The advantages of the site selected for the Tower were considerable, the utilization of the existing Roman wall to form two sides of its bailey, its ditch isolating it from the city, while it was so placed on the river as to command the approach to the Saxon trade harbour at the mouth of the Wallbrook, then literally the port of London, and with easy access to the open country should a retreat become necessary.
It is much to be regretted that London was omitted from the Domesday Survey, for that invaluable record might have furnished us with some information as to the building of the Tower, and perhaps revealed in one of those brief but pithy sentences, pregnant with suggestion, some such ruthless destruction of houses as took place in Oxford and elsewhere[8]
in order to clear a site for the King"s new castle. Unless the site were then vacant, or perhaps only occupied by a vineyard (for these are mentioned in _Domesday Book_ as existing at Holborn and Westminster),[9]
some such clearance must obviously have been made for even the first temporary fortifications of the Conqueror, although contemporary history is silent as to this. The _Saxon Chronicle_ tells us that "upon the night of August the 15th, 1077, was London burned so extensively as it never was before since it was founded,"[10] which may have determined William to replace the temporary eastern fortification by an enlarged and permanent castle, he having then completed the conquest of England and crushed the rebellions of his turbulent baronage.
[Ill.u.s.tration: PLAN OF THE TOWER OF LONDON ABOUT 1597.]
Although the art of the military engineer was then in its infancy, the Conqueror seems to have selected as his architect one already famous for his skill. Gundulf, then just appointed Bishop of Rochester, was no ordinary man. The friend and _protege_ of Archbishop Lanfranc, by whom he had been brought to England in 1070, he had as a young man been on pilgrimage to the Holy Land, and doubtless profited by his travels and the opportunity afforded of inspecting some of the architectural marvels of the Romano-Byzantine engineers. Although Gundulf had rebuilt the cathedral of Rochester, to which he added the large detached belfry tower that still bears his name, built other church towers at Dartford, and St. Leonard"s, West Malling (long erroneously supposed to have been an early Norman castle keep),[11] and founded at the latter place an abbey of Benedictine nuns, his reputation as an architect rests chiefly on his having designed the keep of the Tower of London (probably that of Colchester also), and built the stone wall round the new castle at Rochester for William Rufus. While engaged in superintending the erection of London keep, Gundulf lodged in the house of one Eadmer Anhoende,[12] a citizen of London, probably a friend of the Bishop, for we find his name occurring as a generous donor to Gundulf"s new cathedral at Rochester, where, by his will, he directed his own body and that of his wife to be interred, and to have an obit annually. Gundulf"s work therefore consisted of the great keep (afterwards called the White Tower), which he erected close to the line of the Roman city wall, and some fifteen or twenty feet within it. At first this was probably (like its sister keep at Colchester) only enclosed by a shallow ditch and a high earthen bank, crowned by a stout timber palisade, the city wall forming two sides of its perimeter, and probably broken through where the ditch infringed upon it at either end. With the sole exception of Colchester keep, which, as will be seen from the following table of dimensions, is considerably larger, the tower or keep of the castle of London exceeds in size the great rectangular keep of every other castle in the British Isles. Unfortunately, the two upper stories of Colchester keep have been destroyed, but sufficient remains (coupled with the resemblance of its plan to that of the White Tower) to show that both were designed by the same hand and erected about the same period, while both alike were royal castles.
TABLE OF COMPARATIVE DIMENSIONS
LONDON. COLCHESTER.
Length (North to South) over all 121 feet 170 feet Ditto within b.u.t.tresses 118 " 153 "
Breadth (East to West) over all 100 " 130 "
Ditto within b.u.t.tresses 98 " 115 "
Breadth of Apse 42 " 48 "
Diameter of Apse 21 " 24 "
Length (on South Side) over all 128 " 153 "
Number of Stories 4 now 2 Total Height 92 feet --- Height of Two Lower Stories 42 " 32 feet Thickness of Walls 15 " 14 "
Thanks to the drastic removals of recent years, the White Tower stands to-day very much as when first erected. In plan it is practically rectangular, but the north-east angle is capped by a projecting circular turret containing the great main staircase that ascends from the bas.e.m.e.nt to the roof, serving each floor _en pa.s.sant_, while the south angle of the east face has a large semicircular projection that contains the apse of the chapel. The main staircase terminates in a large circular turret of two stories, that rises some twenty-nine feet above the roof. The other angles terminate in three rectangular turrets about fourteen feet square, and twenty-seven feet high above the roof. The walls are at the base some fifteen feet in thickness, exclusive of the steep battering plinth from which they rise, and which slopes sharply outwards. They diminish by set-offs at each floor. The interior is divided into two unequally sized chambers by a cross-wall ten feet in thickness, running from north to south. Of these, the eastern one is again subdivided by a thick cross-wall at its southern end, which is carried up solid to the roof, while on the upper floors the central wall is perforated by arcades of three, and four perfectly plain semicircular headed arches. To the north and west the bas.e.m.e.nt floor is about sixteen feet below the existing ground level, which falls rapidly along the east side, and on the south it is practically on the ground level, as the ground there has not been artificially raised. The two larger chambers of the bas.e.m.e.nt have a modern plain brick barrel vault. The well, a plain ashlar pipe six feet in diameter, is in the south-western angle of the floor in the western chamber. The south-eastern chamber retains its original stone barrel vault. This forms the sub-crypt of the crypt below St. John"s Chapel, and is lighted, or at least its darkness is made dimly visible, by a single small loop in the east wall. It is now known as "Little Ease," and is said to have served as the prison of Guy Fawkes. The bas.e.m.e.nt chambers have boldly sloped recesses in the walls, with small loops high up in their heads, which afford the minimum of air and light; but as they were only used for stores, this was not of great importance. Ascending by the main staircase to the second floor, the same subdivision into three chambers is continued, but these were lighted by larger loops, that have been converted into larger windows at the time of Sir Christopher Wren"s renovations in 1663. The crypt of the chapel opens from the eastern chamber, and has in its north wall a singular dark cell eight feet wide and ten feet long, in the thickness of the wall, in which Sir Walter Raleigh is said to have once been imprisoned. The western chamber has in its north-west angle a latrine, or garderobe, in the thickness of the wall. At the west end of its south face is a large original opening, with parallel sides, having niches in them. The masonry shows traces of where the arch and door jambs have been torn away and the present large window subst.i.tuted, probably during Wren"s alterations. There is little room to doubt that _this_ was the _original door of entrance_, placed, as is usual, some distance above ground, and probably reached by an external flight of steps, now removed, protected by a similar fore building to that of Rochester keep.[13]
Proceeding by the main stair to the third floor, we enter first what is known as the "Banqueting Hall," which is lighted by four large windows, and has a fireplace in its east wall, with two latrine chambers in its north and east walls. Pa.s.sing through a low doorway in the part.i.tion wall, we enter the great western chamber, which has a fireplace in its west wall, a latrine in its north wall, and is lighted by eight large windows. Two newel staircases in the western angles ascend to the battlements. In the south wall is a doorway leading to a pa.s.sage at the head of a small newel stair, which, rising from a door in the wall on the floor below, formerly afforded a direct communication from the palace to the chapel of St. John upon the third floor, without entering the keep. At the foot of this stair, in the time of Charles II., some bones in a chest were discovered by workmen engaged in repairs, which were said to be those of the murdered Edward V. and his brother the Duke of York. These were transferred; by the King"s instructions, to the vaults of Westminster Abbey.
Ascending to the fourth floor, there are two large rooms separated by the cross-wall, the arcade of which was probably filled in with wooden part.i.tions. The larger or western room is known as the "Council Chamber," and the other as the "Royal Apartments." Neither has any fireplace. Over the vaulting of the chapel, close under the flat, lead roof, there is a curious cell about seven feet high, lighted by small loop windows, which extends the entire length of the chapel. Formerly used as a prison, it must have subjected its miserable inmates to even more trying variations of heat and cold than the famous "_Piombi_" of Venice.
With the exception of the chapel, its crypt, and sub-crypt, which were vaulted throughout, all the floors were originally of wood, and were supported on double rows of stout oak posts, which in their turn sustained the ma.s.sive oak main floor beams.
The forebuilding, on the south face of the keep, was probably added by Henry II. It survived until 1666, as it is shown in a view of the Tower executed by Hollar about that date; but it appears to have been removed prior to 1681.
The chapel of St. John is a fine example of early Norman ecclesiastical architecture. It consists of a nave, with vaulted aisles, having an apsidal eastern termination. It is covered by a plain barrel vault, and on the fourth floor level has a triforial gallery, also vaulted. It is connected by two doors with the gallery in the thickness of the wall that surrounds this floor, from one of the windows of which it is said that Bishop Ralph Flambard effected his remarkable escape.
It is probable that at first (except the chapel, which was covered by its own independent roof) there were two separate high-pitched roofs, one covering each division, and not rising above the battlements, the wall gallery serving as a kind of additional fighting deck, for which reason it was carried round the triforium of the chapel. As the need for this diminished, two large additional rooms were gained by raising the central wall a story, and superposing a flat, lead roof.
The absence of privacy, fireplaces, and sanitary accommodation on this fourth floor, with the cold draughts from the stairways and windows of the wall-gallery, must have been well-nigh intolerable; nor could wooden screens, hangings, or charcoal brasiers have rendered it endurable. It is not surprising, therefore, that under Henry III. the palace was considerably enlarged, or that these chambers were abandoned by him for warmer quarters below, in the Lanthorn Tower "k," and its new turret "J" although the chapel and council chamber continued to be used down to a much later date.
After the siege of Rochester by William Rufus in 1088, Gundulf had built a _stone_ wall round the new castle of Rochester. This probably moved the King to enclose the Tower of London with a similar wall, for the _Saxon Chronicle_ tells us that in 1091 "a stone wall was being wrought about the Tower, a stone bridge across the Thames was being built, and a great hall was being erected at Westminster, whereby the citizens of London were grievously oppressed."[14]
Now, as Gundulf did not die until 1108, it is by no means improbable that, while superintending the erection of these two great towers at London and Colchester,[15] he also constructed the stone wall round the former, for the chronicler says of him that "in opere caementarii plurimum sciens et efficax erat."[16]
As it is on record that the smaller keep of Dover, built by Henry II.
nearly a century later, was upwards of ten years in construction, while some additional time had been consumed--in the collection of materials and workmen--with the preliminary preparation of the site, it does not seem probable that the great Tower of London (honeycombed as its walls are with cells and mural pa.s.sages) could have been erected in a much shorter s.p.a.ce of time. When the ruder appliances of the earlier period are taken into account, such a keep could not have been built in a hurry, for time would be needed to allow the great ma.s.s of the foundation to gradually settle, and for the mortar to set. Although preparations for its erection may have begun as early as 1083, it seems more probable that the White Tower was not commenced much before 1087, or completed before 1097.
Stow, quoting from FitzStephen"s _Description of London_,[17] mentions the White Tower as being "sore shaken by a great tempest of wind in the year 1091," which, as I do not (with the conspicuous modesty of the late Professor Freeman) "venture to _set aside_ the authority of the chronicles"[18] when they have the audacity to differ from my preconceived ideas, seems to me reasonable ground upon which to argue that not only was the White Tower then in course of erection, but that in that year the works were not in a very advanced state. That it must have been completed prior to 1100 is evidenced by the fact that King Henry I., on succeeding to the throne in August of that year, committed to the custody of William de Mandeville, then Constable of the Tower, his brother"s corrupt minister, Ranulph (or Ralph) Flambard, Bishop of Durham. The chronicler exultingly tells us that he was ordered[19] "to be kept in fetters, and in the gloom of a dungeon," which must have been either "Little Ease" or the small dark cell opening from the crypt of St. John"s Chapel, afterwards rendered famous by the imprisonment there of Sir Walter Raleigh.
Although the great fortress-palace was to subsequently acquire a most sinister reputation as a state prison, yet the present is the first recorded instance of the committal of a great and notorious offender to its dungeon cells. Subsequently, however, the severity of the bishop"s imprisonment appears to have been somewhat mitigated, for the King ordered him to be allowed the large sum of two shillings a day for his maintenance; so that, although a prisoner, he was enabled to fare sumptuously.
One day after the Christmas of 1101, a long rope having been secretly conveyed to him, concealed in a cask of wine, by one of his servants, he caused a plentiful banquet to be served up, to which he invited his keepers, and having intoxicated them to such a degree that they slept soundly, the bishop secured the cord to a mullion in one of the double windows of the southern wall-gallery in the keep, and, catching up his pastoral staff, began to lower himself down. Having forgotten to put on gloves, and being a heavy, stout man, the rope severely lacerated his hands, and as it did not reach the ground he fell some feet and was severely bruised. His trusty followers had horses in readiness, on one of which they mounted him. The party fled to the coast, took ship, and crossed over to Normandy to seek refuge with Duke Robert.[20] After some time had elapsed, he contrived to make his peace with Henry, who allowed him to return to England, when he regained his See of Durham, of which he completed the cathedral, and also added to the works of the great castle there. The window from which he is supposed to have escaped is over sixty-five feet from the ground, and his evasion was evidently considered at the time a most audacious and remarkable feat, as more than one contemporary chronicler gives a very detailed and circ.u.mstantial account of it.
It is not until the Edwardian period of our history that we find castles used as places for the secure detention of captives. In the earlier Norman times dungeons were of little use, their policy being one of ruthless extermination, or of mutilation, in order to strike terror into rebellious populations.[21] Only persons of the most exalted rank, such as Duke Robert of Normandy, Bishops Odo, of Bayeux, and Ralph Flambard, of Durham, Earl Roger, the son of William FitzOsbern, with a few distinguished Saxon captives, underwent a prolonged imprisonment.
The Tower of London as it exists to-day has, by a slow process of gradual accretion round the keep as a nucleus, become what is known as a "concentric" castle, or one upon the concentric plan, from the way in which one ward encloses another; and its architectural history falls, roughly speaking, into three chief periods covered by the reigns of William Rufus, Richard I., and Henry III., all the more important additions to the fortress occurring approximately within these periods, as will be seen later on.
Commencing with the building of the great keep (now called the White Tower), and the small inner or palace ward to the south of it, by William the Conqueror, this at first was probably only enclosed by a stout timber palisade on the top of a raised bank of earth, having a ditch at its base. The first recorded _stone_ wall round the Tower was that of William Rufus, already mentioned, and it is not improbable that the wall marked "v" on the plan (only discovered in 1899 during the erection of the new guard house) may have formed part of his work.
But little is known to have been added by Henry I. The sole remaining Pipe Roll of his reign only records a payment of 17 0s. 6d. "in operatione Turris Lundoniae," without any further mention of what these works were, and as the amount is not very large, it is not probable that they included anything of much importance. That the smaller inner or palace ward to the south of the keep was already completed, is shown by a charter of the Empress Maud, dated Midsummer, 1141, which granted to Geoffrey de Mandeville (then Constable of the Tower, and third of his family to hold that important office) the custody of the Tower, worded as follows: "Concedo illi, et heredibus suis, Turris Lundoniae c.u.m "parvo castello" quod fuit Ravengeri";[22] and this "little castle" is the before mentioned inner or palace ward, though how or where this was originally entered from the city nothing now remains to tell us--most probably at or near the point subsequently occupied by the Cold Harbour Gate "u," at the south-west angle of the "turris," or White Tower "r," for it is but seldom that the original entrance gates of castle baileys or courtyards are removed, unless in the case of an entire re-arrangement of the plan, with the consequent rebuilding thereby rendered necessary.
Owing to the state of anarchy that prevailed during the troubled reign of Stephen, and the destruction of all the Pipe Rolls and other records that resulted, it is improbable that any extensive works were in progress during that period.
Although the Pipe Rolls of Henry II. record a total amount expended upon works at the Tower of 248 6s. 8d., but little appears to have been added as to which we can speak with any certainty, unless it be the forebuilding of the keep "y" (long since destroyed), the gatehouse of the inner ward "u," and perhaps the bas.e.m.e.nt of the hall or Wakefield tower "l."
As at first constructed, the White Tower (like its fellow at Colchester) had no forebuilding covering the original entrance, which was at the western extremity of its south front, upon the first floor, then some twenty-five feet above the external ground level. The small doorway leading to the flight of stairs in the south wall which ascends to St.
John"s Chapel, by which visitors now enter the keep, is not, and is far too small in size to have ever been, the original entrance.
On the Pipe Rolls there are frequent entries of sums for the repairs of the "King"s houses in the Tower," probably the great hall "x," with its kitchen and other appendant buildings; "of the chapel" (obviously that of St. Peter, as that of St. John in the keep would hardly be in need of any structural repairs at so early a date); and "of the gaol."
These last doubtless stood in an outer ward added by Henry I., and at first probably only enclosed by the usual ditch and earthen rampart, furnished with stout wooden palisades.
[Ill.u.s.tration: ST. JOHN"S CHAPEL, TOWER OF LONDON.]
It is somewhat difficult to a.s.sign any precise date for the first foundation of the "Chapel of St. Peter ad Vincula apud turrim." It is not probable that it was contemporary with the Chapel of St. John, but was doubtless erected by Henry I. when he enlarged the area of the outer ward of the Tower; as this necessitated a considerable increase to the permanent garrison, St. John"s Chapel in the keep would no longer suffice for their accommodation, and a new chapel would become necessary. If St. Peter"s Chapel had only been parochial (which at its first erection it was not), it might have been possible to ascertain the precise date of its foundation.
In 20 Henry II. (or 1174), Alnod, the engineer, received the sum of 11 13s. 4d. for works at the Tower. Other payments occur for sheet-lead for the repairs of the chapel, the carriage of planks, and timber for the kitchen,[23] the gateway of the gaol (probably Cold Harbour Gate "u"), various repairs to the "King"s houses within the bailey of the Tower,"
and occasionally for the repairs to the "turris" or great keep itself.
This, when first built, was of rough rag-stone, rudely coursed, with very open joints in thick mortar, so that these repairs (consisting, doubtless, of patching and pointing) occur with more or less frequency.
Not until 1663 did the keep receive its final disfigurement, at the hands of Sir Christopher Wren, who cased part of the exterior in Portland stone, rebuilt two of the angle turrets, and "Italianised" all the window openings, thereby obliterating many valuable mediaeval details.
All these outlays are certified by the view and report of two inspecting officials, Edward Blund and William Magnus, the works being carried out by Alnod, while the writs authorising payments were signed by one or other of the justiciars, Ranulph de Glanville and Richard de Lucy, or by the King himself.
The following reign marks a period of great constructive activity at the Tower. The new monarch was one of the foremost military engineers of the age; and when we consider the valuable experience in the art of war which he had already gained, in the decade prior to his accession to the throne, in conducting (while Count of Poitiers and Duke of Aquitaine) various sieges of the castles of his rebellious barons in those provinces, it seems improbable that he would have been satisfied to leave the Tower in the condition it then was, with a keep standing in a small inner ward, enclosed by a plain stone curtain wall, devoid of any projecting towers, unless perhaps the base of the Hall tower, and the Cold Harbour Gate (see plan), and a large outer ward, only enclosed by a wooden palisade and ditch.
Richard must have been well aware of the enormous increase to the power of effective defence conferred by salient or boldly projecting towers flanking with their fire the curtain walls, which in England, at any rate, were then somewhat of a novelty. At this time the Tower was extremely defective in this respect, its great need being not for mere repairs, but for effective modernization as a fortress.
Before embarking upon the hazardous enterprise of the third Crusade, Richard left his trusted Chancellor, William Longchamp, to carry out an extensive series of new works at the Tower, all of which were probably from the designs of the sovereign himself.
In his valuable monograph upon the Tower,[24] the late G. T. Clark, F.S.A., has fallen into a strange error as to the actual amount expended upon works there during the earlier years of the reign of Richard I., which he states "do not show above one or two hundred pounds of outlay." When this rather dogmatic a.s.sertion is tested by reference to the existing doc.u.mentary evidence of the Public Records, its glaring inaccuracy is at once apparent; indeed, it might fitly serve as an ill.u.s.tration of Pope"s well-known lines:
"A little learning is a dangerous thing, Drink deep, or taste not the Pierian spring."
The Pipe Roll of 2 Richard I. discloses an expenditure, "ad operationes turris Lundoniae," amounting to no less than 2,881 1s. 10d., in itself a sufficiently large sum, but one which, when multiplied twenty-fold in order to bring it up to its present-day value,[25] is increased to 57,621 16s. 8d. of our modern money!
The custody of the Tower was entrusted by Longchamp to one of his dependents, William Puinctel, who seems to have acted as Constable and superintendent of the new works, according to the Pipe Roll of 2 Richard I.
It is well known that all the contributions levied in the King"s name do not invariably appear set out in full in the records, and there were certainly other sources of revenue open to the Chancellor, of which he doubtless took the fullest advantage.[26] The difficulty in this case is not so much his raising the funds needed for carrying out these works (which he undoubtedly did), but to account for their rapid completion in so short a time.
If, however, it was possible, only seven years later, for Richard himself to build, in a far more inaccessible situation, the _entire castle_ of Chateau Gaillard in the short s.p.a.ce of a single year, it need not have been so difficult for Longchamp to carry out in two or three years the works we are about to describe, especially when we consider that he had practically _unlimited_ funds at his disposal.[27]