Some Early Norman Castles

The typical Norman castle was introduced into England at the time of the Norman Conquest and persisted, though in a gradually developing form, till its final supersession by an entirely new species at about the beginning of the thirteenth century. It is true, indeed, that it is found in a few Sporadic instances, as at Ewias Harold, Herefordshire, and Clavering, Essex, at a date before the death of the Confessor. But these were wholly due to the French favourites of a monarch who "Normanised" immoderately. The Norman Conquest may therefore be said, in this limited architectural sense, to have begun before the landing of William.

The typical Norman castle was constituted by two main elements; a large, fortified enclosure, or bailey, or a number of such enclosures-there were two at Arundel and Alnwick, and actually three at Clun-and a citadel, or keep. Towards the beginning of the thirteenth century the bailey became so modified as to become virtually a new element, whilst the keep disappeared altogether. During the more than hundred years of its efflorescence on English soil this typical Norman castle was subject, as we have said, to a gradual development, but these two constituent factors remain in essence to the end.

The Norman type of castle as first introduced into England was, with very few exceptions, of a rough-and-ready character-little more, in fact, than a hasty extemporisation, constructed only of earthworks that were crowned by timber stockades. This was the kind of fortress that could be thrown up in a few days by a comparatively small body of men and with the aid of comparatively little skilled labour. It was thus eminently adapted for an invader such as William was at Hastings-who wished to secure his base in the greatest possible hurry; or robber lords of forfeited manors, such as we find by the score in Domesday, anxious to entrench their position in the middle of a hostile population

In the Bayeux Tapestry are rough pictures oi more than one such early stronghold, or more correctly of their "motte" or central keep-a huge, conical earthen mound, with flattened top, the edge of which was surrounded by a wooden fence, and in the centre of which was a wooden tower.

At a date soon after the Norman Conquest England was dotted from end to end with fortifications of this primitive construction. Mrs. Armitage gives a "catalogue raisonne" of the eighty-four that are known from documentary evidence to have existed during the latter part of the eleventh century, and this is certainly only a fraction of the whole. Scattered about the country are a large number of ancient earthworks, the history of which is obscure and the exact origin unknown. Such are the huge embankments in the West Riding of Yorkshire, at Barwick-in-Elmete and at Laughton-en-le-Morthen

That many of these are early Norman "motte-and-bailey" castles is a safe assumption from their structural analogy. Probably they were abandoned at an early stage of their existence for stronger or otherwise better sites. It must be borne in mind, however, that the "motte-and-bailey" type of castle-though undoubtedy imported into England by the Normans-is improbably itself of Norman birth. It possibly originated at about the middle of the tenth century with castles built by Thibault-le-Tricheur, Count of Blois, at Chartres and on the Loire. Of its two distinctive features, the "motte," or artificial earthen mound, is apparently a wholly new device. The bailey is, from its nature, as old as human ingenuity, and traces back to the Neolithic hill-forts, or even to a still remoter past.

The first Norman castles of England were thus extremely simple structures, formed of earth and timber only, and of a very different superficial character from that which they afterwards assumed. Who indeed would dream that the splendid piles of masonry on Windsor hill or Dover cliff are really direct descendants from such primitive banks and ditches as we still find at Great Berkhampstead or Huntingdon? There were, of course, exceptions. Exeter from the beginning, and possibly Tickhill from a fairly early period, were entered by stone gateways; Richmond was apparently walled, but without a "motte," from its foundation; whilst at Colchester and London stone rectangular keeps, of the kind that only made its general appearance in this country at a considerably later period, were built before the close of the eleventh century.

The first step in their development was the translation of earthen mounds and wooden stockades into walls of massive masonry. This, of course, took time, and obviously implies a more settled state of society than was possible immediately after the Conquest. The bailey or courtyard which adjoined the mound or keep thus became surrounded by a solid stone curtain instead of a palisade; was entered by a definite gate-house, and strengthened and enfiladed externally as time went on by a number of square and subsequently circular bastions. A good example of the form that such a curtain-wall might finally assume may be seen at Dover Castle with its twenty-seven towers of various dates on the outer "enceinte." But at first the curtain was very simple.

Very splendid examples of later Norman gatehouses survive at Porchester, Lewes and Newark. At a later period the approach to them was often guarded by elaborate outworks, or barbicans; but these, as at Alnwick, Lewes and Warwick-all very noble instances-were additions of the thirteenth or fourteenth centuries.

Closely connected with this gradual development of the curtain-wall, as times became more tranquil and the castle tended more and more to exchange its purely military for a partly residential character, was the appearance inside its shelter, and often structurally in touch with it, of the domestic apartments or "palace." This would include the hall, where the lord might revel with his followers, his kitchen, and the "solar" for his ladies. Such is the earlier of the two existing halls at Durham-a magnificent example of a highly developed, originally "motte-and-bailey," castle-which was added by Bishop Pudsey in 1153-95, and is entered by what is possibly the most beautiful example of a richly decorated late Norman doorway in the kingdom. Halls like this were raised on vaulted basements, like monastic "fraters," and generally consisted of a single, big apartment. It must be kept in mind, however, that from the strictly military point of view the palace inside the curtain was always more or less of an excrescence. It is the link, in fact, by which the original Norman castle, of wholly war-like emphasis, was finally developed into the purely domestic manor-house of the Elizabethan period.

Side by side with the development of the banks and fences of the bailey, and their translation from earth and woodwork into masonry, proceeded the analogous transformation of the "motte."

The circular stockade and other wooden super-structures on the summit of the "motte" were replaced by a ring of solid masonry, whilst the "motte" itself was suffered to remain. This ring was sometimes oval, or strictly circular, as at Carisbrooke, Arundel and Tonbridge, and sometimes many-sided, as at Lincoln, Tickhill (now mere foundations), Durham (rebuilt in recent years) and Cardiff. Against its inner face would be constructed lean-to buildings, doubtless principally of timber, for which reason they have perished universally; round its crenellated summit was a rampart walk, or allure; and its centre was kept open. This is the type of keep, of purely military import and marking no great step forward in structural growth, that is commonly styled "shell." With the exceptions, already noted, of Colchester and London, the shell-keep probably precedes in date, as a general rule, its more famous "rectangular" rival. It is important, however, to note that possibly, as "rectangular" keeps became more common, the "shell" variety still persisted, and was continued side by side with them. The fine example at Totnes, which retains its original battlements, was perhaps built as late as the dawn of the thirteenth century. The number of shell-keeps now remaining more or less intact in England, is probably less than twenty, but traces exist of others. Their distribution is wide, from Harbottle, in Northumberland, to Trematon, in Cornwall. Normally, a shell-keep is entered by a single gateway, of quite unelaborate character, and reached directly up the steep slope of the mound by a series of steps, as at Carisbrooke and Farnham, though it is impossible to assert with confidence that these approaches are original.

We come at last to that form of Norman keep- the stone rectangular-which, though far less common than its prototype, the early earthen "motte," is perhaps the best known emblem of Norman predominance in England. When we speak of Norman castles it is certainly these grim towers, as at Rochester, Castle Hedingham, Porchester, Dover, Appleby, that leap to the mind's eye, as truly typical. Found at Langeais, on the Loire, as early as a.d. 992, and imported into England, as we have seen, at a date soon after the Conquest at the Tower of London (the White Tower) and Colchester, it only makes its first general appearance in this country in the reign of Henry I, whilst most existing examples are as late as Henry II. Altogether there were once, dispersed all over England, about half-a-hundred instances; and though one of the biggest of these, at Dumel'd, has left nothing but foundations, whilst others, at Nottingham, Bristol, Hastings, have vanished altogether, the bulk remain in ruined grandeur to attest a grand conception.

From the point of view of distribution they fall roughly into groups, of which one, in the north of England, may have been conditioned by the un-neighbourly proximity of Scotland; a second, from Chester down to Chepstow, may have guarded the Welsh Marches; whilst a third, in the eastern and south-eastern counties, had surely a special reference to invasion from the Continent. In the Midlands we find only four occurrences, at Kenilworth, Nottingham, Castleton (very small) and Duffield. Devon has only the trifling example at Lidford, whilst Lincolnshire and Cornwall have none at all. They vary much in size, from the towering piles at Rochester, Castle Hedingham and Corfe, and the huge and sprawling bulks of Colchester, Norwich, Castle Rising and Canterbury, to such insignificant examples as Clitheroe, perhaps the smallest of all, Saffron Walden, and the two diminutive occurrences in Kent, at West Mailing and Sutton Valence.

Subject to this great difference, the Norman rectangular keep, with some minor variations, conforms pretty strictly to type. The angles are often capped by clasping buttresses, which at Colchester and Dover assume almost the dimensions of little towers. Internally it is generally divided vertically, from top to bottom, by a great cross-wall, the meaning of which is not very apparent; and horizontally, according to its height, into a number of different floors-at Norham there were five-of which the bottom one is usually a cellar. The entrance is normally on the first floor, though at Dover and Newcastle on the second, and at Carlisle and Bamburgh from the ground; and is frequently reached by an outside stair, at right angles to the entrance, that is covered and protected by its own independent "fore-building." From the first floor a circular vice, or newel, in one of the angles-at Dover there are two-ascends to the upper floors and battlements and descends into the basement. A well was a necessity, though there is strangely none at Guildford; and though found oddly in some early instances, e.g., London and Castle Rising, placed, at great disadvantage, in the basement. In later keeps, however, the well head is contrived in the topmost storey, and was available thus to defenders who, driven inwards and upwards, maintained themselves there precariously, in a last desperate hope of outside succour. Meant primarily for refuge, the keep in many instances was also residential.

A Norman rectangular keep however, from its very nature, can never have been commodious, though oddly enough, at Castle Hedingham one now serves as a private house. The tendency was to quit it for the more convenient "palace" in the bailey. It often contained a chapel; and in the two abnormally early instances at Colchester and London this was on a scale that was never afterwards rivalled. At Newcastle the chapel is in the basement of the fore-building, whilst at Dover there are two, both also in the fore-building, but respectively on the first and second floor.

At Newcastle and in the lower chapel at Dover the walls are richly arcaded, and there is abundant "zigzag" ornament. At Guildford, one of the smaller keeps, the curious little chapel is contrived in two divisions, at a right angle, in the thickness of two external walls. There was frequently a chapel in the bailey; and at Ludlow, one of the most magnificent of all ruined English castles, this displays the peculiarity of a small twelfth century circular nave. Only six similar "round churches" occur anywhere in England, including mere foundations, or traces, at Garway and Dover.

The Norman rectangular keep, in contrast with the "shell," though it stands in direct succession to the pre-existing "motte," is very seldom built in actual touch with it. There are, as usual, exceptions: at Christchurch, another small example, it is placed boldly on the top, and the same, perhaps, was true of Bishop's Stortford; whilst it is oddly located at Clun and Guildford, as though clinging to the slope. More commonly it stands clear, on virgin soil, in the centre of the bailey, as at Bamborough and Carlisle, or adjacent to the "enceinte," as at Bramber. At Richmond, one of the grandest of all examples, and at Ludlow, one of the smaller, it is strangely raised on the basis of a pre-existing gateway.

From a pictorial point of view the stone rectangular keep is the one outstanding achievement of the Norman military architect. No one who has ever witnessed the haughty dominance of Bamborough, Richmond, Dover, Rochester, Porchester or Corfe, or even the shattered fragments of Bridgnorth, Scarborough or Helmsley, is ever likely to underrate its grand and gloomy significance. It stands the eternal symbol of a vanished mode of life - of its insolence, of its cruelty, of its tyranny, of its fear. It must be borne in mind, however, that the typical Norman fortress outlived by nearly a century the technically Norman dynasty and survived, though in altered shape, till the introduction of a wholly keepless type of castle during the reign of Edward I. This altered shape consists in the substitution - based, no doubt, on the gradually acquired experience that it is easier to defend a circle than a square-of the circular keep, or "Juliet," for the old square pattern, with its indefensible angles.