II.

ANGLO-SAXON BURHS AND EARLY NORMAN CASTLES. BY MRS E. S. ARMITAGE. Communicated by PROFESSOR BALDWIN BROWN, F.S. A. Scot.

I am not aware that any serious attempt has ever yet been made to ascertain what the nature of an Anglo-Saxon fortification was. One of our best archæologists observes that "whatever amount of difficulty may attend our inquiry respecting the domestic buildings of the Saxons, the character of their military edifices is involved in far greater obscurity."1 It is possible that this ignorance is mainly due to not making use of the materials which exist in a scattered and fragmentary condition, and which have never been pieced together. But it cannot be denied that the general absence of interest in questions of English archæology has led to a complete lack of accumulated observations on the subject; and the difficulty of getting information, even about existing remains, can only be appreciated by those who have attempted an inquiry of the

kind.

What is worse is that this lack of interest has left the ground open to assumptions, which are accepted as facts, because no one cares to dispute them. It seems strange that in the nineteenth century any archæologist of reputation should still follow the method of the archæ ologists of a hundred or two hundred years ago, who first guessed at things, and then said they were so. Yet this is certainly the method followed by the late Mr G. T. Clark in his otherwise valuable work on Medieval and Military Architecture. Finding that in several places where the Anglo-Saxon records tell of burhs or strongholds erected by our forefathers, there are still existing round hillocks of earth, surrounded with ditches, he jumped to the conclusion that a burh was a moated hillock, and then proceeded to assert that it was so, without any further inquiry into the literary history of the word. The evidence which he adduced in support of his assumption was chiefly this:-1st,

1 Hudson Turner, History of Domestic Architecture in England, vol. i. p. 18.

that of the fifty burhs mentioned in the Anglo-Saxon Chronicle, there are twenty-two still existing where moated mounds of the kind in question are to be found; 2nd, many of these works are known to have been the centre or caput of great estates in Saxon times. Strange to say, this very scanty and disputable evidence has been accepted without question even by such writers as Freeman and Green, and is adopted by most of the antiquarian books and papers written during the last twenty years.

With the theory that these moated hillocks mark the centre of a private estate in Saxon times, this paper will not attempt to deal, as it can be left in the far more competent hands of Mr J. H. Round, who has clearly expressed his dissent from it. The philological and historical evidence, and the evidence drawn from the actual remains, will be suffi cient for the purpose of this paper. What first led the writer to doubt the truth of Mr Clark's contention that a burh was a conical earthwork, was that on looking through the illustrated Anglo-Saxon MSS. in the British Museum to find a picture of a burh, it was seen that the AngloSaxon idea of a burh, as represented by those pictures, was an enclosure with walls and towers of stone-in other words, a walled town. Not long afterwards, an article on English castles in the Quarterly Review for July 1894, now known to have been written by Mr J. H. Round, led to the conviction that Mr Clark's theory of burhs was simply an archæological delusion. Mr Round's words are: "We hold it proved that

1 Mediaval and Military Architecture, pp. 22, 23.

2 Essex Archæological Society's Transactions, vol. iii. part ii. “The more deeply I have studied the theories of Castle Clark,' the more reason have I seen to doubt his view that these strongholds were intended for the centre and defence of a private estate, for the accommodation of the lord and his household, and for the dwelling of the English lord who succeeded the Roman provincial." In his Feudal England, Mr Round shows that most of the Norman fiefs were wholly new creations, constructed from scattered fragments of Anglo-Saxon estates, p. 260.

3 On p. 29 of the MS. of Prudentius (Cleopatra C. viii.) there is an excellent drawing of a four-sided enclosure, with towers at the angles, and battlemented walls of masonry. The title of the picture is "Virtutes urbem ingrediuntur"; and urbem is rendered in the A.S. gloss as burh.

these fortified mottes were, at least in some cases, erected in the Conqueror's days, and if this is proved of some, it becomes probable of many. Indeed, so far as what we may term private castles are concerned, there is actually, we think, a presumption in favour of this late origin." It is proposed in this paper to carry this contention even further, and to maintain that while the burhs of the Anglo-Saxon Chronicle are almost always walled towns, the moated hillocks scattered so thickly over England and south-western Scotland are the remains of castles built by Normans.

The philological evidence is of considerable importance in this contention. There is not the smallest reason to suppose that the word burh ever meant a hillock, for the history of the word can be clearly traced. Mr Clark had not the advantage of consulting the New English Dictionary, which had not appeared when he wrote, but had he looked into Schmid's Gesetze der Angelsachsen, he would have learned that a burh, which is derived from the same root as the verb bergian, to shelter, meant originally a wall of some kind, whether of earth, wood or stone, built for protection. As in the case of the words tun, yard or garth, and worth or ward, the sense of the word became extended from the protecting bulwark to the thing protected. In this sense of a fortified enclosure, the word was naturally applied by the Anglo-Saxons to the prehistoric and British "camps" which they found in Britain, such as Cissbury, or to similar forts which they constructed themselves, such as Bebbanburh (Bamborough). Sometimes the burh was probably nothing more than a palisade or hedge round a great man's house, if we may judge from the innumerable places whose names end in bury or borough, from which every vestige of bulwark has totally disappeared.

The laws of Ethelbert of Kent, Ine of Wessex, and Alfred, speak respectively of the king's and earl's tun, huse, and healla, and special

1 The dative form burig is the origin of the names ending in bury. “To say nothing of hamlets, we have full 250 parishes whose names end in burgh, bury, or borough, and in many cases we see no sign in them of an ancient camp or of an ex ceptionally dense population." Maitland, Domesday Book and Beyond, p. 184.

punishments are ordained for crimes committed within their precincts.1 It is possible that in two instances in the later laws, the king's burh is used in the same sense. But from the time of the laws of Athelstan the word burh far more commonly means a city or town. Thus he ordains that there shall be a mint in every burh.3 And it appears that already the town has its gemot or meeting. In the laws of Edgar's time and later, the burh has not only its burh-gemot, but its burh-gereja or town-reeve, and its burh-waru or townsmen.5 Burh is contrasted with wapentake as town with country. And in this sense it has descended to our own day as a borough, though because the word borough has so long meant a chartered town or a town with parliamentary representation, we have forgotten its older meaning of a fortified town.

If we turn to Anglo-Saxon literature, we get the same answer. Alfred in his Orosius translates city by burh. The Anglo-Saxon translation of the Gospels (circa A.D. 1000) uses the same word for the civitatem of the Latin version.8 In the Anglo-Saxon Chronicle, the words geweorc or faesten are generally used for a fortress hastily thrown up, and burh is reserved for fortified towns. The word burh, indeed, is seldom used in the Chronicle until we come to the time of Edward the Elder. It is conclusive as to the general meaning of the word that 1 V. Schmid's Gesteze der Angelsachsen, Ethelbert, 5, Ine. p. 22, Alfred, p. 74.

* Thus Edmund (ii. 2) speaks of mine burh as an asylum, the violation of which brings its special punishment (Schmid, p. 176), and Ethelred (iii. 4) ordains that every compurgation and every ordeal shall take place on thaes kyninges byrig. (Schmid, p. 214). A charter of Alfred's time speaks of the hedge of the king's burh. Birch's Cartularium, ii. 305. The word burh does not occur in the laws of Edward the Elder.

4 Professor Maitland says: "In Athelstan's day it seems to be supposed by the legislator that a moot will usually be held in a burh. If a man neglect three summonses to a moot, the oldest men of the burh are to ride to his place and seize his goods." Domesday Book and Beyond, p. 185.

Edgar, iii. 5. Ethelred, ii. 6. Athelstan speaks of the reeves of every burh. I. Preface.

6 Ode on burge, odde on wapengetaece, Edgar, iv. 2.

7 New English Dictionary, Borough.

8 Ib. Matt. xxi. 17.

Florence of Worcester, one of the most accurate of our early annalists, in his account of Edward's reign, regularly translates the burh of the Anglo-Saxon Chronicle by urbs.1

But though we may now feel certain that the general sense of the word burh was a town, its more special sense as an enclosing bulwark does not appear to have been forgotten in Anglo-Saxon times. Thus Athelstan orders that all burhs shall be repaired fourteen days after Rogations; and Cnut, when making a similar provision, expressly defines it as civitatum emendatio.3 Here the word for town is used for the town wall. The same sense appears as late as the reign of William Rufus, in the Anglo-Saxon Chronicle (1092); when relating the restoration of Carlisle by that King, it says:-"He repaired the burh, and ordered the castell to be built."4 And finally, a remarkable charter of Ethelred of Mercia and Ethelfleda his wife states that they have commanded the burh at Worcester to be built as a protection to all the people. Ethelred and his wife were not building a new town, for Worcester already had its churches and its bishop, and possibly the remains of its Roman walls, but they were building or rebuilding a town wall or embankment to protect the city from the Danes.

It is equally clear that a burh was not a castle, in the sense in which we commonly use that word. The word castellum is occasionally used in Anglo-Saxon charters, but when it is used it clearly means a town.

1 Florence of Worcester lived at the end of the eleventh century and beginning of the twelfth, when Anglo-Saxon was still a living language.

22 Athelstan, ii. 13. Schmid, p. 138.

Cnut, ii. 10. Schmid, p. 276.

A passage, by the way, which is fatal to Mr Clark's theory that a burh meant a moated hillock, for there is no such hillock at Carlisle.

Hehtan bewyrcean tha burh at Weogernaceastre eallum tham folce to gebeorge. Birch's Cartularium, ii. 222.

6 Thus a charter of Egbert of Kent, 765, says: “Trado terram intra castelli moenia supranominati, id est Hrofescestri, unum viculum cum duobus jugeribus, adjacentem platea que terminus a meridie hujus terræ," etc. Codex Diplomaticus, i. 138. In two charters of Ethelwulf, Hront castellum is used as an equivalent for Hrofecestre or Rochester. Birch's Cartularium, ii. 48 and 86. In this sense, no doubt, we must interpret Asser's "castellum quod dicitur Werham." Vita Elfredi, 478.