sheep and goats and horses, offering to these symbols even more than they do to the other gods. Of all their enemies that they take alive, they sacrifice one man in every hundred, not according to their fashion of sacrificing sheep and goats, but differently. They pour wine on the men's heads and cut their throats over a vessel; then they carry the blood up on to the pile of sticks and pour it on the scimitar. So they carry the blood aloft, but below by the sacred building they cut off the slain men's right arms and hands and throw these into the air, and presently depart when they have sacrificed the rest of the victims; the arm lies where it has fallen, and the body apart from it.

Ammianus condenses the account considerably in Book 31:2 -

No temple or shrine is to be found among them, not so much as a hut thatched with straw, but their savage custom is to stick a naked sword in the earth and worship it as the god of war, the presiding deity of the regions over which they range

A still later account is found in Jordanes, where we are told of Attila the Hun finding the weapon:

And though his temper was such that he always had great self-confidence, yet his assurance was increased by finding the sword of Mars, always esteemed sacred among the kings of the Scythians. The historian Priscus says it was discovered under the following circumstances: "When a certain shepherd beheld one heifer of his flock limping and could find no cause for this wound, he anxiously followed the trail of blood and at length came to a sword it had unwittingly trampled while nibbling the grass. He dug it up and took it straight to Attila. He rejoiced at this gift and, being ambitious, thought he had been appointed ruler of the whole world, and that through the sword of Mars supremacy in all wars was assured to him.

My Talk with Richard Johnson of O'r Golwg

https://youtu.be/ICS0cyB8bqg?si=svRYaSBheG4p39ZK

THE PROBLEM OF ARTHUR'S 'CITY OF THE LEGION' BATTLE, WHICH IS NOT REALLY A PROBLEM AT ALL!

https://www.youtube.com/watch?v=ICS0cyB8bqg

In response to my friend and fellow Arthurian researcher/author Tony Sullivan's preference for seeing Arthur's City of the Legion battle as the either the Welsh Caerleon or Chester, I posted the following in the comments section of his YouTube channel:

We have a historically attested Artorius at York. We can't independently verify an Arthur at Caerleon or Chester. I suppose I would simplify this by saying "Why bother looking for an Arthur at a Welsh City of the Legion, where Saxon activity is highly doubtful in the sub-Roman period we are discussing, when we already have one at Roman York?" In other words, we don't have to prove the existence of the York Arthur. A Welsh City of the Legion Arthur is, at present, unprovable. 

I suppose a good way to look at this is thusly: as we already have a historically attested Arthur at a legionary fortress, how likely is it that we have another Arthur at another legionary fort? I'd say odds are much against such a coincidence occurring. That's looking at the problem from a purely logical standpoint, of course.

We can instead opt for allowing the city of the legion to be merely an intrusion into the Arthur battle list of the famous Chester battle fought c. 616. Nothing wrong with that, if we accept the premise that the battle list is a fictional creation, in whole or in part.

Now, that's all pretty simple and I would here like to elaborate a bit on the points I made.

First, Arthur's urbs legionis battle must be examined in the context of the other battles.  To fail to do this, to choose to instead see it as some kind of odd outlier, is not a good approach to the problem, in my opinion.

I had pointed out that there are indications in the Welsh sources that York was known to have housed a legion.  These involved the Eliffer/Eleutherius of the Great Retinue (a legion?) who seems to be situated there, based upon the fact that his son Peredur (which I think is from the Latin Praetor, and not from a Celtic *Pritorix) is later made a son of Ebrauc, the Welsh eponym for the city of York. 

But we can see from the maps above that the association of York on Dere Street with the other battles on Dere Street clearly favor "Caer Ebauc" as Arthur's legionary city.  This is common sense from the geographical standpoint. 

I will repeat my strong doubt that either of the Welsh legionary cities could have been the site of a major battle against the Saxons in Arthur's time (supposedly somewhere around 516-537).  To the best of our knowledge, Wales at the time was subjected only to Irish infiltration.  We know that several Irish kingdoms were established in Wales when the Romans withdrew. Some of these kingdoms are of common knowledge.  Not generally accepted (yet) are my findings that show even the great Cunedda came from Ireland (Drumanagh) and not from Manau Gododdin in the extreme North. 

We have absolutely no knowledge of Saxon raids upon Wales during Arthur's supposed floruit.  None. Zero. Zilch.  And until someone produces such evidence, we have no reason to subscribe to the notion that these kinds of raids actually occurred.

Could the English have made a deep, penetrative invasion of either Chester or Caerleon from the east during Arthur's time?  Well, maybe.  It's possible.  But when we look at the range maps of Saxon settlement at the time (shown by the presence of Saxon cemeteries, here from the maps in Higham, drawn from Hines), it's hard to accept as a possibility that the English would have dared to invade or even raid so far into British territory.  

What it comes down to is this: if you are going to accept that we do, in fact, have an Arthur (= Lucius Artorius Castus) at the legionary city of York, why are you trying to force an Arthur into one of the Welsh Cities of the Legions?  Are you not simply doing what Geoffrey of Monmouth did, a writer of fantastic fiction who chose Caerleon for Arthur's plenary court? Is it not more reasonable to allow for the probability that York as Artorius' City of the Legion was relocated to Wales, to either Caerleon or Chester, once York had become English?  Is is not more sensible to see in this relocation of York to a Welsh City of the Legion as one of several known examples of the relocation of places or people from regions long conquered by the enemy to the Celtic Fringe?

In conclusion, I think it is a naive argument to assume that the only legionary cities known to the Welsh are Chester and Caerleon. While this may be true of the time period to which the documents naming the two Welsh legionary cities belong, prior to the confinement of such traditions to Wales (part of the Celtic fringe), it is scarcely likely that there wasn't knowledge of other legionary forts within the conquered region of England. 

Thus the problem of Arthur's 'City of the Legion' battle is not really a problem at all - unless, of course, we want to make it into one.  And I don't think we are justified in doing so.

MY READERS' PREFERENCE: AS EVERYTHING IS GUESSWORK ANYWAY, WHY NOT GO WITH ARTHUR SON OF SAWYL?

           Ribchester Parade Helmet

My regular readers have urged me, almost uniformly, to adopt the Sawyl Benisel of Ribchester = Uther Pendragon theory. I've thought about this for quite awhile, yet am still not sure how I feel about.

My problem with this particular Arthurian reconstruction is nicely summarized in a recent blog piece:

https://mistshadows.blogspot.com/2026/02/eliwlad-flies-again-or-new-solution-to.html?m=1

Basically, there are a couple good alternative purely Welsh etymologies for the personal name Eliwlad that cast some doubt on my deriving that name from a loan-translation via the Irish. Secondly, we are left wondering why it was thought necessary to substitute Uther Pendragon, a presumed poetic metaphor, for the name + epithet Sawyl Benisel. Finally, although Sawyl is the best available emendation for kawyl of the MARWNAT VTHYR PEN, it is absolutely possible the name occurs there merely as a poetic metaphor and that a later storyteller chose to fancifully associate Sawyl Benisel and his son Madog Ailithir with Uther.

None of those problems are in and of themselves insurmountable. But, if we accept Uther as Sawyl Benisel (for no other reason than that we can!), we still have to grapple with a sub-Roman Arthur who originates at Ribchester fighting battles that look suspiciously like Roman military engagements.

               The Arthurian Battles

However, there is a sort of work-around to this problem. The HISTORIA BRITTONUM tells us that Arthur was fighting against the English along the the kings of the Britains. This phrase has been subject to various interpretations. 

The most obvious - given that Arthur was not a king (although he might well have been a royal son who did not outlive his father and hence did not succeed to the throne) - has him serving as a popular and successful war-leader over a force composed of warriors drawn from allied tribes/kingdoms. We might assume in this scenario that Sawyl's kingdom was at the time the most powerful or at least influential in the Northern region.

Support for this view can be found in Alex Woolf's proposition that Bede's "Caedualla rex Brettonum" was Sawyl's grandson Cadwallon Lyw:

https://mistshadows.blogspot.com/2020/01/cadguallaun-liu-grandson-of-sawyl.html?m=1

A variation on this theme would be to see Arthur as a sort of mercenary captain employed by various kingdoms on the front line during the war against the English.

A third option would be to recognize Arthur as a Dark Age attempt to preserve in some form the office of Dux Britannarium, although such a man, one would think, should belong to York. It is true that the Ribchester Roman fort enjoyed a close relationship with York.

I think we can dispense with the Dux Brit. idea pretty quickly. The late Roman period governor of Northern Britain did not fight its enemies with client British kingdoms. The Roman army fought on its own and did not employ native troops in its campaigns.

I've discussed this possibility with Dr. Roger Tomlin and he assures me the description given of Arthur in the HB does not accord well with role and function of the Dux Brit.

A mercenary Arthur has its appeal, I suppose, but necessitates that two conditions held true. First, that the northern British kingdoms found themselves in need of mercenaries. And, second, that a nobly born son of one of those kingdoms would have taken on the mantle of a mercenary captain. I'm not sure either condition1 can be justified.

In my mind, we could have a military leader of a kingdom like that of Sawyl's that held a preeminent position among the Northern tribes. He may well have led an army comprised of what had once been the Brigantian Confederacy and affiliated tribes (like his own Segantii), as well as the Votadini (Gododdin and Manau Gododdin). While I once had great difficulty with the battles on the edge of Manau Gododdin and farther north (as these seemed only credible against a non-Saxon enemy and belonged better with either L. Artorius Castus or Arthur of Dalriada), there is really no reason to deny them to a Segantii general leading allied British troops at or just north of the Antonine Wall. This is especially so as in the next generation we find a Gododdin army coming all the way down to Catterick to fight the English. And a group of allied British kings that included Urien of Rheged, Rhydderch of Strathclyde and Morgan of the Tyne Valley were fighting the same enemy together. Urien was betrayed by Morgan near Lindisfarne.

A huge plus in favor of an Arthur son of Sawyl is the proximity to Ribchester of an excellent Badon candidate, viz. Buxton. And there is nothing wrong with a death at Camlan. We might compare this with the death of Peredur and Gwrci, sons of Eliffer of York, at Carrawburgh in the Wall.

We might compare an Arthur at Ribchester with another theory, i.e. the one that places him at Stanwix, the Uxellodunum Roman fort near the western end of Hadrian's Wall. Although the place was called Arthur's fort in the 18th century, and it was garrisoned by the largest cavalry force in all Britain, we have no genealogical trace putting him there. Ribchester was garrisoned by Sarmatian cavalry. 

I've hypothesized a connection between the Ala Petriana at Stanwix and Petr of horse-loving Dyfed, and a similar association between the Petriana and Alclud/Petra Cloithe and the stone Arthur son of Bicoir used to slay the Irish prince Mongan. I've pointed out the robust equine properties of Dalriada as well.

Sure, Petrianis (a nickname for Uxellodunum) is beautifully situated geographically. It is close to Aballava/Avalona/"Avalon" with its Dea Latis or Lake Goddess, a possible prototypical Grail Castle (Drumburgh/Concavata) and to Camboglanna/Camlan.

BUT... once again, there is no Arthurian genealogical link to Petrianis. 

The only extant non-Galfridian geneaology for the famous Arthur seems to link our hero with Sawyl Benisel of the North. Our only alternative is to accept Uther Pendragon as a figure conjured from the "mil uathmar... chend" of the Irish COMPERT MONGAN. And if we have no real father for Arthur, then we're unable to place the latter anywhere.  Our failure to be able to do that, in turn, adds weight to the theory that there may not have been an original sub-Roman Arthur. 

While there is no evidence the Sarmatians had their own ethnic draco standard (see https://mistshadows.blogspot.com/2024/02/there-is-absolutely-no-evidence-for.html?m=1), the fact that the Alans and Dacians did would favor the presence of a similar totem among the Iazyges in Britain. If this supposition is provisionally accepted, it may, in fact, be the case that Sawyl as a great war-leader was referred to as the Terrible Chief-dragon.

CONCLUSION:

While I suppose it is possible that the Arthurian battles could belong to a son of Sawyl Benisel, the easier and more plausible explanation still remains that the HB list (minus Badon) represents the martial activity of L. Artorius Castus.