convenient arrangement, but one which added greatly to the defensive properties of the stronghold. There are a number of earthworks which have never possessed a bailey, consisting of motte only, of which the North Riding possesses, in Castle Leavington, the finest example known to the writer.

The late Mr. G. T. Clark differentiated these earthworks from others, but propounded the theory that they mark the site of English castles of pre-Conquest date. To Dr. Round must be given the honour of first demolishing (Quarterly Review, 1894) the then universally held opinion that Mr. Clark's theory was correct, and of establishing the highly important fact that these earthworks mark the site of early Norman castles.1 Mr. W. H. St. John Hope, who would appear to have arrived independently at the same conclusion at a later date ("English fortresses and castles of the tenth and eleventh centuries," Arch. Journal, lx, 72-90), says :

a

"Of the three classes of fortresses distinguished in the A.S. Chronicle, namely: (i) the Geworcs or fastnesses thrown up by the Danish invaders or Heathenmen' during the second half of the ninth century; (ii) Burhs, or burgs, builded or wrought by the English in the first half of the tenth as offensive and defensive works against the invaders; and (iii) new form of fortress introduced by the Normans (tempus Edward Confessor) called castels, only the last comprised the moated mounds in question." Mrs. Armitage, in her very painstaking and valuable work, "Early Norman Castles of the British Isles," has proved that these motte and bailey castles are invariably found where the early Norman lord fixed the caput of his fief. Mr. Orpen, in his "Ireland under the Normans," proves that the motte and bailey castles of that country owe their origin to Norman invaders, and Mr. George Neilson, in his "Mottes of Norman Scotland" (Scottish Review, vol. xxxii, 1898), proves that the Scottish examples of these earthworks are confined to those parts of that country influenced by the Anglo-Norman settlement under David I, Malcolm IV, and William the Lion. During the last four or five years an examination by the present writer of over 250 such earthworks -in England (157), Ireland (27), Wales (32), France (35), and Scotland (12), and the expenditure of a considerable amount of time on a study of the original authorities of the Late in the scientific study of English mediaval military architecture.

1 There can be no possible doubt that Dr. Round's discovery is the most important event which has occurred

Saxon and Early Norman periods, leads him to agree with the above-mentioned authorities. After personally examining all the earthworks of this type in the county of York, and after devoting a considerable amount of time to ascertaining, as far as it is possible to do so, the history of each individual example, the writer is in a position to prove that all those earthworks the history of which can be ascertained either from contemporary records or, as is more frequently the case, from historical inferences, were erected by the Norman conquerors, the great majority of them between 1071 and 1145.

The "mottes" or "mounds," the citadels of the great majority of these earth-and-timber fortresses, are alone sufficient to differentiate them from other fortified enclosures, and to put them in a class by themselves. The greater proportion of them are also considerably less in area than are their predecessors, and everything, more especially the self-evident fact that the motte was capable of being defended not only against outside enemies, but also against its own courtyard, points to this particular class of earthwork having been erected by a foreign invader to defend himself, his family, and immediate retainers. These small private castles or individual fortresses are exactly what we should expect would be thrown up by the Norman settlers when they established the feudal tenure of land in a conquered but still hostile country. Such fortresses could be both rapidly and economically erected. It is recorded, for instance, that the erection by the Conqueror, of the castle known as Baile Hill, at York, occupied only eight days, and the cost of the great timber keep at York, erected 3 Richard I, was only £28 13s. 9d.,1 as compared with £1,927 8s. 7d. expended by Henry III on its stone successor, the existing Clifford's Tower. A fortress or private castle was an absolute necessity to a Norman tenant-in-capite or great feudatory; without such a protection neither his life nor his newly-acquired property would have been worth a week's purchase. It was required not only to protect him from attack by the dispossessed natives, but from the aggression of adventurers similar to himself. These castles were essentially the fortresses of individuals and not fortresses erected, as were their predecessors, to shelter

1" In operatione castri, £28 13s. 9d." Pipe Roll, 3 Richard I.

2 Mrs. Armitage's Early Norman Castles of the British Isles, p. 246, which gives a very instructive history of the various towers which succeeded one

another on the motte at York. We may multiply these figures by 20 to get at their equivalent in modern money. Thus we see that a timber keep cost about £560, as compared with about £39,550 expended on a stone keep.

all the folk, "eallum them folc to gebeorge." The defences constructed by the English and Danes were undoubtedly intended to shelter the whole country-side with their flocks and herds; they were essentially-their size and arrangements prove this clearly enough-communal as opposed to individual fortresses or "castels." "castels." No contemporary Saxon chronicler ever mentions a castle erected by an Englishman, for the simple reason that the English never built castles as we understand the meaning of the word. Had they done so the conquest of England by the Normans would possibly never have taken place. Ordericus, a contemporary writer, explicitly states that "the fortresses, which the Gauls call Castella, had been very few in the provinces of England, and on this account the English, although warlike and daring, had nevertheless shown themselves too feeble to withstand their enemies." The individual fortress or castle was a Norman importation; "it was feudalism that built these castles which once covered our soil, and whose remains are now scattered upon it. They are the declaration of its triumph."4

Motte and bailey castles had been in existence in France for some considerable time previous to their introduction into England, but it hardly comes within the scope of this article to enter into the question of their origin. The writer has been long under the impression, and has yet to be convinced of his error, that their originator was Fulk Nerra, the famous Count of Anjou, and certainly the first authentic mention of a motte and bailey castle, undoubtedly the earliest form of individual fortress, occurs in the Chronicles of St. Florent, which state that, in 1010, Fulk and his son Geoffrey, "in occidentali parte montis castellum determinaverunt . . in prospectu monasterii cum

222.

1 Birch's Cartularium Saxonicon, ii,

2 Eddisbury, in Cheshire-which owes its origin to Ethelfleda-may be cited as a typical Anglo-Saxon burh or fortified communal enclosure. It measures some 1,250 feet in length by about 500 feet in width, and is approximately oval in shape, defended by a ditch and a high outer bank. Shoebury, Essex, is a good example of a Danish geweorc, and although about half the enclosure has been washed away by the sea, it measures some 1,600 feet in length by about 700 feet in width.

3 Ord. Vit., Hist. Eccl., iv, 4.

4 Guizot, Histoire de la Civilization en France, iii, 311.

aggerem quoque turre lignea erexerunt." But

Chron. St. Florentii, in Lobineau's Bretagne, ii, 87. Amongst the castles said to have been erected by Fulk were Baugé, Chateaufort, Chéramont, Montboyau, Montrichard, and Montbazon. The earthworks marking the site of the last-mentioned castle, erected c. 991994, still exist-from personal observation-but this is the only one of Fulk's castles the writer has had the opportunity of visiting. M. de Salies, writing in 1874 (Histoire de Foulques Nerra, p. 170), states that Fulk's castle of Montboyau was then in existence. The writer has been unable to ascertain whether this is still the case.

there would seem to exist evidence that Tribault le Tricheur, Count of Blois (932-962), erected such castles.1

However this may be, there is no doubt as to the date of their introduction into England. It is a well-known historical fact that Norman influence began to make itself felt in this country at least a couple of decades previous to the Conquest, and the earliest castles or individual fortresses erected on this side of the Channel undoubtedly owe their origin to Norman favourites of the Confessor. The English nobles lived in twostoried timber houses such as that depicted in the Bayeux Tapestry, in which Harold and his comrades are feasting in the great hall on the first floor, whilst the basement apparently forms a large cellar or store room. That such houses may have been protected from the wolves by an encircling stockade or hedge is probable enough, but they were never intended for defence against man.

The first castle to be erected in England would appear to have been that of Hereford,2 which was built in or about 1048 by Ralph, Count of Hereford, the Confessor's Norman nephew. This is the first "castel or individual fortress mentioned in the Anglo-Saxon Chronicle, and we read that in 1052 Godwin demanded that the Frenchmen of the castle" should be given up by the Confessor, and later that "the Frenchmen of the castle" aided the English to repel a Welsh invasion. The mere fact that Ralph's fortress was known as "the castle" is sufficient to show that to the English it was an innovation, and apparently by no means a popular one. Richard Fitz-Scrob, one of the Confessor's Norman favourites, founded Richard's Castle,3 c. 1050; Osbern, surnamed "Pentecost," founded Ewias Castle in the following year.4

The castle, although entirely unknown to the English, was the great and outstanding symbol of feudalism, the natural

1 See Mrs. Armitage's Early Norman Castles of the British Isles, pp. 74-5.

2 Anglo-Saxon Chronicle, 1048 (Peterborough) and 1052 (Worcester). The motte of this castle has disappeared, but the bailey remains. The motte, however, is mentioned in the Pipe Rolls, 11 Hen. I, p. 100, and 15 Hen. II, p. 140. It is described in a Survey of 1652, and was then crowned by works in masonry (Grose, Antiquities, ii, 18; and Duncombe's History of Hereford, i, 229).

3 This castle is referred to in Domesday (i, 1866) under the name of Avreton, and the "valet ei castellum," T.R.E., is

given at 20 shillings. The motte still remains in good condition, with a halfmoon-shaped bailey two-thirds of an acre in extent.

4 Ewias Castle is a well-preserved and interesting earthwork, and is the only castle mentioned in the Survey as having been rebuilt by the Normans (D.B., i, 186a). Dr. Round (Feudal England, p. 324) is of opinion that it was Pentecost castle," referred to in the Anglo-Saxon Chronicle in 1052, and after visiting the earthwork in 1907 the present writer came to the same conclusion. A half-moon-shaped bailey lies to the south of the motte.

home and base of operations of a feudal lord, and one of the first acts of the Conqueror was to throw up a "castell" at Hastings. The Bayeux Tapestry depicts the construction of this fortress. With that all-inclusiveness for which mediaval pictures are noted, its builders are shown digging the foss or ditch round the motte, ramming the soil of the motte with the flats of their spades, and constructing a palisade round the upper edge. This erection of castles was the leading feature of the Conqueror's modus operandi, wherever he went we read "castellum construxit," arcem condidit," etc. He invariably consolidated his hold on a conquered district by erecting castles to overawe it, and to form the base of further operations. And when he parcelled out the kingdom among his favoured comradesin-arms they followed his example. These castles were absolutely necessary from their founders' point of view, alike to hold their new lands against the disinherited and hostile natives and to protect themselves from attack by other foreign lords who, had such a protection been absent, would not have scrupled, during a time of disorder, to seize such lands and add them to their own estates.

But the castles built during the latter part of the eleventh and during the first half of the twelfth century, i.e. from the time of the Conquest to about 1150, were not those whose enormous keeps and massive curtain walls still form such a picturesque feature of our landscapes. Were it possible for us to transport ourselves into Yorkshire in the year 1154 we should see, in the course of our perambulations, a number of stockaded mounds with their palisaded baileys; a few stockaded promontories with or without mottes; a few stockaded mottes devoid of baileys, but at three castles only, at Richmond, at Scarborough, and at Tickhill, should we find any defences in masonry.

3

1" Dux ibidem (at Pevensey) non diu moratus, haud longe situm, qui Hastinges vocatur, cum suis adiit portum, ibique opportunum nactus locum, ligneum agiliter castellum statuens, provide munivit" (Chron. Monast. de Bello, p. 3, ed. 1846).

The picture showing the erection by the Conqueror of the motte at Hastings represents it as being formed of layers of different materials. When the motte at Carisbrooke was opened in 1903, it was found to be composed of alternate layers of large and small chalk rubble (Stone's Official Guide to the Castle of Carisbrooke, p. 39). The motte still remains at Hastings, but bears no trace

of ever having been crowned by works in masonry. Henry II's stone keep has, in the opinion of the writer, been destroyed by a fall of the cliff. Mr. Harold Sands is of the same opinion.

3 The extent of the stone fortifications existing at Scarborough in 1154 is somewhat difficult to determine. But although the walls of enceinte owe their origin to William le Gros, c. 1136-1140, it is equally certain that both great hall and keep were constructed of timber at the time of the accession of Henry II.

4 This typical motte and bailey castle was founded by Roger de Busli, and is mentioned by Ordericus (xi, ch. iii) as the castle of Blythe. The motte,