Great Zimbabwe, Mashonaland, Rhodesia / An account of two years' examination work in 1902-4 on behalf of the government of Rhodesia

by R. N. Hall

But the most interesting feature of these terraced areas lies in the fact, obvious to anyone who inspects the areas cleared out in July and August, 1902, that these areas were not originally terraces but ordinary enclosures, with floors from 4 ft. to 10 ft. lower than the present reduced summits of the outer walls. Some people of times later than those of the original builders had deliberately taken the blocks from the outer or down-side walls of the enclosures and thrown them inside till the interiors were filled up level with the reduced height of the walls, and over such filling-in had spread a clay floor, and so made these enclosures into terraced platforms and dry vantage ground on which to build their huts, lay their corn-drying, threshing, and winnowing floors, and also their small daka granaries, which, occupying these well-drained and soilless positions, would be free from the ravages of white ants.

By clearing the outer faces of these walls from block débris, which has fallen or rolled down from higher positions on the hill—and these falls have in some instances utterly ruined the walls beneath—and following the curve of wall round to its opposite side on the face of the hill, the rounded entrances into some of these enclosures have been unburied, the floor of such entrances being on an average 5 ft. to 9 ft. below the débris. These entrances, so far as discoveries have been made, are all at the rear of the enclosure and close to the face of the hill. These entrances are narrow, 1 ft. 10 in. and 2 ft. 2 in. being a fair average width, and once the line of floor which is level with the floor of the entrances is cleared, it is seen that there are no steps inside the enclosures, though outside the entrances there are in a few instances steps leading up to the entrances only, but never steps leading upwards from inside the entrances. The floor of the entrance, in each case, is the level of the floor of the enclosure. Inside is seen a mass of dry blocks without soil, and many of these blocks show a face that was once exposed to the weather and become time-eaten if not greatly decomposed, which shows that they once formed part of the face of an ancient wall.

In removing this block débris from these enclosures no pick or spade was necessary. The blocks were picked up and handed outside, and when the enclosures had been emptied there was not 12 in. depth of soil, and what was there was mainly fine granite chippings caused by the throwing in of the blocks, and of decomposed daka, which had formed the floor, and which, being of poor quality, constant rainfalls had practically rotted into sand. It was on these cleared-out floors that portions of large carved soapstone bowls were discovered in July and August, 1902. The “finds” on the bottom and original floors bore undoubted evidences of antiquity, and were totally different in character from the unmistakable Makalanga objects found on the daka floor some feet above.

As before stated, these terraces of enclosures are in some points independent of each other—that is, they extend across the face of the hill most irregularly. The second enclosure may be 20 ft. or 30 ft. in front of the level of its neighbour on either hand, the third 20 ft. or 30 ft. to the rear of the level of the first enclosure, and so on. But the enclosures are built very nearly one behind the other up the face of the hill, the front wall of one being the back wall of the next below. Between these ascending lines of enclosures are narrow sunken passages, the existence of which, until August, 1902, was altogether unsuspected, as such passages showed no outcrop of their side walls. Two of these passages are now known to be at least 350 ft. long, and one of them has been cleared out for this length. They were exceedingly narrow, so that at many points only one person could pass at a time. The widths average 2 ft., and their floors are from 3 ft. to 12 ft. below the present débris-strewn surface of the slope of the hill. Into these passages the entrances to enclosures on either side open.

Evidently the same people who filled in the enclosures and converted them into platforms likewise deliberately filled in the passages for the foundations of old Makalanga huts, and granaries were laid across the filled-in passages. The “finds” on the floors of these passages and those found on their “filled-in” tops showed as great differences in every respect as were presented by the objects found on the original floors of the enclosures and those on the later clay floors above them.

SOUTH TERRACE

This is a natural ledge of ground artificially improved, running east at the foot of the south precipice for 200 yards from the foot of the Rock Passage, and extending beyond the point where the most easterly ruins of the Acropolis are built along the summit cliff. This terrace is supported on its south side by several retaining walls.

Almost all the enclosure walls along this terrace have been demolished by the falls of great masses of walls from the ruins on the summit of the cliff. Barely the foundations of such enclosure walls now remain, the only exception being the enclosure adjoining the bottom end of the Rock Passage.

THE “OUTSPAN RUINS”

These ruins, which lie 75 yds. to the north of Havilah Camp, are on the outspan in the valley which runs along the south side of Zimbabwe Hill, and also on the path leading from the camp to the South-East Ancient Ascent. This block of ruins was entirely cleared out of wall débris and silted soil in September, 1902, and now visitors may walk upon what was practically the ancient floor. The cement which once formed the floor has become decomposed to granite sand, and owing to the enclosures of the ruin having been buried some 3 ft. to 5 ft. under the veld, the enclosures held all the water which in rainy seasons would pour down the side of the hill immediately behind.