Early London: Prehistoric, Roman, Saxon and Norman

by Walter Besant

It is remarkable that the Roman walls of Sens, Dijon, Bordeaux, Bourges, Périgueux, and Narbonne are in the same way partly constructed of old materials, and that they contained the remains of temples, columns, pilasters, friezes, entablatures, sepulchral monuments, altars, and sculptures, all speaking of threatened danger, and the hasty building of a wall for which every piece of stone in the city was seized and used.

Attempts have been made to show that the wall was built in later times. Cemeteries, it is stated, have been found here and there. We have already seen that there were burials in Bow Lane, in Queen Street, Cheapside, in Cornhill, north of Lombard Street, on St. Dunstan’s Hill, in Camomile Street, and at St. Helen’s Church, Bishopsgate. Now interment within the city was forbidden by law. Hence it is inferred that the wall could not have been Roman work. The answer to this inference is very simple. The City of London was the enclosed town or fort on the eastern hillock; all these places of burial lie without the wall of this Citadel—that is to say, without the town.

As for the wall, so many portions of it remained until this century, so many fragments still remain, it is laid down with so much precision on the older maps, and especially on those of Agas and Wyngaerde, that it is perfectly easy to follow it along its whole course. For instance, there were standing, a hundred years ago, in the street called London Wall, west of All Hallows on the Wall, large portions of the wall overlooking Finsbury Circus, with trees growing upon them—a picturesque old ruin which it was a shame to destroy. This part of the wall is shown, with a postern, in the 1754 edition of Stow and Strype. Indeed that edition shows the whole course of the wall very clearly, still standing in its entirety.

Starting from the Tower, it ran in a straight line a little west of north-west to Aldgate; then it bent more to the west and ran in a curve to Bishopsgate; thence nearly in a straight line west by north to St. Giles’s Churchyard, where it turned south; at Aldersgate it ran west again as far as a little north of Newgate, where it turned south once more, crossed Ludgate Hill, and in ancient times reached the river a little to the east of the Fleet, leaving a corner, formerly a swampy bank of the Fleet, which was afterwards occupied by the Dominicans. Fragments of the wall still exist at All Hallows Church, at St. Alphege Churchyard, in St. Giles’s Churchyard, and at the Post Office, St. Martin’s-le-Grand, while excavations have laid bare portions in many other places. For instance, in the year 1852 there was uncovered, in the corner of some building, a very large piece of the wall at Tower Hill. By the exertions of Mr. C. Roach Smith this fragment was saved from destruction, examined carefully and figured. A plate showing that part of it where the ancient facing had been preserved is given in his Illustrations of Roman London. “Upon the foundation was placed a set-off row of large square stones; upon these, four layers of smaller stones, regularly and neatly cut; then a bonding course of three rows of red tiles, above which are six layers of stones separated, by a bonding course of tiles as before, from a third division of five layers of stones; the bonding course of tiles above these is composed of two rows of tiles; and in like manner the facing was carried to the top. The tiles of the third row are red and yellow; and they extend through the entire width of the wall, which is about 10 feet, the height having been apparently 30 feet. The core of the wall is cemented together with concrete, in which lime predominates, as is usual in Roman mortar. Pounded tile is also used in the mortar which cements the facing. This gives it that peculiar red hue which led FitzStephen to imagine the cement of the foundations of the Tower to have been tempered with the blood of beasts.” In the year 1763 there was still standing in Houndsditch part of a Roman tower. It is figured in Roach Smith’sRoman London. The drawing shows that the towers are as square as those still to be traced at Richborough. They were built solid at the bottom, hollow in the middle, and solid again at the top. The middle part contained a room with loopholes for the discharge of missiles and arrows. In the Houndsditch tower a window has taken the place of the loophole. According to FitzStephen, the wall was strengthened by towers at intervals. At the angles, as appears from the bastion in St. Giles’s Churchyard, the towers were circular.

In 1857 excavations at the end of Aldermanbury laid open a remarkable portion of the wall; it was composed of a series of blind arches forming part of the solid masonry.

Nothing is left above ground of the Roman facing; what we see now is the old solid core with perhaps some of the mediæval facing. The uncovering of a large part of the wall at Aldersgate Street is thus described (Archæologia, vol. lii. 609):—

“The Government having determined to erect additional buildings to the General Post Office in St. Martin’s-le-Grand, certain steps were taken in order to ascertain the nature of the ground on which these buildings were to be placed. For this purpose, in the latter part of 1887, shafts were sunk along a line from Aldersgate Street to King Edward Street, some yards south of the old Money Order Office, and parallel to Bull and Mouth Street, a street now swept away. In sinking these pits the workmen came upon the Roman wall, and afterwards, as the process of preparing the site for the new buildings proceeded, a considerable fragment of it was unearthed running east and west, and extending from Aldersgate Street on the one side, to King Edward Street on the other. It was found that the line of buildings and walls forming the southern boundary of the churchyard of St. Botolph, Aldersgate Street, was based upon this wall, and it seems very probable that the churchyard and church above-named partly occupy the ground filling up the original ditch.

The portion of wall exposed, commencing near Aldersgate Street and running westwards, can be well seen for a length of upwards of 131 feet. A considerable length of this has been carefully underpinned, and, I am happy to say, will be preserved. The height remaining varies very considerably, but, measuring from the original ground level, at least eleven or twelve feet of masonry is still standing in places, not in any regular line at the top, but much broken into by the foundations of comparatively modern walls built upon it. Beyond the length named but little of the wall is to be seen, and as it approaches King Edward Street, just beneath the line of the houses in that street, on its eastern side, were discovered the foundations of a semicircular tower, or rather a tower semicircular in plan, with slightly prolonged straight sides. The foundations of this tower—and nothing but foundations remained—did not form any part of the structure of the Roman wall, but came with a butt-joint against it. They were 5 feet 3 inches wide, and composed of rubble-work of Kentish ragstone with some chalk, and a few fragments of old building materials bedded amongst the rubble. The internal measurements of this tower were 17 feet 3 inches by 16 feet, and the foundation of the Roman wall was seen to cross its base. The tower which stood on these foundations is probably an addition to the wall in the mediæval period. It was not Roman, and the foundations had no Roman character. Some pieces of worked stone discovered in them showed traces of Norman mouldings, and of foliage of the early English period. It is possible that the first tower west of Aldersgate, seen in Agas’s map of London 1560, may be the one the remains of which are here described.

Turning now to examine the construction of the Roman wall, it appears that it was built in the following manner:—A trench, from 10 feet 9 inches to 11 feet wide and 6 feet deep, was dug in the natural clay soil, the sides of the trench for a distance of 2 feet from the bottom sloping slightly outwards. The lower part of the trench for the depth of 2 feet was then filled in with puddled clay mixed with flints, and the whole well rammed down. Upon this came 4 feet of rubble foundation, lessening in some places to 2 feet, composed of masses of Kentish ragstone, laid in mortar, the larger pieces being placed with some care in the arrangement, so as to form a solid base for the superstructure of the wall itself.

This wall, between 8 and 9 feet thick, as far as could be ascertained, starts with a bonding course of three rows of tiles at the ancient ground level, which is 6 feet 9 inches below the level of Aldersgate Street. Above this course the body of the wall is composed throughout its height of masses of ragstone, with now and then a fragment of chalk bedded very roughly in mortar which has been pitched in, not run in, sometimes with so little care as to leave occasional empty spaces amongst the stones. The stones are often arranged in a rude herring-bone fashion, perhaps for greater convenience in packing them in, but the layers do not correspond in depth with the facing course.