29 September, 2010

King Arthur: The Bloody Cup, by MK Hume. Book review

Headline, 2010. ISBN 978-0-7553-4871-8. 526 pages. Review copy supplied by publisher.

Set in post-Roman Britain some time in the fifth or sixth century, King Arthur: The Bloody Cup is the third part of a trilogy retelling the Arthurian legend. Many of the main characters are familiar figures from the legends, including Artor (King Arthur), Wenhaver (Guinevere), Gawayne (Sir Gawain), Percivale (Sir Percival), Galahad, Bedwyr (Sir Bedivere), Nimue, Artor’s half-sisters Morgan and Morgause, and Morgause’s son Modred (Mordred). Other characters are fictional, such as Artor’s bodyguard Odin and spy chief Gruffydd, and the villains Pebr and Gronw.

Artor has ruled as High King of Britain from his citadel on Cadbury Tor for many years and is now growing old. He has no legitimate heir as his wife Wenhaver is barren, and his court has grown corrupt and decadent. Artor’s enemies sense that he is growing weak. Three mysterious figures hatch a plot to steal the sacred cup once owned by the saintly Bishop Lucius of Glastonbury, claim that it once belonged to the goddess Ceridwen and use it as a symbol to provoke a rebellion against Artor. The struggle for possession of the cup, and a mysterious spear, threatens Artor’s friends, his kingdom and his life.

I reviewed the previous book in the series, King Arthur: Warrior of the West, in January 2010, and concluded then that it wasn’t for me. The publishers sent me a copy of the third instalment without asking first, and I read it partly out of curiosity to see if the loose ends from Book Two were resolved and partly to see if I got on better with the style on further acquaintance. The answers are ‘sort of’ and ‘no’, respectively. This is still not a series for me.

On the plus side, it was quite fun to spot bits of the medieval legends – e.g. the Trystan-Isolde-King Mark love triangle makes a brief appearance, transplanted to north-east Wales instead of the traditional Cornwall – and the steady attrition of Artor’s friends and potential heirs has a certain poignancy. On the other hand, the corruption and decadence of Artor’s court is so strongly emphasised that it is not obvious why the reader is supposed to be worried when it is threatened. If Artor’s court is full of lies, vanity and backbiting courtiers trying to stab each other in the back while living in the lap of decadent luxury, it’s hard to suppress a niggling thought that a different set up might not be noticeably worse. The tragic grandeur of the Arthurian legend – a good king brought down by lesser men and by his own flaws – seemed to me to be missing from this retelling.

However, the main reason I did not get on with the novel was the same as last time; I found the writing style reminiscent of academic prose. Maybe the intention is to create an archaic flavour (although modern slang such as “gumption”, “sodding thing”, “shite” tends to work against this), but to me it seemed lifeless, especially the dialogue. People speak in grammatically correct complete sentences even when being tortured or when mortally wounded, and everyone sounds much the same.

Place names are a mix of Roman names, e.g. Ratae (modern Leicester), Verterae (modern Brough, Cumbria), and modern names with Old English elements, e.g. Glastonbury, Cadbury. If there is a pattern to the mix it wasn’t clear to me. Similarly, although the ‘Saxons’ are treated throughout as an utterly alien enemy to Artor’s realm (they don’t make an on-stage appearance), some of the characters in Artor’s kingdom have Old English names, such as the saintly Bishop Aethelthred* the Pure of Glastonbury. This struck me as potentially intriguing; does the presence of Old English names indicate that some ‘Saxons’ were acceptable in Artor’s realm, implying a degree of co-operation or integration, and if so, how is this reconciled with the fact that everyone at Artor’s court apparently regards the ‘Saxons’ as the enemy? Was Bishop Aethelthred a ‘Saxon’ immigrant, and if so how did he come to be the revered head of the greatest Christian monastery in Artor’s realm? As far as I could see this was never touched on, unless I missed it somehow, and it contributed to a general impression of unreality about the setting. This perhaps doesn’t matter, since the medieval Arthurian legends are set in a ‘once-upon-a-time’ setting, and this novel is perhaps best read in the same light.

A partial character list at the front of the book helps with keeping some of the large cast straight, and is especially useful for characters who played a role in the earlier books but who are now dead. However, not everyone is listed (e.g. Taliesin and Modred are missing, as are the villains Gronw and Pebr), so I still had to jot down notes. A glossary of place names at the back of the book matches some of the place names in the novel with their modern equivalents, but it is incomplete – I got confused between Salinae and Salinae Minor and looked in the glossary for clarification, but only Salinae is listed. Readers who like to follow the characters’ journeys on a modern map may find they have to keep notes.

Final part of a trilogy retelling the Arthurian legends, but not a book for me.

*I’m not sure if this is a typo for Aethelred, as Aethelthred or Aethelthryth was a female name, or if I have missed something subtle.

22 September, 2010

Kings of Lindsey

Lindsey was an early medieval kingdom south of the Humber estuary, in roughly the area of north Lincolnshire. See sketch map for approximate location. The name was recorded by Bede in his Ecclesiastical History, written in 731, as ‘prouinciae Lindissi’ or ‘the province of Lindsey’ (Book II Ch. 16). What do we know about its rulers?



The genealogy of the kings of Lindsey is given in the Anglian Collection, as follows:

Godwulf Geating
--Anglian Collection genealogies, available online

Each king is given a patronymic derived from the name of his predecessor, e.g. Finn Godwulfing, Frithulf Finning, and so on. (I have missed the patronymics out of the list above for ease of reading). Presumably whoever compiled the genealogy believed, or wished to indicate, father-to-son succession. This may have been accurate, or it may result from a scribe misinterpreting a king list as a genealogy, or a bit of both.

Place name

There is a village called Winteringham, which could mean ‘settlement of the followers of Winta’, on the south shore of the Humber estuary, near the Roman road of Ermine Street. Naturally, even if the place name is derived from the personal name Winta, it need not be the same Winta as the one listed in the kings’ genealogy. However, it may indicate a person of some authority with the right name in the right area, and is thus consistent with the genealogy.


Bede does not mention any kings of Lindsey by name. In his Ecclesiastical History he describes Paulinus, Bishop of Northumbria in 625-633, baptising the ‘prefect’ of the city of Lindocolina (modern Lincoln) and conducting a mass baptism in the River Trent in the presence of King Eadwine (Edwin) of Deira/Northumbria.

Paulinus also preached the word of God to the province of Lindsey, which lies immediately south of the Humber and extends to the sea. His first convert was Blaecca, Reeve of the city of Lincoln, with all his family.
The priest Deda […] told me that one of the oldest inhabitants had described to him how he and many others had been baptised by Paulinus in the presence of King Edwin, and how the ceremony took place at noon in the River Trent…
--Bede, Ecclesiastical History, Book II Ch. 16.

This may suggest that the king of Northumbria had some political authority in Lindsey at this period, since the Northumbrian bishop was active in Lindsey. The nature of the relationship is uncertain; it may represent Northumbrian dominance over Lindsey, or a friendly relationship with the new bishop of Northumbria ministering to an allied neighbouring territory that did not (yet) have a bishop of its own.

At a slightly later period there is clear evidence of hostility and Northumbrian dominance, when the monks of Bardney in Lindsey initially refused to allow (some of) the bones of King Oswald of Northumbria to be buried there after his death in 642.

In the province of Lindsey there is a noble monastery called Beardaneu, which was greatly loved, favoured and enriched by the queen [Osthryth, daughter of Oswy of Northumbria and thus Oswald’s niece] and her husband Ethelred [of Mercia]. She wished that the honoured bones of her uncle be reinterred there. But when the waggon carrying the bones arrived towards evening at the abbey, the monks were reluctant to admit it; for although they acknowledged Oswald’s holiness, they were influenced by old prejudices against him even after his death, because he originally came from a different province and had ruled them as an alien king.
--Bede, Ecclesiastical History Book III Ch. 11



None of the names in the Lindsey kings’ genealogy appear in other sources, so none of them can be independently dated. There may be a couple of slight clues:

  • Biscop = ‘bishop’, and this is a name that is perhaps most likely to have been given after the kingdom’s conversion to Christianity. If so, this would suggest a date some time after Paulinus made his ‘first convert’. The chapter in Bede’s history describing Paulinus’ activity in Lindsey is headed AD 628. As Paulinus came to Northumbria in 625 with Eadwine’s Kentish wife Aethelburh (Bede, Book II Ch. 9), and left Northumbria for good after Eadwine’s death in October 633 (Bede Book II Ch. 20), his activity in Lindsey must fall between these dates even if he made more than one visit. This would place Biscop somewhere after this period.

  • Paulinus’ first convert held a position of authority in the city of Lincoln. Bede describes him as a ‘prefect’, and Leo Sherley-Price translates this as ‘reeve’. He was clearly a person of considerable importance. His name was Blaecca, according to Bede, which alliterates with the names Beda and Bubba in the genealogy. If he was related to these kings, that would be consistent with (but is not proved by) the name alliteration and his position of authority. This would be consistent with dating the B- kings to a period not too far from 625-633, which would also be consistent with the tentative dating for Biscop suggested above.

If we assign Beda, Bubba and Biscop to somewhere around the middle two quarters of the seventh century, at and after Paulinus’ activity in Lindsey, then the admittedly inexact method of counting generations would place their immediate predecessor Caedbaed somewhere in the early seventh century*, Winta somewhere in the early sixth century, and the last name in the list somewhere around the mid-eighth century. It should go without saying that this is very tentative.


Caedbaed contains the common Brittonic name element Cad- or Caed-, meaning ‘battle’. If the name is genuine and represents a genuine ruler of Lindsey, it may indicate Brittonic familial or political connections.

The Bardney incident tells us that Oswald held royal authority over Lindsey, or at least the monks of Bardney considered that he did and resented him for it. Yet we have a separate list of Kings of Lindsey in the Anglian Collection genealogies. How to reconcile these? Three possibilities come to mind:

  • All the kings in the genealogy ruled before the time of King Oswald (and perhaps before his predecessor King Edwin). Since we don’t know the dates of any of the kings in the list, this is possible. However, even if we disregard Woden and the preceding names as mythological, there are 11 names in the list, which is rather a lot to get through before the 620s or 630s;

  • Oswald (and perhaps Edwin) represent an interruption to the rule of Lindsey’s own kings, perhaps reflecting a temporary period of military conquest. This would be consistent with the monks’ hostility to Oswald;

  • Lindsey was at least temporarily a sub-kingdom or client kingdom under the political authority of the Northumbrian kings – voluntarily or by coercion – but retaining a line of sub-kings or client kings of its own.

If Lindsey was (temporarily?) a sub-kingdom of Northumbria, with its own kings who were subject to some degree of political control from the Northumbrian kings, this would be consistent both with Bede’s mention of Northumbrian authority in Lindsey and with the presence of a genealogy for kings of Lindsey. There is no particular reason why a line of sub-kings or client kings could not have continued for several generations under foreign overlordship. Or if the genealogy is in fact a king-list, some or all of the names on it may represent a succession of client kings installed by an over-king. If Lindsey’s status as a sub-kingdom was based partly or wholly on force or threat, rather than voluntary alliance, this would also be consistent with the resentment displayed by the monks of Bardney.

The hostile reaction of the monks at Bardney to King Oswald’s body suggests a strong and definite dislike. This could have been a personal objection to King Oswald himself as a result of some real or perceived insult or injury. However, Bede says that the monks objected to him because he came from a different province and had ruled them as an alien king. Since the monks’ hostility to Oswald was purely territorial according to Bede, it is a reasonable inference that the same attitude would have applied to any Northumbrian king ruling over them. (Caveat that Bede was a Northumbrian himself and a great admirer of King Oswald, so if Oswald had committed some injustice against Lindsey or the Bardney monastery it is possible that Bede might have chosen not to mention it). How far the opinion held by the monks of Bardney was representative of the rest of Lindsey’s population is unknown.

It is worth noting that the monks of Bardney evidently felt confident that they could snub Oswald (and the wishes of Oswald’s niece, Osthryth) without fear of catastrophic reprisals. This is consistent with the incident having occurred shortly after Oswald’s death in 642, when Northumbrian power would have been severely weakened by his defeat. Given that Oswald had been killed in battle by the Mercian king Penda, it is possible that the monks of Bardney were siding publicly with Mercia by insulting the Mercians’ defeated opponent. This could reflect straightforward pragmatism, since Mercian power would have been in the ascendancy in the Midlands region after Penda’s victory over Oswald, and it is usually good politics to be in favour with the winning side. Or it may reflect a preference for the Mercian kings over the Northumbrian kings. Lindsey eventually ended up as part of the later kingdom of Mercia.

The limited evidence seems to me to be consistent with a picture of Lindsey as a client kingdom with its own line of kings but subject to Northumbrian political authority in the early to mid seventh century. If the attitude of the monks of Bardney is representative, Northumbrian political authority over Lindsey may have been obtained by force or threat and resented in Lindsey. As usual, other interpretations are possible.


Anglian Collection genealogies, available online
Bede. Ecclesiastical History of the English People. Translated by Leo Sherley-Price, Penguin Classics, 1968, ISBN 0-14-044565-X

Map links
Winteringham (scroll south to pick up the line of Ermine Street on the map)

*In Paths of Exile I made Caedbaed of Lindsey an older contemporary of Eadwine, ruling in Lindsey in 605 AD. This is the rationale.

15 September, 2010

Hawk of May, by Gillian Bradshaw. Book review

First published 1980. Edition reviewed: Sourcebooks, 2010, ISBN 978-1402240706, 356 pages. Uncorrected advance review copy supplied by publisher.

Hawk of May is the first part of a fantasy trilogy retelling the Arthurian legends, focussing on Gwalchmai as the central character. Gwalchmai translates literally as “Hawk of May”, hence the title, and in later legend he becomes the character Sir Gawain. Other key figures in the legend feature as major characters – Arthur, his evil sorceress sister Morgause, her sons Agravain and Medraut (Mordred), Bedwyr (Sir Bedivere) and Cei (Sir Kay). Gwenhwyfar (Guinevere), not yet Arthur’s wife, gets a walk-on part near the end, and will no doubt reappear in the later books. The historical king of the West Saxons, Cerdic, makes an appearance. So do some other figures from the scanty historical records, such as Maelgwn Gwynedd and Urien Rheged, although they are displaced in time by half a century or more from their actual positions in the mid to late sixth century. The setting for Hawk of May is post-Roman Britain at approximately the end of the fifth century, taking the dates for Arthur’s major battles from Annales Cambriae and for Cerdic’s reign from the Anglo-Saxon Chronicle. However, as the author’s note says, “...since the novel is only partially historical, geography is not that important.” and neither is chronology. The novel works best when read as a story set in the timeless world of “once upon a time”, rather like the medieval Arthurian legends themselves.

Gwalchmai is the second son of Queen Morgause and her husband King Lot of the Orkney Islands. To his father’s disappointment, he shows no noticeable talent as a warrior, although he is a skilled horseman and harpist. Bullied by his elder brother Agravain, Gwalchmai leads a lonely existence until his beautiful mother, whom he worships, offers to teach him reading and, later, black magic. After witnessing some of its cruelties, he comes to fear and hate sorcery, renounces it, and after an adventure in the Otherworld he comes into possession of a magic sword and the skills to wield it. Magically returned to the real world in southern Britain, Gwalchmai sets off to offer his services to Arthur – but Arthur has his own dark reasons to hate and mistrust Morgause’s son. Will Gwalchmai ever persuade Arthur to accept him, and will either escape the shadow of Morgause’s evil magic?

Hawk of May is a fantasy novel, centred on a supernatural conflict between the forces of good (the Light) and evil (the Darkness). Gwalchmai undertakes a supernatural journey on a magic boat to the Otherworld, where he obtains a magic sword and later acquires a fairy horse. He has superhuman strength in battle, and has to physically fight and kill at least one real demon. The magical elements are key to the plot, whereas the approximate historical setting in somewhere in post-Roman Britain is incidental.

Within this fantasy environment, Hawk of May is a coming-of-age story, as the young Gwalchmai has to break free of his mother’s influence, make his own choices and earn a place for himself in the world. The plot mainly follows his upbringing and the circumstances that bring him to Arthur’s warband, so is fairly slight. Perhaps this reflects the book’s position as the first in a trilogy, setting up characters and situations for the novels to come.

Characterisation is effective, with most of the major players clearly drawn as individuals. Gwalchmai is endearingly humble, ever ready to attribute his battle success to supernatural favour rather than to his own prowess as a warrior. He grows from a child to a young man without losing his youthful idealism. Arthur, as portrayed here, is a charismatic battle leader, human enough to win his followers’ affection as well as their admiration. I can see why men would have been drawn to fight and die for this Arthur (something that isn’t always apparent in Arthurian fiction). Among the secondary characters, Cei and Agravain are archetypal ‘Celtic’ warriors, boastful, quarrelsome, flamboyant, cheerful and always ready for a drink or a fight, preferably both. Bedwyr is an intellectual as well as a warrior, with an interest in philosophy and a disinclination to take sides in petty quarrels. Morgause is pretty much pure evil, but given her traditional role in the legend it might have been rather tricky to make her a nuanced character. Gwenhwyfar is attractive and realistic, as far as I can tell from her very brief appearance, which bodes well for the rest of the series (assuming it is going to develop along the traditional lines).

Fantasy retelling of the Arthurian legend from the perspective of Gwalchmai, describing how he came to Arthur’s following as a young man.

12 September, 2010

September recipe: Ratatouille

This is a delectable vegetable stew that makes full use of late summer vegetables – sweet peppers, courgettes, tomatoes, aubergines. Serve it with grilled pork or lamb chops, or add cooked haricot beans and serve with fresh bread for a vegetarian meal. It freezes well, so you can make it in the late summer when the fresh ingredients are plentiful and thaw it for a reminder of summer in the middle of winter.

Ratatouille (serves 4)

1 lb (approx 450 g) courgette* (green or yellow or a mixture)
1 lb (approx 450 g) aubergine**
4 Tablespoons (4 x 15 ml spoon) olive oil
1 onion
3 large cloves garlic
1 red pepper
1 lb (approx 450 g) tomatoes
1 Tablespoon (1 x 15 ml spoon) fresh basil (or other herbs of your choice)

Chop the courgette and aubergine into approximately 2 cm cubes. Sprinkle with salt and leave for 30 minutes. Then rinse the courgette and aubergine in a bowl of cold water to wash off the salt, and drain on kitchen paper.

Peel and chop the onion. Crush the garlic. Remove the seeds from the red pepper and chop.

Chop the tomatoes. You can peel them if you like, but I never do. Tinned tomatoes are not as nice in this recipe as fresh ones, so use fresh tomatoes if at all possible.

Fry the onion and garlic gently in the olive oil in a large saucepan until softened and starting to colour. Add the courgette, aubergine and red pepper.

Put a lid on the pan and simmer over a low heat for about 20 minutes, stirring from time to time.

Add the chopped tomatoes and chopped herbs, and season with salt and black pepper to taste. Stir well.

Simmer for another 30 minutes or so over a low heat until the vegetables are almost disintegrating.

Serve with grilled lamb or pork chops, or add cooked butter beans (or haricot beans or canellini beans) for a vegetarian casserole. It goes particularly well with fresh bread to mop up the juices. If using beans, approximately 1.5 oz (approx 40 g) per person of dried beans is about right, soaked overnight and then cooked for about an hour in boiling water.

*Also called zucchini
**Also called eggplant

03 September, 2010

The Anglian Tower, York

The ‘Anglian Tower’ is a small stone tower in the north-west wall of the Roman legionary fortress at York. It is near the west corner, about 60 m north-east of the surviving Multangular Tower at the west corner. The name ‘Anglian Tower’ was bestowed when the tower was opened for public display after the 1970 excavation, but it owes more to confidence than evidence, since the tower itself is not securely dated. What do we know about it?

Picture from Wikipedia under Creative Commons licence


The Anglian Tower was first discovered in 1842, when a route was being constructed through the city defences to allow access to St Leonard’s Yard (Tweddle et al 1999, p. 216). It had been buried inside an earth bank that had been constructed on top of the original Roman defences. A later re-excavation was carried out in 1970.



The Anglian Tower is rectangular, about 3.6 m by 3.1 m, and the standing remains are about 4.7 m high (Tweddle et al 1999, p. 189). The walls are 0.45 to 0.6 m thick, (Tweddle et al 1999, p. 251). It has two doorways, one in each side, supported by simple arches. The original roof was replaced by a brick vault in the nineteenth century (Ottaway 2004, p.142).

The front of the tower is set into a shallow cut in the top of the Roman fortress wall, and the back wall of the tower stands on the Roman rampart (Ottaway 2004, p.142).

The tower is constructed of roughly cut and shaped blocks of oolitic limestone (Ottaway 2004, p.142). This is a different type of limestone to the magnesian limestone used to construct the Roman walls, and the masonry technique is also different.


The 1970 excavation did not find any direct dating evidence associated with the Anglian Tower (Tweddle et al 1999, p. 189). The date range for the tower’s construction is book-ended by stratigraphy:

  • Since the Anglian tower stands on top of the Roman defences and is inserted into a cut in the wall, it must post-date the construction of the Roman wall;

  • The Anglian Tower was buried inside an earth rampart strengthened with rough stonework, which was constructed on top of the Roman defences. So it must pre-date the construction of this rampart.

The walls of the Roman legionary fortress at York were rebuilt and repaired at several stages in the life of the fortress. Projecting towers like the Multangular Tower are more commonly seen in later Roman military architecture of the later third and fourth centuries, like the Roman shore forts in Britain. So on stylistic grounds York’s Roman walls have sometimes been dated to the early fourth century, perhaps associated with the Emperor Constantine (the Great) who was proclaimed Emperor by his troops in York in 306 AD. However, the abundant pottery in the associated rampart consistently dates to the late second or early third century (Ottaway 2004, p.75). On the basis of the pottery, Patrick Ottaway considers that the defences had been rebuilt in stone and the Multangular Tower constructed by, or shortly after, the reign of Emperor Septimius Severus (193 – 211 AD), perhaps to mark the Emperor’s stay in York in 208-211 (Ottaway 2004, p. 75).

The stone-reinforced earth rampart on top of the Anglian Tower contained a few sherds of eighth- to ninth-century pottery (Ottaway 2004, p.142). A plausible context for its construction could be as a refurbishment of the defences at around the time the Vikings took York in 867 (Tweddle et al 1999, p.189).

So, these indicate that the Anglian Tower was built at some time after the completion of the Roman stone walls in the early third century (or possibly the early fourth century), and at some time before the stone-reinforced earth bank was constructed in approximately the middle of the ninth century.


The Anglian Tower is built of blocks of oolitic limestone. This stone occurs in flat slabs, and is found in the North York Moors north of York and in the Yorkshire Wolds east of York (Ottaway 2004, p.53-54). It was commonly used in Roman York but mainly in the Roman colonia (civilian city) on the west bank of the River Ouse. A different type of limestone, magnesian limestone, was used extensively in the Roman military fortress. Magnesian limestone is a higher quality building stone than oolitic limestone because it can be cut in any direction and to any size and shape, and is the limestone used to make the facing blocks for the surviving stretches of Roman fortress wall. It comes from the Tadcaster area, west of York.

The other notable feature of the stone used to build the Anglian Tower is that it appears to have been freshly quarried. It is not re-used stone from earlier Roman buildings. In this it contrasts sharply with later buildings such as York’s medieval churches, some of which were almost entirely constructed from re-used Roman stone (e.g the tower of St Mary Bishophill Junior, built in the eleventh century; Ottaway 2004, p.151).

Three inferences can be drawn from the stone used to construct the Anglian Tower:
  • At the time it was built, Roman York was not yet regarded as the giant stone quarry it became in the Anglo-Scandinavian and medieval periods. Either the Roman buildings were still in use, or they were still respected, or whoever built the Anglian Tower lacked the technology to demolish them safely;

  • Whoever built the Anglian Tower did not use the same materials or construction techniques as the Roman military had used to build the fortress defences, implying a break in construction methods. This in turn could imply that the necessary knowledge had been lost, or that access to the Tadcaster quarries was no longer possible, or a change in fashion, or some more prosaic reason such as a builders’ merchant happening to have a load of oolitic limestone ready to hand at the time of construction;

  • Whoever built the Anglian Tower had sufficient knowledge of stone construction methods to build a sturdy tower – it hadn’t fallen down by the time it was buried in the overlying ninth-century earth rampart - and sufficient resources available to obtain the stone and carry out the construction. They also regarded York as sufficiently important to be worth the effort of building the tower.

A late Roman date is consistent with the lack of re-used stone in the Anglian Tower, since the fortress buildings would presumably have still been in use and would not have been available as a convenient stone quarry. ‘Consistent with’, rather than definitive proof, since the Roman army did not invariably use freshly quarried stone for military construction; it re-used stone from redundant structures at other locations, such as the walls at Chester and the fort at Birdoswald on Hadrian’s Wall. The rather rough construction technique may argue against a late Roman date, since one might have assumed that the late Roman Army would have been able to produce a more refined tower for its premier northern fortress. However, this could be explained by haste or the absence of the engineering corps elsewhere. There were plenty of crises in Late Roman Britain that could have provided the context for a rushed repair, or for the army to have been stretched very thin and needing to bodge up a repair with any semi-skilled or unskilled labour they had to hand.

A date shortly after the end of Roman administration, perhaps in the fifth or sixth century, is consistent with the coarse building technique and the change in materials, as the engineering skills of the Roman military may well have been partially or wholly lost when Constantine III took most of the army stationed in Britannia to Europe in his (unsuccessful) bid for the top job in 408. There is no reason to assume that Roman York was instantly deserted when Roman government ended in Britain. Quite the reverse; Honorius’ letter in 410 telling the civitas capitals to ‘look to their own defences’ implies that there was still a layer of local or regional government in place. York would be a logical centre, since it had defences and was at the hub of road and river transport networks. It probably also took a while – perhaps several decades – for people to decide that the Empire was not going to come back this time, and during this period the city authorities may well have tried to maintain Roman political and material structures for as long as possible. The Anglian Tower could be regarded as a direct response to Honorius’ command by whoever was in control of post-Roman York, perhaps using a civilian building contractor in the absence of a military engineer. A post-Roman date could also be consistent with the lack of re-used stone. If whoever was in charge in York when the Anglian Tower was built derived their authority – or claimed to derive it – from the preceding Roman military or civil administration, that might contribute to a reluctance to rob stone from Roman buildings, either through respect for Roman structures or through a fear of getting reprimanded for it if/when Roman government was later restored. It’s worth remembering that Britain had been part of breakaway Empires several times in the preceding century or two and that central Imperial control had always been restored sooner or later, frequently to the detriment of those who had been, or were perceived to have been, on the opposing side. In the early to mid fifth century there may well have been no obvious reason to expect it to be any different this time.

A date in the Anglian period, say from the late sixth or early seventh century onwards, is also possible. We know from Bede that York was under the control of the Anglian kings of Deira in 627 when Eadwine of Deira chose to have himself and his court baptised there (Bede Book II Ch. 14). Whether York was a regularly used royal centre or whether the baptism was a one-off visit to a prestigious but little-used ruin is not known. If York was a significant royal centre, repairing the defences would be a logical thing to do. The primary building material in early (‘Anglo-Saxon’) England was timber (which does not mean that buildings were necessarily unsophisticated, as discussed in an earlier post on timber architecture), which at first sight might appear inconsistent with a stone-built tower. However, Bede tells us that Eadwine began to build a stone church in York at some date after 627, and that the church was later completed by his nephew Oswald at some date between 634 and 642 (Book II Ch. 14). So the Anglian kings of the early seventh century in York had access to masonry technology for building churches, and could presumably have used the same technology in other applications, such as repairing the defences. The use of oolitic limestone would also be consistent with this, as the sources for oolitic limestone in the North York Moors and the Yorkshire Wolds are in Deiran territory and may therefore have been familiar to the Deiran kings or their officials – although I wouldn’t read too much into this as the oolitic limestone was also familiar in Roman York. The lack of re-used stone may be an argument against an Anglian date, as there must surely have been at least some derelict buildings in the city by then. However, Bede tells us that Eadwine used a Roman-style standard (Book II Ch. 16). If Roman heritage was seen as a source of authority or legitimacy, there may have been a reluctance to show disrespect by using Roman buildings, however dilapidated, as a stone quarry. There may also have been a reluctance to disturb the ruins either on grounds of superstition (who wants to risk annoying a vengeful ghost?) or practicality (collapsing arches and unstable foundations can make demolition a risky business).

On the basis of the available evidence, the Anglian Tower could have been constructed at any time between the late Roman period and the ninth-century Viking conquest. A reasonable case can be made for late Roman, post-Roman or Anglian construction. All are plausible, none is definitive. You can take your pick.

Bede, Ecclesiastical history of the English people. Translated by Leo Sherley-Price. Penguin Classics, 1968, ISBN 0-14-044565-X.
Ottaway P. Roman York. Tempus, 2004. ISBN 0-7524-2916-7.
Tweddle D, Moulden J, Logan E. Anglian York: a survey of the evidence. Council for British Archaeology/York Archaeological Trust, 1999. ISBN 1-902771-06-0.

Map links
North York Moors