IN preparing this detailed description of the ruins of Great Zimbabwe—the first given to the world in modern times—the author has aimed at permitting the actual ruins themselves to relate their own story of their forgotten past unweighted by any consideration of the many traditions, romances, and theories which—especially during the last decade—have been woven concerning these monuments.

The only apology offered for this apparently lengthy Preface is the mention of the fact that the operations at Great Zimbabwe were carried on for six months after the text of this volume had been sent to the publishers in England. The Preface, therefore, thus affords an opportunity of bringing down the results of these operations to a recent date.

RUINS’ AREA

The recent examination of the district surrounding the ruins now shows the Ruins’ Area to be far larger than either Mr. Theodore Bent (1891) or Sir John Willoughby (1892) supposed. Instead of the area being confined to 945 yds. by 840 yds., it is now known to be at least 2 miles by 1¼ miles, and even this larger limit is by no means final, as traces of walls and of walls buried several feet under the veld have been discovered, not only in Zimbabwe Valley, but in the secluded valleys and gorges and on the hillsides which lie a mile and even two miles beyond the extended area. Huge mounds, many hundred feet in circumference, with no traces of ruins, covered with large full-grown trees and with the remains on the surface of very old native huts, on being examined have been found to contain well-built ruins in which were unearthed small conical towers, gold ornaments, a few phalli, and in one instance a carved soapstone bird on a soapstone beam 4 ft. 8 in. high, which is more perfect and more ornate than any other soapstone bird on beam yet found at Zimbabwe. The examination of such spots and of all traces of walls which lie at the outer edge of the extended Ruins’ Area would, even with a large gang of labourers, occupy almost a lifetime.

Mr. Bent spoke of Zimbabwe as a “city,” and recent discoveries show the employment of this title to be fully justified, for not only is the Ruins’ Area vastly extended, but the formerly conjectured area can now be shown by recent excavations to have been much more crowded with buildings than could possibly have been seen in 1891. For instance, 2,300 ft. of passages have recently been discovered within the heart of the old Ruins’ Area buried some feet under the silted soil below the veld in spots where the siltation is rapid, the existence of which structures had been altogether unsuspected. In some instances the native paths, used by visitors inspecting the ruins, crossed these passages from 3 ft. to 5 ft. above the tops of the passage walls. The enormous quantity of dÉbris, evidencing occupations in several periods, scattered over both the old and the extended area, is simply astonishing, and judging by the value of “finds” made during the recent work, it seems quite possible that further exploration would, in the intrinsic value of relics as relics, largely reimburse the expense of its continuance, while securing the opening up of fresh features of architecture and probably some definite clues as to the original builders of the numerous periods of occupation respectively; would bring an immense addition to scientific knowledge, while the more important ruins themselves, having been cleared of silted and imported soils and wall dÉbris, are now ripe for the further examination for relics.

BURIAL-PLACES OF THE OLD COLONISTS

The secluded valleys, and also the caves in hills, for a distance of six miles, and in some cases as far as ten miles, from Zimbabwe have been systematically searched in the hope of discovering the burial place of the old gold-seekers. The neighbourhood of Zimbabwe contains several extensive ranges of granite hills each enclosing many secluded and Sinbad-like valleys and gorges, where natives state white men had never previously entered. Such spots on the whole of the Beroma Hills to the east of Zimbabwe, the south end of the Livouri Range to the west, the Bentberg Range to the south, and several hills in the Nini district, as well as several parts in the Motelekwe Valley, have been systematically searched without avail, though there are in certain of these secluded places traces of walls and artificially placed upright stones and other signs of human presence which require some explanation. The siltation of soil from the steep hillsides of many of these most romantically situated valleys has been very extensive. These searches could only be carried on after veld fires had swept the district of the rank grass which here grows to a height of 12 ft. Mr. Bent and other writers have shown that the old Arabians religiously preserved their dead, burying them in secluded spots at some considerable distance from any place of occupation. The writer is not without hope that these burial-places may yet be found. The population of Zimbabwe at several different periods must have been immense, and, judging by the remains found near some of the oldest types of ruins in other parts of the country where the amount of gold ornaments buried with each corpse ranged from 1 oz. to 72 oz., the discovery of such places in the Zimbabwe district would yield important results, especially as, for many reasons, Zimbabwe undoubtedly appears to have been the ancient metropolitan capital and the centre of gold-manufacturing industry of the original and later Arab gold miners, and the place so far has yielded the richest discoveries of gold in every form.

The writer is now perfectly assured that no burial-places of the original builders will be found under the interior of the Elliptical Temple or within 30 yds. of the exterior. Holes have been sunk at regular intervals within the temple and immediately outside the walls, and boring-rods have been systematically employed, and the position and lie of the formation rock ascertained throughout, so that sections and levels have been made of the soil and rock under the temple. All the results gained from each hole and boring are recorded. But beyond discovering buried foundations at the higher level, only virgin soil, never before disturbed, was gone through. French and German archÆologists who visited Zimbabwe during the operations confirmed what British scientists have affirmed, that no burials of people of Semitic stock would be found within or near to any building so frequently in use as the great temple must have been. The severe restrictions with regard to cleanliness and sanitation, especially as to the dead, are among the most notable features of the old Semitic nations.

ABSENCE OF INSCRIPTIONS

No ancient writing has been discovered, though close attention has been paid to all stones and pottery likely to bear it, and notwithstanding that the interiors of some of the more ancient portions of the ruins have been cleared down to the old floors where, if any existed, they might reasonably have been expected to be found. Post-Koranic lettering was found on highly glazed pottery, also on glass, but all such specimens are of a fragmentary character; but experts such as Mr. Wallace Budge, the Head Keeper of Egyptian and Assyrian Antiquities at the British Museum, state that the glass and other “finds” of pottery are not older than the thirteenth or fourteenth century of this era. Other pottery thickly covered with dull-coloured glazes—mainly purples, greens, and browns—is thought to be somewhat older than that on which the lettering was found. Still, as such a very large portion of what may be considered as the more ancient of the ruins remains to be examined, it may yet be possible to unearth older specimens of Arab writing.

TWO PERIODS OF GOLD MANUFACTURE

Gold in a manufactured form is found on the lowest and original floors of the most ancient portions of the Zimbabwe ruins. In several ruins this was found as thickly strewn about the cement floors as nails in a carpenter’s shop. Gold ornaments discovered at this depth, in some instances from 3 ft. to 5 ft. below any known native floors, were always found in association with the oldest form of relics yet unearthed at Zimbabwe. Such gold articles are of most delicate make, and are doubtless of an antique character, and expert opinion recently obtained in England confirms this conclusion.

But there are other gold articles which are ruder in design and make, and these by no means are entitled to claim such antiquity. In fact, expert opinion declines to recognise them as being in any sense ancient; for instance, beaten gold of irregular shape showing the rough hammer marks of some very crude instrument, and with holes round the edges of such plates very rudely cut—or rather torn—and placed in imperfect rows altogether in a haphazard style. This form of gold plates is identical in every detail with the copper sheathing with which it is always found associated. The same remarks apply equally to the gold beads also found with this class of plates which betoken crude workmanship, as well as to the iron instruments decorated with small gold knobs.

With regard to the location of the later-period gold articles there is ample evidence that these are of very old native origin. Such ornaments are commonly met with on the floors of, or in close proximity to, the old native huts of the types of Nos. 2 and 3 (see Architecture, s.s. Native Huts found in Ruins, pp. 154, 155, post), and also in the cement huts with small radiating walls on levels several feet above any ancient floorings. In every instance such gold ornaments are found in association with articles of old native make—such as double iron gongs, copper sheathing, and copper assegai- and arrow-heads.

NORTH ENTRANCE

In 1902 the floor of the North Entrance to the temple was exposed to a depth of 5 ft. below the surface, as shown in Mr. Bent’s book (p. 106), while a flight of steps in perfect condition leading up to the entrance from the exterior was discovered at a depth of 9 ft. below the old surface. This entrance, showing a bold conception and admirable construction, is now considered as one of the principal show features at Zimbabwe. Further, it is the oldest form of entrance and steps as well as the finest of any yet discovered in Rhodesia. A quantity of gold was found on the floor and steps of this entrance, which were once covered with fine granite cement, also a few true phalli.

PARALLEL PASSAGE

This has been cleared throughout to a depth of at least 3 ft., and in one place 7 ft. Cement floors were exposed, and these were found to be divided into small catchment areas with a drain from each passing outwards through the main wall. Five additional drains were discovered in this passage. Here were found eight ornate phalli, a portion of a gold bangle, some beaten gold and gold tacks of microscopic size, and fragments of carved soapstone beams.

SACRED ENCLOSURE

This was cleared out to a depth of 4 ft. throughout its whole area, and a few phalli of unmistakable form were found, and old granite cement floors and steps were uncovered. Explorers and relic hunters had worked in this enclosure, and had double trenched it from end to end.

A remarkable discovery was made here of distinct traces of granite cement dadoes, 7 ft. high, round the interior faces of the walls of this enclosure. In some other enclosures the remains of dadoes can still be seen.

The small conical tower in this enclosure has during the last ten years been seriously damaged by the large trunk of a tree pushing over the summit of the cone. Photographs of this small tower taken in 1891 show that it was then almost intact.

PLATFORM AREA

This open area, lying to the west and north of the Conical Tower and the Platform, corresponds to the open areas immediately in front of the altars in old Grecian temples. This was Mr. Bent’s opinion, and possibly it answered at Zimbabwe a similar purpose of accommodating the worshippers. The area, some 120 ft. by 60 ft., has been cleared out of large trees, and of about 6 ft. of soil throughout, and floors—both cement and clay—were disclosed, also a fine circular structure of excellent granite cement, and ascended by two steps. On and close to this structure were found fragments, mainly bases, of carved soapstone beams of slender appearance, also some phalli and gold. This platform lies slightly off the north line between the Conical Tower and the Main North Entrance.

Some of the walls surrounding this area on the west and north sides, once considered to be ancient, can now be seen to cross over very old native clay huts and native copper and iron-smelting furnaces. The soil contained some phalli, which had been converted by the natives into amulets, also some Arabian glass—thirteenth and fourteenth centuries—Venetian beads, gold wire-work, beaten gold, gold scorifiers of native pottery, iron pincers, and fragments of carved soapstone bowls with geometric designs.

ENCLOSURES 6, 7, AND 10

Gold-smelting operations must have, at some late period, been extensively carried on in these enclosures, for on removing from each enclosure all dÉbris and fallen stones to a depth of from 4 ft. to 7 ft., there were found burnishing stones of fine grain and still covered with gold, gold scorifiers with gold in the flux, cakes of gold, gold furnace slag, beaten gold, and gold dust.

At a still lower depth in No. 6 Enclosure a quantity of granite clay crucibles, showing gold richly, were met with, and these are undoubtedly of older type than the native pottery scorifiers, also some ingot moulds of soapstone of the double claw-hammer or St. Andrew’s cross pattern.

CENTRAL AREA

This area is only partially excavated, it being covered with old native-built walls which cross over bone and ash dÉbris, old native huts, an iron furnace, and rich black mould in which the vegetable matter was still undecayed. Experimental holes and boring-rods showed that some very old foundations ran below the soil upon which the later and poorer walls are built. However, a key has now been found which will enable further excavations to be made within this area without injury to the upper walls.

SUMMIT OF MAIN EAST WALL

Along the summit of the east main wall, and only over the chevron pattern which faces east, have recently been discovered the traces of foundations of small circular towers, both on the inner and outer edges of the wall. These correspond in measurement and relative position to the small conical towers on the west wall of the Western Temple at the Acropolis Ruins, which is decorated with monoliths. Some of the best-known surveyors and practical builders in Rhodesia are prepared to certify as to the traces of these foundations. This is entirely a new discovery, as is also the fact that at one time the summit of the wall, only over the chevron pattern, bore beautifully rounded soapstone monoliths, the bases being found displaced under the ruck of loose blocks which runs along the centre of the summit of this part of the main wall. Some carved splinters of these monoliths were found at the bases of the wall. A collection of these “finds” has been sent to the Salisbury Museum.

PROBABLE AGES OF THE WALLS OF THE ELLIPTICAL TEMPLE

All the walls of the Elliptical Temple are not ancient; that is, not ancient in the sense applying to the suggested SabÆo-Arabian occupation of Rhodesia and also to that of the Solomonic gold period. The evidences pointing to this conclusion, and now for the first time available, are so obvious and general, and the ocular demonstration so positive, that one of the many popular myths concerning Great Zimbabwe must, even at the risk of committing a vandalism on cherished romantic theories and beliefs, go by the board. The writer prefers that the ruins should tell their own story, and this can now be read in the walls, in the dÉbris heaps, and in the relics and their associated “finds” and locations.

The oldest walls of the temple for which great antiquity may be claimed are—the main east wall from north to south, the Conical Tower, the Platform, portions of the inner wall of the Parallel Passage (reconstructions are present here), and some adjoining walls, and some buried walls and foundations, and possibly some other walls on the south side, concerning which some doubt exists, as also the west wall of the West Passage, a well-built structure which once was extended at either extremity. As to the question of obviously much later walls, this is involved in the following section of this preface.

WEST WALL CONTROVERSY

The writer is fully convinced that the original west wall of the temple once extended outwards further west, and that the present west wall extending towards the south is of much more recent construction and is built on a shorter curve, also that most of the structures of the central and western portions of the building are also of much later construction, and this for many substantial reasons, some of which are here briefly stated:—

(a) The west wall is considered by all practical builders and architects to be far slighter, much inferior in construction, fuller of defects, and to contain to a greater extent ill-shaped stones than the main wall on the east side, while the foundations are at many points far more irregular, and the batter-back of the interior face of the west wall is less severe than is the case of the east side. Lengths of 25 ft. each of both walls have been examined and compared and photographed, and the number of defects of construction recorded. The number of false and “straight joints,” false and disappearing courses, and stones supported at their corners by granite chips, which the west wall contains, is roughly about forty odd to every one of such defects in the east wall, which is the architectural marvel for symmetry, grand proportion, true courses of most carefully selected and assorted blocks (some of which have been dressed with metal tools) of any other ancient architectural features at Zimbabwe. All this is an ocular demonstration, and is commented upon by the most casual visitor to these ruins. This, too, is very patent when seen from the summit of Zimbabwe Hill, the view looking down upon the temple revealing most obviously the different characters of the walls.

(b) In 1903 the writer cleared the soil away from the gap between the older and later walls, and found that they were widely different in construction; that the later and narrower wall approached the older and well-built and wider wall at an oblique angle; and that the end of the older wall is broken and not finished off as are other ends of ancient walls. In a trench made at a distance of twelve yards west of the gap, and on the curve the older wall, if continued, would have passed, a mass of buried masonry, which might have been a portion of the old wall, was disclosed.

(c) Dr. Hahn, the leading expert in South Africa in chemical metallurgy, analysed the soil underlying the foundation of the west wall, and pronounced it to be composed of disintegrated furnace slag and ashes containing gold and iron. The ground to the west of the west wall has always been the spot at which gold prospectors have washed the soil for gold, and here gold crucibles and scorifiers are to be found. This soil contains 73 per cent. of silica, and would make an excellent foundation for walls, and the west wall is built right along this bed of furnace slag, which is about 2 ft. in depth, many yards wide, and extends from north to south.

(d) At a few feet from the exterior of the west wall, and at a depth of four feet below the level of its foundation, and extending as shown in trenches and cross-cuts for at least thirty yards from north to south, is a floor of granite cement laid on the formation rock, hiding its irregularities and making a perfectly level surface. The full extent of this flooring has not yet been ascertained. For two feet between the level of this cement flooring and the furnace-slag soil under the foundations of the west wall is fine silted soil. Evidently the later wall was erected at a very considerable period subsequently to the laying of the cement flooring and after the siltation of the soil, and also after the gold-smelting operations had been extensively carried on for a long period.

(e) No single relic of any great antiquity has been found by any explorer or prospector in the western portion of the temple, while the eastern portion has yielded at depth great quantities of phalli and of every relic believed to be associated with the earliest occupiers.

The oldest “find” in the western half of the building is pronounced by Dr. Budge to be of a period dating from between the thirteenth and fifteenth centuries of this era, and other “finds” relate to the same and later periods.

WRITER’S CONCLUSIONS

The writer is now and for the above and further considerations, and after two years’ residence within the ruins, perfectly convinced of the following:—

(1) That on the departure of the ancient builders and occupiers the temple became a ruin, and remained as such for some centuries, the west wall disappearing in the meantime (as explained later); (2) that some organised Arab people, possibly a split of the numerous Arab colonies and kingdoms which existed down the East African coast, possibly of the Magdoshu kingdom, who, according to De Barros, reached Sofala (1100 a.d.), exploited the gold mines, and formed a mixed population between the Arabs and natives, or possibly the Arabs of Quiloa, who secured as suzerain power Sofala and the kingdom of the Monomotapa (Rhodesia). One of these peoples is believed to be responsible for the ruins of Inyanga, which the writer after examining these remains does not consider to be ancient in the fullest sense of the term. One of these peoples are also believed to be responsible for making the “old workings,” the distinction between which and the “ancient workings” must always be kept in mind, a distinction which the late Mr. Telford Edwards always pointed out and insisted upon, and concerning which recent investigations prove him to have been correct; (3) that these Arabs made Zimbabwe their headquarters, to which the washed gold dust was brought to be converted into ingots for transport; (4) that these Arabs carried on extensive gold-smelting operations at the west end of the temple in the shelter of the massive walls, which would protect them against the prevailing winds and drifting rains; (5) that after carrying on these gold-smelting operations extensively and for a considerable period, they built a wall across the open space and upon their furnace-slag beds, possibly employing native labour (the Makalanga being notorious for their skill in wall building); and (6) that these Arabs also built several of the enclosures in the central and western parts of the temple to suit their special convenience, and altogether regardless of the buried foundations of the ancient builders.

DESTRUCTION OF THE ORIGINAL WEST WALL

It may be asked what caused the destruction of the original west wall. Its disappearance may be accounted for as follows. The south and west walls have for centuries borne the full brunt of all the torrential rain and storm water which rushes to these points from the Bentberg Kopjes, which lie close to the temple on the south side. This accounts for the great depth of silted soil which buries the old cement flooring. This must have washed the lower portions of the walls till the cement foundations decomposed and brought down the structure as it has done at other ruins at Zimbabwe. The writer at the commencement of his first rainy season at Zimbabwe found a large pool about 30 yds. in length, 15 yds. in breadth, and 2 ft. in depth up against the present west wall, towards which all surface water from the higher ground rushed unchanged. This had been going on every rainy season for many generations, with the result of forming large cavities under the foundations, and of keeping the wall in a constant drip with damp even at noontide, and of causing the spread of large moss over the walls, while shrubs and small trees grew out of the walls at some height from their base. Trenches and runs-off and banks soon cured this evil, and now the walls have changed from being black with damp to being grey with dryness. The moss has naturally flaked off, and the trees and shrubs in the walls are dead, owing to lack of moisture.

WESTERN TEMPLE

Operations in this temple since the description of the earlier work was embodied in the text of this volume have been carried on to June, 1904. Soil to a depth of from 3 ft. to 5 ft. was removed from the whole of the eastern portion of this area. The excavations showed several layers of native clay floors one above another. The “finds” were those known to be of native origin, though not made by natives of to-day. The later or native period of gold manufacture was greatly in evidence, beaten gold, gold tacks, and gold wire being frequently met with in association with copper sheathing, copper assegai- and arrow-heads, the copper containing no alloy.

A trial hole sunk to a depth of 6 ft. below this cleared portion of the temple area, or 9 ft. below the surface as it appeared in 1903, showed in its sides the lines of several clay floors and the side of a Kafir clay hut, now quite decomposed and soft. At the bottom of the pit a rough pavement of closely-fitting stones of irregular shape and size was come upon, and the articles found were identical with those discovered at a higher level.

The clearing of the area also disclosed clay sides of huts with the remains of short walls of stone radiating from the sides of the huts. The wall which Mr. Bent considered might have been the “altar” was found to be the radiating wall of a similar hut built upon a higher level. These small radiating walls are a general feature of exceedingly old native huts found at several places at Zimbabwe.

A large circular platform of granite cement was also disclosed. This spot yielded beaten gold of native make.

A ZIMBABWE REVIVAL

The writer believes that between the thirteenth and fifteenth centuries, or slightly earlier, a great influx of people took place at Zimbabwe, and that the majority of the minor ruins in the Valley of Ruins were built about this period. This is shown by the number of walls built across exceedingly old dÉbris heaps of native origin, by the “finds” of Arabian articles on their lowest floors, and by the fact that no relic of greater age than that period has been found. Two or three of the better-built minor ruins have the appearance of greater age, and some of the relics found in this class of ruins are of the oldest type. No one who had not spent considerable time at Zimbabwe could have any possible conception of the immense population present here at a period of but a few centuries ago. The remains of their stone walls are scattered thickly over the valleys and hillsides of Zimbabwe. The Makalanga state these are all Makalanga of generations long passed away. Some are constructions by indigenous peoples, and certainly they are not ancient, though largely built of stones quarried from the ancient ruins, and the “finds” are those of old native type, including Arab articles.

PRESERVATION OF RUINS

The thanks of all scientific circles, and of South Africans generally, are due to Sir W. H. Milton, Administrator of Rhodesia, whose great interest in the preservation of the ancient monuments in these territories is well known, and to whose direction is due the recent and timely preservation work at Great Zimbabwe. The author desires to express his personal indebtedness to Sir William Milton for the adequate arrangements made by him while engaged in his recent researches at the Great Zimbabwe.

PLAN OF ELLIPTICAL TEMPLE

The clearing of the Elliptical Temple and its vicinity has enabled Mr. Franklin White, m.e., Bulawayo, to prepare the latest and so far the most perfect plan of that building, and this he has kindly placed at the service of the author.

Indebtedness is also expressed to Professor A. H. Keane, ll.d. (author of The Gold of Ophir), for the contribution of the Introduction to this volume; to Mrs. Theodore Bent for generously permitting the use in this volume of illustrations from The Ruined Cities of Mashonaland; to Mr. Gray, Chief Veterinary Surgeon, Salisbury, Mr. H. S. Meilandt, Government Roads Inspector, Bulawayo, and Trooper Wenham, b.s.a.p., Victoria, for permission to reproduce certain photographs of the ruins, and also to the Directors of the British South Africa Company for permission to include the map of Rhodesia in this work.

---